Was waiting for this video! As a software engineer that dabbles in electronics/embedded and who deals with unnecessary and unbearable complexity in his work, it's nice to see a simple signalling mechanism used. Less to go wrong, cheap/easy to implement (as you showed), and doesn't need reliable checksums/retransmission/ack and other nonsense to happen when a tool is rattling away in terrible conditions.
The activation pulse seems unnecessarily complicated, though. Don't see why it needs to be a pulse of a certain time, rather than just J1 high brings J2 high. The consequences of J2 accidentally being activated are just that the battery consumes a few microamps for ~3min. There's some nice things you can do with a full digital protocol, like having the battery transmit its max current, continuous current, max temp, et cetera. Then tools that can control the load (blowers, vacuums, pumps, lights), can adjust to avoid overheating the battery. I know that Makita XGT has the batteries communicate digitally with the tools, and the tool won't run without this communication, which does make them less robust. It will be interesting to see if they're doing useful things with it, or if it's just there to make it harder for 3rd party makers to copy their batteries and tools.
The old(er) M28 batteries would show flash codes when the battery level button is held down. So, I'd think the M18 has something equivalent. Maybe there is a variation on the signal to ask the battery to output diagnostic data. Milwaukee may have a rig for warranty/qc to get an idea of what kind of life the battery had: Number of over temp events, max current draw, charge cycles, etc... The higher end "one key" tools could be the ones to look at if they communicate with the battery since they have "smarts" and use Bluetooth for tool settings. It doesn't make much sense to have the communication if there is nothing useful to do. Thanks for the great videos!
I have 2x M18 batteries that give those flash codes, although I haven't had time to look at it yet. Someone posted a fairly good guess of what they were in a forum or reddit (something like manufacture year, week of manufacture, etc). They're both 2013 batteries, whereas all my other batteries are 2015+ and don''t give any flashing led codes. Good thinking on the one-key tools. I was going to try and send queries to the battery based on what seemed logical from the charger messages, but a one-key tool could save a lot of effort there.
@rossradford5767 that is interesting. Can you put a quick vid of the sequence on youtube? You should be able to post youtube links in the comments here, but any external link will block your comment without telling you. If you can't post the link, post the title so I can search for it.
I love the depth of information you provide in all this - not to mention how well you detailed what you did Doing Australia proud and keeping the “fuck you, because I can” mentality thriving
We have discussed this briefly before, but EGO 56V packs do have communication with the tools through the D-terminal. The tools will run for around 10secs or so without the D-terminal, but it seems a watchdog will shut it down after the 10 secs with no communication between the two. In case you wanted to see a pack with communication. The BMS has 2 microcontrollers and is quite advanced with temp monitoring boards in between the pack halves that uses transistors PN junction for temp monitoring and 2 NTC's as well. The board even has a calendar chip for storage mode I presume. I wanted to look more into the communication protocol, but have not had any time to go back to it. I threw all this on a video called "What is the D-terminal?" Since so many viewers asked about it. I also shared that someone else could take the information and go further than I did. I don't have the best equipment at this time to go much further. I hope you get time to look into it. Great video!
Interesting. I'll have to watch your video. I'm happy for you to direct link any of your videos. Don't think I'll be getting any Ego anytime soon. I do have some Makita XGT and my initial tests show that the tool won't even run without digital communication. A user called Malvineous on github has figured out some of the XGT protocol between battery and charger. Will be interesting to see what it's saying to the tool. As with you, it's a matter of finding the time!
Here's the video for anyone else: ruclips.net/video/H-hggUXgYlU/видео.html Search showed me the T-terminal video first, so I watched that thinking you made a typo!
Massive thank you for the info you're sharing, mate. I found your channel while trying to figure out an issue (well, ONE of the issues lol) with my M18 12Ah batteries. I have tinkered/hacked/repaired many m18 batteries over the years, including my 2 12ah. The reason i watched this video is that my 12ah batteries have changed the way they 'behave'. Presently, the seem to function as the units you show in your video. However, this was definitely not always the case. I used to be able to connect them to other branded tools and they would 'wake up' and worked fine. One adapternthat was always quirky was one to my dyson v7. Older milwaukee m18s wiuld work fine, but with the 12Ah the vacuum would initially pulse, but i discovered that flicking the trigger several times would get it to power up. Then one day a year or so ago, both my 12ah batteries mould not work through any adapeters, but worked fine on milwaukee. Some troubleshooting with a multimeter showed the 'activation' happened by to tool ao my work around has been to have a milwaukee tool activate the nattery then do a quick change over to my diy adapted tool. It works as long as i keep it 'awake' but, you can imagine how much fun THAT process is!! 😂😵💫. Anyhooo, I have searched for answers as to how the behaviour of these batteries could have changed, without much luck. The only thing i can think of is that i have a one-key impact wrench, and could Milwaukee have pushed an OTA update to the batteries via the one-key bluetooth network??!
The 12Ah is the only battery with mosfets in it (actually, the 6Ah forge does as well). Their's 2 of them under the brass heatsink. So there could be something going on with those. I haven't got around to testing what conditions make those mosfets turn off, though.
Here are some other things that I'm certain have changed in the way BOTH M1812Ah batteries from when i purchased them: -When connected to a tool with with an led light, eg drills/impact drivers, the light either stays on constantly, apart from flickering every 8 seconds (I remember one of the things your investigations found was the 8 second 'activation' time of the 12ah). I only first took notice of this because my drill ( with an almost fully charged 12ah battery connected ) was left I. The boot of my car, a week or two later, I was surprised to find the battery completely dead. When I charged the battery and reconnect to the drill I payed more attention and noticed the change in behaviour. No other M18 or M12 batteries do this other than BOTH my 12ah m18. -when initially connected to any tool, there is a delay in delivering power. Sometimes I have to flick the trigger a couple of times before the tool will work. -With my Milwaukee chainsaw, however, things get REALLY strange. After connecting either of the 12ah batteries, the saw will generally work first time. However, when I release the trigger/throttle things get a little confusing, as MOST of the time, when the trigger is re-engaged the chainsaw does not start but hitting the trigger several times the saw fires up again. THIS OROCESS IS so Anti intuitive, as the number of times the trigger has to be toggled is sometimes 4, 5 or even more! I first noticed this behaviour after having the chainsaw warranty repaired, and my initial thoughts were the trigger was failing. My first remedy was to take the battery out and reconnect it which seemed to work. Over time though, Ive narrowed it down to the battery, and toggling the trigger ALWAYS works. Part of me theorises it to be a safety measure by Milwaukee incase of accidental pressing of the trigger. I have confirmed that when paired with a regular, older, M18 5ah or 6ah, the chainsaw fires up at every trigger press, and up until I received it back from the warranty repair, the trigger worked first time every time with 12ah also. All signs point to software update in the 12ah, but nowhere have I found any information/documentation regarding this, or whether this is even possible via Bluetooth?? Sorry for my long post, and thank you in advance for any light you might be able to shed? Cheers
@@neddraeger2013 it sounds like the internal mosfets are getting turned on and off, which is something I haven't seen. If it were only 1 battery I'd say your PCB is faulty, but unlikely that 2 would do the same thing. Do you think you could have damaged the PCBs when you repaired them? Maybe they can detect when something has gone wrong (like cells removed) and the battery goes into a safe state where it keeps the mosfets off until needed.
I hadn't considered that it could be the result of damage. I suppose I can't rule that out as a possibility however, I would be staggered if it were, for a few reasons. 1. Like you mentioned, the fact that it's the same with both 12ah, and ONLY the 12ah. 2. I bought the batteries around two years apart, yet the issue/change in behaviour appeared to happen to bother pretty much at the same time (when I first had the issue with one of the 12ah, the first thing I did was swap batteries to see if the issue was the tool. 3. Bother batteries had been opened countless times before for cell balancing etc without incident. it's definitely a head scratcher
Another awesome video packed with great info and testing! Can't wait to see you do this for xgt packs if that is the plan :) Also, Milwaukee and horrible bms engineering with false marketing, wow that was a surprise /s
Again great info also was sort of waiting for this video. This is something I’ve been curious about since I started using these batteries to power projects. Having a sheet that breaks down information about battery communication protocols as well as well as recommended wiring and a pcb / gerber file would be very appreciated if you have the bandwidth. Even just running list of everything you think you know would be useful to have in one place.
It is sorta funny how little seems to be actually going on in the communication protocol though lol especially when these companies really play up the secret sauce in these batteries
Petty Cool! I enjoy geeking out on stuff like this from time to time. As your channel grows, I'd like to see maybe how to make adapters for M18 to M12 or Flex 24 to M18 and or M12.
The circuit you built with mosfets is similar to whats in the greenworks 40v packs i have.i havent layed out a schematic but the layout and function is really similar.
Greenworks have low-voltage cutoff inside the battery packs? That's interesting. Never even seen a greenworks tool in person, so I have no knowledge of them. Only ever seen them in a few comparison reviews.
I think they have a voltage signal cutoff inside the pack. There is 3 relatively beefy mosfets and an aluminium heat sink. If you measure the + and - terminals you get 0v when the battery isn't connected to a tool/charger. You can short the pack and nothing happens. One of mine got put on a charger wrong and discharged to 9.8v....thats a 10s pack.
I was wrong. The pack was 9.8v last time I checked....the cells are all now between 4.11 and 4.13v and the whole pack is 41.2v. There is 3.6v across the main power terminals of the pack. This is really low current stuff so some shenanigans are happening but there is 35.2v between the - output on the board and the B- into the board. The resistance is not equal in both directions either
Don't have any Dewalt's yet. Powerstack is the only one that cuts power. It's most likely some simple connection on ID or Th pin, otherwise powerstack wouldn't work with older tools.
Great demonstration. Needed that. I hate those hidden "features." It should be VERY clear on a device or a battery where the limits are and how to re-create it with other parts, if we can't buy the originals any more, but have a great tool hidden to spring into life instead of throwing a lot away. It applies to more devices than you think.
Milwaukee think they can fool the public until they encounter you Mr. Scientist, wow 👍. Your analysis is truly out of this world..! I also have a few M18 & M12 batteries (cell from Korea & Singapore) and very tired of their LOW LOW quality cells where the charger shows red/green flashing. My M18 battery is fully charged and still shows 3 bar status (very common). My M12 cell pack loses voltage daily and a month later, each pack will show 3.7+4.1+4.1 causing charger to flash red/green. I am not a contractor so I don't use my battery often..! Recently I found an adapter that can convert Ryobi battery into Milwaukee tool interface and it's been working so far b/c I don't see the value of Milwaukee batteries. If this adapter performs well, I will buy Ryobi battery on sale and use adapter for Milwaukee tools to save money !
Some of their packs are good. Anything with 3000mah 18650s (M12 3Ah & 6Ah) or 4000mah 21700s (M18 8Ah and 12Ah) are prone to failure. In M18, their 5Ah and 6Ah HO are the most reliable.
Be careful of FET ratings! If you are unlucky, the FET you mention (IRFZ44) will when 3V is applied to the gate, pass 50uA or so. Vgs(th) is never a measurement you should use, look at Rds(on) voltage gate spec, this is at 10V. At best when using this FET, you need to verify that the Rds(on) at 3V vgs is low enough to not promptly burn up. High current FETs rated for 3V gate voltage are nonexistant in through-hole. Look at 'drive voltage' on digikeys selector, this shows 12 FETs from 30-170A and 3V or under gate drive option with Vds >21V. CSD16321Q5C is probably one of the least bad ones. They are all annoying to mount. A single transistor 3V->12V booster will fix this FET selection problem and makes the FETs you mentioned just fine, rather than a crapshoot on if they meet the high end of the spec (won't work at all), low end (work fine) or burn up a few minutes in.
I'm not using the IRFZ44, I'm using IRLZ44N. In the datasheet, Fig 1 & 3 show that it is rated to just under 20A at Vgs = 3V, hence why I said 10A for that simple circuit. My high-powered circuit (which I blurred to avoid inexperienced people from building) is using a booster setup to drive my 2x IRLB3813 at Vgs = 17V (dropping to around 10V as the battery depletes)
@@toolscientist Oops on the part number. I note you can't quite trust Fig 1 and 3, Vgs(th) is defined as varying by a volt maximum, typical is undefined. Id can be as low as 250uA at 2V Vgs, if Vgs(th) is worst-case at 2V. The Rds(on) spec @4V/21A anchors this@0.035R .735V. I am surprised! This is the first datasheet I have checked on which typical seems close to worst case, as Fig 1 seems very close to .735 at 21A/4V. This has unfortunately bitten me before.Expecting 10A easily due to Fig 1 on a different datasheet, and finding 10mA due to the part having a worst case Vgs(th) spec. Great work.
All good. You are right, though. The IRLZ44n isn't intended to be used at Vgs = 3V, it just happens to work decently there. So yeah, variance could be quite high between samples.
I was permanently banned from a Milwaukee subreddit for saying exactly what's in this video about 9 months ago by a mod who was claiming all sorts of advanced Milwaukee communication.
I think I know that subreddit, and that mod. Is it the smaller subreddit - MilwaukeePowerTools? Until I did this test, I thought there was at least some communication, even if it wasn't as fancy as Milwaukee tried to claim. I was fairly surprised that it didn't even differentiate between batteries as it's common for people to claim that the tools can detect high-output batteries.
Yep! That's the one...lol. It was weird because he was claiming the charger was doing stuff I've never seen it do and I monitor the voltage and current of every battery I charge. I simply said he was wrong (didn't know he was even a mod at the time) Immediate permanent ban.
Yeah, I've seen him make some very strange claims in a few subreddits. If you've seen my power testing video, he's the guy I got into an internet argument about the drill being 1500W. I think he has some issues, so I just leave him alone. Interesting on your charging tests. Have you got a jig, or do you wire the terminals up each time and measure current/voltage from the leads? Any interesting discoveries?
@@toolscientist I use a non Milwaukee charger that displays that info now. Before that I built a little jig from an old dead battery pack I cut out the terminal section (connected to charger) and the bottom of a dead drill (I put the battery to be charged in) connected with wires I would have access to. It's not really necessary for legit Milwaukee batteries, but I like to buy the knockoffs just out of curiosity a lot so like to keep an eye on what's actually happening. The only real useful information I've seen is that charging at the rapid charger rate very often leaves the cells unbalanced sometimes...but you have to open up the pack to check that after anyway.
Microsoft (the African Americans of software thanks to them having more oxygen and blood flowing to their aggression sector than the rest of their brains) are the same way. They've practically dominated Mojang and corrupted them in more was than one, and their horrible design tactics are quite literally killing Minecraft in thousands of screwed up ways.
J2 is a safety feature. I know guys who have broken wrists from the tools attachment, like a hole saw, catching in material and the entire drill spinning. It eliminates the need to have a secondary Handle. As someone who works in commercial construction It’s actually a pretty nice feature to have.
J2 is a safety feature for the battery to tell the tool that it has overheated or reached its low voltage cutoff. The anti-kickback that you're talking about is only in a few tools and is only in the tool. The tool has no need to tell the battery that it wants to stop, the tool just stops. You can somewhat see this when I stall the grinder. The grinder just stops with no change on J2
The good thing about running high amps through cheap alligator clamps...is it makes cheap, stiff cables feel soft, fkexible, and high quality for a few moments before they completely melt.
Awesome video! It's fun to poke at Milwaukee engineers about Redlink whenever they swing by. From what they have told me directly, redlink is very dumb so they can maintain backwards compatibility with all their tools. Only some tools and I believe the forge batteries will actually throttle the current (by reducing the voltage with a mosfet) and this is done with some type of check right after startup.
Yeah, I've got 2 batteries from 2013 that have standard Redlink and not Redlink PLUS. They use basic pulses with the charger rather than serial data. Maybe it's just the one-key tools that communicate? The 12ah has a large mosfet that might be able to throttle power, but I didn't see any different behaviour between tool and the 12ah.
Just curious about how you're talking shop to Milwaukee engineers. Do you sub-contract for them? I understand if you're not allowed to say anything. If you can talk, I'd love to hear any inside gossip :-)
@@toolscientist Haha no I do not subcontract with them, but I have a couple friends and coworkers who work/have worked with Milwaukee. I'll poke one of them for which batteries have the power reduction on them, my guess is it's the HO batteries with 21700 cells. Also some fun gossip is they had a hell of a time getting pouch cells working, they were having issues with them puffing from extreme loading so they were late to release. Dewalt beat them to it and they get a little upset if you mention it.
@jackphilp7057 yeah, it's hard to beat a cylinder as a pressure vessel. Whatever weight you save from going to pouch cells, you have to put back on the enclosure. I think the best use of them is making sizes that aren't achievable with 18650s and 21700s. The Dewalt 1.7ah seems like the only one so far that offers anything over cylindricals.
Very cool. I found your videos by trying to see how Milwaukee TICK trackers work technically speaking, surprisingly nobody's covered it yet. I guess it's Bluetooth and broadcasts a unique identifier somehow. I thought I could make a separate microcontroller to detect them instead of using my phone.
I’ve been using parkside for a while and it seems those do have some sort of communication you can set max amps in battery and also change some tool settings in the parkside app
I could see them at least doing a basic handshake, and if that fails, the tool uses a lower overload current. In the grinder tests at 4:39, I'm pretty sure the 12ah and 6ah are the tool cutting out when it hits ~160A, or maybe overheating the ESC. So if they did have even basic comms, I could see the tool lowering its stall current for weaker batteries. The batteries should be pulling J2 low when they detect that they've exceeded their current limit, but it seems Milwaukee relies on the batteries tripping low voltage or overtemp rather than looking for overcurrent.
So since there's no useful communication from battery to tool, do you think i could frankenstein a a cable from an M18 tool to a 12v set of alligator clamps? I keep the M18 inflator in my car but worry the desert heat will destroy the batteries, and I've always wondered if it'd possible to run it off my car battery in a pinch. Of course, i know it'd run at a reduced speed (if at all) given it's 12-14v rather than 18-20.
Definitely. Any M18 tool can run off external power. All you have to do is put > 0.72 volts on the J2 pin, but you can use 3V, 5V, 12V, etc. The highest voltage I put on J2 was I think 25V, but not sure if it would cause problems if left like that. The J2 pin on the tool is the one closest to the ground pin, and the ground pin is the longest pin. Easiest way is a 100kOhm (or more, my fingers are something like 3 MegaOhm) resistor from the positive pin to J2. Just be mindful that running at reduced voltage could burn it out as the slower RPM will mean the cooling fan isn't running as fast. It shouldn't be terrible as 18650 based M18 batteries run down to 10V (although they fall off pretty quickly and only spend around 30sec under 12.5V with a 18A load)
@@toolscientist this was far and above what I expected in response. I think the only thing you could've done more for me would be building the wiring harness and shipping it to my door! 🤣 Thank you so much! I certainly hadn't thought about the cooling fan, that's a very good point. I'll make note to consider it's duty cycle halved when running on 12v. Just gotta go out and build the adapter now hah! This will be a huge boon for not having to store a second compressor while offroading! Thank you again!
@@toolscientist you rock. I'm so glad I'm subscribed haha The first Milwaukee video with the charger is what I saw first and instantly subscribed, specifically hoping for a video like this. :D
All hail! Our Australian Backyard Milwaukee Brethren has done it yet again! Eagerly awaiting what you come up with after this discovery...🤔🤔😈 Well done sir! 👏
@@toolscientist nothing special just fixed an impact wrench but it didn't have any batteries to test it only my bench power supply. And using the finger to j2 trick managed to prove it working.
What i found curious is that a Ridgid (AEG) battery to M18 tool adapter i just bought has a 1Kohm resistor bridging + and - to J1 and J2 respectively. Seems to do the trick for enabling the tool. No extra components on the PCB besides that. ( It came with one those contacts slightly damaged, my tool refused to work until I bent it back into place)
You should be able to get an M18 tool to work with any resistor (from 0 to whatever my fingers are [I think 3+MOhm]) between + and J2. Shouldn't need anything on J1.
Any chance you will ever do Ryobi? Specifically their HP line that has the two extra contacts on the back of the battery stem? Word on the street is an HP tool signals the battery it can handle more amps than a standard tool. The HP battery will then allow more amps to flow. Both battery and tool have to be HP line.
I have a Ryobi 36/40V mower and battery, so I'll be doing something with that. Don't have any Ryobi 18V, though. My suspicion is that the extra contacts on HP tools/batteries allow them to bypass the low-voltage cutoff mosfets in the battery.
@@toolscientist Ryobi made the same claims that those extra pins enable communication. In my limited testing, it seems those pins enable the brushless tool to draw more power than a battery without them. So perhaps the tool normally operates with an amperage limit, but with the extra pins that enables a higher amp limit or perhaps a constant power mode. I'm really surprised nobody has looked at this tool communication stuff before, good work.
Are there any plans for testing batteries from other companies, or testing if protocols are consistent for different brands within the same company? For example, will a Craftsman battery communicate in the same way as a Porter Cable, or a DeWalt?
I've got some Makita and Dewalt that I'll eventually get to. I don't think Porter Cable and Craftsman are available in AU, so I'm unlikely to test them. I've heard that you can cut part of the battery off and it will work across Porter Cable, Craftsman and Dewalt, so likely using the same system.
Don't have any OneKey. Someone else mentioned OneKey and I can't believe I didn't think of it. There could be useful info being transmitted and it would be fairly easy to intercept it. Might have to get my hands on a OneKey tool
No, the tool stops when voltage goes below 8.2V or when the voltage across the thermistor ("Th" pin) goes below a threshold (which indicates overheat). The batteries have no disconnect whatsoever. Original M12 batteries only had resistors in them. One from ~2019 onwards have a chip in them, but I'll need to desolder one to see what it does.
@@stevenmullin3616 i should just sacrifice one of mine to the science gods. Maybe I can find a cheap/dead one on a 2nd hand site. There looks to be an i2c interface on the PCB, so if I can find out what chip it is, I might be able to talk to the chip on a live one, which will be fun.
I find it odd that the PC fan you're using for the test has a QR code sticker on its hub. Must be some kind of aftermarket fan 🤔 (I could easily be wrong about that assumption)
Do you see any issues with only using the cheap LVDs for small current draws on the milwauke batteries ? (EG 10-15 amps) We use them a lot for camping situations. lighting, charging other devices, camp fridge etc.. via 18v-12 step down.
Nah, that's pretty much the perfect situation for them. Because the smaller M18 let them discharge down to 10V (2V/cell), you're actually better off using a 13-16V cutoff at low-ish amps. All the low-powered M18 tools (anything with only 2 pins), do exactly this. Only thing is I'd avoid using the 1.5Ah and 2Ah at 10-15A. The cells are rated at 20A, but because they are in a pack, they trigger overheat at around 15A. Also, try measuring the off/standby current of your low-voltage module. I've heard some of them are a few milliamps, which can be a problem. That's obviously easily fixed by taking the battery off when not in use.
@@toolscientist Thanks mate for the speedy reply, that's a big help. Generally use the 4 and 5a/h units with cut off at 16v. good to know i can drop it back a little. I always run a cut off switch so it disconnects the battery from the LVD for that exact reason. (or pull the battery out)
Great video! Do you have any faulty batteries? I've tried repairing batteries by replacing the cells, sometimes they come back to life, but sometimes they continue to flash even with all cells fully charged. I've always wondered if a command could be sent to the battery to start the battery again.
There's an instructable by Noloxs on Makita LXT that covers the gist of it. The 3rd pin is similar to the J2 pin on the M18, except instead of being 3V, it's 0.5V below pack voltage and cuts to zero when a fault is detected. This makes driving a mosfet with it very easy - there's no need to boost the signal like I had to do to get the M18 to drive a high-powered mosfet. There's presumably digital communication on the yellow connector as Makita sells a tool that lets you read battery temp, charge cycles, et cetera.
Hello ,help please 🙏 i buy an 5s bms i put on battery .My battery have 4 pin negative, positive ,ds,cs i dont know last two what means .Only Ds have a pin on battery cs dont have I power up the tools work 3 seconds and stop .What can i do ?
Very lovely video! Learned a ton and helped with troubleshooting a broken tool. Specifically, it looks like the J1 pulse from my tool isn't getting to that 12.7 V threshold. The tool I'm working on is gen 2 blower. I was lucky enough to buy 5 broken units for $50. To breathe some life back into them, I'm trying to figure out this J1 pulse from the tool but don't have a single shot scope to capture the pulse. What I do see is J1 will go high for as long as the trigger is pulled - but the broken blower only goes to 12.5 V then holds steady at 11.5 V. Comparing to the good blower, it initial hits 18.3 V and holds at 17.5 V. These voltages are valid for both benchtop supply and m18 battery. My question to you is - have you any ideas what might be causing that J1 pulse to be so low? Is this going to be buried in the epoxy block they hide their magic? Failing figuring out wtf is going on - I was going to wire in a momentary switch and do the button mash to wake up J1. Maybe add a LED that shows the status of J2 so I know I mashed enough without having to try to fire the tool. I could hotwire J2 from +ve, but I would assume this would disable over-temperature and under-voltage protection since J2 wouldn't be able to go low. Thoughts?
Good debugging! Yes, hotwiring J2 will remove any protection. J1 likely has a fault inside the epoxy, probably a short that's causing extra load and therefore voltage drop on J1. J1 should be at B+ when trigger is pulled (I think). You won't be able to run an LED off J2, the LED will draw too much current. You can connect J2 to a mosfet which powers an LED, though.
@toolscientist thanks for the reply. I realized afterwards that you had an entire section on the max draw of J2 not including the comments lol. I was afraid of the black magic box fault. So either pulse the J1 through external switch, hotwire J2, or boost signal to J1. I'm not overly versed with buck converters, but would that be a valid option to boost J1?
@@toolscientist thank you very much for taking the time to do that up. The cct itself is clear, what I would ask about is how critical the value of the mosfets are. It looks like they are mostly just showing the voltage and there is next to no current demands. I just have to go through my stash of dead boards to see if I have both types of mosfets kicking around.
Question for ya. would using a m12 battery with just a positive and negative adapter to power up components on vehicles like a window motor as an example affect the battery in a negative way ?
The M12 batteries are 12V at full charge. Their average voltage is 10.8V. Milwaukee cuts power when voltage goes below 8.2V and refuses to start if voltage is below 9.6V. Depending on the pack, they'll be good for 15-20A for the compact and 30-40A for the XC packs. So just make sure you're not putting too much current on them and that you're not dropping the voltage too low.
Not a clue. Maybe the lights turn on randomly, but you obviously only notice when you're near. You'll have to setup a camera to see if they're going full Toy Story on you and getting up to mischief when you're not watching.
Hmm, very good video. In the future I plan on both making portable appliances run using an M18 battery, and on making Milwaukee tools run without their battery, so this is exactly what I need. Though maybe you could release a concise document/video with all the key information right there. While I don’t think it’s intrinsically useful to power something while the BMS protections are live, getting power from J2 might be a cheap way of running an MCU without needing a DC-DC converter. Milwaukee dev-boards when?
Hello, does the Milwaukee BMS detect overcurrent and change the state on J2? Will a 2Ah battery have a different overcurrent threshold than, for example, a 5Ah battery? Thank you.
It doesn't directly detect overcurrent (maybe some of the bigger one's do, but I can't test above 160A). It indirectly has overcurrent protection as high currents cause voltage sag, which triggers cutoff from low voltage. 2Ah cuts out around 50-60A, 5Ah cuts out around 80-100A. As the cutoff happens from low voltage, the exact cutoff current changes with age of battery and state of charge
Love this. Always looking for more ways to use the batteries I have. Maybe you'd be one to ask.... Any way to get "higher amperage" (than those 12 converters) to use 18 volt batteries in 12 volt curcuits? By no means at your level of electronics understaning (thou I can follow directions), but the mechanic in me would love to use an 18 volt battery to test/run 15-20-30 amp 12 volt items.
There's a couple of those "15-40V to 12V" buck converters in an aluminium heatsink that claim to do 30A, and lots that claim to do 10A. You may be able to run them in parallel to get even more current. Ideally you'd look at the output on a scope to make sure there's no craziness. If you've only got a multimeter, then you can check the Hz and ripple voltage of a single unit, then compare those results to putting two in parallel. If you don't get crazy frequencies or large ripple voltage, then it's prob ok.
@@toolscientist Thank you for the info, I'll take a look into those and check them with my meter. So many uses I can come up with for a portable 12 volt source, I am surprixed they don't offer something off the shelf. Gtanted, that is old school, a lead acid will do that, but oz to oz.... tool packs are easier to move around.
12ah is 5s3p Samsung 40T. 6ah is 5s2p Samsung 30T. 5ah is 5s2p Samsung 25R. 2ah is 5s1p Samsung 20R. 12ah and 6ah have a physical fuse. Project Farm short circuited an M18 6ah a few weeks ago and got a few hundred amps for a few seconds before the fuse melted. Theoretical continuous draw for 6ah and 12ah is ~70A, but I'm pretty sure it would overheat due to cells being tightly packed and enclosed in plastic.
@@toolscientist I can't speak for other people, but this was exactly what I was hoping for. Thanks for the staggering amount of effort you put into this. That whole aspect of different batteries having different thresholds for things must have been a fairly irritating complication.
Definitely! I said a few hours, but I think that aspect was closer to 10 hours of testing time for something that isn't super important to most people. I was pretty thorough with the 2ah, so I'm pretty sure 700mA is its threshold. I got bored on the 5ah and left it at 2amps, when it's threshold is probably around 1.5 amps. 1 amp kept it awake for 10m40s. The testing time just becomes exponential as each increment in current increases the time that it keeps the battery awake. I nearly had a heart attack when the 12ah behaved completely different, but luckily it was the only one. I was also really dumb and did a lot of testing with resistors before I realised that I had a battery tester that is essentially a programmable load!
Whats the J2 output from the battery to the tool? I tried to connect the Parkside battery to M18 grinder, putting 4k resistor on + wire and connecting it to J2 but its not working. Btw. great video! cheers!
It's supposed to be 3V, but should work with anything above ~0.72V. It should work when connected to B+. You definitely connected to J2? J2 is the one closest to B-
@@toolscientist haven’t tried it cause now when i connect multimeter to + and -, i see they are connected without pushing switch. Maybe i shorted something
I don't think we can get Einhell in AU, but apparently Ozito is just relabelled Einhell. I'm currently doing Makita, and then Dewalt. Might get to Ozito/Einhell next year.
The charging side is a bit more complicated. But you could design your DIY batteries to have a JST balance connector and then just use a hobby charger.
Dude I think your on to something about the current not being considered in regards to the low voltage cutoff as I’ve killed two 8ah batteries with my top handle chainsaw 2826-20, I feel that the tool is either pulling too much current overloading something on the board or along the current pathway as the strings are within .3v and the overall pack voltage is at 18 but throws the red and green flashing light and will not charge ? Any thoughts ?
0.3V is fairly imbalanced. Not enough to prevent charging, but enough to suggest that cells are failing. The Samsung 40T in the 8ah and 12ah had fairly high failure rates. You could try measuring voltage of each bank of cells under load to see if one is dropping voltage more than the others. When you put the battery on the charger, does it make a hum that gets higher pitched in about 4 steps? It might even repeat this 2 or 3 times before giving the flashing green and red. If so, I think that's the charger detecting too much resistance. Only other thing to try is buying a logic analyser and recording the signalling between battery and charger, then compare your data to what I get in my [004] video.
I don't think we can get Parkside in AU. We get Ozito, which is identical to Einhell. I see that Parkside is owned by Einhell so are they compatible with either of those?
A question: could there be some wireless communication happening? I ask this because newer M18 batteries have a OneKey symbol on the service tag and what appears to be a unique address, I've done some fiddling but never could get the OneKey app or anything else to recognize it. That said, even if there is some wireless communication happening, as you've demonstrated it still doesn't result in any changes in the battery's output.
Do you think it's possible to charge more than one battery with one charge? I was thinking about trying to do some wiring wizard to see if it would work.
Please man. This is amazing. You are bringing truth to the power tool industry. U have to do this analyses to makita since makita is the king of knockoff Chinese random tools and battery
I should be getting to Makita soon. Makita will be hard as they like to brick their batteries. I might have to start with the knock-offs first to avoid bricking too many batteries.
@eduardoramiro4846 I'm going off what I've heard, but apparently if a LXT battery is put on a charger and gives an error, it records it. If you get 3 errors in a row, the battery bricks itself and won't charge ever again and can't be repaired. So doing crazy stuff like I've done to my M18 packs (undervolting and overtemp) might trigger the battery to brick itself.
I don't have any One-key. Other people mentioned it and I looked into it. From what I can see, it's tool only. The only thing I saw that might be battery is a screenshot showing battery capacity, but they could be getting that from pack voltage. If you can see battery temperature, manufacture date, charge cycles, or similar in one-key, then that would indicate that there is digital comms between battery and one-key tools.
You should check the Dewalt battery BMS. there's basically no protection. The protection comes from the tool. I had 3 batteries burst in flames after running it on an old 18v dewalt circular saw.
Yeah, they're the most basic of the major brands. They're interesting to me as you have direct access to the thermistor and cell voltages, so lots of scope to make DIY adapters that don't overheat or overdischarge
Yes, but very unlikely. The tool will keep going as long as it receives > 0.725V on J2. Using a power supply, I've run these tools as low as 5V (although they barely run at that voltage). However, a connection issue on J2 will just make J2 at the tool 0V, and the tool will stop. The only way it could happen would be a short circuit between + and J2. J2 is closest to the ground pin, so it is very unlikely that this could happen.
Looks like their previous gen was licensed M18. Their website even says they partnered with TTI. M-Cube looks to have same pinout as M18, but different physical interface. Do you have these tools?
Hi i would like to know how to rest the M18 bms after removing it from the battery? I hade two batteries, one with with broken cells and one with broken bms , so i remeved the broken bms and put it to the battery that had the good cells but when you press the button of the bms you get nothing . Is there a way to reset the bms without puuting it on the original charger ?
The tool coms cant use complex coms because the interference would just make the tool unreliable and damage brand reputation. Charging on the other hand, in a controlled electrical environment means they can use digital coms to charge optimally and also prevent cheap chargers from destroying their brand reputation.
