Ruining mosfets with a transistor tester and tweezers

[VSV_electron]

Hello!

I've never had any experience with MOSFETs before so I bought myself two kinds of mosfets (or "MOSFETS" if you like) on AliExpress.
I know, I know - those are not the real thing and they are not up to specs (if working at all) but the majority of  those who are aware of what they buy on Ali say: "These are valid devices with the 1/2nd or 1/3rd of the Drain current rating". Others, expecting too much cry in despair in their 1-star reviews: "Fakes-fakes-fakes!".

Sorry for the lengthy intro, I'll try to be as concise as possible (not my thing really...).

- I bought a bunch of 2n7000 signal E-MOSFETs (50 pcs).
- I bought 10 IRLZ44N "power" E-MOSFETs (parenthesis intentional)

I as a always do: put on my antistatic wrist strap (checked and working at 1 MOhm), got my 2 antistatic bags of 2n7000 and IRLZ44N out (transistors never come in ESD-safe bags from Ali but I always repack them hoping for the best case). Got my simple and common GM328A transistor tester out and ready.

Note: Before starting the test procedure I read a few instructions on the Web on careful handling of the MOSFETs. Well, one thing is to read something and another is to get it all right and not to slip somewhere.

- I started testing the mosfets from the less expensive 2n7000. The tester identified them correctly as an E-MOSFET and a half a dozen randomly picked out of the bag were all fine.

What I could do WRONG, please comment: after the tester finished the identification of each 2n7000, I removed them from the ZIF socket and with the tweezers shorted all the three leads together (just to make sure the Gate is discharged). I actually did that for a few more times in a row to one 2n7000 and after 3 or 4 times (tester + tweezers) it was still identified as the valid E-MOSFET.

- After that I moved on to IRLZ44N and then there was frustration waiting for me:
Once I inserted the IRLZ44N into the tester ZIF socket it was identified as an E-MOSFET. I waited until the tester LCD dimmed (at that point there is no testing voltage being applied to the DUT; I checked that with a MM). I pulled out the IRLZ44N, did the same thing as I did to the 2n7000 - shorted all the 3 leads with the tweezers to discharge the gate. (Note: My ESD wrist strap is on me all the time...).

On the second insertion of the IRLZ44N into the tester it was already a dud. I didn't believe it so I did all the same thing with the second IRLZ44N... dud again!
I went on to the 3rd IRLZ44N and... it was already a dud out of the box.

So I got out the 4th one and tested it in the tester and it was fine, valid E-MOSFET. This time, however I decided to stay on the safe side and didn't short the leads of the 4th E-MOSFET. I waited for the tester to finish its procedure and got the E-MOSFET out
On the second insertion into the tester ZIF socket... it was already a dud. I didn't do anything special to it that time.

After wasting 3 of the counterfeits (not counting the 1 out of the box dud) with the simple tester and tweezers manipulation I stopped. I still had 6 from the batch untouched so I feared to ruin what I still got left. I got my UNI-T UT61E+ MM set on 'Resistance' and checked the D-S resistance of the destroyed IRLZ44Ns first. The Red lead on the Drain and black on the Source gave me around +/- 1 kOhm for each of the duds; the same with reverse.
I did that the same to the remaining potentially still good 6 IRLZ44N. All of them showed around 3.5 to 3.9 MOhm Drain to Source (Red-Black) and the '0' reading for reverse. So they all went back into the bag this time leaving me in the a little upset condition. I decided to stop there completely for today.

So - is that explainable that I did the same thing to the 2n7000 and IRLZ44N and the former was fine after several insertions into the tester ZIF with the subsequent shorting of the leads and they survived all the mild "abuse" and the latter were duds for no apparent reason ? What did I do wrong?

[thm_w]

Shorting all the leads together with tweezers is not necessary. But I don't see that damaging any common FETs like we have here. And if anything IRLZ44 should be more robust than the smaller part, which is a bit suspicious. Reading 1k Drain to Source with the gate discharged is definitely bad.

See if you can get a hold of some genuine power FETs, maybe from salvaged electronics boards.
Maybe also measure the voltage your GM328 tester is applying to the part with a multimeter, and see when the GM328 is done if it leaves any voltage on the part or not. I assume it should discharged all outputs to near 0V.

Your wrist strap is connected back to mains earth? What power source is supplying the GM328?

BTW if you want cheap genuine parts, can just buy from LCSC.

[VSV_electron]

...
Maybe also measure the voltage your GM328 tester is applying to the part with a multimeter, and see when the GM328 is done if it leaves any voltage on the part or not. I assume it should discharged all outputs to near 0V.

Your wrist strap is connected back to mains earth? What power source is supplying the GM328?
...

To be honest I hesitated to buy any "power" stuff in the TO220 package off Ali for a long time. I knew Chinese signal BJTs of large variety and 2n7000 (but not JFETs) were fine according to the reviews but reviews of their power BJTs and MOSFETs were always doubtful.

I would have to connect a scope to the GM328 to see what it supplies to the DUT but with the MM it shows cycling through some voltage ranges from 0V up to maybe 5V which is expected according to the tester's algorithms.
My wrist strap is connected to the mains earth.
GM328 is powered by a wall adapter rated at 12V. GM328 confirms that the "battery 12V is OK" and something like "5.05V is OK" - I think that's about the regulated voltage to the Atmega328P.

I went back to Ali and found a popular listing with the cloned IRFZ44N (should be close to my logic level IRLZ44N). Except for 1 undelivered review the listing is almost 100% 5-star reviews with people reporting good tests on all received units.
https://aliexpress.com/item/1005003080494688.html

That was in line with my understanding of what to expect: don't drive them too hard and they should be OK for hobby projects. My situation frustrates me as I blame myself for destroying potentially fine units. Well, that's actually stupid and I probably just need to refine my approach but you see how desperately ESD-safe I am?
I never handle any stuff I receive off Ali directly without a wrist strap; Never handle any chips by their leads, etc.

[DimitriP]

What did I do wrong?

so I bought myself two kinds of mosfets (or "MOSFETS" if you like) on AliExpress.

know, I know - those are not the real thing and they are not up to specs (if working at all)

I rearranged it a bit so it's easier to follow

[VSV_electron]

...
I rearranged it a bit so it's easier to follow

Oh, please, I understand your humor but it's inappropriate here. Be helpful, not sarcastic.
I found the listing of the shop I bought my IRLZ44Ns - people were happy with what they bought for their needs. Their MOSFETs are not exactly up to specs but useable and they successfully tested them with their testers.
I'm trying to figure out where I went wrong with my handling and/or testing of the units.

[VSV_electron]

I agree with what thm_w said above and I probably shouldn't blame my GM328 tester because if it was harmful it would first destroy the 2n7000, not the IRLZ44N. Makes sense?

[DimitriP]

I was demonstrating what you "did wrong" using your words.
So I don't end up using any of my own words. Sometimes they don't lead to a conclusion with the same degree of humor.
I save those for special circumstances.

Humor  is always apppopriate except maybe in funerals - depending on local customs

You took a gamble, and partially lost.
Enjoy the "good" parts.
We are sorry this happened to you. And please don't do it again.

[VSV_electron]

Here is a typical handling scenario of the MOSFETs in Ali reviews which as you probably understand adds more frustration to my case.

[VSV_electron]

I was demonstrating what you "did wrong" using your words.
...
We are sorry this happened to you. And please don't do it again.

I tell you once again: people buy those Ali clones and they are good enough for DIY projects. No one reported ruining them with the tester or even by handling them with their hands as shown above.
As such your humor is out of place here.
What's up with you? Are you here to make laughs or to help people?

[Kim Christensen]

Unless your GM328 tester has the high voltage zener test option, it shouldn't harm your MOSFETs since the test voltages and currents are too low.
Must be a bad batch of MOSFETs.

[sleemanj]

The 2N7000 has significantly higher Gate-Source absolute maximum than the IRLZ44

Your tester might possibly be busting that and killing the IRLZ44 as a result.  One datasheet  has Absolute Maximum Gate-Source as just +/- 10v ( https://www.vishay.com/docs/91328/irlz44.pdf ), half that of the 2N7000.

I got my UNI-T UT61E+ MM set on 'Resistance' and checked the D-S resistance of the destroyed IRLZ44Ns first.

A floating gate could give you any results.  Not to mention there is the body diode when you measure the reverse.

Setup a simple test circuit with a fet, a load, and a pull down/up resistor on the gate and see if it's your testing that is the problem or if they really are dead.

[Ian.M]

My first suspicion would be mains leakage current from the PSU powering the GM328A tester, + you supplying a ground by handling the 'IRLZ44N' MOSFETs by their tab (drain) while inserting/removing them.

Switched mode wall warts without ground pin are not suitable for such an application due to their Y capacitor leakage current, and ones with a ground pin must be in a properly grounded socket.

[VSV_electron]

The 2N7000 has significantly higher Gate-Source absolute maximum than the IRLZ44

Your tester might possibly be busting that and killing the IRLZ44 as a result.  One datasheet  has Absolute Maximum Gate-Source as just +/- 10v ( https://www.vishay.com/docs/91328/irlz44.pdf ), half that of the 2N7000.
...

Thanks much! I'll give it some thought.

[Kim Christensen]

My first suspicion would be mains leakage current from the PSU powering the GM328A tester,

That's also a highly likely scenario.

[magic]

My first suspicion would be mains leakage current from the PSU powering the GM328A tester,

That's also a highly likely scenario.

Another vote for this. I have blown some MOSFETs while playing with them with a wall-wart PSU.

Try grounding your wrist strap to the tester's ground instead. Or ground both to mains earth.

edit
A typical 1nF capacitor between live mains and secondary ground in such PSU has 3MΩ reactance at 50Hz. This plus 1MΩ in your wrist strap appears in series with the MOSFET between mains and earth when you insert one pin into the tester and grab it by another pin (e.g. the metal tab) with your hand. If the pin which makes first contact with the tester is the gate, you are out of luck. If the FET's gate capacitance is another 1nF, that's another 3MΩ impedance in the series chain and it gets almost half the mains voltage across it.

It's possible that your 2N7000 is not entirely a "true" 2N7000 and has built-in gate protection.

[VSV_electron]

Would you all say then that it was my stupid mistake to power the GM328 tester from the mains adapter instead of a battery block? Would it be just fine if I didn't wear the wrist strap?
If what you suggest regarding the mains leakage current is true how would I avoid that? Honestly I haven't really relized the situation yet and that is sad because it's telling me that I'm not ready for even simple things like this one yet.
I have to re-read the mains leakage scenario couple times.

[magic]

Battery power should help. It would probably also help if you disconnect the strap, grab the MOSFET by the drain and insert it into the tester while touching the tester's ground with the other hand.

The circuit is a series chain like this:

mains -- 1nF or thereabouts in the PSU -- PSU output "ground" = tester ground -- MOSFET gate capacitance -- you -- 1MΩ -- earth

So it works like a voltage divider, with each element getting a voltage drop proportional to its impedance. The highest impedances are the PSU Y capacitor and the MOSFET gate, both in the vicinity of 1nF, so they get most of the voltage. The resistance of "you" might be a few kΩ and cables are even less.

[shapirus]

Would it be just fine if I didn't wear the wrist strap?

Not necessarily. There can be more than enough coupling to your body even if not grounded to kill the transistors.

Try powering your transistor tester from battery only, don't connect it to mains voltage.

[VSV_electron]

My first suspicion would be mains leakage current from the PSU powering the GM328A tester,

That's also a highly likely scenario.

Another vote for this. I have blown some MOSFETs while playing with them with a wall-wart PSU.

Try grounding your wrist strap to the tester's ground instead. Or ground both to mains earth.

edit
A typical 1nF capacitor between live mains and secondary ground in such PSU has 3MΩ reactance at 50Hz. This plus 1MΩ in your wrist strap appears in series with the MOSFET between mains and earth when you insert one pin into the tester and grab it by another pin (e.g. the metal tab) with your hand. If the pin which makes first contact with the tester is the gate, you are out of luck. If the FET's gate capacitance is another 1nF, that's another 3MΩ impedance in the series chain and it gets almost half the mains voltage across it.

It's possible that your 2N7000 is not entirely a "true" 2N7000 and has built-in gate protection.

The 2n7000 whatever it is has no metal tab unlike the IRLZ44N in which it is the drain terminal, right? I inserted 2n7000 holding it with the tweezers by the plastic head (no mains voltage across the gate?). Of course the most convenient place to hold the TO220 by is the tab... that might explain it all.
I must admit I don't get the "half of the mains voltage" across the gate part due to my lack of circuit analysis knowledge. Could you please try to elaborate on that? In that case do I also get half of the mains voltage across my wrist between the point of the wrist strap contact on the wrist and the tips of my fingers holding the ESD-safe stainless steel tweezers (coated)? How to understand the physics of that scenario in question? Sorry to admit that I'm dumb and can't easily grasp the concept, even if you already tried to explain it.

[VSV_electron]

Let me ask you please to extend the whole thing with the PSU and general breadboarding beyond the MOSFET testing in the GM328A.
Should I NOT power the ubiquitous "Arduino" (ESP32, Raspberry Pi Pico etc. etc.) from the common floating wall SMPS PSU? Is that a bad practice in itself?
Or should I simply NOT wear the wrist strap in that case?

[VSV_electron]

Battery power should help. It would probably also help if you disconnect the strap, grab the MOSFET by the drain and insert it into the tester while touching the tester's ground with the other hand.

The circuit is a series chain like this:

mains -- 1nF or thereabouts in the PSU -- PSU output "ground" = tester ground -- MOSFET gate capacitance -- you -- 1MΩ -- earth

So it works like a voltage divider, with each element getting a voltage drop proportional to its impedance. The highest impedances are the PSU Y capacitor and the MOSFET gate, both in the vicinity of 1nF, so they get most of the voltage. The resistance of "you" might be a few kΩ and cables are even less.

Actually I missed this post of yours. Perhaps you don't have to say anything else to explain it to me on a deeper level. It shoud be enough for me for the analysis, otherwise I shouldn't touch any electronic devices at all.

[VSV_electron]

I have to give more thought to safety issues related to SMPS and its place on the amateur work bench.
If it was not the SMPS but a traditional linear PSU I would have no such issues related to the mains current leakage, right? - Because the secondary of the transformer would be truly isolated from the mains?
I didn't see it anywhere emphasized about the dangers of using SMPS PSUs on the workbench in relation to sensitive electronics devices. I understand that it's obviously my fault not equalizing the ground of the DUT and myself when using the SMPS as the PSU but man... why didn't I even remotely imagine the situation?

[shapirus]

Because the secondary of the transformer would be truly isolated from the mains?

Nope, there would still be capacitive coupling between the transformer's windings. If you measure the potential (simply using a 10 MOhm voltmeter; most DMMs will do) between any of the secondary winding's terminals and earth ground, or even your own body (especially when grounded), you will see values sufficiently high to damage the MOSFET gate's isolation. And that's even at a 10 MOhm load. If you increase the effective load resistance further by creating a, say, x0.5 voltage divider with another 10 MOhm resistor (and the higher the ratio, the better), using the DMM as the other part of the divider, then you'll see an even higher potential, when you multiply the resulting DMM reading by 2.

It's all about potential difference, and you can always measure it. It's zero (or near-zero) between ground and ground, but it can be way above zero between ground and anything else, especially if that "anything" is powered from mains and is not grounded. Yes, current will typically be very small (microamps) when you close the loop, but in case of killing MOSFETs we care about potentials, not current.

[Ian.M]

Let me ask you please to extend the whole thing with the PSU and general breadboarding beyond the MOSFET testing in the GM328A.
Should I NOT power the ubiquitous "Arduino" (ESP32, Raspberry Pi Pico etc. etc.) from the common floating wall SMPS PSU? Is that a bad practice in itself?
Or should I simply NOT wear the wrist strap in that case?

Not wearing a wrist strap, especially in a northern climate that can be very low humidity in the winter, risks ESD damage, so don't stop wearing your wrist strap.  Do however check its lead has >1 Meg resistance, and check that the ESD grounding point you use also has >1 Meg resistance to mains PE ground.  This limits the leakage current from any SMPSU that can flow through your wrist strap to under 100uA which most devices except 'bare gate' MOSFETs will withstand with no issues.

You need to give the leakage current somewhere safe to go if you are working on the circuit.  This may be achieved by grounding the PSU's output 0V to mains PE ground.  However as your soldering iron bit is probably grounded for similar reasons, you must be careful to fully disconnect the circuit so its floating whenever you need to solder a backup battery positive terminal.

When its in use, the suitability of an ungrounded SMPSU varies with the application.  If its self-contained in an insulating case with non-metallic controls, it probably doesn't matter if its floating at half mains voltage if nothing is grounding the leakage current.  However if its got extensive external wiring or electrically connected metal parts that people touch it may be a problem.

I have to give more thought to safety issues related to SMPS and its place on the amateur work bench.
If it was not the SMPS but a traditional linear PSU I would have no such issues related to the mains current leakage, right? - Because the secondary of the transformer would be truly isolated from the mains?
I didn't see it anywhere emphasized about the dangers of using SMPS PSUs on the workbench in relation to sensitive electronics devices. I understand that it's obviously my fault not equalizing the ground of the DUT and myself when using the SMPS as the PSU but man... why didn't I even remotely imagine the situation?

Even linear PSUs aren't free from leakage current.  Unless there is a grounded screen between the transformer primary and secondary, there will be *some* interwinding capacitance and thus some leakage current.  However its typically orders of magnitude lower than from a SMPSU, and usually that's small enough to ignore.  In sensitive applications it may be helpful to check it before and after reversing the linear PSU's transformer primary connections, and use the way round that give a lower leakage current.  This is because nearly all the interwinding capacitance isn't to the whole primary but only to the layer of it that is nearest to the secondary, so in countries that maintain mains Neutral near ground, it helps to have that end of the primary connected to Neutral to reduce the voltage driving the leakage current.

Why didn't you think of it?  Well how many SMPSUs have you repaired or designed and built in your career?  Mostly its due to lack of experience with the internals of what many of us tend to think of as 'black boxes' with only their external interface of interest. 

Shapirus is correct that we care about potentials on unprotected MOSFET gates, but the MOSFET's gate capacitance is also a factor - since the leakage current is AC, to do damage it must be large enough to charge the MOSFET gate up to its gate oxide breakdown voltage in under half a mains cycle.

[shapirus]

but the MOSFET's gate capacitance is also a factor - since the leakage current is AC, to do damage it must be large enough to charge the MOSFET gate up to its gate oxide breakdown voltage in under half a mains cycle.

Yes good point. I did kill a few 2n7000/2n7002's with my homemade lab PSU powered by a linear step-down toroidal transformer followed by a diode bridge and a filter cap. It was easy: connect the output of the PSU to the circuit assembled on a breadboard which the transistor is a part of, then touch the gate of the poor transistor intended to be used as a switch (unless protected with clamping diodes, TVS, zener etc.) -- done, it's dead. At the same time, beefier transistors, the "power" ones, with 2 nF or more gate capacitance, survived the same without issues, or at least no immediately noticeable issues.

To give an idea about what numbers we're speaking of, I measured the potential between the secondary winding (actually between any of the output terminals of the DC/DC converter that follows the transformer and rectifier) and my own body via the 10 MOhm resistance of a DMM.

It measured 9V RMS (12.7V peak) when I was floating, and 23V RMS (32.5V peak) when I was grounded (to mains PE). The latter is enough to kill a mosfet (at least a smaller one), and the former is also enough, since it's the reading under a relatively, for the current in question, low load resistance: if I measure the same with two voltmeters connected in series (~20 MOhm total resistance), then, as expected, each reads about 8V RMS, meaning 16V RMS total (or 22.6V peak), and it will continue to increase further as the load resistance is increased (up to a certain point, of course).

BTW, knowing that, we can calculate the current flowing in this scenario. Let's take 10V RMS across 10 MOhm load for simplicity, this yields 1 uA RMS.

Now, we can use it for a very rough estimation (I'm not sure we can apply this directly to the process of the gate capacitance charging, but at least it should be fine to understand the order of magnitude) of the lowest capacitance that will not be able to reach 20V during one mains voltage half-cycle.

Let's assume 50 Hz. One half-cycle is then 10 ms. Then, 1 uA (again, assuming constant current for simplicity) during 10 ms creates a charge of I*t = 1e-6 A * 10e-3 s = 1e-8 C. To have this amount of charge, the capacitor charged to 20V must have a capacity of C = Q/U = 1e-8 C / 20 V = 500 pF. This is somewhere in between smaller and bigger mosfets and agrees with what I have observed in practice.