I'm working on a project and I have a bunch of DN2540s (both TO220 and TO92) and IRF9610s new from Mouser in protective packaging. Last week I got a Peak DCA75 tester and I was surprised to find that several of the MOSFETs read as defective.
With a DMM, the most telling sign is the bad DN2540s read kilohms or tens of kilohms from gate to drain while the good ones read open-circuit.
The more I play with them, the more failures I have. It's cold and dry here right now. I have a Velleman antistatic pad on the bench but no wrist strap. I try to sit still and keep my hands on the mat and the antistatic bags but I guess that's not enough.
The weird thing is, I can hook the DN2540s up with a gate-to-source resistor -- making a current source -- and they bias up at the expected current.
I assume I've blown the gates with ESD to a sufficient degree to upset the component tester, but not badly enough to completely fry the transistors. I plan to bin the parts, but I'm just curious about the mechanism. In particular, I worry about the remote possibility there's something about the tester and my PC that's causing the failures. The tester otherwise seems to work perfectly. I do understand electromagnetics and the fundamentals of MOSFET construction.
Mainly I'm wondering if people have had similar experiences and what the best way is to prevent this from happening in the future. Because the transistors seem to function after a fashion, I'm worried about ending up with a degraded circuit that doesn't appear obviously broken from basic test measurements.
-Henry
With a DMM, the most telling sign is the bad DN2540s read kilohms or tens of kilohms from gate to drain while the good ones read open-circuit.
The more I play with them, the more failures I have. It's cold and dry here right now. I have a Velleman antistatic pad on the bench but no wrist strap. I try to sit still and keep my hands on the mat and the antistatic bags but I guess that's not enough.
The weird thing is, I can hook the DN2540s up with a gate-to-source resistor -- making a current source -- and they bias up at the expected current.
I assume I've blown the gates with ESD to a sufficient degree to upset the component tester, but not badly enough to completely fry the transistors. I plan to bin the parts, but I'm just curious about the mechanism. In particular, I worry about the remote possibility there's something about the tester and my PC that's causing the failures. The tester otherwise seems to work perfectly. I do understand electromagnetics and the fundamentals of MOSFET construction.
Mainly I'm wondering if people have had similar experiences and what the best way is to prevent this from happening in the future. Because the transistors seem to function after a fashion, I'm worried about ending up with a degraded circuit that doesn't appear obviously broken from basic test measurements.
-Henry
A. Get a wrist strap.
B. Is your anti-stat pad grounded?
C. Get a humidifier if you're going to be working with MOSFETs or CMOS ICs. At least the ICs have ESD protection. The MOSFETs don't. The only thing that can save them is their relatively large gate capacitance. It's not large enough for winter static, though.
C. Don't handle MOSFETs unless you need to.
D. You might find if you test the MOSFETs again with proper ESD gear and a humidifier in the room, that some of them are not damaged. Accumulated charge can cause them to seem damaged.
B. Is your anti-stat pad grounded?
C. Get a humidifier if you're going to be working with MOSFETs or CMOS ICs. At least the ICs have ESD protection. The MOSFETs don't. The only thing that can save them is their relatively large gate capacitance. It's not large enough for winter static, though.
C. Don't handle MOSFETs unless you need to.
D. You might find if you test the MOSFETs again with proper ESD gear and a humidifier in the room, that some of them are not damaged. Accumulated charge can cause them to seem damaged.
Thanks, Russell. My pad isn't grounded. I figured if I kept my hands on the pad, touched the computer and tools before removing the devices, didn't move around, etc., that it would keep everything in the local environment at low enough relative potential. I'll look into getting a humidifier and wrist strap, and connecting the pad to ground.
Last night I carefully soldered the five (of six needed) good devices I have onto my PC board and they apparently survived so should be safe now. More parts are coming next week.
In the future, I think I will use the trick of poking the device leads through aluminum foil to keep them shorted while I handle them during installation.
-Henry
Last night I carefully soldered the five (of six needed) good devices I have onto my PC board and they apparently survived so should be safe now. More parts are coming next week.
In the future, I think I will use the trick of poking the device leads through aluminum foil to keep them shorted while I handle them during installation.
-Henry
Working with good old school Mosfets, think IRFP250-240-450-630-820-3055-etc. I have never taken special precautions, usually treat them like bipolars, and never ever had a problem.
That said, Buenos Aires is NOT a dry place, quite the contrary.
It might be different in very dry ones.
That said, Buenos Aires is NOT a dry place, quite the contrary.
It might be different in very dry ones.
