Over on YouTube our hacker [VIP Love Secretary] shows us how to make a simple MOSFET tester.
This is a really neat, useful, elegant, and simple hack, but the video is kind of terrible. We found that the voice-over constantly saying “right?” and “look!” seriously drove us to distraction. But this is a circuit which you should know about so maybe do what we did and watch the video with subtitles on and audio off.
To use this circuit you install the MOSFET you want to test and then press with your finger the spare leg of each of two diodes; in the final build there are some metal touch pads attached to the diodes to facilitate this. One diode will turn the MOSFET off, the other diode will turn the MOSFET on, and the LED will show you which is which.
Apparently this works through stray capacitance, an explanation which makes sense to us. We were so curious that we ran over to the bench to build our own version (pictured with the schematic above) just to see if it worked as advertised, and: it did!
We tested it with a faulty MOSFET and when the MOSFET under test is faulty then the LED won’t turn on and off like it should when the MOSFET works. Also, if you build one of these, you want to feed in a two or three volt supply (it will depend on the specs of the LED you use); it’s not mentioned in the video but two volts is what we used that worked best for us.
Thanks to [Danjovic] for writing in to let us know about this one. If you’re interested in MOSFETs maybe it’s time to learn the truth about them.
Nice video. A better explanation is that the stray capacitance of the high-impedance gate should retain whatever charge is imparted upon it to keep the LED turned on or off. The human acts as an antenna for lots of stray EMF and the diodes are acting as peak detectors for the antenna voltage. Touching the cathode drags the gate down to the peak negative voltage (minus the diode drop), while touching the anode raises the gate up to the peak positive voltage (again, minus the diode drop).
Be cautious using this as ESD from touching the diodes could destroy your MOSFET. If you add a ~10V zener diode from gate to source, it’ll protect it without affecting the tester functionality.
Agree, but a typical zener will do little to protect against ESD; and many MOSFET gates can blown to smitthereens by HBM ESD.
Oh, bummer. Is there anything to be done to make this circuit work safely?
A series resistor – zener shunt to source – series resistor – gate configuration is pretty robust, and this application is so insensitive to turn-on/turn-off time that you don’t have to think much about picking resistors. Grab something in the kiloohms, and make the first resistor something with a decent voltage rating (if you have no datasheets because you’re fishing around the junk box, physically larger resistors are usually better).
Junk box parts and haphazard layout won’t be perfect, and if you scuff your feet across the carpet enough or stand up from a fabric chair fast enough you will still be able to kill your FETs, but a circuit like that will be fine if you exercise even a tiny bit of care.
Thanks for your advice! I will revisit this circuit next week.
FYI the circuit I plan to implement next is this one. I would be happy to have any feedback on it! Thanks! https://www.inthelabwithjayjay.com/file/2025-09-14-235202/MOSFET-tester-2025-09-14-235154.png
You can remove D6 (Zener can also conduct)
You have also drawn a wire though D6. It’s shorted now.
Remove R1 (Why use two resistors?)
Increase R1 to 100k to a few Mega Ohm (You only have to charge the gate capacitance)
And as I wrote in that other post, TVS diodes are better, because they react faster. But with a 100k or more series resistor, the RC time (with the gate capacitance) is probably more then the reaction time of the zener.
Also, the usage of the Diodes implies that you pick up noise (If I touch my scope probe, I can often see >80V mains voltage hum.
A variant of this circuit: touch the gate with one finger, and then alternately touch GND or the 12V power supply. Your body will conduct enough to turn the FET on or off. (For this the power supply needs to be higher then the gate voltage).
This also works with darlington transistors, you may have to add a 3rd BJT for more amplification of the picked up noise.
Thanks paulvdh. The latest circuit revision can be found at the following link. I assume I want a unidirectional TVS diode and not a bidirectional one? Will a 6.8V TVS diode suffice? https://www.inthelabwithjayjay.com/file/2025-09-15-023319/MOSFET-tester-2025-09-15-023308.png
Thanks for the tip! I have some 9.1V Zeners on hand, will one of those do?
Don’t use regular zeners, but use TVS diodes.
In a class at school they were troubled by a lot of blown FET’s (Used on breadboards), so they started soldering TVS diodes directly from Gate to Drain.
Most FET’s have their threshold voltage around 4V and 9V is enough In that class they used something like 16V or 18V TVS diodes. It interferes less with the gate driver, while still being below the (most often 20V allowable gate voltage. (Some high power fets have a 30V gate voltage these days). For small signal fets, it can be less.
I assume what’s going in with the narration is that some kind of AI is doing a “too literal” translation of the creator’s original words. The final published result is not outright awful, but not excellent.
I guess we’re going to see this more and more as opposed to the hordes of “word-less workshop” “howto” videos that has plagued YouTube in the past.
Some of the “wordless” videos were actually useful, while others were obvious BS (if they involve chucking a AA battery in a drill and unbending a paper clip to unlock all channels worldwide on your smart TV, they are obvious trash). And then there was a vast middle ground, that I could go on and on about.
What we need is for YouTube (or a YouTube competitor) to bring back down-doots — let’s face it, someone is out of office and 81m(dot)com is now an archived defunct website. There is no reason not to go back to ratings sanity.
Lacking that, I’m daydreaming about an add-on or something that would let me mark/ban bad or deceptive channels from the videos that YouTube recommends. Not only the AA-battery-in-the-drill-chuck videos, but the channels that make fake “free energy” videos and other fake, time-wasting content. I do have a name for the program picked out already, I would call it: “Remove_Van_de_Paar”
(if anyone wants to remake this video with a better dialog and an explanation of what is going on, I’d watch and bookmark it. OTOH, I’d recommend one of the multi-function testers (various names like TC1) with the ZIF sockets and a color screen that’s running an unknown version of Karl-Heinz Kübbeler’s freeware firmware. You can’t make the tester cheaper than what you can buy off of of Ebay and it works extremely well.)
Although all those ridiculous ‘tech’BS videos are annoying what I find more annoying is that I get all those political and war propaganda videos suggested with 90% of the normal videos I watch.
And no its not cookies, those are not kept.
I don’t want to hear fantasies about tr*mp or about random people who got shot in the US or imaginary tales about ongoing wars.
I wonder what percentage of YT is pure BS clickbait at this point, must be over 50% I imagine right? Or at least the suggested content is over 50% (maybe up to 70%) clickbait.
I better not ask what percentage of regular news is clickbait BS fantasies though..
Dropping cookies is part of your problem… If they know about your interests, they’ll try to pick something decent, and more ads will be targeted. If not, they go for rage-bait recommendations and spam-the-world advertisers.
How many people that drop cookies watch ads you think?
And the interest they know about would be the video you are watching.
Problem is that they now do suggestions based on what the interest of the video-creator are and what he/she watched and subscribed to, and a percentage of what other viewers of the video are interested in. And the problem there is that an awful lot of people fall for those stupid videos.
But there’s the issue of them getting paid by those annoying youtubers too of course, now that YT takes money to promote channels.
On a related note, I just heard a youtuber tell how a video he made that has the most negative comments, on the subject of the video, is actually the video that is making him the most money. Seems people love to find videos that they can b|tch about.
I’m guilty of it too in a sense, I sometimes go to stupid videos to see the comments, while not allowing it to play. But is not allowing it to play really stopping the google metrics from marking the video as popular though? Is it views only or is simply clicking enough to cause an unwanted boost? I hope not, but I fear it might.
Not a big fan of circuits that rely on EM background noise pickup via stray capacitances, other than for demonstrating that this noise is everywhere. They can be irritatingly random, and can behave very differently when being powered from a wall wart vs a battery.
Oh, interesting, thanks for your remarks. I might give my circuit a go on a battery and see if I can see any changes in behavior.
“This is a really neat, useful, elegant, and simple hack, but the video is kind of terrible.”
To me this seems to be more crude than elegant…
Here is an idea…
Same circuit, however build a low power signal source instead of using a finger…. Now power the whole thing off a low voltage solar panel; that puts the word elegant back into the conversation.
Thanks for the suggestion. That would be an interesting mod. I will see if I can find some time to implement it!
First line of defense of the MOSFET gate is an ESD wrist strap grounded to the test circuit ground. This is industry standard operating procedure.
Sound good [Harry] but..
in actuality in 90% of the repair videos I’ve seen the guy (never see women for some reason) doing the repair has no ground wrist strap. Even from big outfits.
And of course when working with mains grounding yourself can be pretty dangerous, so you’d have to unhook yourself constantly depending on the repair, and not mess up.
That’s why the wrist straps usually have a large series resistor between strap and actual ground. Allows ESD to bleed off while preventing dangerous currents from mains.
To keep both hands free, the wrist strap can also be attached to ankle.