Aw shucks, thanks for the mention. I like to leave the mistakes in my videos and openly discuss them because it reassures people that they're not the only ones who make mistakes.
@@pickholder6189 , every time I let the smoke out of an electronic device because of some stupid mistake I made, I have to mentally remind myself of all the previous times that I said to myself, " I gotta remember to not ever do that again!"
Ah, the paper clip test! Many years ago I was a video engineer in the BBC studio shooting an episode of Dr Who (Tom Baker!). Our cameras were EMI 2001s which had 4 pickup tubes which had a 525V DC supply fed down the camera cable. There were 4 seperate supplies, one for each tube. About 20 minutes into the recording session in the middle of a tack, one camera suddenly lost one of the colours and then went off. I went to check the power supply in the rack, and sure enough it had tripped it's breaker. To cut a long story short, I went onto the studio floor while a colleague Eric stayed in the equipment room. I unplugged the camera from the wall socket, and plugged it into a different socket while Eric changed the PSU. We blew the breaker on the spare! I looked at the cable and found that it had arced over. This was a 101 way cable! We changed the cable put everything back to the way it started and reset breakers. Again the PSU breaker tripped. We found that both wall sockets had arced over. After changing the camera head, the cables and using a new wall socket we had things working. When the recording session was finished I went and got the camera head moved up to the studio workshop so we could check it out. The problem was that a production person had put a script on top of the camera, and when they took it off the paperclip had come off, fallen into a ventilation slot and shorted out one of the 525V supplies! It took me the best part of a day to change out components in the camera head that had been blown. Eric had a much easier time with the PSUs as only some fuses had been blown. The cable ends and wall sockets were a much bigger problem as they were all potted and had to be broken open. It took a couple of wiremen several days to replace the wall sockets and the cables got sent back to the manufacturer. Those were the days! Just for the record in the early 70s a single EMI 2001 camera system (camera head, PSU, Control Unit and monitoring) cost the best part of £100K! About $200K. We had 5 plus a spare, not to mention the video and audio control and distribution kit plus about 4MW of lighting. At the time I owned a 3 bed house which had cost me £4,250. How times have changed. Where I live in SW London, a 3 bed house would cost at least £400K, and I could by a good Pro-sumer 4K video camera with lots of bells and whistles for £5K. At the time I took all the studio gear I was working with for granted. Great video Jeri. Nice to see you back.
Sadiq Mohamed, the moral of that story is that there is no end to the amount of trouble in the world caused by paper-pushers! I liked the Tom Baker episodes of Doctor Who best of all. And I do really miss Sarah. I never really warmed up to the actor who played the doctor after Tom Baker, but I sure liked Nyssa and Tegan wasn't half-bad either! Tom Baker was also excellent as Rasputin in the miniseries about Nicholas and Alexandra the tsar and czarina of Russia.
Jeri, I first saw your Commodore based bass years ago and only recently dedicated the time to research your accomplishments and watch some of your video. I hope It's not too far to tell you that you are brilliant. You are the type of strong leader that current generations desperately need. I wish you the very best in life and I hope I get the opportunity to show my son new videos of yours, someday.
In 1974 Radio Australia incorrectly connected the phases of their new 3 Mega watt standby generator to the Grid, during the test the generator picked up the engine and it all spun around like a 10 tonne top destroying the building except the control room. Miraculously no one was injured and the whole thing was kept secret from the public.
We all make mistakes. I believe it even was you who told us in an old video that "to learn and succeed you need to make *Lots* of mistakes". And tbh you just showed me protections that are in devices all around me, in clear human-speak even.
+FroggyMosh -- I keep a running tally of published errors by Nobel prize winning scientists (Including two for Pauling, since he got two Nobels) for times like these
I've heard the 'Explody diode' called a crowbar circuit before. I've used that with a PTC fuse with good results. A P-channel FET is easier in some cases, though. I do a lot of work on high-end programmable LED hula hoops, which have some interesting design constraints. Everything has to be small and tolerant of board flex, but for practical reasons (like being able to assemble boards on our little in-house pick and place machine and to be able to repair things by hand) component count ends up often being a bigger concern than cost.
Hello. "Explody diode" at 3:44 is well-known protection method however "PMOS diode protection" at 12:00 is new and good method that you disclose. Thanks for your good explanation on the video.
It makes me really happy to see that you're posting again Jeri. I've always enjoyed your videos for your strong and unique character as well as the insightful curiosity of your content. Welcome back and thank you.
I just got finished listening to your story, (restarting the video),for the second time. I'm amazed by all you have gone through. Your an amazing woman!
Jeri, you're a real inspiration to people. We all make mistakes, don't worry about it. You can even just edit the videos into something that is "perfect". You don't have to get it all perfect in one take, just do what makes you happy :)
Even though Big Clive and Dave do teardowns, there's a serious difference in how valuable the things they tear down are. Big Clive : I got this novelty light from Poundland for £1 and I'm tearing it down EEVBlog : My neighbours threw away another $100,000 workstation. I'll tear it down, replace a few capacitors and get it working.
EEVBlog guy gives me a headache. I have to be in the right mood. Definitely has better devices. Julian , sometimes he does amazing stuff. No headaches either. None of the Nerd guys have talked about him. At least that I've noticed, and some have had a list of about 15. He should be on the top of every list. Big Clive got his format from Julian. ruclips.net/channel/UCmHvGf00GDuPYG9DZqQKd9A
@@brandonlewis2599 Would you be talking about Distillation ? I ask because a sailboat channel I have been watching for yeers , originally because they got tired of $100 dollars for a $10 doallar bottle In some of the places they went to, started distilling. The started with the announcement, then, the still off later. .ruclips.net/video/l4bAJAh9r0A/видео.html At one time we had this still for sale in Redding Ca ! If you google you find a lot of copper ones. All very cool. I'd probably make brandy. I have lots of grapes and 8 gallons of wine needing to be bottled right now ! Cool to run into you dude lol !
Filmed in South Africa I believe. At least they picked up one of the girls there, and the other guy is a visitor on the boat. The guy without the beard.
Hello Jeri. I came across this clip while searching for OC protection on my breadboard+PSU combo unit. I have now started watching some of your 2009 clips, very retro for me. Thanks for the calm and precise delivery. Best regards Mark (UK)
Thanks for coming back! So happy you're making videos again. Your channel was making me so sad, because all your old videos were super cool. But seeing a bunch of new videos in the last few weeks, it's really a great surprise!
Thanks Jeri .Your videos are EASY to understand.I was a Part-time Engineer at K-40 Electronics.I was one of the 7 Engineers.I was not that bright but,seeing my neighbor's Lamborghini(the older C.H.I.P.S. radar),with my product in it made me call my old Boss,at K-40 .Being Autistic is a challenge(as well as the Parkinson's) but,I LOVED electronics. Here's a KOOL note ;; The 1st girl that I,ever,kissed listens to my Global Station,1000 Miles from my the Studio(Panbeara-KFUY)!!! One of these day,I'll be able to afford a Kenwood,like the one on your bench.Thanks again for your videos. This is KE0DNQ.......73
You have a Robie! I had one when I was a kid. I took him apart and springs and gears went flying everywhere. I was sweating bullets for over an hour trying to get him put back together. But I eventually got him back to working. And I learned a lot about mechanical things. Fun!
I always add polarity protection into my designs if there is a chance that a user will connect polarity incorrectly, but I usually don't bother with resetting current protection. My thought is that if overcurrent occurs, it's almost certainly due to a downstream fault, and therefore a resetting current limit doesn't save the product from needing repair. The original fault will still exist. So I generally fit fuses that are designed to blow to prevent the fault progressing to a fire or serious damage occurring. The exception to this is if there are external ports, e.g. USB ports. In which case, for those, I generally use current-limited load switches. But the DC input is still only protected by a fuse.
I have dropped a scope probe onto a circuit, and shorted it to destruction. HDMI cables are excellent for this too. I have never placed a PCB onto loose change on the desk. I have to try it! Thanks for the suggestion.
You have to be my soul mate. After watching your interview im blown away. Your so smart it makes my heart flutter. I could listen to you all day. Earlier today i came across your video for making an x ray scanner with mouse. Helped me figure out what 5g is. So thank you for that. Its a giant xray grid. Probably the all seeing eye, but you made my day even after that.
Why have I not come across your channel before? That was a beautifully presented summary of low power reverse polarity protection options! Definitely gonna look at more of your content... Thank you!
Haha, the SW-engineer situation is so spot on. A few months ago I was assigned to build a cubernetis demo cluster. Within the hour of the hardware arriving; Delivery packet wasn't assigned in my name unfortuanately so my lovely fellow SW- colleagues had put all of the 10+ rpi's in a pile (yes, like a big mikado-styled heap! Cables, PCBs flash drives and routers all laying ontop of another)... And ofcourse they powered them up and tried to setup the server... I was dumbfounded, these are after all MSc CS people... Pie's arn't that fancy nor expensive, but it was so unnecessary and leadtime can't be replaced.
Very happy to see you producing videos again! Your topics always teach me something! Although I don’t always understand everything, I deeply appreciate the pure genius behind it! Keep up the great work! 73 de Seth - W8FG
Jeri, I love what you do. Your way of understanding and explaining electronics is just brilliant. I guess that's because you figured most of this out by yourself, as opposed to "being taught". I hope you will inspire lots of girls and boys to play around with electronics - not as a career, but as a way of life. As a father of three girls I know how badly we need inspiring personalities like you.
I’ve used almost that circuit in automotive engine controllers. The difference is that I arrange the low side resistor and voltage clamp so if the engine controller is in standby, the circuit is shut off, reducing quiescent current. In this mode the body diode supplies power to just the voltage regulator and it has OVP via a 8A27 TVS. The battery voltage usually makes the gate zener conduct. Because of this, I had to split the voltage limiter. It supplies a limited to 35V gated power feed to all other circuits, the OV clamp is powered between the two MOSFETs, and the gates are connected together so they share one zener, and the pull down and OVP have a single N mosfet with Vdd going to the gate. The junction of the two MOSFETs have the TVS that clamps at 45V (with about 100A load dump) to ground and that feeds the voltage regulator with a standby mode. I use a 20P06L FET, it can handle most load dump surges (verify with testing). The other MOSFET with a voltage limiter clamps to 35 volts and feeds the H bridges and other circuits that blow up over 40 volts, and turns off after the ECM goes to sleep, further reducing quiescent current. Only one more part needed from your component count, one small signal 60V logic level MOSFET. Note that to deal with reverse spikes, the gate pull down impedance needs to be low enough to discharge the gate in a couple microseconds so maybe 5-10k or so which is a fairly high load. Does that all make sense? With a battery, gating the pull down can save battery life. In my case, it’s battery life of a 60Ah, 12V lead acid battery, but at a parking lot for an extended couple month vacation. Those mAh add up. BTW plugging into an automotive 12V (or 24V) power supply bus is plugging in to a world from hell. Love your channel!
This video is proof why everybody is so glad you are back!!! So are you telling us to go to big Clive to see the exploding diode? Lastly, are you telling us this is not a perfect world so design as such? Great video - keep it up!!!
Great video, I'm glad I've found your channel. Incidentally I once worked with someone who showed me that the protection circuit could be built "the other way up" using an n channel device. The advantage of this was the slightly better performance of the FET. The problem is of course that the positive rail runs from battery to load and the FET is in the negative rail, so battery negative isn't at ground. This really only matters if the battery negative is run outside the equipment for anything other than an isolated charger. Incidentally I'm now usually the embedded software geezer and I DON'T leave loose change on the bench. :=*)
I am delighted you are back posting again Jeri. I have really enjoyed all your videos. You bring an interesting perspective to our amazing hobby by thinking outside the box too. 73 from MI6GNP and Merry Christmas from Northern Ireland with best wishes for a great 2019.
Nice! I have also used a single n channel mosfet in the negative input leg. N channel mosfets generally have a lower on resistance than p channel. I had never thought of your circuit for current limiting also. Thanks for that additional insight!
This is really good for me, I'm doing a Ben Eater 6502 build, and the other day I tied power to both VCC and VSS on the processor. I was sure I'd cooked it. I hadn't, but circuit suggestions like these could save me some sweat and tears, down the road, particularly the reverse polarity part.
I made something similar but used the Rdson as my shunt. Yes i know that the Rdson isn't super accurate and will change vs current, but often a limiter is just being used to make sure nothing really bad happens. I think you can even combine the two fets. Of course there are dedicated IC's doing all of this in a small package.
Another approach I learned about recently is TI's LM74610-Q1 "smart diode" controller, which uses an N-channel MOSFET as the pass device, which I believe has some advantages in terms of efficiency. Interestingly, it does not use a ground reference, so there's no quiescent current in the device.
No resistor across the E-C of the PNP is really needed. When I do this, I usually put a series resistor in the base leg of the PNP. If the input is from something like a car battery, the MOSFET may pass enough briefly to make the drop in the series resistor big. You can also put the sense resistor between the pair of MOSFETs.
The resistor across the pnp transistor is performing the same function as the zener diode in the simpler circuit. It prevents the pmosfet gate from being pulled so low that it exceeds the maximum Vgs for the device. Of course, it's not 50Ω as claimed, but (for example) 50K as a resistive divider with the 10K resistor to ground would keep a 12V supply from pulling the gate below 2V, thus keeping Vgs to 10V maximum. You'd obviously have to recalculate for a different supply voltage and maximum Vgs, and clearly it's only needed if Vsupply > Vgs(max).
The simplest reverse polarity protection is a Schottky diode. The drop is below 400mV. Your PMOS solution works fine. It‘s also possible to use a NMOS solution putting the protection circuit into the GND line. NMOS On-resistance is lower at same cost as a PMOS.
Hi, nice video! "Final circuit": I've done this many times, but always put the reverse polarity first, complete with a 2 way tranzorb and used a PTC for "Shunt" that I call "Sense" for a better fold-back function. This type of circuit is very helpful for fault condition coverage of circuits that go into hazardous environments.
Excellent circuit to protect against end user mistakes ....now, can you figure out how to keep the purchasing dept. from buying the cheapest components in their mis-guided efforts to save pennies while disregarding life endangering risk ? Retorical question ...this is a question without an answer. Glad you came back to youtube ...and that you're in sync with the best; Big Clive and EEblog.
If you're doing a non battery powered device, a series Schottky diode should work just fine. I'm designing an automotive device at the moment and I find Schottky diodes to be the most cost effective solution, especially since they can withstand all the weird inductive pulses in the vehicle's electrical installation.
@@mikeguitar9769 Even in the mA range, thr leakage current for a Schottky diode is too low to do any harm in case of a reverse poalrity event. Their big downside is the power dissipated through the voltage drop.
I use series diodes when input voltages is high ( >= 9V ) and reverse diodes and fuses for low voltages. If the voltage drop is important, I use Schottky diode . For over current protection, usually I use a fuse or reversible fuse (resettable fuse).
The big advantage of the explodey diode is that it can be added even though you forgot all about reverse polarity protection until the end of the project. Also, not so explodey if it is only likely to connect to a power source that can only deliver a few amps.
You might also consider the use of a re-settable fuse. The Bell Fuse Inc 0zcj0005ff2e (surface mount) has a hold current rating of 50MA and a trip rating of 150MA, 3.6 ohms cold, 50 oms when tripped, cost in quantity is 5.6 cents. Many varieties are available.
I have a vintage 1979 Chess Challenger 7 portable computer that I wanted to protect from reverse polarity (it's using a 9VDC small external psu). I simply put a diode bridge between the power input jack and the device inside. That way, it works no matter what the polarity is. Very simple to do: a diode bridge and a few wires. Cost: a few cents. The device already has a 7805 voltage regulator inside, so over-voltage is not an issue.
It's interesting that this mix of 3 transistors ends up being a good design. The BJT makes a better voltage reference compared to the FET (lower voltage and lower tolerance). The FET's insulated gate makes a more efficient pass transistor, even though 2 are needed to get bipolar switching.
If manufacturers decided that they cared enough to make their products somewhat closer to foolproof, maybe there would be an incentive for the semiconductor manufacturers to build circuits like this into one all-encompassing, easy to install and design with, package. But I wouldn't hold my breath.
Very good video! Thank you. I like to see a well made discrete design instead of picking a fancy chip to solve all problems whithout knowing what's going on in detail. Please more....
Oh to have had something like this on an expensive wireless microphone at my old work. It had the same barrel jack as a PoE switch that was part of the same kit of equipment. It didn't like having 48v AC shoved in where 12v DC was expected. The series diode wasn't much help. Pretty much every semiconductor on the board was toast...
Coolest mistake I remember making was mistaking an LM335 comparator for a Darlington array. Er, the light was a bit dim at the time, the Darlington array was being used to switch on a capacitor discharge thing to switch some points, but, er, well, the socket I plugged the wrong thing into had 8 pins. The fascinating thing was, instead of the magic smoke just quickly puffing out, or lazily curling out, no, this was way more impressive - the Darlington array simply exploded into shards. Needless to say, I quickly switched the system off. No serious damage done. I have used servos to switch points since then - piano wire is very effective as the stuff that goes from the servo horn to the hole where the point is. Only other thing is, best to switch the servos off after using them, due to interference due to the locomotive motors.
Hey Jeri, Thanks heaps for sharing! Using the PMOS transistor in the linear region for current limiting is an application I'll have to keep in mind next time I need some cheap and cheerful protection from... less than carefully chosen peripherals.
Wait why does this video look exactly like franlab? Is there a common lady engineer camera angle thing going on? BRB, now imagining Dave Jones from this camera angle instead of making the short fella look big. Or Big Clive at this angle from the knees down only.
It's difficult to get my fat head into the shot and have the camera pointing at the bench unless its pointing down. I like making 0 edit videos if possible because its challenging.
So glad ur back on puting videos on youtube. I love listening to the interviews you give about your life and story such a cool story sure ur saying yeah it hasent been all cool :)
My new favorite lady! Subbed! Where have you been all my life? Your vids are helping me refresh my memory on electronics. You can't fire yourself...ever!
Nice video Jerry. Happy to see you again ! I love the calm way you explain and in contrary to the other videos you keep it to the point and short. Hope to see more videos of yours.
Good circuit. The pain always is the power dissipation in the current limiting fet. How about a bistable or maybe monostable one which does not have that problem ? Using only a few components ?
I really enjoy these videos! They’re also a great public service! Electronics for me is a hobby and these videos have really educated me and have been super entertaining. It’s also great that Jeri has been self taught and can present things in a way a self taught student can understand.
I really enjoyed this video. (I have some knowledge of electronics and almost zero hands on circuit building) Your video breaks down your project in a way I understand which inspires me to dig the breadboard out and visit the world of circuit construction again. I also like how you provide links to other cool videos of what other thunder nerds are up to. Thanks!!
Very useful circuit. I still tought the 'explody diode' was still the way to do it, but my knowledge has aged. I'll start experimenting with 'pmos diodes'.
As with all things that limit at some constant current - be sure your pass device can handle the required dissipation from full current at the limit against a dead short at your device end.
I'm hoping for a part two where you discuss the considerations for running the circuit routinely in limiting mode and best practices for making the current limiting adjustable.
I'd like to see devices like these with an integral two or three color low current LED to tell the consumer at a glance if something is wrong with the way the device is being powered.
For myself I don't think I'm doing odd enough things to make a realistic chance to get to discoveries. I just prefer to do a good job using well known approaches.
@@SE45CX I encounter about once a month or so a situation where a well-known procedure just refuses to work right, or where a seemingly minor change in a parameters has great consequences. Have been a CNC machinist in a job shop for about a year, since quitting a dead-end career in IT...
I always try out a simple circuit using the free LTSpice circuit simulator first to see how it works then go to the bench to verify...I'm lazy /;>) P.S. The models are accurate.
This looks good! I think I would use a poly-fuse instead of the current-limiter circuit-- and if you are REALLY feeling like you need to go cheap-- just a Christmas tree flasher lamp (the kind with a built-in bi-metal switch that shorts the lamp) used in series. The flasher lamp will have very low impedance until there is enough current to turn it off (and then it will cycle). If using the poly-fuse or flasher-lamp, you could also use a back-connected uni-directional transorb instead of the PFET. This has the advantage of also protecting against over-voltage (up to the rating of the poly-fuse or flasher-lamp). Obviously, it would take some experimentation to find the right flasher-lamp-- and it's current limit would be slightly temperature dependent-- but if you are going for cheap... Fun stuff! Endless possibilities!
A motor is another interesting current limiter, especially for AC. A large single phase induction motor without a start capacitor approximates a short circuit at standstill, but draws comparatively little current once moving.
Another method is an over current latch circuit, that hard cuts the output upon seeing a fault condition until reset. Better for higher current loads where linear regulation would prove infeasible. Maybe you could just bond a thermistor to the FET of the linear circuit, producing a positive feedback loop, though using an actual comparator latch circuit is probably better for precision.
This was the first I'd seen of you Jeri. In watching others too I'm like, wow this girl has some serious equipment, is this for real? .... After google, i see, of course, makes sense. Protection circuits will be NB for those mass market games! Luck! And thanks for sharing.
![I.N "HALLUCINATION" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/n5B5q1Hwt_U/mqdefault.jpg)
Aw shucks, thanks for the mention. I like to leave the mistakes in my videos and openly discuss them because it reassures people that they're not the only ones who make mistakes.
Mistakes are how we all learn. Today I put a 450v capacitor the incorrect way in a valve amp.... been doing these amps for YEARS !
It's always nice to see the mistakes. A good opportunity to demonstrate how to not panic.
I'm sure that she's your Favourite shunt. Anyone tell her about the C64mini?
@@pickholder6189 , every time I let the smoke out of an electronic device because of some stupid mistake I made, I have to mentally remind myself of all the previous times that I said to myself, " I gotta remember to not ever do that again!"
You rock Clive...
Great to see you back making videos Jeri, and thanks for the shoutout!
"Bad systems are designed to work properly Good systems are designed to go wrong properly." - Keep up the good work Jeri.
who said this? It's a gem.
@@physiqueDrummond That's one of my own - feel free to steal it.
😄
We don't care if you messed up. All we care about is that you are back on RUclips.
Ah, the paper clip test! Many years ago I was a video engineer in the BBC studio shooting an episode of Dr Who (Tom Baker!). Our cameras were EMI 2001s which had 4 pickup tubes which had a 525V DC supply fed down the camera cable. There were 4 seperate supplies, one for each tube. About 20 minutes into the recording session in the middle of a tack, one camera suddenly lost one of the colours and then went off. I went to check the power supply in the rack, and sure enough it had tripped it's breaker. To cut a long story short, I went onto the studio floor while a colleague Eric stayed in the equipment room. I unplugged the camera from the wall socket, and plugged it into a different socket while Eric changed the PSU. We blew the breaker on the spare! I looked at the cable and found that it had arced over. This was a 101 way cable! We changed the cable put everything back to the way it started and reset breakers. Again the PSU breaker tripped. We found that both wall sockets had arced over. After changing the camera head, the cables and using a new wall socket we had things working. When the recording session was finished I went and got the camera head moved up to the studio workshop so we could check it out.
The problem was that a production person had put a script on top of the camera, and when they took it off the paperclip had come off, fallen into a ventilation slot and shorted out one of the 525V supplies! It took me the best part of a day to change out components in the camera head that had been blown. Eric had a much easier time with the PSUs as only some fuses had been blown. The cable ends and wall sockets were a much bigger problem as they were all potted and had to be broken open. It took a couple of wiremen several days to replace the wall sockets and the cables got sent back to the manufacturer. Those were the days! Just for the record in the early 70s a single EMI 2001 camera system (camera head, PSU, Control Unit and monitoring) cost the best part of £100K! About $200K. We had 5 plus a spare, not to mention the video and audio control and distribution kit plus about 4MW of lighting. At the time I owned a 3 bed house which had cost me £4,250. How times have changed. Where I live in SW London, a 3 bed house would cost at least £400K, and I could by a good Pro-sumer 4K video camera with lots of bells and whistles for £5K. At the time I took all the studio gear I was working with for granted.
Great video Jeri. Nice to see you back.
For comparison for a foreigner, what was the average salary back then compared to now in London ?
Great story! Thank you for sharing it.
I don't believe a word of it. If you were working with Dr Who you could just fly back to the past and prevent the paper clip from falling in. ;-)
I remember those cameras, they had just about reached end of life when I joined the TV business.
Sadiq Mohamed, the moral of that story is that there is no end to the amount of trouble in the world caused by paper-pushers!
I liked the Tom Baker episodes of Doctor Who best of all. And I do really miss Sarah. I never really warmed up to the actor who played the doctor after Tom Baker, but I sure liked Nyssa and Tegan wasn't half-bad either! Tom Baker was also excellent as Rasputin in the miniseries about Nicholas and Alexandra the tsar and czarina of Russia.
I am so glad you are back posting videos Jeri !
Jeri, I first saw your Commodore based bass years ago and only recently dedicated the time to research your accomplishments and watch some of your video. I hope It's not too far to tell you that you are brilliant. You are the type of strong leader that current generations desperately need. I wish you the very best in life and I hope I get the opportunity to show my son new videos of yours, someday.
In 1974 Radio Australia incorrectly connected the phases of their new 3 Mega watt standby generator to the Grid, during the test the generator picked up the engine and it all spun around like a 10 tonne top destroying the building except the control room. Miraculously no one was injured and the whole thing was kept secret from the public.
We all make mistakes.
I believe it even was you who told us in an old video that "to learn and succeed you need to make *Lots* of mistakes".
And tbh you just showed me protections that are in devices all around me, in clear human-speak even.
+FroggyMosh -- I keep a running tally of published errors by Nobel prize winning scientists
(Including two for Pauling, since he got two Nobels) for times like these
I've heard the 'Explody diode' called a crowbar circuit before. I've used that with a PTC fuse with good results. A P-channel FET is easier in some cases, though.
I do a lot of work on high-end programmable LED hula hoops, which have some interesting design constraints. Everything has to be small and tolerant of board flex, but for practical reasons (like being able to assemble boards on our little in-house pick and place machine and to be able to repair things by hand) component count ends up often being a bigger concern than cost.
Hello. "Explody diode" at 3:44 is well-known protection method however "PMOS diode protection" at 12:00 is new and good method that you disclose. Thanks for your good explanation on the video.
How am I only NOW discovering a) this channel, and b) nerd thunder?
The reverse diode across the supply with a fuse is the DC version of the "crowbar protection circuit" replacing the SCR with a diode.
When I read the title of the video I immediately thought about Big Clive. His teardown and reverse engineering videos are golden.
It makes me really happy to see that you're posting again Jeri. I've always enjoyed your videos for your strong and unique character as well as the insightful curiosity of your content. Welcome back and thank you.
I just got finished listening to your story, (restarting the video),for the second time. I'm amazed by all you have gone through.
Your an amazing woman!
Jeri, you're a real inspiration to people. We all make mistakes, don't worry about it. You can even just edit the videos into something that is "perfect". You don't have to get it all perfect in one take, just do what makes you happy :)
Even though Big Clive and Dave do teardowns, there's a serious difference in how valuable the things they tear down are.
Big Clive : I got this novelty light from Poundland for £1 and I'm tearing it down
EEVBlog : My neighbours threw away another $100,000 workstation. I'll tear it down, replace a few capacitors and get it working.
EEVBlog guy gives me a headache. I have to be in the right mood. Definitely has better devices. Julian , sometimes he does amazing stuff. No headaches either. None of the Nerd guys have talked about him. At least that I've noticed, and some have had a list of about 15. He should be on the top of every list. Big Clive got his format from Julian. ruclips.net/channel/UCmHvGf00GDuPYG9DZqQKd9A
@@brandonlewis2599 Would you be talking about Distillation ? I ask because a sailboat channel I have been watching for yeers , originally because they got tired of $100 dollars for a $10 doallar bottle In some of the places they went to, started distilling. The started with the announcement, then, the still off later. .ruclips.net/video/l4bAJAh9r0A/видео.html At one time we had this still for sale in Redding Ca ! If you google you find a lot of copper ones. All very cool. I'd probably make brandy. I have lots of grapes and 8 gallons of wine needing to be bottled right now ! Cool to run into you dude lol !
Filmed in South Africa I believe. At least they picked up one of the girls there, and the other guy is a visitor on the boat. The guy without the beard.
Jeri, darling, every one makes mistakes. You admitted yours and corrected it. Thanks, for the video and the honesty.
Hello Jeri. I came across this clip while searching for OC protection on my breadboard+PSU combo unit. I have now started watching some of your 2009 clips, very retro for me. Thanks for the calm and precise delivery. Best regards Mark (UK)
I'm so glad that she is back, always thought her videos where among the most informative electronics videos on RUclips
Thanks for coming back! So happy you're making videos again. Your channel was making me so sad, because all your old videos were super cool. But seeing a bunch of new videos in the last few weeks, it's really a great surprise!
Thanks Jeri .Your videos are EASY to understand.I was a Part-time Engineer at K-40 Electronics.I was one of the 7 Engineers.I was not that bright but,seeing my neighbor's Lamborghini(the older C.H.I.P.S. radar),with my product in it made me call my old Boss,at K-40 .Being Autistic is a challenge(as well as the Parkinson's) but,I LOVED electronics. Here's a KOOL note ;; The 1st girl that I,ever,kissed listens to my Global Station,1000 Miles from my the Studio(Panbeara-KFUY)!!! One of these day,I'll be able to afford a Kenwood,like the one on your bench.Thanks again for your videos. This is KE0DNQ.......73
You have a Robie! I had one when I was a kid. I took him apart and springs and gears went flying everywhere. I was sweating bullets for over an hour trying to get him put back together. But I eventually got him back to working. And I learned a lot about mechanical things. Fun!
Ahh, the halcyon days of Radio Shack & my youth...
I always add polarity protection into my designs if there is a chance that a user will connect polarity incorrectly, but I usually don't bother with resetting current protection.
My thought is that if overcurrent occurs, it's almost certainly due to a downstream fault, and therefore a resetting current limit doesn't save the product from needing repair. The original fault will still exist. So I generally fit fuses that are designed to blow to prevent the fault progressing to a fire or serious damage occurring.
The exception to this is if there are external ports, e.g. USB ports. In which case, for those, I generally use current-limited load switches. But the DC input is still only protected by a fuse.
This person is a REAL electrical engineer.
I have dropped a scope probe onto a circuit, and shorted it to destruction.
HDMI cables are excellent for this too.
I have never placed a PCB onto loose change on the desk.
I have to try it!
Thanks for the suggestion.
N-channel, p-channel, who cares. The best channel is the Jeri-channel.
That was something haha.
Nince once, and btw, Jeri should use JFETs just because of "J" ;-P
Man who even cares about electronics jery channel is amazing 😅🤣🤣😂😂😂lol.
Has the best body...diode....
About that though. For NPN and PNP, what polarity is placed on the gates?
You have to be my soul mate. After watching your interview im blown away. Your so smart it makes my heart flutter. I could listen to you all day. Earlier today i came across your video for making an x ray scanner with mouse. Helped me figure out what 5g is. So thank you for that. Its a giant xray grid. Probably the all seeing eye, but you made my day even after that.
Why have I not come across your channel before? That was a beautifully presented summary of low power reverse polarity protection options! Definitely gonna look at more of your content... Thank you!
Haha, the SW-engineer situation is so spot on. A few months ago I was assigned to build a cubernetis demo cluster. Within the hour of the hardware arriving; Delivery packet wasn't assigned in my name unfortuanately so my lovely fellow SW- colleagues had put all of the 10+ rpi's in a pile (yes, like a big mikado-styled heap! Cables, PCBs flash drives and routers all laying ontop of another)... And ofcourse they powered them up and tried to setup the server... I was dumbfounded, these are after all MSc CS people... Pie's arn't that fancy nor expensive, but it was so unnecessary and leadtime can't be replaced.
Very happy to see you producing videos again! Your topics always teach me something! Although I don’t always understand everything, I deeply appreciate the pure genius behind it! Keep up the great work! 73 de Seth - W8FG
Jeri, I love what you do. Your way of understanding and explaining electronics is just brilliant. I guess that's because you figured most of this out by yourself, as opposed to "being taught".
I hope you will inspire lots of girls and boys to play around with electronics - not as a career, but as a way of life. As a father of three girls I know how badly we need inspiring personalities like you.
I’m so happy to see more uploads from you, my sister and I need more role models like yourself. :)
Check out angela white
I’ve used almost that circuit in automotive engine controllers. The difference is that I arrange the low side resistor and voltage clamp so if the engine controller is in standby, the circuit is shut off, reducing quiescent current. In this mode the body diode supplies power to just the voltage regulator and it has OVP via a 8A27 TVS. The battery voltage usually makes the gate zener conduct. Because of this, I had to split the voltage limiter. It supplies a limited to 35V gated power feed to all other circuits, the OV clamp is powered between the two MOSFETs, and the gates are connected together so they share one zener, and the pull down and OVP have a single N mosfet with Vdd going to the gate. The junction of the two MOSFETs have the TVS that clamps at 45V (with about 100A load dump) to ground and that feeds the voltage regulator with a standby mode. I use a 20P06L FET, it can handle most load dump surges (verify with testing). The other MOSFET with a voltage limiter clamps to 35 volts and feeds the H bridges and other circuits that blow up over 40 volts, and turns off after the ECM goes to sleep, further reducing quiescent current. Only one more part needed from your component count, one small signal 60V logic level MOSFET. Note that to deal with reverse spikes, the gate pull down impedance needs to be low enough to discharge the gate in a couple microseconds so maybe 5-10k or so which is a fairly high load. Does that all make sense? With a battery, gating the pull down can save battery life. In my case, it’s battery life of a 60Ah, 12V lead acid battery, but at a parking lot for an extended couple month vacation. Those mAh add up. BTW plugging into an automotive 12V (or 24V) power supply bus is plugging in to a world from hell. Love your channel!
This video is proof why everybody is so glad you are back!!! So are you telling us to go to big Clive to see the exploding diode? Lastly, are you telling us this is not a perfect world so design as such? Great video - keep it up!!!
Thank you.
I worked as a chemist.
I was able to survive my mistakes and those of others.
Great video.
Great video, I'm glad I've found your channel. Incidentally I once worked with someone who showed me that the protection circuit could be built "the other way up" using an n channel device. The advantage of this was the slightly better performance of the FET. The problem is of course that the positive rail runs from battery to load and the FET is in the negative rail, so battery negative isn't at ground. This really only matters if the battery negative is run outside the equipment for anything other than an isolated charger.
Incidentally I'm now usually the embedded software geezer and I DON'T leave loose change on the bench. :=*)
I am delighted you are back posting again Jeri. I have really enjoyed all your videos. You bring an interesting perspective to our amazing hobby by thinking outside the box too. 73 from MI6GNP and Merry Christmas from Northern Ireland with best wishes for a great 2019.
I love your current limiting pull up design. I’ve got it burned in to my head after rewinding the video 10 times.
It would be really nice to see a full screen-shot of the circuit as you are pointing out its features and the test setup. Thanks and great video.
Nice! I have also used a single n channel mosfet in the negative input leg. N channel mosfets generally have a lower on resistance than p channel. I had never thought of your circuit for current limiting also. Thanks for that additional insight!
This is really good for me, I'm doing a Ben Eater 6502 build, and the other day I tied power to both VCC and VSS on the processor. I was sure I'd cooked it. I hadn't, but circuit suggestions like these could save me some sweat and tears, down the road, particularly the reverse polarity part.
I made something similar but used the Rdson as my shunt. Yes i know that the Rdson isn't super accurate and will change vs current, but often a limiter is just being used to make sure nothing really bad happens.
I think you can even combine the two fets.
Of course there are dedicated IC's doing all of this in a small package.
There is foldback current limiting to limit dissipation on short cct. Also a thyristor can latch like a circuit breaker.
Another approach I learned about recently is TI's LM74610-Q1 "smart diode" controller, which uses an N-channel MOSFET as the pass device, which I believe has some advantages in terms of efficiency. Interestingly, it does not use a ground reference, so there's no quiescent current in the device.
No resistor across the E-C of the PNP is really needed.
When I do this, I usually put a series resistor in the base leg of the PNP. If the input is from something like a car battery, the MOSFET may pass enough briefly to make the drop in the series resistor big.
You can also put the sense resistor between the pair of MOSFETs.
The resistor across the pnp transistor is performing the same function as the zener diode in the simpler circuit. It prevents the pmosfet gate from being pulled so low that it exceeds the maximum Vgs for the device. Of course, it's not 50Ω as claimed, but (for example) 50K as a resistive divider with the 10K resistor to ground would keep a 12V supply from pulling the gate below 2V, thus keeping Vgs to 10V maximum.
You'd obviously have to recalculate for a different supply voltage and maximum Vgs, and clearly it's only needed if Vsupply > Vgs(max).
The simplest reverse polarity protection is a Schottky diode. The drop is below 400mV. Your PMOS solution works fine. It‘s also possible to use a NMOS solution putting the protection circuit into the GND line. NMOS On-resistance is lower at same cost as a PMOS.
Hi, nice video! "Final circuit": I've done this many times, but always put the reverse polarity first, complete with a 2 way tranzorb and used a PTC for "Shunt" that I call "Sense" for a better fold-back function. This type of circuit is very helpful for fault condition coverage of circuits that go into hazardous environments.
Excellent circuit to protect against end user mistakes ....now, can you figure out how to keep the purchasing dept. from buying the cheapest components in their mis-guided efforts to save pennies while disregarding life endangering risk ? Retorical question ...this is a question without an answer.
Glad you came back to youtube ...and that you're in sync with the best; Big Clive and EEblog.
If you're doing a non battery powered device, a series Schottky diode should work just fine. I'm designing an automotive device at the moment and I find Schottky diodes to be the most cost effective solution, especially since they can withstand all the weird inductive pulses in the vehicle's electrical installation.
Too high of a leakage current for batteries?
www.planetanalog.com/author.asp?section_id=3389&doc_id=564739
@@mikeguitar9769 Even in the mA range, thr leakage current for a Schottky diode is too low to do any harm in case of a reverse poalrity event. Their big downside is the power dissipated through the voltage drop.
Look up "super barrier" diodes from diodes.com. They dont leak as much and take higher voltage.
I use series diodes when input voltages is high ( >= 9V ) and reverse diodes and fuses for low voltages. If the voltage drop is important, I use Schottky diode . For over current protection, usually I use a fuse or reversible fuse (resettable fuse).
The big advantage of the explodey diode is that it can be added even though you forgot all about reverse polarity protection until the end of the project. Also, not so explodey if it is only likely to connect to a power source that can only deliver a few amps.
You might also consider the use of a re-settable fuse. The Bell Fuse Inc 0zcj0005ff2e (surface mount) has a hold current rating of 50MA and a trip rating of 150MA, 3.6 ohms cold, 50 oms when tripped, cost in quantity is 5.6 cents. Many varieties are available.
Your content is calming to watch. Which is nice when I’m trying to learn something and I get confused. Thank you 🙏
I am now caught in the nerd thunder tornado. And pretty happy about that.
Vortex of Nerdly!!!!
Nice Video Jeri! Can't wait to see what other gems you have to offer... i see a backlog binge session in my near future.
Nice circuit Jeri. Well familiar with the exploding diode method. My truck driver brother used to bring me regular examples of it to repair.
I have a vintage 1979 Chess Challenger 7 portable computer that I wanted to protect from reverse polarity (it's using a 9VDC small external psu). I simply put a diode bridge between the power input jack and the device inside. That way, it works no matter what the polarity is. Very simple to do: a diode bridge and a few wires. Cost: a few cents. The device already has a 7805 voltage regulator inside, so over-voltage is not an issue.
Jeri, great video and design using the PMOS diode(s). Thanks for sharing and welcome back!
It's interesting that this mix of 3 transistors ends up being a good design. The BJT makes a better voltage reference compared to the FET (lower voltage and lower tolerance). The FET's insulated gate makes a more efficient pass transistor, even though 2 are needed to get bipolar switching.
If manufacturers decided that they cared enough to make their products somewhat closer to foolproof, maybe there would be an incentive for the semiconductor manufacturers to build circuits like this into one all-encompassing, easy to install and design with, package. But I wouldn't hold my breath.
Very good video! Thank you. I like to see a well made discrete design instead of picking a fancy chip to solve all problems whithout knowing what's going on in detail. Please more....
Jeri! I am so glad you are making videos again. This stuff is GOLD. Keep it up!
It takes some guts to admit a mistake in public. Well done - and it happens.
Oh to have had something like this on an expensive wireless microphone at my old work. It had the same barrel jack as a PoE switch that was part of the same kit of equipment. It didn't like having 48v AC shoved in where 12v DC was expected. The series diode wasn't much help. Pretty much every semiconductor on the board was toast...
It would be amazing to start seeing you posting videos more often!
I'd love to share them through Tefen's Tech to others in the PC Modding Industry.
Coolest mistake I remember making was mistaking an LM335 comparator for a Darlington array. Er, the light was a bit dim at the time, the Darlington array was being used to switch on a capacitor discharge thing to switch some points, but, er, well, the socket I plugged the wrong thing into had 8 pins. The fascinating thing was, instead of the magic smoke just quickly puffing out, or lazily curling out, no, this was way more impressive - the Darlington array simply exploded into shards. Needless to say, I quickly switched the system off. No serious damage done. I have used servos to switch points since then - piano wire is very effective as the stuff that goes from the servo horn to the hole where the point is. Only other thing is, best to switch the servos off after using them, due to interference due to the locomotive motors.
Hey Jeri, Thanks heaps for sharing! Using the PMOS transistor in the linear region for current limiting is an application I'll have to keep in mind next time I need some cheap and cheerful protection from... less than carefully chosen peripherals.
Wait why does this video look exactly like franlab? Is there a common lady engineer camera angle thing going on?
BRB, now imagining Dave Jones from this camera angle instead of making the short fella look big. Or Big Clive at this angle from the knees down only.
It's difficult to get my fat head into the shot and have the camera pointing at the bench unless its pointing down. I like making 0 edit videos if possible because its challenging.
I do not understand much in English, but I like the fact that a beautiful woman tells about electronica. Respect!
So glad ur back on puting videos on youtube. I love listening to the interviews you give about your life and story such a cool story sure ur saying yeah it hasent been all cool :)
I love it when RUclips actually suggests a video I enjoy. You rock. New subscriber earned!
My new favorite lady! Subbed! Where have you been all my life? Your vids are helping me refresh my memory on electronics. You can't fire yourself...ever!
I missed you so much, welcome back.
Thanks. Glad to be back.
Nice video Jerry. Happy to see you again ! I love the calm way you explain and in contrary to the other videos you keep it to the point and short. Hope to see more videos of yours.
Good circuit. The pain always is the power dissipation in the current limiting fet. How about a bistable or maybe monostable one which does not have that problem ? Using only a few components ?
I really enjoy these videos! They’re also a great public service!
Electronics for me is a hobby and these videos have really educated me and have been super entertaining.
It’s also great that Jeri has been self taught and can present things in a way a self taught student can understand.
Could you please show the schematics larger by bringing them closer the camera. Or even better to do an insert of scan. Thanks.
Agreed. The diagrams can be a little tough even if I can't this to a TV.
I really love the old content you put out there :)
I really enjoyed this video. (I have some knowledge of electronics and almost zero hands on circuit building) Your video breaks down your project in a way I understand which inspires me to dig the breadboard out and visit the world of circuit construction again. I also like how you provide links to other cool videos of what other thunder nerds are up to. Thanks!!
Very useful circuit. I still tought the 'explody diode' was still the way to do it, but my knowledge has aged. I'll start experimenting with 'pmos diodes'.
As with all things that limit at some constant current - be sure your pass device can handle the required dissipation from full current at the limit against a dead short at your device end.
I'm hoping for a part two where you discuss the considerations for running the circuit routinely in limiting mode and best practices for making the current limiting adjustable.
I'd like to see devices like these with an integral two or three color low current LED to tell the consumer at a glance if something is wrong with the way the device is being powered.
I used to use MOSFETs with the current sense output. I haven't seen any lately. They had no output voltage drop, pretty cool.
To make mistakes is human, and sometimes that's how discoveries are sometimes made.
Teflon was also discovered by accident.
Can't make it through a single day without a discovery.
For myself I don't think I'm doing odd enough things to make a realistic chance to get to discoveries. I just prefer to do a good job using well known approaches.
@@SE45CX I encounter about once a month or so a situation where a well-known procedure just refuses to work right, or where a seemingly minor change in a parameters has great consequences. Have been a CNC machinist in a job shop for about a year, since quitting a dead-end career in IT...
@@SE45CX , cyanocrylate super glue was also a happy accident.
I wondered where that went! Thanks for fixing it and thanks for being Jeri!
I'm just glad you are back, making mistakes is human
I'm so happy to see that you're back.
I can't describe how much I like your videos. Thank you :)
We could also spice model the circuit, assuming there are accurate models for the FET's and transistors
Sure, with this you aren't going to need exotic FET model behavior.
I always try out a simple circuit using the free LTSpice circuit simulator first to see how it works then go to the bench to verify...I'm lazy /;>) P.S. The models are accurate.
Hope to QSO someday. You, Leo, Joe Walsh, etc. You're in my top list of cards to get! Take care!
I haven't seen Jeri in my suggested playlist for years. I forgot she existed.
This looks good! I think I would use a poly-fuse instead of the current-limiter circuit-- and if you are REALLY feeling like you need to go cheap-- just a Christmas tree flasher lamp (the kind with a built-in bi-metal switch that shorts the lamp) used in series. The flasher lamp will have very low impedance until there is enough current to turn it off (and then it will cycle). If using the poly-fuse or flasher-lamp, you could also use a back-connected uni-directional transorb instead of the PFET. This has the advantage of also protecting against over-voltage (up to the rating of the poly-fuse or flasher-lamp). Obviously, it would take some experimentation to find the right flasher-lamp-- and it's current limit would be slightly temperature dependent-- but if you are going for cheap... Fun stuff! Endless possibilities!
A motor is another interesting current limiter, especially for AC. A large single phase induction motor without a start capacitor approximates a short circuit at standstill, but draws comparatively little current once moving.
Hi Jeri, thanks for all your work!! Just trying to learn something about AR/VR glasses! Will check all your videos and make loads of questions!!
Another method is an over current latch circuit, that hard cuts the output upon seeing a fault condition until reset. Better for higher current loads where linear regulation would prove infeasible. Maybe you could just bond a thermistor to the FET of the linear circuit, producing a positive feedback loop, though using an actual comparator latch circuit is probably better for precision.
At about 10:00 , despite digital scope and digital power supply, an old-fashioned analog ammeter comes into play. I love it!
This was the first I'd seen of you Jeri. In watching others too I'm like, wow this girl has some serious equipment, is this for real? .... After google, i see, of course, makes sense. Protection circuits will be NB for those mass market games! Luck! And thanks for sharing.
This is a lesson everyone needs reminding of.
You were the person who did say to make a lot of mistakes, and learn from them.
Hey Jeri. Loving this "relaunch" or whatever it is!