Yeah, I originally had a section of the vid discussing the noise in J1 and J2 during operation and saying that was the reason for no digital comms. I left it out as I'm not sure if that noise could be filtered, so didn't want to categorically rule it out. I know that Makita XGT tools use digital communication and the tool won't run unless it receives that communication. I'm pretty sure they have a separate isolated ground for comms to get around the noise issue. I have heard from one person of having problems with their XGT tools and having a rep come and clean out the batteries for him multiple times. In the end they gave up and went back to the older LXT system.
This compared to Dewalt batteries which... directly connect the cell terminals to the battery leads. Nothing stops the tool from working from any power supply. Nothing stops the battery from being charged or discharged.
Yeah, Dewalt is the best battery for the apocalypse! Dewalt tools are slightly harder to trick as you need to trick the ID pin and Th pin, M18 just needs 3V on J2. Dewalt is obviously much easier to charge.
@@toolscientistYou sure that Ryobi isn’t the best for the apocalypse? NiCd compatibility, low voltage cutoff in the battery…much less communication than most other brands.
Haven't seen anything, but I haven't really looked, either. I thought about doing it, but don't really have the time, and I figured it'd just get copied and I'd be undercut. I should finish my updated circuit and publish it, then someone might make them.
@@toolscientist After putting some more thought into it and spending a lot of time looking at mosfets and mosfet circuit design, I realized a couple things. 1. I really can just use a relay/contactor to switch high power loads. Which is exactly what I do now except using a tiny non-adjustable LVD circuit. This can fit entirely inside the case of a particular M18 battery adapter so it's a super clean install. However, the circuit cuts out way too early though so I'm not getting anywhere near full capacity. Hence why this subject is so interesting to me. 2. For my use case (repowering childrens toys) it dawned on me that since I've added speed control boxes that use low current key switches to turn on and off, I really just need to interrupt that line with a relay/low current mosfet instead of having to fully shut off the battery. Now the hard part is probably that none of the aftermarket battery adapters include provisions for the J1 and J2 contacts. And I really don't want to deal with 3D printing... I smell epoxy in my future.
@@toolscientist Exactly my point, I described it poorly. My realization is that I don't need to worry about a high power mosfet circuit. I can use a very compact and simple circuit.
That's actually a pretty good solution. I always neglect relays as they're more expensive and wear out, but in this case it makes for a very simple circuit with only 2 components (3 if you add the pulldown resistor for the mosfet, maybe 4 if the relay needs a current limiting resistor)
Wait, Milwaukee has not released a toaster oven? 😆 (There's a video of someone using a heat gun toasting toast though. Interesting there's a heat gun, that has to be massive current!)
All of my high-output ones do, and most of my 18650 packs, too. Seems any battery newer than 2016 has it. Do you have any one-key tools? Can it tell you any information about the battery? Best I can find is the radius light will tell you remaining capacity of the batteries, but they can just guess that from the voltage.
The 12ah does have mosfets on the ground rail. @ThriftyToolShed showed them in a video here: ruclips.net/video/TphIrEWvW68/видео.html It's possible that they could throttle the power with this mosfet. PWM would likely play havoc with the tools, unless Milwaukee knows that all their tools have sufficient input capacitors. They could also run the mosfet in the ohmic region, but that would make a lot of heat, although they do have a sizeable heatsink on it. They'd need to add ~20mOhm to make the 12ah look like a 5ah, which would create 50W of heat at 50A. Doable, but seems unlikely. The 6ah high-output is similar resistance to the 12ah and it doesn't have any power mosfets. I think it's most likely a power cutoff. I'm hoping I can get some probes on the gate pin and see what it's doing.
Probably going to do the big 3 first (Milwaukee, Dewalt, Makita), then get on to others. Bosch has a pretty small presence in AU, so might be hard to find 2nd hand stuff.
@@toolscientist Okay, looking forward to watch those videos as well :) In Europe the three brands you mentioned + Boch are the most common. For a suggestion for future videos it had been awesome with some videos on the DIY battery kits that are sold, to see if the bms used in those kits are ok etc. A video making your own kit had also been awesome ;) I have made my own power tool batteries for a couple of years now and im really happy with them. Way better performance (because of better cells) and a much lower price is making it well worth it.
Here it's probably Hikoki, then Bosch. I think Metabo might be below Bosch - they probably do better in EU. In the future I can hopefully get people to loan me their tools for a week, otherwise I'm going to end up with a very expensive and barely used tool collection! Yeah, I'd like to test some of the 3rd party batteries and DIY kits to see if they've implemented the same charge/discharge/temp limits as the manufacturer. Problem is they can change fairly regularly.
i just wathed all your Vid's , i EXSTREAMLY liked the -jump starting one , and i think your testing method is GRATE and Newb-friendly while be a advanced subject .. i think you should keep making them , if you need ideas , im full of them lol .. but to get you the viewers, do some thing like this- with Auido equitment , like amplifyers, get a Hot-ticket one, that dosent brake the bank , (like a Scar-auido RP 2000.1, ) you would be the only person doing pratical test on them.. most of those auido-guys are dumb as rocks... also a good viewer-bass is doing those E-bike and E-scooter crowed , they never doing good testing like what your doing.. wish you the best .. PS, i want to see your High-power diagram , you shouldent hide them becusye of stupid people. .. LIke moma-allways said "stupid is what stupid dose" , and a book that is sencored is not a woth-reading book, (like the antarkest cook book, you cant buy the Real thing thire in the EU/UK, - its almost 2x thicker in the USA/Mexico/Canada - even blue-prints for a AR-full auto trigger asimibly in both .22-LR & the .223 chamberd
Still working on vids. Things just got slowed down a lot over Xmas. Hoping to have my Makita LXT vid finished this week. I'll probably stick to tool related tests. If I had an escooter I'd definitely look at modding it to work with power tool batteries. Although I think that's been done, and I want to avoid rehashing stuff. High powered circuit was mostly not shown because it's a bad circuit that can lock on are cause pulsing. I'm working on a better circuit which I will make public.
Mate i repair power tool Batteries for a living ! Just randomely found you So subbed and liked now hoping that i am going to search your vids and find something on the worst of the worst BMS mentioning no names like MAkIT& what a complete Rip Off Bollocks Bullshit BMS that is. Cheers mate.
Ok mate just checked and nothing there. You may allready be on this one i dont know. But buy a new Makita battery and a duff one. Work out the terminal address lines and monitor as i can see that you are well capable of doing. Now change one or maybe two duff cells in the bad battery, you will get the battery monitor blinking from side to side. Balance ( manually ) all of the cell pairs to 4.1 or 2 and and try charging on a makita charger...it will still say NO whatever you do the BMS locks out and the only thing you can do is chuck it in the bin and fit a new one then you are good to go again. They have the nerve to sell a £300 tester that tells you that you have charged your battery 97 times, or that you charged it once when it was really cold or even three times when it was really hot, it even will tell you what is wrong with your battery ? Maybe or say oh sorry computer says No. There is a memory chip on the Makita BMS that is like no other and when it locks you out thats it ! All the best and good luck Kev
I am currently working on LXT. They even have fuses that the BMS can burn and physically disconnect the B+ pin. I'm about to get to the part where I start overheating it to see its response. I bet there's a limit where if I overheat it too many times, it'll lock itself.
I'm sure other brands are doing lazy/stupid stuff, too. Milwaukee is just the first to go under my microscope. Letting the battery disable and enable the tool every 50ms is pretty high on the stupid scale, though. I hope none of the other brands are doing stuff like that.
Try monitoring the entire run cycle of the vacuum cleaner. I used my M12 vacuum to compare lifetimes of used batteries. Maybe four or five times across a 15 minute runtime there were audible changes in the motor note, I was curious if that was an artifact of redlink communications. There was zero load for the entire test. M12 tools show the charge indicator on the tool rather than the pack.
M12 tools don't have any digital communication. It's all done with analog pins. The tool is just reading temperature directly from the T-pin, which is a thermistor. I've done a few rundown tests with the blower, but that's a brushed tool that's not likely to have anything fancy in it. Can't say I've ever heard any sudden changes in motor speed from the backpack vac, but maybe there is something.
![[010] Why do M18 batteries charge to 3 bars?](/img/1.gif)
This is the kind of thing that makes RUclips great. Thank you for sharing your findings with fellow scumbags of the world!
Scumbags unite!
Wow ... 😳
I wasn't expecting the sheer lack of communication
Thank you for all the research you've publicized! 😁
Was waiting for this video! As a software engineer that dabbles in electronics/embedded and who deals with unnecessary and unbearable complexity in his work, it's nice to see a simple signalling mechanism used. Less to go wrong, cheap/easy to implement (as you showed), and doesn't need reliable checksums/retransmission/ack and other nonsense to happen when a tool is rattling away in terrible conditions.
The activation pulse seems unnecessarily complicated, though. Don't see why it needs to be a pulse of a certain time, rather than just J1 high brings J2 high. The consequences of J2 accidentally being activated are just that the battery consumes a few microamps for ~3min.
There's some nice things you can do with a full digital protocol, like having the battery transmit its max current, continuous current, max temp, et cetera. Then tools that can control the load (blowers, vacuums, pumps, lights), can adjust to avoid overheating the battery. I know that Makita XGT has the batteries communicate digitally with the tools, and the tool won't run without this communication, which does make them less robust. It will be interesting to see if they're doing useful things with it, or if it's just there to make it harder for 3rd party makers to copy their batteries and tools.
A new RUclips channel popping up with this quality of videos straight away is so cool to see. You could be huge if you keep at it!
The old(er) M28 batteries would show flash codes when the battery level button is held down. So, I'd think the M18 has something equivalent. Maybe there is a variation on the signal to ask the battery to output diagnostic data. Milwaukee may have a rig for warranty/qc to get an idea of what kind of life the battery had: Number of over temp events, max current draw, charge cycles, etc... The higher end "one key" tools could be the ones to look at if they communicate with the battery since they have "smarts" and use Bluetooth for tool settings. It doesn't make much sense to have the communication if there is nothing useful to do. Thanks for the great videos!
I have 2x M18 batteries that give those flash codes, although I haven't had time to look at it yet. Someone posted a fairly good guess of what they were in a forum or reddit (something like manufacture year, week of manufacture, etc). They're both 2013 batteries, whereas all my other batteries are 2015+ and don''t give any flashing led codes.
Good thinking on the one-key tools. I was going to try and send queries to the battery based on what seemed logical from the charger messages, but a one-key tool could save a lot of effort there.
@rossradford5767 that is interesting. Can you put a quick vid of the sequence on youtube? You should be able to post youtube links in the comments here, but any external link will block your comment without telling you. If you can't post the link, post the title so I can search for it.
@@rossradford5767 I've saved it. Haven't had a chance to view it yet.
That pulse duration trick had me dying. Literally could've never thought of something so simple to generate a varying duration signal haha
scumbag wave generator amazing
I love the depth of information you provide in all this - not to mention how well you detailed what you did
Doing Australia proud and keeping the “fuck you, because I can” mentality thriving
We have discussed this briefly before, but EGO 56V packs do have communication with the tools through the D-terminal. The tools will run for around 10secs or so without the D-terminal, but it seems a watchdog will shut it down after the 10 secs with no communication between the two. In case you wanted to see a pack with communication. The BMS has 2 microcontrollers and is quite advanced with temp monitoring boards in between the pack halves that uses transistors PN junction for temp monitoring and 2 NTC's as well. The board even has a calendar chip for storage mode I presume. I wanted to look more into the communication protocol, but have not had any time to go back to it. I threw all this on a video called "What is the D-terminal?" Since so many viewers asked about it. I also shared that someone else could take the information and go further than I did. I don't have the best equipment at this time to go much further. I hope you get time to look into it. Great video!
Interesting. I'll have to watch your video. I'm happy for you to direct link any of your videos.
Don't think I'll be getting any Ego anytime soon. I do have some Makita XGT and my initial tests show that the tool won't even run without digital communication. A user called Malvineous on github has figured out some of the XGT protocol between battery and charger. Will be interesting to see what it's saying to the tool.
As with you, it's a matter of finding the time!
Here's the video for anyone else: ruclips.net/video/H-hggUXgYlU/видео.html
Search showed me the T-terminal video first, so I watched that thinking you made a typo!
@@toolscientist Thanks man!
Massive thank you for the info you're sharing, mate. I found your channel while trying to figure out an issue (well, ONE of the issues lol) with my M18 12Ah batteries. I have tinkered/hacked/repaired many m18 batteries over the years, including my 2 12ah. The reason i watched this video is that my 12ah batteries have changed the way they 'behave'. Presently, the seem to function as the units you show in your video. However, this was definitely not always the case.
I used to be able to connect them to other branded tools and they would 'wake up' and worked fine. One adapternthat was always quirky was one to my dyson v7. Older milwaukee m18s wiuld work fine, but with the 12Ah the vacuum would initially pulse, but i discovered that flicking the trigger several times would get it to power up.
Then one day a year or so ago, both my 12ah batteries mould not work through any adapeters, but worked fine on milwaukee. Some troubleshooting with a multimeter showed the 'activation' happened by to tool ao my work around has been to have a milwaukee tool activate the nattery then do a quick change over to my diy adapted tool. It works as long as i keep it 'awake' but, you can imagine how much fun THAT process is!! 😂😵💫.
Anyhooo, I have searched for answers as to how the behaviour of these batteries could have changed, without much luck. The only thing i can think of is that i have a one-key impact wrench, and could Milwaukee have pushed an OTA update to the batteries via the one-key bluetooth network??!
The 12Ah is the only battery with mosfets in it (actually, the 6Ah forge does as well). Their's 2 of them under the brass heatsink. So there could be something going on with those. I haven't got around to testing what conditions make those mosfets turn off, though.
Here are some other things that I'm certain have changed in the way BOTH M1812Ah batteries from when i purchased them:
-When connected to a tool with with an led light, eg drills/impact drivers, the light either stays on constantly, apart from flickering every 8 seconds (I remember one of the things your investigations found was the 8 second 'activation' time of the 12ah). I only first took notice of this because my drill ( with an almost fully charged 12ah battery connected ) was left I. The boot of my car, a week or two later, I was surprised to find the battery completely dead. When I charged the battery and reconnect to the drill I payed more attention and noticed the change in behaviour. No other M18 or M12 batteries do this other than BOTH my 12ah m18.
-when initially connected to any tool, there is a delay in delivering power. Sometimes I have to flick the trigger a couple of times before the tool will work.
-With my Milwaukee chainsaw, however, things get REALLY strange. After connecting either of the 12ah batteries, the saw will generally work first time. However, when I release the trigger/throttle things get a little confusing, as MOST of the time, when the trigger is re-engaged the chainsaw does not start but hitting the trigger several times the saw fires up again. THIS OROCESS IS so Anti intuitive, as the number of times the trigger has to be toggled is sometimes 4, 5 or even more! I first noticed this behaviour after having the chainsaw warranty repaired, and my initial thoughts were the trigger was failing. My first remedy was to take the battery out and reconnect it which seemed to work. Over time though, Ive narrowed it down to the battery, and toggling the trigger ALWAYS works. Part of me theorises it to be a safety measure by Milwaukee incase of accidental pressing of the trigger.
I have confirmed that when paired with a regular, older, M18 5ah or 6ah, the chainsaw fires up at every trigger press, and up until I received it back from the warranty repair, the trigger worked first time every time with 12ah also.
All signs point to software update in the 12ah, but nowhere have I found any information/documentation regarding this, or whether this is even possible via Bluetooth??
Sorry for my long post, and thank you in advance for any light you might be able to shed?
Cheers
@@neddraeger2013 it sounds like the internal mosfets are getting turned on and off, which is something I haven't seen. If it were only 1 battery I'd say your PCB is faulty, but unlikely that 2 would do the same thing. Do you think you could have damaged the PCBs when you repaired them? Maybe they can detect when something has gone wrong (like cells removed) and the battery goes into a safe state where it keeps the mosfets off until needed.
I hadn't considered that it could be the result of damage. I suppose I can't rule that out as a possibility however, I would be staggered if it were, for a few reasons.
1. Like you mentioned, the fact that it's the same with both 12ah, and ONLY the 12ah.
2. I bought the batteries around two years apart, yet the issue/change in behaviour appeared to happen to bother pretty much at the same time (when I first had the issue with one of the 12ah, the first thing I did was swap batteries to see if the issue was the tool.
3. Bother batteries had been opened countless times before for cell balancing etc without incident.
it's definitely a head scratcher
@@toolscientist-I thought I should add that I haven't removed/ replaced any cells, only tested/remedied any voltage discrepancies.
Another awesome video packed with great info and testing! Can't wait to see you do this for xgt packs if that is the plan :)
Also, Milwaukee and horrible bms engineering with false marketing, wow that was a surprise /s
Got one more project before XGT packs. From what Malvineous posted, XGT is about 20x more complicated than M18, so could take a while.
@@toolscientist yeah I bet it will be a doozy but I believe in you 👍😁
@@toolscientist I want to see Flexvolt batteries too please!
Love your impulse generation solutions
So much info to take in. I'll have to watch a few times I think.
Again great info also was sort of waiting for this video. This is something I’ve been curious about since I started using these batteries to power projects. Having a sheet that breaks down information about battery communication protocols as well as well as recommended wiring and a pcb / gerber file would be very appreciated if you have the bandwidth. Even just running list of everything you think you know would be useful to have in one place.
It is sorta funny how little seems to be actually going on in the communication protocol though lol especially when these companies really play up the secret sauce in these batteries
Yeah, I should do a written version of all my tests. Video is not the best format for a lot of this stuff, but it has much better reach.
You made a very good video with a detailed explanation.❤️
Well done my Friend🙂🤝
Thanks! I enjoy your tool videos, too.
Petty Cool! I enjoy geeking out on stuff like this from time to time. As your channel grows, I'd like to see maybe how to make adapters for M18 to M12 or Flex 24 to M18 and or M12.
Thanks for this valuable contribution to the tool anarchists of the world! I especially like the mechanical function generators
The circuit you built with mosfets is similar to whats in the greenworks 40v packs i have.i havent layed out a schematic but the layout and function is really similar.
Greenworks have low-voltage cutoff inside the battery packs? That's interesting. Never even seen a greenworks tool in person, so I have no knowledge of them. Only ever seen them in a few comparison reviews.
I think they have a voltage signal cutoff inside the pack. There is 3 relatively beefy mosfets and an aluminium heat sink. If you measure the + and - terminals you get 0v when the battery isn't connected to a tool/charger. You can short the pack and nothing happens. One of mine got put on a charger wrong and discharged to 9.8v....thats a 10s pack.
I was wrong. The pack was 9.8v last time I checked....the cells are all now between 4.11 and 4.13v and the whole pack is 41.2v. There is 3.6v across the main power terminals of the pack. This is really low current stuff so some shenanigans are happening but there is 35.2v between the - output on the board and the B- into the board. The resistance is not equal in both directions either
I’d be very curious to see this with Dewalt’s packs. I’ve only been able to draw 5A from a 5Ah power stack before it shuts down
Don't have any Dewalt's yet. Powerstack is the only one that cuts power. It's most likely some simple connection on ID or Th pin, otherwise powerstack wouldn't work with older tools.
@@toolscientist Would love to see a video if/when you do!
Great demonstration. Needed that. I hate those hidden "features." It should be VERY clear on a device or a battery where the limits are and how to re-create it with other parts, if we can't buy the originals any more, but have a great tool hidden to spring into life instead of throwing a lot away. It applies to more devices than you think.
Hopefully there'll be some standardisation in the future. It'll probably require the Brussels Banhammer to make it happen, though
Milwaukee think they can fool the public until they encounter you Mr. Scientist, wow 👍. Your analysis is truly out of this world..!
I also have a few M18 & M12 batteries (cell from Korea & Singapore) and very tired of their LOW LOW quality cells where the charger shows red/green flashing. My M18 battery is fully charged and still shows 3 bar status (very common). My M12 cell pack loses voltage daily and a month later, each pack will show 3.7+4.1+4.1 causing charger to flash red/green. I am not a contractor so I don't use my battery often..!
Recently I found an adapter that can convert Ryobi battery into Milwaukee tool interface and it's been working so far b/c I don't see the value of Milwaukee batteries. If this adapter performs well, I will buy Ryobi battery on sale and use adapter for Milwaukee tools to save money !
Some of their packs are good. Anything with 3000mah 18650s (M12 3Ah & 6Ah) or 4000mah 21700s (M18 8Ah and 12Ah) are prone to failure. In M18, their 5Ah and 6Ah HO are the most reliable.
@toolscientist you are 100% correct, my failed battery is M12 6Ah & M18 8Ah !
Brilliant work. All hail the algorithm!!
The algorithm giveth, and algorithm taketh. All hail our algorithm overlord!
Be careful of FET ratings! If you are unlucky, the FET you mention (IRFZ44) will when 3V is applied to the gate, pass 50uA or so. Vgs(th) is never a measurement you should use, look at Rds(on) voltage gate spec, this is at 10V. At best when using this FET, you need to verify that the Rds(on) at 3V vgs is low enough to not promptly burn up. High current FETs rated for 3V gate voltage are nonexistant in through-hole. Look at 'drive voltage' on digikeys selector, this shows 12 FETs from 30-170A and 3V or under gate drive option with Vds >21V. CSD16321Q5C is probably one of the least bad ones. They are all annoying to mount. A single transistor 3V->12V booster will fix this FET selection problem and makes the FETs you mentioned just fine, rather than a crapshoot on if they meet the high end of the spec (won't work at all), low end (work fine) or burn up a few minutes in.
I'm not using the IRFZ44, I'm using IRLZ44N. In the datasheet, Fig 1 & 3 show that it is rated to just under 20A at Vgs = 3V, hence why I said 10A for that simple circuit. My high-powered circuit (which I blurred to avoid inexperienced people from building) is using a booster setup to drive my 2x IRLB3813 at Vgs = 17V (dropping to around 10V as the battery depletes)
@@toolscientist Oops on the part number. I note you can't quite trust Fig 1 and 3, Vgs(th) is defined as varying by a volt maximum, typical is undefined. Id can be as low as 250uA at 2V Vgs, if Vgs(th) is worst-case at 2V. The Rds(on) spec @4V/21A anchors this@0.035R .735V. I am surprised! This is the first datasheet I have checked on which typical seems close to worst case, as Fig 1 seems very close to .735 at 21A/4V. This has unfortunately bitten me before.Expecting 10A easily due to Fig 1 on a different datasheet, and finding 10mA due to the part having a worst case Vgs(th) spec. Great work.
All good. You are right, though. The IRLZ44n isn't intended to be used at Vgs = 3V, it just happens to work decently there. So yeah, variance could be quite high between samples.
I was permanently banned from a Milwaukee subreddit for saying exactly what's in this video about 9 months ago by a mod who was claiming all sorts of advanced Milwaukee communication.
I think I know that subreddit, and that mod. Is it the smaller subreddit - MilwaukeePowerTools? Until I did this test, I thought there was at least some communication, even if it wasn't as fancy as Milwaukee tried to claim. I was fairly surprised that it didn't even differentiate between batteries as it's common for people to claim that the tools can detect high-output batteries.
Yep! That's the one...lol. It was weird because he was claiming the charger was doing stuff I've never seen it do and I monitor the voltage and current of every battery I charge.
I simply said he was wrong (didn't know he was even a mod at the time) Immediate permanent ban.
Yeah, I've seen him make some very strange claims in a few subreddits. If you've seen my power testing video, he's the guy I got into an internet argument about the drill being 1500W. I think he has some issues, so I just leave him alone.
Interesting on your charging tests. Have you got a jig, or do you wire the terminals up each time and measure current/voltage from the leads? Any interesting discoveries?
@@toolscientist I use a non Milwaukee charger that displays that info now.
Before that I built a little jig from an old dead battery pack I cut out the terminal section (connected to charger) and the bottom of a dead drill (I put the battery to be charged in) connected with wires I would have access to.
It's not really necessary for legit Milwaukee batteries, but I like to buy the knockoffs just out of curiosity a lot so like to keep an eye on what's actually happening.
The only real useful information I've seen is that charging at the rapid charger rate very often leaves the cells unbalanced sometimes...but you have to open up the pack to check that after anyway.
Microsoft (the African Americans of software thanks to them having more oxygen and blood flowing to their aggression sector than the rest of their brains) are the same way. They've practically dominated Mojang and corrupted them in more was than one, and their horrible design tactics are quite literally killing Minecraft in thousands of screwed up ways.
J2 is a safety feature. I know guys who have broken wrists from the tools attachment, like a hole saw, catching in material and the entire drill spinning. It eliminates the need to have a secondary Handle. As someone who works in commercial construction It’s actually a pretty nice feature to have.
J2 is a safety feature for the battery to tell the tool that it has overheated or reached its low voltage cutoff. The anti-kickback that you're talking about is only in a few tools and is only in the tool. The tool has no need to tell the battery that it wants to stop, the tool just stops.
You can somewhat see this when I stall the grinder. The grinder just stops with no change on J2
The good thing about running high amps through cheap alligator clamps...is it makes cheap, stiff cables feel soft, fkexible, and high quality for a few moments before they completely melt.
Awesome video! It's fun to poke at Milwaukee engineers about Redlink whenever they swing by. From what they have told me directly, redlink is very dumb so they can maintain backwards compatibility with all their tools. Only some tools and I believe the forge batteries will actually throttle the current (by reducing the voltage with a mosfet) and this is done with some type of check right after startup.
Yeah, I've got 2 batteries from 2013 that have standard Redlink and not Redlink PLUS. They use basic pulses with the charger rather than serial data. Maybe it's just the one-key tools that communicate? The 12ah has a large mosfet that might be able to throttle power, but I didn't see any different behaviour between tool and the 12ah.
Just curious about how you're talking shop to Milwaukee engineers. Do you sub-contract for them? I understand if you're not allowed to say anything. If you can talk, I'd love to hear any inside gossip :-)
@@toolscientist Haha no I do not subcontract with them, but I have a couple friends and coworkers who work/have worked with Milwaukee. I'll poke one of them for which batteries have the power reduction on them, my guess is it's the HO batteries with 21700 cells.
Also some fun gossip is they had a hell of a time getting pouch cells working, they were having issues with them puffing from extreme loading so they were late to release. Dewalt beat them to it and they get a little upset if you mention it.
@jackphilp7057 yeah, it's hard to beat a cylinder as a pressure vessel. Whatever weight you save from going to pouch cells, you have to put back on the enclosure.
I think the best use of them is making sizes that aren't achievable with 18650s and 21700s. The Dewalt 1.7ah seems like the only one so far that offers anything over cylindricals.
Very cool. I found your videos by trying to see how Milwaukee TICK trackers work technically speaking, surprisingly nobody's covered it yet. I guess it's Bluetooth and broadcasts a unique identifier somehow. I thought I could make a separate microcontroller to detect them instead of using my phone.
Also, you are the stuff of legends. Keep it up in all aspects.
On a tangential note, I just found that Makita batteries support timed lockout, apparently even quite old ones. Makita BPS01
I’ve been using parkside for a while and it seems those do have some sort of communication you can set max amps in battery and also change some tool settings in the parkside app
Fist video of yours I have seen. Hit subscribe within a minute. Great work can't wait to see the rest of your projects!
It’s obvious that there is no performance communication when you look at the way the tools perform. Thanks for confirming!
I could see them at least doing a basic handshake, and if that fails, the tool uses a lower overload current. In the grinder tests at 4:39, I'm pretty sure the 12ah and 6ah are the tool cutting out when it hits ~160A, or maybe overheating the ESC. So if they did have even basic comms, I could see the tool lowering its stall current for weaker batteries.
The batteries should be pulling J2 low when they detect that they've exceeded their current limit, but it seems Milwaukee relies on the batteries tripping low voltage or overtemp rather than looking for overcurrent.
@@toolscientist That’s my finding as well. The 2Ah will droop voltage and overheat in a blower before it ever cuts. Tragic lol.
Yes! Great information. Thanks for the follow up on this topic.
This is super intelligent! Send the battery a pulse, get power. 😆
So since there's no useful communication from battery to tool, do you think i could frankenstein a a cable from an M18 tool to a 12v set of alligator clamps?
I keep the M18 inflator in my car but worry the desert heat will destroy the batteries, and I've always wondered if it'd possible to run it off my car battery in a pinch. Of course, i know it'd run at a reduced speed (if at all) given it's 12-14v rather than 18-20.
Definitely. Any M18 tool can run off external power. All you have to do is put > 0.72 volts on the J2 pin, but you can use 3V, 5V, 12V, etc. The highest voltage I put on J2 was I think 25V, but not sure if it would cause problems if left like that. The J2 pin on the tool is the one closest to the ground pin, and the ground pin is the longest pin. Easiest way is a 100kOhm (or more, my fingers are something like 3 MegaOhm) resistor from the positive pin to J2.
Just be mindful that running at reduced voltage could burn it out as the slower RPM will mean the cooling fan isn't running as fast. It shouldn't be terrible as 18650 based M18 batteries run down to 10V (although they fall off pretty quickly and only spend around 30sec under 12.5V with a 18A load)
@@toolscientist this was far and above what I expected in response. I think the only thing you could've done more for me would be building the wiring harness and shipping it to my door! 🤣
Thank you so much! I certainly hadn't thought about the cooling fan, that's a very good point. I'll make note to consider it's duty cycle halved when running on 12v. Just gotta go out and build the adapter now hah! This will be a huge boon for not having to store a second compressor while offroading! Thank you again!
I'm always interested in people hacking their tools. This one was also a fairly simple answer, too!
@@toolscientist you rock. I'm so glad I'm subscribed haha
The first Milwaukee video with the charger is what I saw first and instantly subscribed, specifically hoping for a video like this. :D
All hail!
Our Australian Backyard Milwaukee Brethren has done it yet again!
Eagerly awaiting what you come up with after this discovery...🤔🤔😈
Well done sir! 👏
So much win! subbed
Treat video! Thank You for all the work!
I love that you're doing this 👍
Wow exactly what i was looking for - thanks
What shenanigans are you trying to do with M18 batteries?
@@toolscientist nothing special just fixed an impact wrench but it didn't have any batteries to test it only my bench power supply. And using the finger to j2 trick managed to prove it working.
What i found curious is that a Ridgid (AEG) battery to M18 tool adapter i just bought has a 1Kohm resistor bridging + and - to J1 and J2 respectively. Seems to do the trick for enabling the tool. No extra components on the PCB besides that. ( It came with one those contacts slightly damaged, my tool refused to work until I bent it back into place)
You should be able to get an M18 tool to work with any resistor (from 0 to whatever my fingers are [I think 3+MOhm]) between + and J2. Shouldn't need anything on J1.
My brain hurts but I love it🙌
Any chance you will ever do Ryobi? Specifically their HP line that has the two extra contacts on the back of the battery stem? Word on the street is an HP tool signals the battery it can handle more amps than a standard tool. The HP battery will then allow more amps to flow. Both battery and tool have to be HP line.
I have a Ryobi 36/40V mower and battery, so I'll be doing something with that. Don't have any Ryobi 18V, though. My suspicion is that the extra contacts on HP tools/batteries allow them to bypass the low-voltage cutoff mosfets in the battery.
@@toolscientist Ryobi made the same claims that those extra pins enable communication. In my limited testing, it seems those pins enable the brushless tool to draw more power than a battery without them. So perhaps the tool normally operates with an amperage limit, but with the extra pins that enables a higher amp limit or perhaps a constant power mode. I'm really surprised nobody has looked at this tool communication stuff before, good work.
Great work, thanks for sharing your knowledge.
No wuckers!
Are there any plans for testing batteries from other companies, or testing if protocols are consistent for different brands within the same company?
For example, will a Craftsman battery communicate in the same way as a Porter Cable, or a DeWalt?
I've got some Makita and Dewalt that I'll eventually get to. I don't think Porter Cable and Craftsman are available in AU, so I'm unlikely to test them. I've heard that you can cut part of the battery off and it will work across Porter Cable, Craftsman and Dewalt, so likely using the same system.
Would be very interesting to test other tools. Makita for example: they have an order of magnitude more advanced batteries...
Yeah, Makita will be interesting. Also a lot more effort to crack...
@@toolscientist yeah, their tools are way more picky about their "meals"
Did you manage to test any one key tools to see if they are different and do talk to the battery??? I didnt notice any
Don't have any OneKey. Someone else mentioned OneKey and I can't believe I didn't think of it. There could be useful info being transmitted and it would be fairly easy to intercept it. Might have to get my hands on a OneKey tool
This channel needs to work with torque test channel on some Milwaukee mods
Opposite sides of the planet, unfortunately. He's welcome to use any of this info to power M18 impacts with a custom battery.
Great work, mate!
Do the M12 batteries use the same J1 J2 communication system as their M18 counterparts? Or do they have an integral disconnect in the battery?
No, the tool stops when voltage goes below 8.2V or when the voltage across the thermistor ("Th" pin) goes below a threshold (which indicates overheat). The batteries have no disconnect whatsoever. Original M12 batteries only had resistors in them. One from ~2019 onwards have a chip in them, but I'll need to desolder one to see what it does.
@@toolscientist This is great! Thanks for the info. Looking forward to another video when you tear into one of the newer M12 batteries!
@@stevenmullin3616 i should just sacrifice one of mine to the science gods. Maybe I can find a cheap/dead one on a 2nd hand site. There looks to be an i2c interface on the PCB, so if I can find out what chip it is, I might be able to talk to the chip on a live one, which will be fun.
I find it odd that the PC fan you're using for the test has a QR code sticker on its hub. Must be some kind of aftermarket fan 🤔 (I could easily be wrong about that assumption)
Do you see any issues with only using the cheap LVDs for small current draws on the milwauke batteries ? (EG 10-15 amps)
We use them a lot for camping situations. lighting, charging other devices, camp fridge etc.. via 18v-12 step down.
Nah, that's pretty much the perfect situation for them. Because the smaller M18 let them discharge down to 10V (2V/cell), you're actually better off using a 13-16V cutoff at low-ish amps. All the low-powered M18 tools (anything with only 2 pins), do exactly this. Only thing is I'd avoid using the 1.5Ah and 2Ah at 10-15A. The cells are rated at 20A, but because they are in a pack, they trigger overheat at around 15A.
Also, try measuring the off/standby current of your low-voltage module. I've heard some of them are a few milliamps, which can be a problem. That's obviously easily fixed by taking the battery off when not in use.
@@toolscientist Thanks mate for the speedy reply, that's a big help. Generally use the 4 and 5a/h units with cut off at 16v. good to know i can drop it back a little.
I always run a cut off switch so it disconnects the battery from the LVD for that exact reason. (or pull the battery out)
Great video! Do you have any faulty batteries? I've tried repairing batteries by replacing the cells, sometimes they come back to life, but sometimes they continue to flash even with all cells fully charged.
I've always wondered if a command could be sent to the battery to start the battery again.
Haven't tried repairing batteries yet. There's a few good repair channels on YT that might have good info.
Looking forward to experimenting with my Makita batteries to see if they have any smarts
There's an instructable by Noloxs on Makita LXT that covers the gist of it. The 3rd pin is similar to the J2 pin on the M18, except instead of being 3V, it's 0.5V below pack voltage and cuts to zero when a fault is detected. This makes driving a mosfet with it very easy - there's no need to boost the signal like I had to do to get the M18 to drive a high-powered mosfet.
There's presumably digital communication on the yellow connector as Makita sells a tool that lets you read battery temp, charge cycles, et cetera.
Hello ,help please 🙏 i buy an 5s bms i put on battery .My battery have 4 pin negative, positive ,ds,cs i dont know last two what means .Only Ds have a pin on battery cs dont have I power up the tools work 3 seconds and stop .What can i do ?
would be interesting how much power you can pull out of each battery with a MPPT
Very lovely video! Learned a ton and helped with troubleshooting a broken tool. Specifically, it looks like the J1 pulse from my tool isn't getting to that 12.7 V threshold.
The tool I'm working on is gen 2 blower. I was lucky enough to buy 5 broken units for $50. To breathe some life back into them, I'm trying to figure out this J1 pulse from the tool but don't have a single shot scope to capture the pulse. What I do see is J1 will go high for as long as the trigger is pulled - but the broken blower only goes to 12.5 V then holds steady at 11.5 V. Comparing to the good blower, it initial hits 18.3 V and holds at 17.5 V. These voltages are valid for both benchtop supply and m18 battery.
My question to you is - have you any ideas what might be causing that J1 pulse to be so low? Is this going to be buried in the epoxy block they hide their magic?
Failing figuring out wtf is going on - I was going to wire in a momentary switch and do the button mash to wake up J1. Maybe add a LED that shows the status of J2 so I know I mashed enough without having to try to fire the tool.
I could hotwire J2 from +ve, but I would assume this would disable over-temperature and under-voltage protection since J2 wouldn't be able to go low.
Thoughts?
Good debugging! Yes, hotwiring J2 will remove any protection. J1 likely has a fault inside the epoxy, probably a short that's causing extra load and therefore voltage drop on J1. J1 should be at B+ when trigger is pulled (I think). You won't be able to run an LED off J2, the LED will draw too much current. You can connect J2 to a mosfet which powers an LED, though.
@toolscientist thanks for the reply. I realized afterwards that you had an entire section on the max draw of J2 not including the comments lol.
I was afraid of the black magic box fault. So either pulse the J1 through external switch, hotwire J2, or boost signal to J1. I'm not overly versed with buck converters, but would that be a valid option to boost J1?
@grrrrwolf you could use a level shifter to boost J1 voltage. Can't find a good example so I might have to draw you one and post the link
@@grrrrwolf tell me if this link works (and makes sense) tinyurl.com/27ytflv4
@@toolscientist thank you very much for taking the time to do that up. The cct itself is clear, what I would ask about is how critical the value of the mosfets are. It looks like they are mostly just showing the voltage and there is next to no current demands. I just have to go through my stash of dead boards to see if I have both types of mosfets kicking around.
And then there’s me in college. Using paper clips and paper to light a candle with a Bosch 12v battery. All my neighbor asked for was a lighter
Question for ya. would using a m12 battery with just a positive and negative adapter to power up components on vehicles like a window motor as an example affect the battery in a negative way ?
The M12 batteries are 12V at full charge. Their average voltage is 10.8V. Milwaukee cuts power when voltage goes below 8.2V and refuses to start if voltage is below 9.6V. Depending on the pack, they'll be good for 15-20A for the compact and 30-40A for the XC packs.
So just make sure you're not putting too much current on them and that you're not dropping the voltage too low.
Can ypu please explain why the tools know when I'm near?
Sometimes the lights come on when I'm reaching into the box and sometimes when I walk by.
Not a clue. Maybe the lights turn on randomly, but you obviously only notice when you're near. You'll have to setup a camera to see if they're going full Toy Story on you and getting up to mischief when you're not watching.
Hmm, very good video. In the future I plan on both making portable appliances run using an M18 battery, and on making Milwaukee tools run without their battery, so this is exactly what I need. Though maybe you could release a concise document/video with all the key information right there.
While I don’t think it’s intrinsically useful to power something while the BMS protections are live, getting power from J2 might be a cheap way of running an MCU without needing a DC-DC converter. Milwaukee dev-boards when?
You can't get much power from J2. I think 10uA drops it to 2V. Around the halfway point of the video, I show my plot of voltage vs current for J2.
Hello, does the Milwaukee BMS detect overcurrent and change the state on J2?
Will a 2Ah battery have a different overcurrent threshold than, for example, a 5Ah battery?
Thank you.
It doesn't directly detect overcurrent (maybe some of the bigger one's do, but I can't test above 160A). It indirectly has overcurrent protection as high currents cause voltage sag, which triggers cutoff from low voltage. 2Ah cuts out around 50-60A, 5Ah cuts out around 80-100A. As the cutoff happens from low voltage, the exact cutoff current changes with age of battery and state of charge
Thank you for your explanation.
Love this. Always looking for more ways to use the batteries I have.
Maybe you'd be one to ask.... Any way to get "higher amperage" (than those 12 converters) to use 18 volt batteries in 12 volt curcuits?
By no means at your level of electronics understaning (thou I can follow directions), but the mechanic in me would love to use an 18 volt battery to test/run 15-20-30 amp 12 volt items.
There's a couple of those "15-40V to 12V" buck converters in an aluminium heatsink that claim to do 30A, and lots that claim to do 10A. You may be able to run them in parallel to get even more current. Ideally you'd look at the output on a scope to make sure there's no craziness. If you've only got a multimeter, then you can check the Hz and ripple voltage of a single unit, then compare those results to putting two in parallel. If you don't get crazy frequencies or large ripple voltage, then it's prob ok.
@@toolscientist Thank you for the info, I'll take a look into those and check them with my meter. So many uses I can come up with for a portable 12 volt source, I am surprixed they don't offer something off the shelf. Gtanted, that is old school, a lead acid will do that, but oz to oz.... tool packs are easier to move around.
What cells do they use? Whats the upper limit of current draw?
12ah is 5s3p Samsung 40T. 6ah is 5s2p Samsung 30T. 5ah is 5s2p Samsung 25R. 2ah is 5s1p Samsung 20R.
12ah and 6ah have a physical fuse. Project Farm short circuited an M18 6ah a few weeks ago and got a few hundred amps for a few seconds before the fuse melted. Theoretical continuous draw for 6ah and 12ah is ~70A, but I'm pretty sure it would overheat due to cells being tightly packed and enclosed in plastic.
I'm actually putting off playing Alan Wake 2 to watch this first.
Thanks! Hope you enjoy it. It's a different video to my last one, so not sure how it will go.
@@toolscientist I can't speak for other people, but this was exactly what I was hoping for. Thanks for the staggering amount of effort you put into this. That whole aspect of different batteries having different thresholds for things must have been a fairly irritating complication.
Definitely! I said a few hours, but I think that aspect was closer to 10 hours of testing time for something that isn't super important to most people. I was pretty thorough with the 2ah, so I'm pretty sure 700mA is its threshold. I got bored on the 5ah and left it at 2amps, when it's threshold is probably around 1.5 amps. 1 amp kept it awake for 10m40s. The testing time just becomes exponential as each increment in current increases the time that it keeps the battery awake. I nearly had a heart attack when the 12ah behaved completely different, but luckily it was the only one. I was also really dumb and did a lot of testing with resistors before I realised that I had a battery tester that is essentially a programmable load!
Whats the J2 output from the battery to the tool? I tried to connect the Parkside battery to M18 grinder, putting 4k resistor on + wire and connecting it to J2 but its not working.
Btw. great video! cheers!
It's supposed to be 3V, but should work with anything above ~0.72V. It should work when connected to B+. You definitely connected to J2? J2 is the one closest to B-
@@toolscientist yeah, the closest to the B-. Would 12v on J2 be too much?
@@ivankraljevic2845 does it work when using your fingers like I do at 2:03?
@@toolscientist haven’t tried it cause now when i connect multimeter to + and -, i see they are connected without pushing switch. Maybe i shorted something
@@ivankraljevic2845 maybe 4k is too small
Great video, keep it up.
I love that type on content!
Can you test Einhel Charging communication?
I don't think we can get Einhell in AU, but apparently Ozito is just relabelled Einhell. I'm currently doing Makita, and then Dewalt. Might get to Ozito/Einhell next year.
Imagine making own 18v12Ah packs diy that can deliver 120A currents almost continuous and costing 50$ + bms
The charging side is a bit more complicated. But you could design your DIY batteries to have a JST balance connector and then just use a hobby charger.
Dude I think your on to something about the current not being considered in regards to the low voltage cutoff as I’ve killed two 8ah batteries with my top handle chainsaw 2826-20, I feel that the tool is either pulling too much current overloading something on the board or along the current pathway as the strings are within .3v and the overall pack voltage is at 18 but throws the red and green flashing light and will not charge ? Any thoughts ?
0.3V is fairly imbalanced. Not enough to prevent charging, but enough to suggest that cells are failing. The Samsung 40T in the 8ah and 12ah had fairly high failure rates. You could try measuring voltage of each bank of cells under load to see if one is dropping voltage more than the others.
When you put the battery on the charger, does it make a hum that gets higher pitched in about 4 steps? It might even repeat this 2 or 3 times before giving the flashing green and red. If so, I think that's the charger detecting too much resistance. Only other thing to try is buying a logic analyser and recording the signalling between battery and charger, then compare your data to what I get in my [004] video.
any video for parkside ?
I don't think we can get Parkside in AU. We get Ozito, which is identical to Einhell. I see that Parkside is owned by Einhell so are they compatible with either of those?
Fantastic video.
A question: could there be some wireless communication happening? I ask this because newer M18 batteries have a OneKey symbol on the service tag and what appears to be a unique address, I've done some fiddling but never could get the OneKey app or anything else to recognize it. That said, even if there is some wireless communication happening, as you've demonstrated it still doesn't result in any changes in the battery's output.
Definitely no wireless chips in any of my batteries. My newest battery is a 5ah from 2021, though. Newest high output battery is from 2019.
Very interesting thank you
Do you think it's possible to charge more than one battery with one charge? I was thinking about trying to do some wiring wizard to see if it would work.
Not at the same time. You could make a system that detected when one battery had finished and switched to the next one
Please man. This is amazing. You are bringing truth to the power tool industry. U have to do this analyses to makita since makita is the king of knockoff Chinese random tools and battery
I should be getting to Makita soon. Makita will be hard as they like to brick their batteries. I might have to start with the knock-offs first to avoid bricking too many batteries.
@@toolscientist what do you mean brick? If they are messed they no longer work anymore?
@eduardoramiro4846 I'm going off what I've heard, but apparently if a LXT battery is put on a charger and gives an error, it records it. If you get 3 errors in a row, the battery bricks itself and won't charge ever again and can't be repaired.
So doing crazy stuff like I've done to my M18 packs (undervolting and overtemp) might trigger the battery to brick itself.
What if you used one of those tools enabled with onekey? I know you’re able to lock tools and stuff through an app if you have onekey enabled tools.
I don't have any One-key. Other people mentioned it and I looked into it. From what I can see, it's tool only. The only thing I saw that might be battery is a screenshot showing battery capacity, but they could be getting that from pack voltage. If you can see battery temperature, manufacture date, charge cycles, or similar in one-key, then that would indicate that there is digital comms between battery and one-key tools.
You should check the Dewalt battery BMS. there's basically no protection. The protection comes from the tool. I had 3 batteries burst in flames after running it on an old 18v dewalt circular saw.
Yeah, they're the most basic of the major brands. They're interesting to me as you have direct access to the thermistor and cell voltages, so lots of scope to make DIY adapters that don't overheat or overdischarge
Is it possible for the tool to over discharge battery when there is a connection issue with the battery or an internal fault?
Yes, but very unlikely. The tool will keep going as long as it receives > 0.725V on J2. Using a power supply, I've run these tools as low as 5V (although they barely run at that voltage).
However, a connection issue on J2 will just make J2 at the tool 0V, and the tool will stop. The only way it could happen would be a short circuit between + and J2. J2 is closest to the ground pin, so it is very unlikely that this could happen.
@@toolscientist Ok, thanks!!
Would be great if you can do this with würth m cube battery and tools
Looks like their previous gen was licensed M18. Their website even says they partnered with TTI. M-Cube looks to have same pinout as M18, but different physical interface. Do you have these tools?
I have Würth brushless angle grinder but i don't have M-Cube battery.
Hi i would like to know how to rest the M18 bms after removing it from the battery? I hade two batteries, one with with broken cells and one with broken bms , so i remeved the broken bms and put it to the battery that had the good cells but when you press the button of the bms you get nothing . Is there a way to reset the bms without puuting it on the original charger ?
I haven't managed to reset one yet. Only way I've heard is with the genuine charger.
The tool coms cant use complex coms because the interference would just make the tool unreliable and damage brand reputation. Charging on the other hand, in a controlled electrical environment means they can use digital coms to charge optimally and also prevent cheap chargers from destroying their brand reputation.
Yeah, I originally had a section of the vid discussing the noise in J1 and J2 during operation and saying that was the reason for no digital comms. I left it out as I'm not sure if that noise could be filtered, so didn't want to categorically rule it out.
I know that Makita XGT tools use digital communication and the tool won't run unless it receives that communication. I'm pretty sure they have a separate isolated ground for comms to get around the noise issue. I have heard from one person of having problems with their XGT tools and having a rep come and clean out the batteries for him multiple times. In the end they gave up and went back to the older LXT system.
This compared to Dewalt batteries which... directly connect the cell terminals to the battery leads. Nothing stops the tool from working from any power supply. Nothing stops the battery from being charged or discharged.
Yeah, Dewalt is the best battery for the apocalypse! Dewalt tools are slightly harder to trick as you need to trick the ID pin and Th pin, M18 just needs 3V on J2. Dewalt is obviously much easier to charge.
@@toolscientistYou sure that Ryobi isn’t the best for the apocalypse? NiCd compatibility, low voltage cutoff in the battery…much less communication than most other brands.
Can you do makita since I currently use those now.
This post might get you excited. Still a few weeks away, though. www.patreon.com/posts/expanding-92383338
How to crack Festool battery communication protocol?
Been a while since this came out now, wondering if anybody has spotted any commercial products making use of this information yet?
Haven't seen anything, but I haven't really looked, either. I thought about doing it, but don't really have the time, and I figured it'd just get copied and I'd be undercut. I should finish my updated circuit and publish it, then someone might make them.
@@toolscientist After putting some more thought into it and spending a lot of time looking at mosfets and mosfet circuit design, I realized a couple things.
1. I really can just use a relay/contactor to switch high power loads. Which is exactly what I do now except using a tiny non-adjustable LVD circuit. This can fit entirely inside the case of a particular M18 battery adapter so it's a super clean install. However, the circuit cuts out way too early though so I'm not getting anywhere near full capacity. Hence why this subject is so interesting to me.
2. For my use case (repowering childrens toys) it dawned on me that since I've added speed control boxes that use low current key switches to turn on and off, I really just need to interrupt that line with a relay/low current mosfet instead of having to fully shut off the battery.
Now the hard part is probably that none of the aftermarket battery adapters include provisions for the J1 and J2 contacts. And I really don't want to deal with 3D printing... I smell epoxy in my future.
You prob need a mosfet. Relays need 100-500mA through the coil to hold the switch closed. The J2 pin can only do
@@toolscientist Exactly my point, I described it poorly. My realization is that I don't need to worry about a high power mosfet circuit. I can use a very compact and simple circuit.
That's actually a pretty good solution. I always neglect relays as they're more expensive and wear out, but in this case it makes for a very simple circuit with only 2 components (3 if you add the pulldown resistor for the mosfet, maybe 4 if the relay needs a current limiting resistor)
Wait, Milwaukee has not released a toaster oven? 😆 (There's a video of someone using a heat gun toasting toast though. Interesting there's a heat gun, that has to be massive current!)
It's doable. I think I'd prefer a sandwich press, though. They're 1000W, so you'd only get about 10min runtime on a 12ah.
Do a makita one next 😊
Soon ;-)
Do your batteries have the One-Key symbol?
All of my high-output ones do, and most of my 18650 packs, too. Seems any battery newer than 2016 has it.
Do you have any one-key tools? Can it tell you any information about the battery? Best I can find is the radius light will tell you remaining capacity of the batteries, but they can just guess that from the voltage.
I mean with some mosfets and diodes they could do pwm or a buck converter it could but they don't
The 12ah does have mosfets on the ground rail. @ThriftyToolShed showed them in a video here: ruclips.net/video/TphIrEWvW68/видео.html
It's possible that they could throttle the power with this mosfet. PWM would likely play havoc with the tools, unless Milwaukee knows that all their tools have sufficient input capacitors.
They could also run the mosfet in the ohmic region, but that would make a lot of heat, although they do have a sizeable heatsink on it. They'd need to add ~20mOhm to make the 12ah look like a 5ah, which would create 50W of heat at 50A. Doable, but seems unlikely. The 6ah high-output is similar resistance to the 12ah and it doesn't have any power mosfets.
I think it's most likely a power cutoff. I'm hoping I can get some probes on the gate pin and see what it's doing.
150 amps! I didn't think M18 batteries could put out anywhere near this much.
And that's the tool limit. The battery was still at 16V, so it still had some juice left (cutoff is 12.5V)
Bosch batteries hack next? 😉
Probably going to do the big 3 first (Milwaukee, Dewalt, Makita), then get on to others. Bosch has a pretty small presence in AU, so might be hard to find 2nd hand stuff.
@@toolscientist Okay, looking forward to watch those videos as well :) In Europe the three brands you mentioned + Boch are the most common.
For a suggestion for future videos it had been awesome with some videos on the DIY battery kits that are sold, to see if the bms used in those kits are ok etc. A video making your own kit had also been awesome ;)
I have made my own power tool batteries for a couple of years now and im really happy with them. Way better performance (because of better cells) and a much lower price is making it well worth it.
Here it's probably Hikoki, then Bosch. I think Metabo might be below Bosch - they probably do better in EU. In the future I can hopefully get people to loan me their tools for a week, otherwise I'm going to end up with a very expensive and barely used tool collection!
Yeah, I'd like to test some of the 3rd party batteries and DIY kits to see if they've implemented the same charge/discharge/temp limits as the manufacturer. Problem is they can change fairly regularly.
@@toolscientistOkay, looking forward to new content :)
I should make an adapter with some 4 gauge lugs.
Do it! 😅
i just wathed all your Vid's , i EXSTREAMLY liked the -jump starting one , and i think your testing method is GRATE and Newb-friendly while be a advanced subject ..
i think you should keep making them , if you need ideas , im full of them lol .. but to get you the viewers, do some thing like this- with Auido equitment , like amplifyers, get a Hot-ticket one, that dosent brake the bank , (like a Scar-auido RP 2000.1, ) you would be the only person doing pratical test on them.. most of those auido-guys are dumb as rocks...
also a good viewer-bass is doing those E-bike and E-scooter crowed , they never doing good testing like what your doing..
wish you the best ..
PS, i want to see your High-power diagram , you shouldent hide them becusye of stupid people. .. LIke moma-allways said "stupid is what stupid dose" , and a book that is sencored is not a woth-reading book, (like the antarkest cook book, you cant buy the Real thing thire in the EU/UK, - its almost 2x thicker in the USA/Mexico/Canada - even blue-prints for a AR-full auto trigger asimibly in both .22-LR & the .223 chamberd
Still working on vids. Things just got slowed down a lot over Xmas. Hoping to have my Makita LXT vid finished this week.
I'll probably stick to tool related tests. If I had an escooter I'd definitely look at modding it to work with power tool batteries. Although I think that's been done, and I want to avoid rehashing stuff.
High powered circuit was mostly not shown because it's a bad circuit that can lock on are cause pulsing. I'm working on a better circuit which I will make public.
Mate i repair power tool Batteries for a living !
Just randomely found you So subbed and liked now hoping that i am going to search your vids and find something on the worst of the worst BMS mentioning no names like MAkIT& what a complete Rip Off Bollocks Bullshit BMS that is.
Cheers mate.
Ok mate just checked and nothing there.
You may allready be on this one i dont know.
But buy a new Makita battery and a duff one.
Work out the terminal address lines and monitor as i can see that you are well capable of doing.
Now change one or maybe two duff cells in the bad battery, you will get the battery monitor blinking from side to side.
Balance ( manually ) all of the cell pairs to 4.1 or 2 and and try charging on a makita charger...it will still say NO whatever you do the BMS locks out and the only thing you can do is chuck it in the bin and fit a new one then you are good to go again.
They have the nerve to sell a £300 tester that tells you that you have charged your battery 97 times, or that you charged it once when it was really cold or even three times when it was really hot, it even will tell you what is wrong with your battery ? Maybe or say oh sorry computer says No.
There is a memory chip on the Makita BMS that is like no other and when it locks you out thats it !
All the best and good luck Kev
I am currently working on LXT. They even have fuses that the BMS can burn and physically disconnect the B+ pin. I'm about to get to the part where I start overheating it to see its response. I bet there's a limit where if I overheat it too many times, it'll lock itself.
Keep us all in the loop mate !
I'm happy I do not have one of those crappy milwookie... 😀
Next step: battery charging subscription.
And this, ladies and gentlemen, is why AvE, our lord and saviour, calls Milwaukee "Milfuckee".
I'm sure other brands are doing lazy/stupid stuff, too. Milwaukee is just the first to go under my microscope. Letting the battery disable and enable the tool every 50ms is pretty high on the stupid scale, though. I hope none of the other brands are doing stuff like that.
Man most of this is above my pay grade lol
I do go a bit too fast sometimes. Feel free to ask any questions.
Try monitoring the entire run cycle of the vacuum cleaner. I used my M12 vacuum to compare lifetimes of used batteries. Maybe four or five times across a 15 minute runtime there were audible changes in the motor note, I was curious if that was an artifact of redlink communications. There was zero load for the entire test. M12 tools show the charge indicator on the tool rather than the pack.
M12 tools don't have any digital communication. It's all done with analog pins. The tool is just reading temperature directly from the T-pin, which is a thermistor. I've done a few rundown tests with the blower, but that's a brushed tool that's not likely to have anything fancy in it. Can't say I've ever heard any sudden changes in motor speed from the backpack vac, but maybe there is something.
@@toolscientist ty gl