Like you, I've always assumed I can get away with casual handling. The thing I'm most interested in understanding is how the device can be in a state where gate leakage is evident to a meter and component tester, yet it still apparently functions as a 150mA current source. I would have expected the damage at least to change the observed bias current for a given resistor value. One of the things I like about electronics is there's always something new to learn.
-Henry
-Henry
Two possible answers.
These vertical MOSFETs are big devices, and ESD may have caused a small hole in one localized area of the gate. If the leakage is not too bad, and the circuit can drive the leakage, it can still seem to work.
Accumulated charge can also cause the devices to measure oddly, but most testers will deplete the accumulated charge, so the first explanation is more likely. They are only a little broken. My guess is that you will end up having to throw these away.
These vertical MOSFETs are big devices, and ESD may have caused a small hole in one localized area of the gate. If the leakage is not too bad, and the circuit can drive the leakage, it can still seem to work.
Accumulated charge can also cause the devices to measure oddly, but most testers will deplete the accumulated charge, so the first explanation is more likely. They are only a little broken. My guess is that you will end up having to throw these away.
Wearing man made fibres can generate a lot of static. Nylon carpets are bad too.
I have a skin complaint so can only wear cotton which doesnt have that problem.
I cant remember ever blowing up a semiconductor by touch in 40 years.
I have blown up DN2540 by applying more than 20v gate voltage.
So I suspect they are prone to static discharges by touch.
I have a skin complaint so can only wear cotton which doesnt have that problem.
I cant remember ever blowing up a semiconductor by touch in 40 years.
I have blown up DN2540 by applying more than 20v gate voltage.
So I suspect they are prone to static discharges by touch.
With large devices, yes, but as already mentioned, small, low-cap devices are much more sensitive.
Note that you don't necessarily need all the usual antistatic protective measures and gear: awareness of what you do and touch, in which order, what you touch simultaneously is generally sufficient.
For a chain-worker who doesn't understand the problem, all the measures are required, but for an educated DIYer doing a one-off, it is possible to dispense with them.
I have had a few accidents in many years, but it is generally OK.
When you handle precious, unique, unprotected devices, I recommend using all the standard measures
A damaged device might seem to operate correctly in certain conditions, but it is a Damocles sword: the damaged spot is a small region where unintended effects like tunnel conduction take place, and these effects can progress over time, causing a catastrophic failure after weeks, months or even years
Sometimes static damage can be minor and the device works ok or maybe not quite strictly within specs.
Its good not to touch the pins if possible. I hold the plastic part while inserting into a pcb.
A lot of equipment I worked on was in a metal case so simply discharging my hands onto the metal case is a good practice.
Its good not to touch the pins if possible. I hold the plastic part while inserting into a pcb.
A lot of equipment I worked on was in a metal case so simply discharging my hands onto the metal case is a good practice.
Last edited:
I have had trouble with some Chinese mosfets.
Soldered them into a pcb.
PCB worked fine for about 10 minutes until I turned off my soldering iron.
Glitch on mains blew up the mosfets.
I can only assume the glitch pushed mosfets into breakdown region and shorted them out.
I replaced them with mosfets from RS Components and they have survived well over a year despite lots of mains switching.
Soldered them into a pcb.
PCB worked fine for about 10 minutes until I turned off my soldering iron.
Glitch on mains blew up the mosfets.
I can only assume the glitch pushed mosfets into breakdown region and shorted them out.
I replaced them with mosfets from RS Components and they have survived well over a year despite lots of mains switching.
hpasternack's location is "New England". I'm there (fringe). Indoor relative humidity is 24%, despite a cup of water on every hot air outlet (yankee humidifier). In this weather, I don't work with anything more sensitive than a 6L6 without first boiling a big pot of pasta. And of course take off any silk or nylon lingerie. And be very careful the order I handle devices and objects. Once! If I need repeated handling, that can wait for Spring.
EDIT: BA humidity at this moment is 30%. Very nearly same as here! That is outdoor, but at 76/24 degrees you won't be heating it much.
Last edited:
23% RH here in the man cave today. I use a lot of mosfets in my designs. I have a grounded ESD mat, grounded solder irons, wrist strap and a humidifier I turn on when I am about to handle mosfets, jfets and digital transistors (ones with integrated resistor networks and are very ESD sensitive)
The most important rules with MOSFETs is no nylon clothing/carpets and _never_ handle the gate pin. (And of couse live somewhere damp)
Test them in circuit with a 1m resistor from gate to negative rail.
If they still provide the correct current your meter is incorrect
They won't work at all if the gate is destroyed.
If they still provide the correct current your meter is incorrect
They won't work at all if the gate is destroyed.
Last edited: