Installing a FET

Thread Starter

JulesP

Joined Dec 7, 2018
383
Hi all,

I'm completing a circuit on a PCB to drive an electrolyser that uses a FET (IXTP62N15P) [Pic 1].

To check that it is not damaged by the heat from the soldering iron I put a mini tester onto each of the three leads before installing it (Pic 2) and then did the same after soldering (Pic 3).

The readout looks rather different despite my taking a lot of care during soldering including using heat sink clips and minimising the touch time to about 4 secs.

Can anyone say if the FET is still likely to work given the second readout?

Thanks

Jules

Pic 1.JPGPic 2.jpegPic 3.jpeg
 

dl324

Joined Mar 30, 2015
16,839
uses a FET (IXTP62N15P) [Pic 1].
There are many types of FETs. The one you're using is a MOSFET and you don't have enough solder on the leads.
clipimage.jpg
Can anyone say if the FET is still likely to work given the second readout?
MOSFETs are static sensitive. Did you handle the device properly?

As noted, in the general case, you can't measure a transistor in circuit.
 

Thread Starter

JulesP

Joined Dec 7, 2018
383
There’s plenty of solder on the reverse side where the pins come through. I wasn’t earthed but was not likely carrying much static.
 

dl324

Joined Mar 30, 2015
16,839
There’s plenty of solder on the reverse side where the pins come through.
If you had heated the joint properly, solder would have flowed through the hole. If you just melt solder on the lead, you might not have a good electrical connection.

I've had cases where I used insufficient solder paste when soldering surface mount components. Even after checking the joints with a magnifying glass, I couldn't find defects. The bad joints would fail after hours of use. