Andy: I also purchased two of the Seplos 280Ah DIY Kits in April and May. I really liked the idea and they did not seem expensive, although the shipping was $190 per kit. My cells were only a few months old and came from Docan Technology. I was thoroughly dissatisfied with the cells as very few test to the alleged capacity at a very low C rate. The second set of cells also came swollen. Docan suggested I just compress the cells back to the original size. I chose not to do this and will not buy more from Docan. Both kits came packaged as yours did, with loose parts all over the place. The PCB boards had broken leds that had been sheared off by the loose parts. I soldered the leads back and requested a new board from Seplos which was sent quickly. One thing I did receive you did not was a parts list. It was not completely accurate, but it really helped, especially with the screws, took me several hours to sort the screws out. The one thing that made me decide not to buy another Seplos kit was the customer service. One of the bus bars, S-F-3, that connect the BMS P+ to the fuse block was bent backwards. The BMS sits lower than the fuse post. Normally the bar is bent so the short section goes over the top of the long section. This causes the short section to the fuse block to sit higher than the long section. This way it will sit on the BMS and fuse block very well. I contacted Seplos and explained the problem, they refused to replace it. I sent pictures of the problem, they refused to replace it. I sent pictures of their other S-F-3 I had that was correct and the mismade bus bar, they refused to replace it. They told me to take the mismade bar and just bend it back over. I told them the kit would then have been modified by me and the warranty would be void. Again I was told to just bend it. I told them it was not possible to take a piece of thick metal that had already been bent and reverse the bend, even if it was a lot of thin pieces, and expect it to work well. Again I was told to just bend it. I asked them if losing a good customer for want of a $2 part was worth it to Seplos. I was told that the part was not a "qualified" part and would not be replaced. Eventually I removed the shrink wrap and did bend the bar back so it would fit. It is an ugly bend and it barely fits. But it was my first option other than making a cable, my second option. Consequently I will move on to other commercial batteries. I have also not been able to get the Mason 280 kit connected to the computer and would like to know how you did it. I have not seen a video of the Mason 280 kit connected yet if you have done so already.
Great vid & info thanks Andy, I think u may have helped me break the ice with my 3 x Seplos Suntorque Batteries I have had since march and 2 of them have not wanted to charge over 72% max while the 3rd battery & last one delivered still charges to 97% max. I have had a discussion with the sellers techs who advised I needed to get some communication happening between the batteries and 5kw MPPT All-in-one inverter. I have been sidetracked with other issues, but I did find this all rather intimidating as I dont have a working PC presently, but it has been on my list of stuff to do, and you have made this all so much less confronting for me now, so it is time to get the old PC going again and fix this BMS situation I am having. Cheers
Great video Andy, you really got me on the cliff hanger near the end 😆 Maybe you could mount a key switch on the side to activate the active balancer addition.
@@OffGridGarageAustralia Another function for the Raspberry Pi or an Arduino? Actually a simple BMS could be easily implemented in either. When the "BMS" is initially turned-on individual CELL charging and PACK discharge is disabled. begin routine. Check each CELL's voltage against the upper and lower limits and when the CELL is below the UPPER limit enable charging for that CELL. If the CELL is below the lower limit disable PACK discharge. (Do for each cell) If after checking the cells and they are ALL above the lower limit enable pack discharge. Rest a short time. repeat routine. A physical structure would need to be created to limit charging current (Amperage) for each cell as well as the entire pack. Since cells stacked in series increase voltage of the PACK but not amperage the discharge current for the entire pack could be the same as each CELL's capacity. IF anyone decides to try to build their own BMS, they and they ALONE are responsible for any results. I assume no responsibility since this is only a theoretical example of how such a device MIGHT be constructed and NOT a detailed example of how to actually do so. For what it's worth, my career was spent in large mainframe programming and systems design at both national and international companies. I designed and programmed systems handling millions of dollars monthly in accounts receivable and payable as well as insurance and manufacturing systems.
Great research and testing Andy, I love it, and one hour went by pretty fast. From what your research has already shown most of us guessed the active balancer was going to be a necessary addition. Perhaps Mason should perhaps include an active balancing option and compensate you for your research and marketing.
I think that the behavior could be in fact appropriate. As this BMS has really small balancing current, it needs much longer to fix a cell. If it re-enabled charging immediately, the cell would be still off by a lot and it would turn off again almost instantly.
That is right, but without any charging at all, the other cells will never catch up and only the highest cell gets discharged a tiny bit every time this happens. Without a proper top balancing and matched cells, this concept will not work.
@@OffGridGarageAustralia Makes total sense. My JBD smart BMS will cut off charging if one cell is high and then the balancer turns on for 2 hours or so and then when the cell drops the bms will start up the charging again for a few minutes and it cycles until its balanced... fully automatic!
Thanks for all that went into making this video. I've been corresponding with Seplos for the past 6 month on the Mason DIY kit for a planned expansion (and possible conversion) to two home made 16s Overkill EVE LF304 cell batteries. I never pulled the trigger on this Seplos Mason expansion project because I could never find any comprehensive information on the Seplos BMS including integration in Victron DVCC. DVCC is also suppose to provide intelligent charging integration with Victron smart MPPTs, but I'm not sure exactly what that is. While I like the Overkill BT app for looking at the cell information, my installation is remote and I'd like to be able to see the individual cell voltages and deviations for each battery on VRM or the remote console. If the Seplos BMS balancing function is struggling with 135Ah cells it's probably going to have real problems with 280/304Ah EVE cells. I'll keep an open mind, but I think you've help make my decision. Looking forward to the 280Ah or 304Ah Mason DIY kit build. Thanks again for all you do.
Thanks, Brian. Your absolutely right and we have to get the remote access to the BMS data working. I have tested all day yesterday to get any information or of the Seplos BMS without any success. I've ordered some other adapters now which should work and give us the info on the VRM. So stay tuned. You already have the Cerbo so it should be fairly easy to use CAN protocol to get the information from the Seplos to your VRM. Thanks for sending these amazing pictures of your off grid observatory a while back. That looks really fantastic.
Here in Lebanon Growatt and Deye inverters are widely used and Seplos support them with CANBus. we install Seplos + heltec balancer with a switch for maintenance. If you use canbus with growatt for example, the bms will tell the growatt to charge with CV, only one voltage. A good advantage of Communication with canbus, is that our customers will see the accurate percentage SOC from the inverter/App directly. Any big error from the BMS will be informed to the inverter, will lead to safety shutdown and more features.
HI Andy ... nice review of Seplos battery package ... i need to say that i have (just) 2 seplos BMS on my own 48V 230ah package and all working fine, as you said, you need little bit time with software but then all goes right way. Also communication with cerbo gx works fine.
Hi Jaroslev, how did you get coms to work with the Cerbo GX? Did you use Louis van der Walt's driver? And did you have to downgrade the Cerbo's firmware? Ciao, Rik
@@OffGridGarageAustralia Hi Andy, in software you can setup CAN port for plyontech,victron or other protocols. Plyontech CAN works like battery (set up SOC, charging from grid etc), Victron works like battery monitor just see actual statuses
@@jaroslavchytil5732 thanks for your response. Could you please elaborate a bit more, or was just a matter o setting it up to use the Victron protocol and use the right cables? Could you just go by without using the CerboGX ? (my plans include VenusOS in a RPi , but not a GX) Thank you very much for all your insights!
NOW I see where tomorrow went!!! LoL. Took a while to get it done. Sorry for all the hard work Andy. Still all and all, I think I like the home built, peace by peace. Choose the components I want and understand everything that is going on with it. But you will never know what is out there unless someone like you tests them. See ya on the next video. ( tomorrow) ??? Lol
Thank a lot Alan. Yeah, it is a lot of time going into these videos but I'm not complaining, I really enjoy this kind of work, editing and forming a story out of all these single video snippets I recorded over the last 3-4 days. I may have another video tomorrow actually... 😉
It's crazy the amount of parameters it has. Also, it seems like the one cell is messing up everything. Does SEPLOS let you purchase individual cells for replacement purposes? Great Video as always!
Thank you. Seplos has cells on their website, but how do you know they are better than the ones they have already sent? It was part of the test to let one cell go really high to test the BMS's function. Usually you would do a top balancing of all the 16 cells before installing them so this extreme voltage difference would not occur. But even of having the pack balanced with the active balancer, as soon as I disconnected that, the voltage difference was on the rise again. I have to ask Seplos how their balancer works and how to check...
@@OffGridGarageAustralia Please ask, I have a feeling the balancer turns on the bleed resistor for a cell at 3.5 judging by the FULL pack voltage over voltage setting. The point of a BMS is to just monitor and that means you should set your charger to charge the pack at less than 3.5 per cell. I think that makes sense for lifep04 packs because they have such a weird curve at the top (above 3.45v) so its a good idea to stay away from that area.
Hi Andy and thank you for all your testings about Seplos I use Vertical Kit from Seplos to make 16S 300Ah Battery Pack BMS inside is 10E, with bluetooth I use with a Solis inverter, everything looks good now, only that i just can't figure out what is the warning / parameter that make BMS to ask inverter to limit charge current to 10Amps Sometimes happend to see inverter charging at 70Amps ( maximum set by me ) and finish charging without lowering current and sometime is lowering to 10Amps and stay there for a while and finish charging Thank you again
If you set your charger to ±100mA for the last part of the charging process, you allow the bms to discharge the "full" cells until the lowest one creeps up
I know, in theory thsi should work, right? It would just take 1000h+ maybe. The active balancer has done it in 30min. I'm not even sure if the passive balancer of the BMS is working or not 🤷♂️
@@OffGridGarageAustralia that is my biggest question too cause you barely see it happening and I guess that I will need a BMS too like you had mentioned for a kind of quarterly maintenance. But not sure what to get for this task cause the Neey seems to be a big gun for a 8 kWh battery.
Hi Andy, really glad you did this deep dive on the Seplos BMS and it's software. And you're right, the mobile app is alot easier but I guess where the Seplos will shine is that once you make the change once, the software will send the updates to all the BMS in your battery bank. That'll be very helpful if you have a battery bank with 3 or more batteries each with their own BMS. With the JK, which is excellent, you'll need to log into each of the JKs and update parameters on each of them. I guess it's a trade off somewhat. My question to you though, let's take the 4th gen NEEY smart balancer for example, is that something you would leave permanently connected with the Seplos so that balancing would occur automatically and you wouldn't have to look back at it?
Thanks Henderson. Yeah, I guess we will have more fun with the BMS soon. I have ordered a while bunch of adapters now to make it talk to the Victron system (not just the inverter). We will see how this plays out and works. I agree with you, as good as the BT app seems to be but from a maintenance perspective it is very cumbersome and takes a long time to go through all the parameters individually. I hope this is really something the Seplos BMS will shine and make us happy. The Neey is certainly something you can easily integrate into the Seplos Mason system. There is heaps of room at the front and some Velcro will install the balancer nicely. Once programmed, it will help balancing these cells.
Andy... Have you checked the bms PCB to see if there is an ujustment pot for the shunt and voltage calibration. Is it possible it's holding back the charging until the balance delta improves ? Hmmmm. Is the current firmware up to date ? If it was my battery I would top ballance all the cells in parallel before instalation. May be a good idea as you say to check DC ripple or any inducted AC ! Stu.
Yes, top balancing is a must, as I said in the video. This was part of the test to see how the BMS reacts and how the balancer works. There is no pot on the BMS as far as I could see. This is all software based.
@@OffGridGarageAustralia That's a shame. It's allways hard to get shunts to read accurately. They are simply not highly precise devices. One question if I may..... You had the shunt on the floor. I was wondering if you had the meter voltage reference wire connected at the battery terminal as a lower voltage will give a different reading as you know. With the meter your using I prefer to power the meter with a separate supply and common the ground connection. Great episode as always 🙂
@@OffGridGarageAustralia If you wanted to use the balancer you could always use a zener diode with a pot connected to a relay via a mossfet. Then use the normally open terminals to close the run jumper on the balancer. Just an idea but not something you expect to be doing with a brand new battery and bms !!!!
For your efficiency you should calculate the A X V difference from the supplying battery to the A x V to the receiving battery - not just the current. You have correction in both the inverter and probably your charger.
Hi Andy, did you already figured out how to connect the Seplos BMS to a raspberry pi? With the Cerbo it's easy but I'm not sure how to do it with the Rpi?
if you want the charging switch to turn on again, you don't have to discharge until 96% but you simply turn of the bms and turn it on again (aka restart).. 😁 I have been using this bms for over 1 year now. The only thing I dislike is the balancing function, it is really really need an upgrade.. 😆 Anyway.. great video as always. 😄👍🏼
Yes. I use that in the SPAT Calibration Centre installation (Battery 3.0). I also use DVCC in my Off-Grid Garage for limiting the charge current without any smart BMS connected.
Thanks for the solid content this is exactly what I was looking for. I already had one of the new JK BMS‘s on-hand but was going to order one of these Seplos BMS to go with the 280ah case I ordered ... going to stick with the Jk bms + battery box only
The JK seems great and provides all function we need. I test the communication of the Seplos now but doubt it will have any advantages in comparison to any other BMS out there.
Now I am so excited to see what happens with the DIY 280. Will Andy incorporate the active charger from the start knowing the internal balancer will take 365 years to fully balance his pack? How will he run the wires for the external active balancer? I have an ideal , run the cables to the outside. As the active balancers are seldom used, When the need arises just plug in the balancer from the 135. After balanced unplug, put on the shelf until something needs balanced. If the wires are already externally located then it would take a matter of seconds to connect. Just a thought and not telling you how to do your battery. As for the temp sensor, I personally un soldered and re soldered the connections. As the connections at my astonishment were very poorly done. Not at all like the rest of the board. Good luck and stay creative!
Hello from Germany, if you know the Seplos BMS, you must have thought about balancing, many have an active 5A balancer from Alibaba and don't know whether you always leave it on or just plug it in every month and then look so that the cells are well balanced. On the bottom of the boards there is a soldering point or several where there is also soldering for Lifepo batteries. this big pad covers 2 contacts and if you solder a wire and switch you can turn the balancer on and off. look closely, on the circuit board at the edge. I like seplos and the service also
The BMS will be working with a lot of canbus inverters like goodwe or growatt and sma, lot others more. That is there biggest benefit, cause no other BMS supported Growatt SPH or SPF inverters , even not the JK BMS as canbus version which I had bought trusting them that they could fit and run.
@@OffGridGarageAustralia that will work too, cause our american relatives are using Victron too , I guess the Raspberry pi is managing that for them, but I know about cause they also have growatt. Great progress even though lot of points that are not as straight forward as the seplos bms could be.
I have another question that maybe others have dealt with and may have found a solution. Since I operate several packs in parallel and the individual cell voltages(and many other values) are unfortunately not transmitted to the Cerbo and then to the VRM, I use the option of being able to check all values on the desktop in the office via RS485 and USB using the SEPLOS software if necessary. With only one pack, this is not a problem, since the DIP adressing all remain at 0 and the parallel connection CAN & RS485 does not cause any problems. However, if you use several packs and adjust the DIP switches - so far I have unfortunately not been able to find a setting where ALL packs including the master pack are displayed in the SEPLOS software. This would be PACK 0. Maybe another one already has this great one missing the possibility of being able to read out all packs remotely - WITH a parallel connection via CAN bus.
I think you have to leave DIP 5 off for that to work but your packs will be in a different order than. Pack01 will be the master than. I heard this from a friend. (Thanks Raphael). I will test all this once we have the second pack build and connected.
@@OffGridGarageAustralia Many thanks Andy for the info - but unfortunately this doesn't work - or I misunderstand it. I've also tested almost infinite variants to work around problems that I don't think can be enforced, but without success. The only problem is that when all packs are connected and the master enables communication with the Cerbo - this master pack is then no longer displayed in the SEPLOS software. If you then address the DIP switches to just one pack - all packs( since still addressed with DIP 1-4) displayed in the software but the Cerbo only sees the one "master pack"
Watched your video on charge controller settings and you mentioned using parameters set by cell manufacturer, do you have any info on the higee cells and max per cell charge?
Good video and very informative, thanks again. I bought the Seplos BMS to make a battery pack to use with the Sofar ME3000SP ac charger in the U.K as the charger needs to communicate with the BMS. This BMS appears to be the same that is used in the Pylontech batteries that Sofar also use. The batteries (230ah) came 2 days ago from Docan Technology in China in less than a month which surprised me and I wasn’t ready to set anything up! All the cells are at 3.3v but I’m going to top balance them before I build it. Is it not correct though that the cells will all adjust to a very similar voltage over a period of time with the BMS and, the mv difference doesn’t really matter because the total capacity of the battery is high enough already? I’m sure that you won’t ever discharge your battery by 100% in the real world so it will never need the cells to be all the same voltage. There is always a variation in anything that is manufactured and a small amount of tolerance should be expected. Just my thoughts and I’m not an expert by any stretch of the imagination but I found this test very informative. Thanks.
The big advantage of the BMS is that it can talk to the charger/inverter. You ask why? For example your issue with the balancing. If the SOC reaches 100% the BMS asks the inverter to slow down the charging amps. This then allows the BMS to balance the battery. You see the same effect in your video when you manually lower the charge amps when it seems that the balancing is doing it's thing. Of course due to the large difference caused by the 4x cell test you did it isn't doing a great job but it will eventually reach a good balance. I have a similar setup running 6 months now. No initial top balance but the cells are still in good balance at 100% SOC. It just keeps them nice and perfect in balance when you just let him do it's job over time.
@@danielp5208 True but normally that would never happen. I you might have a runaway cell you better need to check why that is and remove the cause for that. Balancing (150mA per cell) is ok in this BMS because of the slow top charge request. Only if you don't use communcation to the inverter and do a manually charge this slow balancing might be an issue. But why use a BMS like this then? People buy the Seplos for this reason. If you really need active balancing you could add it afterwards but in 6 months I didn't need it and doing a full charge/discharge everyday.
Hello Andy - one more question - I had copper washers in the boxes next to the aluminum bus bars. You also included them? and if so - you didn't install them? I followed the recommendation but I'm not sure if this makes sense, because it could be that the teeth of the nuts could then not function?
We are racing motorcycles .. We are LEEEEEEEEEANING on lithium ion chemistry .. I am learning quit a lot from you.. But your chemistry is different from mine.. It is difficult to keep our chemistry balanced using it like we do... I like it when you share the difference of what you are using to lithium ion.. How you are using your chemistry is like taking a nap on the couch compared to how we are using ours..
Hi Andy, and thank you again for this long video...😁 In fact I didn't think they used this software, I thought it was another🤔 The PBMS Tool. Yes it's true this one I find less good, but it does the job, that's the main thing. These Seplos battery packs for me are fine. Thanks again Andy, and until next time...👍
Wow. So glad you did this. As a novice I would of not known what to do had I bought this kit. Until mason sort this balance issue and make an app I will stay away. It’s a shame as I was talking to a salesperson from them yesterday about a possible purchase. I wanted to build a 5kwh for a first project. I have no engineering or electronic background so would have been relying on artisans in the Caribbean to help me 🤷🏾♂️😱
Thank you for your testing of this battery set. I think trying to sqeeze out the last milliamp is like installing a solar tracking system instead of a fixed one. For me it makes more sense to keep it simple and install just more to get the same output. I would not need to be able to check the battery at any time too, because I trust that it will work on it's own without supervising every day😊
I always wonder if the capacity as measured is because of the small but real losses in the bus bars and the flexible connections to the BMS. Also there must be measurement errors on the clamp meter and the shunt. Proof of this I guess when you do the high discharge tests next? A very interesting and informative video Andy,. Thank you!
Great review of this battery setup. I agree the balancing seems to be lacking, but you made a workaround and got it done. These reviews are forcing you to learn new things and share it with your viewers. Thanks Andy.
Hi Andy. I really appreciate your channel and your are funny too. One question: do I have to let the active balancer always connected, or should I connect it only sometimes? Best greetings from hot, most northern Germany.
Thanks for the feedback. As we have tested in previous videos. It is not recommended leaving it connected as it will balance 24/7 then. Once you fully charge again, the cells will be totally out of balance. I would recommend either using it occasionally for maintenance purposes or turning it on only above 3.4/3.45V. Most northern D can only be north of Flensburg...
I wanted to know if you plan on buying more of them. Is this basically going to be your battery-3.0 setup. and how is the total price for one vs if you made it yourself like battery-2.0. I only saw you mention side handles, but it kind of looks like it might be rack mountable in size. Did it have holes and/or rack mount ears to rack mount it?
Can I successfully connect the seplos BMS 14s to my energy storage consisting of the LG Chem NCM (Li-Ion) 12S battery? What should I do with the 2 free bms cables?
I don't think this will work with Seplos BMSes. They only work with the specified number of cells unlike many other where you have 8-24s support for example.
I'm looking to replace a 48v 400ah Lead acid battery with lithium can you recommend a comparable size ah setup and a BMS to go with it. Love the videos thanks
@@OffGridGarageAustralia Not sure. Discharge is set at 50 amps, but the other day at low SOC it had changed to 25a. Maybe it disables the inverter. I've never tested what it does. Where can I send screenshot?
I have watch your seplos videos several times and I thought you said the BMS shuts off of no power is going in or out, did I misunderstand what you were saying or did I completely imagine it?
Can you clarify were the cells supplied to you unbalanced? With 100mA balance current you are always going to want to be building a pack with balanced cells. This BMS only has value if you have it networked to the Inverter/Charger, otherwise it's just a pain in the arse. I appreciate it disconnecting if the inverter is trying to charge past 100% because that is not a nominal condition, and I'd much rather it just shuts off if something becomes unusual, caution over everything else. There's probably a toggle switch to disable these protection functions? I had an issue with it rarely shutting down because of a false short-circuit detection - I disabled the feature (I have a fuse for that condition) and the problem was solved. Love your videos please keep up the good work!
This was part of the test. I have explained this in the video: the last 4 cells were capacity tested and therefore were on a different SOC s the other ones. I wanted to see how the BMS handles such a situation. Usually, you would do a full top balance in parallel. As I said in the video 😉
Andy, Fantastic job. Congratulations. I am having the same charging problem with two batteries. To balance with active. Question: the active balancer, do you use it with the battery connected but without extracting energy to the house? Cheers
That totally depends on your project. The Seplos Mason system is easy to setup and if you don't want to wait weeks and months for all your parts to arrive, go for it.
Thanks Andy, very very interesting video. So, it seems that overall the BMS is doing its job correctly except for balancing the cells. I think for that it would be interesting to use the NEEEEY active balancer as a companion. It balances efficiently with 4A, it is possible to program the start and stop conditions, and as a bonus you can monitor a lot of battery pack information directly via Bluetooth. It would be interesting to see if it fits inside the Seplos box.
Thank you. Yes, the BMS seems to work as programmed with all its safety features. Adding an active balancer for the occasion to balancing or even a Neey seems to be the way to go...
@@OffGridGarageAustralia there is a programable relay for heater function but maybe we can use it controll the 5a active balance. Maybe explore it for future video
@@OffGridGarageAustralia there is an unused header built-in on the BMS board. I think is only logic level relay but we can connect to a bigger relay if we need to control a bigger load there is even a heater toggle on the software. We maybe need a modded firmware to use of other purpose.
@andy I was wondering why the BMS didn't turn on charging limit switch? On my BMS the charging limit switch turn on (10 amp) when high-voltage warning occurred. Did you turn it off or something?
No, I didn't change any settings in this perspective. 10A was the current anyway I charged with so would not made a difference in this test, I believe.
@@OffGridGarageAustralia the charging process wouldn't be different but the indication on the software should show. I found the limit switch is quite nice when the battery approaches 100% and it will recalibrate the soc to 100% after it reaches top balance voltage.
This video is very valuable material to watch, for SW developer, how NOT to code BMS. BMS have to send date / ESP MESH, RS485, MQTT, CAN / to another HUB or gadget / inverter , MPPT / but, if BMS will switch of battery, it will kill my MPPTs or my inverters and MPPT and inverters has higher value, then on cell in LFP battery. Within 9 yeras, I have no need to switch off my MPPTs or inverters, because I have not BMS in my NMC, NCA, LFP and Pb flooded cells. But now, when my LFP cells are old, I see different AC impedance, so I am building my DIY BMS with 2 4 8 10 and 16 AMPS balance currents. ESP32 BMS + MPPT_0 + MPPT_1 + MPPT_2 wil communicate, without cable / no need for data surge protection /, without WiFi - no need for router and WiFi NET, and if DIY BMS tells MPPTs, that "I need 6 AMPS", MPPT 0 to 2 will decide, with respect to Sun, panels orientation, weather conditions, teperature of MPPTs toroid coils and MPPTs FETs, if all MPPTs will give 2 + 2 +2 AMPS equally, or 6 + 0 + 0 and so on. I have also ESP32 thermocam, so thermocamera will see if there is problem with cells or bus bars or PV DC fuses, and with help of libs, like TensorFLow will apply solution. Also my DIY BMS and MPPTs communicate with me with speech commands, and give me info with sound like: OV of cell_x is y VOLTs, temp of cell y is z *C, connection 6-7 is hot etc.
Hello Andy, finally... I'm glad that you finally bought a Mason box with the Seplos BMS. After a few months of fine-tuning the Seplos settings, I have to smile when I see that I'm similar had to learn all the individual problem cases in your video. You are absolutely right that the Seplos software still needs some attention. Just like many in the DIY forums, I can't understand why the Sepos Guys don't make full use of the possibilities of a good BMS. I now run 6 packs 280Ah in parallel on my Victron system. These communicate with the Cerbo. Again here, I had to use some tricky settings to get the maximum performance. But well... after a lot of hours everything is working fine. I have also retrofitted an active balancer in all packs .. which is switched on externally depending on the voltage.
@@patrickmartin4482 Hi Patrick, the communication Seplos with the Cerbo works without any problems via Can Bus. At Victron I have activated DVCC, which is now used to control the MPPTs externally (at least in part) by the SEplos. Here I had to adjust the settings of the parameters a bit. On the one hand, the maximum charging current so that this is normally over 400A, since my two MPPTs each deliver 200A and are otherwise already regulated below my final charging voltage. Likewise, the charging end voltage itself, since I only charge up to 3.40 volts, corresponds to 54.4 for the 16s pack. In the DVCC I set 54.40 in the Seplos, but 54.45. Victron always takes the lower value. But since I want to have 54.40 as a 100% SOC, I have to do the trick with the 54.45 Volt in the Seplos, because at 54.40 the SOC shows too early. Due to a few load fluctuations in the system, the voltage briefly reaches 54.45 volts at around 85-90% - which leads to Voltage Protect in the Seplos and thus to calibration at 100% SOC. if you don't do this... after a while the SOC gets out of hand like with the victron BMV. There are many opportunities for improvement in the Seplos software, but unfortunately Seplos itself is overwhelmed by it. It starts with the problem of the precharge... the large capacitors are not charged far enough in just 3 seconds so that the seplos always goes into short-circuit protection when it is first started. So manually precharge and as soon as one pack is connected... the others work without any problems -in the normal case, at least one should always be connected to the BUS - otherwise there would be a bigger problem anyway
Hi Andy, I have just built a 48V 280Ah battery and I have a Seplos BMS 100A 10E. I ‘ve noticed that despite the inverter being setup to charge at 3000W, it only charges at 2500W maxing out between 45-49A even at low state of charge. I should be able to see around 60A going into the battery but for some reason the BMS is telling the inverter not to exceed 50A. I have checked the settings and I don’t see anything that could explain this. Do you have any idea?
Thanks again ! You’re Teaching is helpful , I have the Seplos Stand up 15s battery case & the BMS I ordered was a Seplos 7ss 24v 150A , my inverter is a Solar inverter 3000W 24v MPPT china it was less costly hopefully that does not come back & BIT ME !
Yes, I figured that as well. The translation is partly correct using the word voltage but then again pressure. Very easy to miss something or not being able to find the right setting in this chaos.
Hi Andy what amps is the balancer your using capable of putting in to the batteries to balance them I'm thinking I need to buy one too,and do u have a link for it.cheers from cloudy brrrrrrrrr.victoria,Rene.
Hallo Andy! Kannst du mir bitte erklären wie du das 16S Protokoll reingeladen hast? Bei mir kommt da immer wieder eine Fehlermeldung, also kein Verbinden möglich. Hat jemand das gleiche Problem? Lg aus Österreich
Re. BMS stop charging, this is exactly why you need communications to your victron gear. When you have comms, it just slows down your charge current when your battery approaches the high voltage per cell so that it never goes ABOVE that and trips the BMS. When it slows the charge current, it gives your battery time to balance instead of just shutting off your charger (when using chargers with comms like your Victron MPPTs and Multi). Also helps to have a real top balance before assembly ;)
While that theory sounds ok, i think Im with Andy on this one. He slowed the charge current down to 2A was it? And it still was having problems. I think Seplos definitely need to have a more active balancer in this system
Like apples and pears, I would say. The JK is great for a single battery and has exceptional balancing features. But it is a consumer grade BMS. The Seplos, Tiangbangda and Pace BMS are far more robust, can be programmed in far more details (which can be very confusing), have additional safety features and can communicate amongst each other and your system/inverter.
To get the BMS to register 96% SOC so it will charge to higher cell V, you should set the capacity higher, like 150 Ah. I think you set it lower than what it was before. If my logic is right, the BMS should then charge further, and if it reaches 3.6 V at say 140 Ah, it would record 140 as FCC (Full Charge Capacity).
Hello, I can't get a connection from the Pc (BatteryMonitor V2.1.9 )) to the Seplos BMS (Model 48150-1101-10C-SP15-16S) . Connection is always "Time out". What am I doing wrong ? Thanks ( Wenn möglich auf Deutsch dann ist es einfacher für mich)
I have those batteries also. When I charged them they alwas stopped around 96% more or less. I Always let a second charger connect connect to them and never worried myself about the 100% percent charged the way you do. Than I come to check them they were all 100 % charged. But the inverter ever really charged them at 100 % only the other Aims charger did that job. All you have to do put a 15 Amp charger connect to it and the charge them to 100% without you knowing it.
Seplos recommend to charge the battery pack with 100A and discharge 100A to the bottom or when the BMS switches it off. Then charge again at 100A to the top to reach 100% or the Ah rating of the cell modules is reached. That is what I was recommend by their tech team.
My 200A Seplos BMS with approx. 15 cycles keeps my 320Ah cells perfectly balanced. It is connected and managed by the Sofar inverter where I set the discharge depth to 90%.
Great. I didn't do an initial top balancing to test the BMS. As it is now top balanced, we will see how it performs the next time we fully charge it. I don't even know if the balancer works...
@@OffGridGarageAustralia And this is possibly one of the problems. No top balancing done… Another problem could be to stress the cells with charging it almost to 100%. This could be hard to manage for many BMSs, that have low balancing current.
Great video! Too bad that there is a problem with Temp1. I have 3 Seplos BMS 16s200A but with original sensors. 2 are currently in operation. No problems with the sensors. In my opinion, the BMS is foolproof! So suitable for me. I'am not need a extra balancer , Made TopBalancing before...
Yes, if you do a initial top balance, this would not occur. I wanted to test 😉 Yeah, the settings are actually quite good for a pre-set BMS and will work out of the box right away without the need to change anything.
Woascheinlich is es trotzdem sinnvoll, an aktiven balancer einzubaun, folls des päck amol über de joa aus der balance kummt. Sonst muas man olls wieda aussanond baun....
@@OffGridGarageAustralia I have used LFP batteries. That's why I only charge and discharge the cells between 20% and 90%. Your 130ah batteries are definitely better than mine. At 2.9V I have to stop discharging mine otherwise the delta will be too big. Between 20% and 90% I usually have 2-10mv delta. Yes , it does not have an active balancer . If a cell breaks out in the 20-90 range despite top balancing, I'll swap it out - Cell6 🤬. But I don't think it's a bad idea for maintenance purposes. Maybe I'll buy one(active balancer) :-), it's your fault!🙃
The intermittent power supply waiting is a great function, really great function. It protect your inverter from damage. When single cells reach high voltage protect, then this function working. Only when soc less than 96%, then can charge the battery again. Then why need this function? As battery is communicate with victron inverter through CANbus, when charging is cutoff as full charging, then this current will go back to inverter( has chance to damage inverter). So we can not allow this things always happerened. That is why need this function. When battery is full charging, then stop charging the battery. Otherwise you will find that the battery is always charging, stop charging and charging, stop charging….. and this current always back to inverter. So with intermittent power supply waiting, only when soc less than 96% start charging again, it helps a lot with inverter.
And about the Clamp ammeter, LCD and Software current shows is different, it is not a problem. As you working with inverter, the discharging current is sine wave. The software is update current every second, so it changed every second. The software need exact current number, so when problem, easy solved. But LCD is shows to all clients, can not let clients see the sine wave current and changed very quick, so it will shows a average value per 8 second. But clamp ammeter is more slow update. So you will find the clamp ammeter is always like 25A. If you working with stable discharging load, then you will find the problem solved.
About the battery really capacity, you need do one cycle to help the bms to learn the battery really capacity. Full charging the battery to single cells high voltage protect or total voltage high voltage protect, then discharging the battery to single cells low voltage protect or total voltage low voltage protect, then the BMS will get the battery really capacity( this is a very important step when doing aging test). But as it is home energy storage system, so it is hard to always reach this situation, but still have chance to reach it, when reach the situation, then BMS will learn the battery really capacity again. The SOH is depends on the really capacity decrease. For example a 100ah really capacity cells, our warranty is 3000 cycles at 80% DOD. So after 3000 cycle, the soh start working. It remains 80ah, then soh is 100%, if 60ah, then soh is 60/80=75%.
Das hilft mir sehr! Ich werde mir noch einen aktiven Balancer bestellen! Ich habe meine Zellen jetzt alle auf 3,65v mit 0,001A geladen, von daher habe ich wahrscheinlich bessere Voraussetzungen! Aber schaden kann der balancer nicht, wenn man ihn zum top-balanicing verwendet! 👍
Very interesting, thanks for your extensive tests! I guess if you don't need the display function you could probably ditch the BMS and throw a JK in there with custom cables instead of the supplies bus bars. Maybe the JK display even fits the cutout... But then a JK is more expensive than a smart balancer so that's probably the way to go.
The active balancer should only be used when the voltage are 3.45V/cell or higher with LifePo4. Otherwise it will do the opposite and unbalance your battery pack.
Hi Andy I am a distributor for a Lithium Battery that uses the seplos bms. They set it to 80% dod. I said I need the soc to go down to 15% to lower my electric bill to almost 0. by doing this I reduce the battery life by 5% said the China Engineer. What can you recommend that I discharge my batteries to? with a 20% soc they claim I will get a 6000 cycle life for these CATL 15s 100Ah battery
Thanks to your channel I also bought different BMS. and I have them in the closet now. No smart Daly BMS, ANT BMS 0,2a balancr, JBD relay BMS 0,15a balancer. all scrap. now I have two JK 2A... BMS in use and this is really the BMS you want. and this milliampere balancing can be done with 18650 cells. ps. I made a copy of your battery shelf. but without the mega bus bars. Grüße Alex
Another ‚disrespectful‘ battery test - thank‘s a lot. I‘m looking forward to see what’s wrong with the temperature sensor, or is it the BMS itself, that does not take the correct measurement?
Heya, oke a good battery pack except for 5wh less but maybe after a couple of charge and discharge cycle it will even get the extra 5 wh. still intressed in more info about the bms but I gues that will come soon
Great question, I really don't know. The cell voltages decreased a bit during that time but it could be the voltage is slowly settling anyway without charger. Could be due to the balancer. Could be both... 🤷♂️
andy , the "runner" cell, isnt that one you tested, and thus may have caused the imbalance yourself ? i must confess i topbalance my cells before i put them in my kit, but i am not seeing must deviation. furthermore , when you connect the seplos to you victron setup, charging amps will be lowered at the top of the curve, preventing cell imbalances
Yes, 13, 14, 15 and 16 were the cells I tested and as I said in the video, this massive imbalance was home made. I wanted to see if how the BMS handles that... Usually that would not occur and you would have done a proper top balancing with all cells in parallel before installing them in the case.
@@OffGridGarageAustralia Would you though? They seem to indicate that you install the cells as they arrive, no mention of balancing beforehand. If they come roughly balanced at mid SoC, maybe the balancer can handle that? That would have been good to see. Want to discharge the cells to 50% SoC and reset the BMS and test the theory? 😉
@@OffGridGarageAustralia sorry andy, but kind of feel like you setup the bms to fail, especially since you knew it was passive balancing only. i hope you'll still give it a fair chance in your 280ah kit , and you'll do a over time review. i do agree seplos needs to start looking into adding active balancing into their bms, as voltage difference will become bigger as cells age
@@houseofancients it was part of the test. I knew about the small passive balancing the BMS has and we have seen this same result with many other BMS's in this situation before. When we build the large MASON system, I'll do a full top balancing of the batteries before we assemble it as one would do. I still believe, the BMS will struggle with even a very small deviation in these large capacity cells.
![Blox Fruits Dragon Rework Update [Full Stream]](http://i.ytimg.com/vi/EqDAp8udhm0/mqdefault.jpg)
That hour literally flew by 👍
I enjoyed going on that journey of discovery with you all 😎
Thank you. That was my longest video yet but I felt, splitting it in two parts wasn't going to work. I need to tell the story, right?! 😉
Andy:
I also purchased two of the Seplos 280Ah DIY Kits in April and May. I really liked the idea and they did not seem expensive, although the shipping was $190 per kit. My cells were only a few months old and came from Docan Technology. I was thoroughly dissatisfied with the cells as very few test to the alleged capacity at a very low C rate. The second set of cells also came swollen. Docan suggested I just compress the cells back to the original size. I chose not to do this and will not buy more from Docan. Both kits came packaged as yours did, with loose parts all over the place. The PCB boards had broken leds that had been sheared off by the loose parts. I soldered the leads back and requested a new board from Seplos which was sent quickly. One thing I did receive you did not was a parts list. It was not completely accurate, but it really helped, especially with the screws, took me several hours to sort the screws out.
The one thing that made me decide not to buy another Seplos kit was the customer service. One of the bus bars, S-F-3, that connect the BMS P+ to the fuse block was bent backwards. The BMS sits lower than the fuse post. Normally the bar is bent so the short section goes over the top of the long section. This causes the short section to the fuse block to sit higher than the long section. This way it will sit on the BMS and fuse block very well. I contacted Seplos and explained the problem, they refused to replace it. I sent pictures of the problem, they refused to replace it. I sent pictures of their other S-F-3 I had that was correct and the mismade bus bar, they refused to replace it. They told me to take the mismade bar and just bend it back over. I told them the kit would then have been modified by me and the warranty would be void. Again I was told to just bend it. I told them it was not possible to take a piece of thick metal that had already been bent and reverse the bend, even if it was a lot of thin pieces, and expect it to work well. Again I was told to just bend it. I asked them if losing a good customer for want of a $2 part was worth it to Seplos. I was told that the part was not a "qualified" part and would not be replaced. Eventually I removed the shrink wrap and did bend the bar back so it would fit. It is an ugly bend and it barely fits. But it was my first option other than making a cable, my second option. Consequently I will move on to other commercial batteries.
I have also not been able to get the Mason 280 kit connected to the computer and would like to know how you did it. I have not seen a video of the Mason 280 kit connected yet if you have done so already.
Thanks for the detailed info!
Great vid & info thanks Andy, I think u may have helped me break the ice with my 3 x Seplos Suntorque Batteries I have had since march and 2 of them have not wanted to charge over 72% max while the 3rd battery & last one delivered still charges to 97% max. I have had a discussion with the sellers techs who advised I needed to get some communication happening between the batteries and 5kw MPPT All-in-one inverter. I have been sidetracked with other issues, but I did find this all rather intimidating as I dont have a working PC presently, but it has been on my list of stuff to do, and you have made this all so much less confronting for me now, so it is time to get the old PC going again and fix this BMS situation I am having. Cheers
Great video Andy, you really got me on the cliff hanger near the end 😆 Maybe you could mount a key switch on the side to activate the active balancer addition.
Haha, great. Couldn't resist with the premature ending of the video🤡
Yeah, a little switch somewhere could be easily implemented. We will see...
@@OffGridGarageAustralia Another function for the Raspberry Pi or an Arduino?
Actually a simple BMS could be easily implemented in either.
When the "BMS" is initially turned-on individual CELL charging and PACK discharge is disabled.
begin routine.
Check each CELL's voltage against the upper and lower limits and when the CELL is below the UPPER limit enable charging for that CELL. If the CELL is below the lower limit disable PACK discharge.
(Do for each cell)
If after checking the cells and they are ALL above the lower limit enable pack discharge.
Rest a short time.
repeat routine.
A physical structure would need to be created to limit charging current (Amperage) for each cell as well as the entire pack. Since cells stacked in series increase voltage of the PACK but not amperage the discharge current for the entire pack could be the same as each CELL's capacity.
IF anyone decides to try to build their own BMS, they and they ALONE are responsible for any results. I assume no responsibility since this is only a theoretical example of how such a device MIGHT be constructed and NOT a detailed example of how to actually do so.
For what it's worth, my career was spent in large mainframe programming and systems design at both national and international companies. I designed and programmed systems handling millions of dollars monthly in accounts receivable and payable as well as insurance and manufacturing systems.
Belle analyse et recherche ANDY, merci pour le développement de tes idées.
Great research and testing Andy, I love it, and one hour went by pretty fast. From what your research has already shown most of us guessed the active balancer was going to be a necessary addition. Perhaps Mason should perhaps include an active balancing option and compensate you for your research and marketing.
I think that the behavior could be in fact appropriate. As this BMS has really small balancing current, it needs much longer to fix a cell. If it re-enabled charging immediately, the cell would be still off by a lot and it would turn off again almost instantly.
That is right, but without any charging at all, the other cells will never catch up and only the highest cell gets discharged a tiny bit every time this happens. Without a proper top balancing and matched cells, this concept will not work.
@@OffGridGarageAustralia Makes total sense. My JBD smart BMS will cut off charging if one cell is high and then the balancer turns on for 2 hours or so and then when the cell drops the bms will start up the charging again for a few minutes and it cycles until its balanced... fully automatic!
Thanks for all that went into making this video. I've been corresponding with Seplos for the past 6 month on the Mason DIY kit for a planned expansion (and possible conversion) to two home made 16s Overkill EVE LF304 cell batteries. I never pulled the trigger on this Seplos Mason expansion project because I could never find any comprehensive information on the Seplos BMS including integration in Victron DVCC. DVCC is also suppose to provide intelligent charging integration with Victron smart MPPTs, but I'm not sure exactly what that is. While I like the Overkill BT app for looking at the cell information, my installation is remote and I'd like to be able to see the individual cell voltages and deviations for each battery on VRM or the remote console. If the Seplos BMS balancing function is struggling with 135Ah cells it's probably going to have real problems with 280/304Ah EVE cells. I'll keep an open mind, but I think you've help make my decision. Looking forward to the 280Ah or 304Ah Mason DIY kit build. Thanks again for all you do.
Thanks, Brian. Your absolutely right and we have to get the remote access to the BMS data working. I have tested all day yesterday to get any information or of the Seplos BMS without any success. I've ordered some other adapters now which should work and give us the info on the VRM. So stay tuned. You already have the Cerbo so it should be fairly easy to use CAN protocol to get the information from the Seplos to your VRM.
Thanks for sending these amazing pictures of your off grid observatory a while back. That looks really fantastic.
Here in Lebanon Growatt and Deye inverters are widely used and Seplos support them with CANBus. we install Seplos + heltec balancer with a switch for maintenance. If you use canbus with growatt for example, the bms will tell the growatt to charge with CV, only one voltage. A good advantage of Communication with canbus, is that our customers will see the accurate percentage SOC from the inverter/App directly. Any big error from the BMS will be informed to the inverter, will lead to safety shutdown and more features.
That's good info here, thanks a lot for sharing!
Are you familiar with Felicity and its EDMS settings? If so I want to ask you some questions if it is ok
@@esmailalkassir1565 sorry, i don't use Felicity products.
Hello Andy, many thanks for your test, question, can you use the NEEY on this battery ? Great Video as always !!!!!
HI Andy ... nice review of Seplos battery package ... i need to say that i have (just) 2 seplos BMS on my own 48V 230ah package and all working fine, as you said, you need little bit time with software but then all goes right way. Also communication with cerbo gx works fine.
Thanks for your feedback. You just use CAN communication to the Cerbo?
Hi Jaroslev, how did you get coms to work with the Cerbo GX? Did you use Louis van der Walt's driver? And did you have to downgrade the Cerbo's firmware? Ciao, Rik
@@OffGridGarageAustralia Hi Andy, in software you can setup CAN port for plyontech,victron or other protocols. Plyontech CAN works like battery (set up SOC, charging from grid etc), Victron works like battery monitor just see actual statuses
@@jaroslavchytil5732 thanks for your response. Could you please elaborate a bit more, or was just a matter o setting it up to use the Victron protocol and use the right cables?
Could you just go by without using the CerboGX ? (my plans include VenusOS in a RPi , but not a GX)
Thank you very much for all your insights!
@@TheShadowVortX it just about correct cable, then cerbo find BMS and all communication works fine
NOW I see where tomorrow went!!! LoL. Took a while to get it done. Sorry for all the hard work Andy. Still all and all, I think I like the home built, peace by peace. Choose the components I want and understand everything that is going on with it. But you will never know what is out there unless someone like you tests them. See ya on the next video. ( tomorrow) ??? Lol
Thank a lot Alan. Yeah, it is a lot of time going into these videos but I'm not complaining, I really enjoy this kind of work, editing and forming a story out of all these single video snippets I recorded over the last 3-4 days.
I may have another video tomorrow actually... 😉
It's crazy the amount of parameters it has. Also, it seems like the one cell is messing up everything. Does SEPLOS let you purchase individual cells for replacement purposes? Great Video as always!
Thank you.
Seplos has cells on their website, but how do you know they are better than the ones they have already sent?
It was part of the test to let one cell go really high to test the BMS's function. Usually you would do a top balancing of all the 16 cells before installing them so this extreme voltage difference would not occur. But even of having the pack balanced with the active balancer, as soon as I disconnected that, the voltage difference was on the rise again. I have to ask Seplos how their balancer works and how to check...
It is the same with PC software from QUCC…quite confusing and no explanations on what the parameter does. Some are simply not intuitive at all.
@@OffGridGarageAustralia Please ask, I have a feeling the balancer turns on the bleed resistor for a cell at 3.5 judging by the FULL pack voltage over voltage setting. The point of a BMS is to just monitor and that means you should set your charger to charge the pack at less than 3.5 per cell. I think that makes sense for lifep04 packs because they have such a weird curve at the top (above 3.45v) so its a good idea to stay away from that area.
@@OffGridGarageAustralia We eagerly await seplos response.😉
Hi Andy and thank you for all your testings about Seplos
I use Vertical Kit from Seplos to make 16S 300Ah Battery Pack
BMS inside is 10E, with bluetooth
I use with a Solis inverter, everything looks good now, only that i just can't figure out what is the warning / parameter that make BMS to ask inverter to limit charge current to 10Amps
Sometimes happend to see inverter charging at 70Amps ( maximum set by me ) and finish charging without lowering current and sometime is lowering to 10Amps and stay there for a while and finish charging
Thank you again
If you set your charger to ±100mA for the last part of the charging process, you allow the bms to discharge the "full" cells until the lowest one creeps up
I know, in theory thsi should work, right? It would just take 1000h+ maybe. The active balancer has done it in 30min.
I'm not even sure if the passive balancer of the BMS is working or not 🤷♂️
@@OffGridGarageAustralia that is my biggest question too cause you barely see it happening and I guess that I will need a BMS too like you had mentioned for a kind of quarterly maintenance. But not sure what to get for this task cause the Neey seems to be a big gun for a 8 kWh battery.
@@OffGridGarageAustralia 166 hours to balance 5000mah @30ma balance current
@@OffGridGarageAustralia Since the balance current of the Seplos BMS is about 150mA to 180mA, it takes about 6h to balance 1Ah difference.
Hi Andy, really glad you did this deep dive on the Seplos BMS and it's software. And you're right, the mobile app is alot easier but I guess where the Seplos will shine is that once you make the change once, the software will send the updates to all the BMS in your battery bank. That'll be very helpful if you have a battery bank with 3 or more batteries each with their own BMS. With the JK, which is excellent, you'll need to log into each of the JKs and update parameters on each of them. I guess it's a trade off somewhat. My question to you though, let's take the 4th gen NEEY smart balancer for example, is that something you would leave permanently connected with the Seplos so that balancing would occur automatically and you wouldn't have to look back at it?
Thanks Henderson. Yeah, I guess we will have more fun with the BMS soon. I have ordered a while bunch of adapters now to make it talk to the Victron system (not just the inverter). We will see how this plays out and works. I agree with you, as good as the BT app seems to be but from a maintenance perspective it is very cumbersome and takes a long time to go through all the parameters individually. I hope this is really something the Seplos BMS will shine and make us happy.
The Neey is certainly something you can easily integrate into the Seplos Mason system. There is heaps of room at the front and some Velcro will install the balancer nicely. Once programmed, it will help balancing these cells.
@@OffGridGarageAustralia ok great thanks Andy. Looking forward to your follow up videos.
Andy...
Have you checked the bms PCB to see if there is an ujustment pot for the shunt and voltage calibration.
Is it possible it's holding back the charging until the balance delta improves ?
Hmmmm.
Is the current firmware up to date ?
If it was my battery I would top ballance all the cells in parallel before instalation.
May be a good idea as you say to check DC ripple or any inducted AC !
Stu.
Yes, top balancing is a must, as I said in the video. This was part of the test to see how the BMS reacts and how the balancer works.
There is no pot on the BMS as far as I could see. This is all software based.
@@OffGridGarageAustralia That's a shame. It's allways hard to get shunts to read accurately. They are simply not highly precise devices.
One question if I may.....
You had the shunt on the floor. I was wondering if you had the meter voltage reference wire connected at the battery terminal as a lower voltage will give a different reading as you know. With the meter your using I prefer to power the meter with a separate supply and common the ground connection.
Great episode as always 🙂
@@OffGridGarageAustralia If you wanted to use the balancer you could always use a zener diode with a pot connected to a relay via a mossfet. Then use the normally open terminals to close the run jumper on the balancer.
Just an idea but not something you expect to be doing with a brand new battery and bms !!!!
For your efficiency you should calculate the A X V difference from the supplying battery to the A x V to the receiving battery - not just the current. You have correction in both the inverter and probably your charger.
Hi Andy, did you already figured out how to connect the Seplos BMS to a raspberry pi? With the Cerbo it's easy but I'm not sure how to do it with the Rpi?
Awesome, time to see the results, get to compare your results with mine
if you want the charging switch to turn on again, you don't have to discharge until 96% but you simply turn of the bms and turn it on again (aka restart).. 😁
I have been using this bms for over 1 year now. The only thing I dislike is the balancing function, it is really really need an upgrade.. 😆
Anyway.. great video as always. 😄👍🏼
Nice video Andy, looking forward to more tests on this unit.
Did somebody say “DC ripple”?
Hi Andy. Are you using the DVCC function in VenusOS so that the current limits are controlled by Seplos BMS?
Yes. I use that in the SPAT Calibration Centre installation (Battery 3.0).
I also use DVCC in my Off-Grid Garage for limiting the charge current without any smart BMS connected.
Thanks for the solid content this is exactly what I was looking for. I already had one of the new JK BMS‘s on-hand but was going to order one of these Seplos BMS to go with the 280ah case I ordered ... going to stick with the Jk bms + battery box only
The JK seems great and provides all function we need.
I test the communication of the Seplos now but doubt it will have any advantages in comparison to any other BMS out there.
@@OffGridGarageAustralia Does the JK BMS have cell balancing while "charge" and while "static?" I wish all smart BMS's would enable that!
@@uhjyuff2095 Yes
@Kadeem Nameman>>> Kindly share the details of your battery box/case and where you ordered it from with me.
@@uhjyuff2095 I don't even know if it balances at all 🤷♂
Now I am so excited to see what happens with the DIY 280. Will Andy incorporate the active charger from the start knowing the internal balancer will take 365 years to fully balance his pack? How will he run the wires for the external active balancer? I have an ideal , run the cables to the outside. As the active balancers are seldom used, When the need arises just plug in the balancer from the 135. After balanced unplug, put on the shelf until something needs balanced. If the wires are already externally located then it would take a matter of seconds to connect. Just a thought and not telling you how to do your battery. As for the temp sensor, I personally un soldered and re soldered the connections. As the connections at my astonishment were very poorly done. Not at all like the rest of the board. Good luck and stay creative!
I assume he’ll top balance the pack first which is best practice anyway.
Hello from Germany, if you know the Seplos BMS, you must have thought about balancing, many have an active 5A balancer from Alibaba and don't know whether you always leave it on or just plug it in every month and then look so that the cells are well balanced. On the bottom of the boards there is a soldering point or several where there is also soldering for Lifepo batteries. this big pad covers 2 contacts and if you solder a wire and switch you can turn the balancer on and off. look closely, on the circuit board at the edge.
I like seplos and the service also
The BMS will be working with a lot of canbus inverters like goodwe or growatt and sma, lot others more.
That is there biggest benefit, cause no other BMS supported Growatt SPH or SPF inverters ,
even not the JK BMS as canbus version which I had bought trusting them that they could fit and run.
I want the BMS to work with the Victron OS, not just the inverter. Working on it...
@@OffGridGarageAustralia that will work too, cause our american relatives are using Victron too , I guess the Raspberry pi is managing that for them, but I know about cause they also have growatt.
Great progress even though lot of points that are not as straight forward as the seplos bms could be.
@andy
Just wondering how much do you generate in such a blue sky ?....
Here in Dar es Salaam, I have been seeing 70A from 5kw solar since 9am.
Amazing
Hard to compare as we still have winter. You can see all my solar data here: vrm.victronenergy.com/installation/100328/share/2e028071
@@OffGridGarageAustralia Asking for login details
I have another question that maybe others have dealt with and may have found a solution. Since I operate several packs in parallel and the individual cell voltages(and many other values) are unfortunately not transmitted to the Cerbo and then to the VRM, I use the option of being able to check all values on the desktop in the office via RS485 and USB using the SEPLOS software if necessary. With only one pack, this is not a problem, since the DIP adressing all remain at 0 and the parallel connection CAN & RS485 does not cause any problems. However, if you use several packs and adjust the DIP switches - so far I have unfortunately not been able to find a setting where ALL packs including the master pack are displayed in the SEPLOS software. This would be PACK 0. Maybe another one already has this great one missing the possibility of being able to read out all packs remotely - WITH a parallel connection via CAN bus.
I think you have to leave DIP 5 off for that to work but your packs will be in a different order than. Pack01 will be the master than.
I heard this from a friend. (Thanks Raphael).
I will test all this once we have the second pack build and connected.
@@OffGridGarageAustralia Many thanks Andy for the info - but unfortunately this doesn't work - or I misunderstand it. I've also tested almost infinite variants to work around problems that I don't think can be enforced, but without success. The only problem is that when all packs are connected and the master enables communication with the Cerbo - this master pack is then no longer displayed in the SEPLOS software. If you then address the DIP switches to just one pack - all packs( since still addressed with DIP 1-4) displayed in the software but the Cerbo only sees the one "master pack"
@@lincolnator885 do you use the victron firmware or pylon ?
@@sergea1138 have choosen victron
Watched your video on charge controller settings and you mentioned using parameters set by cell manufacturer, do you have any info on the higee cells and max per cell charge?
Good video and very informative, thanks again. I bought the Seplos BMS to make a battery pack to use with the Sofar ME3000SP ac charger in the U.K as the charger needs to communicate with the BMS. This BMS appears to be the same that is used in the Pylontech batteries that Sofar also use. The batteries (230ah) came 2 days ago from Docan Technology in China in less than a month which surprised me and I wasn’t ready to set anything up! All the cells are at 3.3v but I’m going to top balance them before I build it. Is it not correct though that the cells will all adjust to a very similar voltage over a period of time with the BMS and, the mv difference doesn’t really matter because the total capacity of the battery is high enough already? I’m sure that you won’t ever discharge your battery by 100% in the real world so it will never need the cells to be all the same voltage. There is always a variation in anything that is manufactured and a small amount of tolerance should be expected. Just my thoughts and I’m not an expert by any stretch of the imagination but I found this test very informative. Thanks.
I sent an rfq to seplos and I will look at your video an decide if I purchase it. I will tell you my decision…
The big advantage of the BMS is that it can talk to the charger/inverter. You ask why? For example your issue with the balancing. If the SOC reaches 100% the BMS asks the inverter to slow down the charging amps. This then allows the BMS to balance the battery. You see the same effect in your video when you manually lower the charge amps when it seems that the balancing is doing it's thing. Of course due to the large difference caused by the 4x cell test you did it isn't doing a great job but it will eventually reach a good balance. I have a similar setup running 6 months now. No initial top balance but the cells are still in good balance at 100% SOC. It just keeps them nice and perfect in balance when you just let him do it's job over time.
The Seplos BMS only balances with 150mA, it makes very little difference if you have runaway cells like Andy displayed.
@@danielp5208 True but normally that would never happen. I you might have a runaway cell you better need to check why that is and remove the cause for that. Balancing (150mA per cell) is ok in this BMS because of the slow top charge request. Only if you don't use communcation to the inverter and do a manually charge this slow balancing might be an issue. But why use a BMS like this then? People buy the Seplos for this reason. If you really need active balancing you could add it afterwards but in 6 months I didn't need it and doing a full charge/discharge everyday.
Nicely done professor Andy…… as per usual!
I was impressed with how David Poz calibrated his shunt by cutting a notch into it
Hello Andy - one more question - I had copper washers in the boxes next to the aluminum bus bars. You also included them? and if so - you didn't install them? I followed the recommendation but I'm not sure if this makes sense, because it could be that the teeth of the nuts could then not function?
We are racing motorcycles .. We are LEEEEEEEEEANING on lithium ion chemistry .. I am learning quit a lot from you.. But your chemistry is different from mine.. It is difficult to keep our chemistry balanced using it like we do... I like it when you share the difference of what you are using to lithium ion.. How you are using your chemistry is like taking a nap on the couch compared to how we are using ours..
Hi Andy, and thank you again for this long video...😁
In fact I didn't think they used this software, I thought it was another🤔
The PBMS Tool.
Yes it's true this one I find less good, but it does the job, that's the main thing.
These Seplos battery packs for me are fine.
Thanks again Andy, and until next time...👍
Wow. So glad you did this. As a novice I would of not known what to do had I bought this kit. Until mason sort this balance issue and make an app I will stay away. It’s a shame as I was talking to a salesperson from them yesterday about a possible purchase. I wanted to build a 5kwh for a first project. I have no engineering or electronic background so would have been relying on artisans in the Caribbean to help me 🤷🏾♂️😱
Thank you for your testing of this battery set.
I think trying to sqeeze out the last milliamp is like installing a solar tracking system instead of a fixed one. For me it makes more sense to keep it simple and install just more to get the same output.
I would not need to be able to check the battery at any time too, because I trust that it will work on it's own without supervising every day😊
I always wonder if the capacity as measured is because of the small but real losses in the bus bars and the flexible connections to the BMS. Also there must be measurement errors on the clamp meter and the shunt. Proof of this I guess when you do the high discharge tests next? A very interesting and informative video Andy,. Thank you!
Sharing you’re hard worked knowledge !
Great review of this battery setup. I agree the balancing seems to be lacking, but you made a workaround and got it done. These reviews are forcing you to learn new things and share it with your viewers. Thanks Andy.
Hi Andy. I really appreciate your channel and your are funny too. One question: do I have to let the active balancer always connected, or should I connect it only sometimes? Best greetings from hot, most northern Germany.
Thanks for the feedback. As we have tested in previous videos. It is not recommended leaving it connected as it will balance 24/7 then. Once you fully charge again, the cells will be totally out of balance. I would recommend either using it occasionally for maintenance purposes or turning it on only above 3.4/3.45V.
Most northern D can only be north of Flensburg...
@@OffGridGarageAustralia thanks a lot Andy. You are right with Flensburg.
I wanted to know if you plan on buying more of them. Is this basically going to be your battery-3.0 setup. and how is the total price for one vs if you made it yourself like battery-2.0. I only saw you mention side handles, but it kind of looks like it might be rack mountable in size. Did it have holes and/or rack mount ears to rack mount it?
So far, I have two of these systems to play around with. If I will order more... time will tell. More testing needed.
Great video mate, could u just leave the active balancer connected permanently?
Can I successfully connect the seplos BMS 14s to my energy storage consisting of the LG Chem NCM (Li-Ion) 12S battery? What should I do with the 2 free bms cables?
I don't think this will work with Seplos BMSes. They only work with the specified number of cells unlike many other where you have 8-24s support for example.
Andy, would you mind sharing what app you use to read the QR codes on LIFEPO4 batteries to tell how old, who made and capacity ?
I'm looking to replace a 48v 400ah Lead acid battery with lithium can you recommend a comparable size ah setup and a BMS to go with it. Love the videos thanks
Thanks, Scott. I would get a 280Ah battery bank with a JK-BMS like I have shown here: ruclips.net/video/edvKvxFpkec/видео.html
After connecting Seplos BMS with Victron inverter you turn your ESS to next level - Victron can change charge current during charging for example.
Also with daly, jk etc. I have UART to venus from daly, and reduced charge current works. Also reduced discharge at low SOC.
How do you reduce discharge with low SOC? That is not possible.
@@OffGridGarageAustralia Not sure. Discharge is set at 50 amps, but the other day at low SOC it had changed to 25a. Maybe it disables the inverter. I've never tested what it does. Where can I send screenshot?
@@OffGridGarageAustralia It only works in a ESS system with a grid connection.
@@OffGridGarageAustralia By taking more from grid so in your off grid situation impossible indeed.
I have watch your seplos videos several times and I thought you said the BMS shuts off of no power is going in or out, did I misunderstand what you were saying or did I completely imagine it?
Can you clarify were the cells supplied to you unbalanced? With 100mA balance current you are always going to want to be building a pack with balanced cells.
This BMS only has value if you have it networked to the Inverter/Charger, otherwise it's just a pain in the arse. I appreciate it disconnecting if the inverter is trying to charge past 100% because that is not a nominal condition, and I'd much rather it just shuts off if something becomes unusual, caution over everything else. There's probably a toggle switch to disable these protection functions? I had an issue with it rarely shutting down because of a false short-circuit detection - I disabled the feature (I have a fuse for that condition) and the problem was solved. Love your videos please keep up the good work!
This was part of the test. I have explained this in the video: the last 4 cells were capacity tested and therefore were on a different SOC s the other ones. I wanted to see how the BMS handles such a situation.
Usually, you would do a full top balance in parallel. As I said in the video 😉
In one video you built in a neey bms and you installed a switch to it. I can't find the video, could you tell me wich one it was, thanks a lot.
Andy, Fantastic job. Congratulations. I am having the same charging problem with two batteries. To balance with active.
Question: the active balancer, do you use it with the battery connected but without extracting energy to the house?
Cheers
Thank you. Would you recommend this Setup? Or a Very Basic set up with 130ah batteries\cells for a newbee?
That totally depends on your project. The Seplos Mason system is easy to setup and if you don't want to wait weeks and months for all your parts to arrive, go for it.
Thanks Andy, very very interesting video. So, it seems that overall the BMS is doing its job correctly except for balancing the cells. I think for that it would be interesting to use the NEEEEY active balancer as a companion. It balances efficiently with 4A, it is possible to program the start and stop conditions, and as a bonus you can monitor a lot of battery pack information directly via Bluetooth. It would be interesting to see if it fits inside the Seplos box.
Thank you. Yes, the BMS seems to work as programmed with all its safety features. Adding an active balancer for the occasion to balancing or even a Neey seems to be the way to go...
@@OffGridGarageAustralia there is a programable relay for heater function but maybe we can use it controll the 5a active balance. Maybe explore it for future video
@@Pey5531 the relay function is within the Seplos BMS or an external, additional component?
@@OffGridGarageAustralia there is an unused header built-in on the BMS board. I think is only logic level relay but we can connect to a bigger relay if we need to control a bigger load there is even a heater toggle on the software. We maybe need a modded firmware to use of other purpose.
@@Pey5531 I'll ask the Seplos engineers. That would be a great feature for this BMS.
@andy I was wondering why the BMS didn't turn on charging limit switch?
On my BMS the charging limit switch turn on (10 amp) when high-voltage warning occurred. Did you turn it off or something?
No, I didn't change any settings in this perspective. 10A was the current anyway I charged with so would not made a difference in this test, I believe.
@@OffGridGarageAustralia the charging process wouldn't be different but the indication on the software should show.
I found the limit switch is quite nice when the battery approaches 100% and it will recalibrate the soc to 100% after it reaches top balance voltage.
This video is very valuable material to watch, for SW developer, how NOT to code BMS. BMS have to send date / ESP MESH, RS485, MQTT, CAN / to another HUB or gadget / inverter , MPPT / but, if BMS will switch of battery, it will kill my MPPTs or my inverters and MPPT and inverters has higher value, then on cell in LFP battery. Within 9 yeras, I have no need to switch off my MPPTs or inverters, because I have not BMS in my NMC, NCA, LFP and Pb flooded cells. But now, when my LFP cells are old, I see different AC impedance, so I am building my DIY BMS with 2 4 8 10 and 16 AMPS balance currents. ESP32 BMS + MPPT_0 + MPPT_1 + MPPT_2 wil communicate, without cable / no need for data surge protection /, without WiFi - no need for router and WiFi NET, and if DIY BMS tells MPPTs, that "I need 6 AMPS", MPPT 0 to 2 will decide, with respect to Sun, panels orientation, weather conditions, teperature of MPPTs toroid coils and MPPTs FETs, if all MPPTs will give 2 + 2 +2 AMPS equally, or 6 + 0 + 0 and so on. I have also ESP32 thermocam, so thermocamera will see if there is problem with cells or bus bars or PV DC fuses, and with help of libs, like TensorFLow will apply solution. Also my DIY BMS and MPPTs communicate with me with speech commands, and give me info with sound like: OV of cell_x is y VOLTs, temp of cell y is z *C, connection 6-7 is hot etc.
Andy, might it be possible to use the Mason components less the case on higher capacity cells, and if so can you buy them less the case?
Ich steuer mit meinen Victron BMV 712 Smart ein Relaisausgang bei SOC über 90% und aktiviere damit auch ein Activ Balancer.
Perfect, that's one way of doing it.
Will you also try to connect it to the Victron Venus OS?
He needs a Victron GX-device with CAN-Communication port...he does not have one jet...
Yes, I will do that but using the RS485 if possible.
@@OffGridGarageAustralia other way would be the CAN-Bus?
@@OffGridGarageAustralia , NO, you can't with RS485 and this seplos bms, only CAN.
Hello Andy, finally... I'm glad that you finally bought a Mason box with the Seplos BMS. After a few months of fine-tuning the Seplos settings, I have to smile when I see that I'm similar had to learn all the individual problem cases in your video. You are absolutely right that the Seplos software still needs some attention. Just like many in the DIY forums, I can't understand why the Sepos Guys don't make full use of the possibilities of a good BMS. I now run 6 packs 280Ah in parallel on my Victron system. These communicate with the Cerbo. Again here, I had to use some tricky settings to get the maximum performance. But well... after a lot of hours everything is working fine. I have also retrofitted an active balancer in all packs .. which is switched on externally depending on the voltage.
Hey, as i am switching from SMA to 3x victron Multiplus, may i ask for your experience and settings ? Using 3 Seplos Mason Kits :)
@@patrickmartin4482 Hi Patrick, the communication Seplos with the Cerbo works without any problems via Can Bus. At Victron I have activated DVCC, which is now used to control the MPPTs externally (at least in part) by the SEplos. Here I had to adjust the settings of the parameters a bit. On the one hand, the maximum charging current so that this is normally over 400A, since my two MPPTs each deliver 200A and are otherwise already regulated below my final charging voltage. Likewise, the charging end voltage itself, since I only charge up to 3.40 volts, corresponds to 54.4 for the 16s pack. In the DVCC I set 54.40 in the Seplos, but 54.45. Victron always takes the lower value. But since I want to have 54.40 as a 100% SOC, I have to do the trick with the 54.45 Volt in the Seplos, because at 54.40 the SOC shows too early. Due to a few load fluctuations in the system, the voltage briefly reaches 54.45 volts at around 85-90% - which leads to Voltage Protect in the Seplos and thus to calibration at 100% SOC. if you don't do this... after a while the SOC gets out of hand like with the victron BMV. There are many opportunities for improvement in the Seplos software, but unfortunately Seplos itself is overwhelmed by it. It starts with the problem of the precharge... the large capacitors are not charged far enough in just 3 seconds so that the seplos always goes into short-circuit protection when it is first started. So manually precharge and as soon as one pack is connected... the others work without any problems -in the normal case, at least one should always be connected to the BUS - otherwise there would be a bigger problem anyway
@@lincolnator885 thanks a lot !
Hi Andy, I have just built a 48V 280Ah battery and I have a Seplos BMS 100A 10E. I ‘ve noticed that despite the inverter being setup to charge at 3000W, it only charges at 2500W maxing out between 45-49A even at low state of charge. I should be able to see around 60A going into the battery but for some reason the BMS is telling the inverter not to exceed 50A. I have checked the settings and I don’t see anything that could explain this. Do you have any idea?
Is this BMS branded as seplos or other name when it sold separately
What is the charging overvoltage protection set to 63V? 63/16=3.94; does it mean cells can go up to 3.94V while charging?
Thanks again ! You’re Teaching is helpful , I have the Seplos Stand up 15s battery case & the BMS I ordered was a Seplos 7ss 24v 150A , my inverter is a Solar inverter 3000W 24v MPPT china it was less costly hopefully that does not come back & BIT ME !
Funny, yet interesting how they translated 'cell' as 'monomer' :) Guess a battery from now on will be called a 'polymer'!
Yes, I figured that as well. The translation is partly correct using the word voltage but then again pressure. Very easy to miss something or not being able to find the right setting in this chaos.
Hahaha…was thinking something similar.
Yes, monomer is a bit of a misnomer. :)
Hi Andy what amps is the balancer your using capable of putting in to the batteries to balance them I'm thinking I need to buy one too,and do u have a link for it.cheers from cloudy brrrrrrrrr.victoria,Rene.
He has all the links on his website ;)
And he has made videos about the Heltec active balancer, and others.
Yep, the links are below the video or on my website for easy access:
off-grid-garage.com/battery-management-systems-bms/
(Thanks Raphael!)
Hallo Andy!
Kannst du mir bitte erklären wie du das 16S Protokoll reingeladen hast?
Bei mir kommt da immer wieder eine Fehlermeldung, also kein Verbinden möglich.
Hat jemand das gleiche Problem?
Lg aus Österreich
So you think you need to top charge and balance before you install the batteries in the case?
Re. BMS stop charging, this is exactly why you need communications to your victron gear. When you have comms, it just slows down your charge current when your battery approaches the high voltage per cell so that it never goes ABOVE that and trips the BMS. When it slows the charge current, it gives your battery time to balance instead of just shutting off your charger (when using chargers with comms like your Victron MPPTs and Multi). Also helps to have a real top balance before assembly ;)
I'm working on this right now. We will see. I doubt slowing down charging will actually prevent this from happening.
While that theory sounds ok, i think Im with Andy on this one. He slowed the charge current down to 2A was it? And it still was having problems. I think Seplos definitely need to have a more active balancer in this system
How does the seplos bms stack up against JK bms? I'm interested in 8S configuration
Like apples and pears, I would say. The JK is great for a single battery and has exceptional balancing features. But it is a consumer grade BMS. The Seplos, Tiangbangda and Pace BMS are far more robust, can be programmed in far more details (which can be very confusing), have additional safety features and can communicate amongst each other and your system/inverter.
To get the BMS to register 96% SOC so it will charge to higher cell V, you should set the capacity higher, like 150 Ah. I think you set it lower than what it was before. If my logic is right, the BMS should then charge further, and if it reaches 3.6 V at say 140 Ah, it would record 140 as FCC (Full Charge Capacity).
I didn't set any capacity manually in the BMS settings at all. It did this all by itself. But you could be right. I'll test this out at some stage.
Good idea!
Hello, I can't get a connection from the Pc (BatteryMonitor V2.1.9 )) to the Seplos BMS (Model 48150-1101-10C-SP15-16S) . Connection is always "Time out". What am I doing wrong ? Thanks ( Wenn möglich auf Deutsch dann ist es einfacher für mich)
Do you know how to connect this battery to the Seplos website?
I have those batteries also. When I charged them they alwas stopped around 96% more or less. I Always let a second charger connect connect to them and never worried myself about the 100% percent charged the way you do. Than I come to check them they were all 100 % charged. But the inverter ever really charged them at 100 % only the other Aims charger did that job. All you have to do put a 15 Amp charger connect to it and the charge them to 100% without you knowing it.
Seplos recommend to charge the battery pack with 100A and discharge 100A to the bottom or when the BMS switches it off. Then charge again at 100A to the top to reach 100% or the Ah rating of the cell modules is reached. That is what I was recommend by their tech team.
Is the JK BMS the best so far?
Thanks Andy! Nice one!
My 200A Seplos BMS with approx. 15 cycles keeps my 320Ah cells perfectly balanced. It is connected and managed by the Sofar inverter where I set the discharge depth to 90%.
Great. I didn't do an initial top balancing to test the BMS.
As it is now top balanced, we will see how it performs the next time we fully charge it. I don't even know if the balancer works...
@@OffGridGarageAustralia And this is possibly one of the problems. No top balancing done… Another problem could be to stress the cells with charging it almost to 100%. This could be hard to manage for many BMSs, that have low balancing current.
They should sell it as a 130Ahr battery! A passive balancer of 150mA is just a joke for such a big battery.
Great video! Too bad that there is a problem with Temp1. I have 3 Seplos BMS 16s200A but with original sensors. 2 are currently in operation. No problems with the sensors. In my opinion, the BMS is foolproof! So suitable for me. I'am not need a extra balancer , Made TopBalancing before...
Yes, if you do a initial top balance, this would not occur. I wanted to test 😉
Yeah, the settings are actually quite good for a pre-set BMS and will work out of the box right away without the need to change anything.
Woascheinlich is es trotzdem sinnvoll, an aktiven balancer einzubaun, folls des päck amol über de joa aus der balance kummt. Sonst muas man olls wieda aussanond baun....
@@OffGridGarageAustralia I have used LFP batteries. That's why I only charge and discharge the cells between 20% and 90%.
Your 130ah batteries are definitely better than mine. At 2.9V I have to stop discharging mine otherwise the delta will be too big.
Between 20% and 90% I usually have 2-10mv delta.
Yes , it does not have an active balancer .
If a cell breaks out in the 20-90 range despite top balancing, I'll swap it out - Cell6 🤬.
But I don't think it's a bad idea for maintenance purposes.
Maybe I'll buy one(active balancer) :-), it's your fault!🙃
The intermittent power supply waiting is a great function, really great function. It protect your inverter from damage. When single cells reach high voltage protect, then this function working. Only when soc less than 96%, then can charge the battery again. Then why need this function? As battery is communicate with victron inverter through CANbus, when charging is cutoff as full charging, then this current will go back to inverter( has chance to damage inverter). So we can not allow this things always happerened. That is why need this function. When battery is full charging, then stop charging the battery. Otherwise you will find that the battery is always charging, stop charging and charging, stop charging….. and this current always back to inverter. So with intermittent power supply waiting, only when soc less than 96% start charging again, it helps a lot with inverter.
And about the Clamp ammeter, LCD and Software current shows is different, it is not a problem. As you working with inverter, the discharging current is sine wave. The software is update current every second, so it changed every second. The software need exact current number, so when problem, easy solved. But LCD is shows to all clients, can not let clients see the sine wave current and changed very quick, so it will shows a average value per 8 second. But clamp ammeter is more slow update. So you will find the clamp ammeter is always like 25A. If you working with stable discharging load, then you will find the problem solved.
About the battery really capacity, you need do one cycle to help the bms to learn the battery really capacity. Full charging the battery to single cells high voltage protect or total voltage high voltage protect, then discharging the battery to single cells low voltage protect or total voltage low voltage protect, then the BMS will get the battery really capacity( this is a very important step when doing aging test). But as it is home energy storage system, so it is hard to always reach this situation, but still have chance to reach it, when reach the situation, then BMS will learn the battery really capacity again. The SOH is depends on the really capacity decrease. For example a 100ah really capacity cells, our warranty is 3000 cycles at 80% DOD. So after 3000 cycle, the soh start working. It remains 80ah, then soh is 100%, if 60ah, then soh is 60/80=75%.
What is the activ balancer where can I bay it?
Are the issues solved nowadays? (intermittent power supply waiting)
Well, where do I start...
Seplos quoted a very high price for this kit. It has a higher $/Ah than the available rack packs in N.A.
Shipping is not cheap unfortunately.
@@OffGridGarageAustralia base kit without shipping was still a far bit higher.
Das hilft mir sehr! Ich werde mir noch einen aktiven Balancer bestellen! Ich habe meine Zellen jetzt alle auf 3,65v mit 0,001A geladen, von daher habe ich wahrscheinlich bessere Voraussetzungen! Aber schaden kann der balancer nicht, wenn man ihn zum top-balanicing verwendet! 👍
Anybody knows where can I get a Seplos BMS in the UK, US or Europe without having to import it?
Very interesting, thanks for your extensive tests!
I guess if you don't need the display function you could probably ditch the BMS and throw a JK in there with custom cables instead of the supplies bus bars. Maybe the JK display even fits the cutout...
But then a JK is more expensive than a smart balancer so that's probably the way to go.
Isn't there a way to use a RS485 to wifi module and remotely controller the seplos BMS ? hmm
Why are you disabling the active BMS for daily use? Why not use the active balancer and disable the seplos balancing feature?
The active balancer should only be used when the voltage are 3.45V/cell or higher with LifePo4. Otherwise it will do the opposite and unbalance your battery pack.
@@OffGridGarageAustralia sure, this is obvious, but it can be used/ installed permanent
@@patrickmartin4482 With some sort of additional switch, yes.
@@OffGridGarageAustralia just a "start voltage" and daily use is not recommended? Ordered yesterday the neey
Hi Andy I am a distributor for a Lithium Battery that uses the seplos bms.
They set it to 80% dod. I said I need the soc to go down to 15% to lower my electric bill to almost 0.
by doing this I reduce the battery life by 5% said the China Engineer.
What can you recommend that I discharge my batteries to?
with a 20% soc they claim I will get a 6000 cycle life for these CATL 15s 100Ah battery
Thanks to your channel I also bought different BMS. and I have them in the closet now. No smart Daly BMS, ANT BMS 0,2a balancr, JBD relay BMS 0,15a balancer. all scrap. now I have two JK 2A... BMS in use and this is really the BMS you want. and this milliampere balancing can be done with 18650 cells.
ps. I made a copy of your battery shelf. but without the mega bus bars.
Grüße Alex
Another ‚disrespectful‘ battery test - thank‘s a lot. I‘m looking forward to see what’s wrong with the temperature sensor, or is it the BMS itself, that does not take the correct measurement?
Heya, oke a good battery pack except for 5wh less but maybe after a couple of charge and discharge cycle it will even get the extra 5 wh. still intressed in more info about the bms but I gues that will come soon
So their program is just a modbus reader.
You could hook this up with esphome or whatever monitoring software you want.
To be honest, don't bother trying to use their software. There are so many great modbus programs.
Thank for the info. That is helpful...
Are the modus registers for Seplos documented somewere ?
@@DrCoolZic look for the seplos document "Battery Management System (BMS) for Communication MODBUS-ASCII Communication Protocol"
EXCELLENT EXCELLENT Evaluation. Thank You.. CCHHEERRSSSSS..... Your Mascot Cat has grown blue whiskers. 36:22😊
Thanks a lot. Yes, hahaha. it really looks like it... 😹
What is the best bms for a 14s 5p Lion pack (E-Bike diy battery)?
When it's in that "power supply" warning state, is it still balancing or does it only balance while it is accepting charge current?
Great question, I really don't know. The cell voltages decreased a bit during that time but it could be the voltage is slowly settling anyway without charger. Could be due to the balancer. Could be both... 🤷♂️
andy , the "runner" cell, isnt that one you tested, and thus may have caused the imbalance yourself ?
i must confess i topbalance my cells before i put them in my kit, but i am not seeing must deviation.
furthermore , when you connect the seplos to you victron setup, charging amps will be lowered at the top of the curve, preventing cell imbalances
Yes, 13, 14, 15 and 16 were the cells I tested and as I said in the video, this massive imbalance was home made. I wanted to see if how the BMS handles that... Usually that would not occur and you would have done a proper top balancing with all cells in parallel before installing them in the case.
@@OffGridGarageAustralia Would you though? They seem to indicate that you install the cells as they arrive, no mention of balancing beforehand. If they come roughly balanced at mid SoC, maybe the balancer can handle that? That would have been good to see.
Want to discharge the cells to 50% SoC and reset the BMS and test the theory? 😉
@@OffGridGarageAustralia sorry andy, but kind of feel like you setup the bms to fail, especially since you knew it was passive balancing only.
i hope you'll still give it a fair chance in your 280ah kit , and you'll do a over time review.
i do agree seplos needs to start looking into adding active balancing into their bms, as voltage difference will become bigger as cells age
@@houseofancients it was part of the test. I knew about the small passive balancing the BMS has and we have seen this same result with many other BMS's in this situation before.
When we build the large MASON system, I'll do a full top balancing of the batteries before we assemble it as one would do.
I still believe, the BMS will struggle with even a very small deviation in these large capacity cells.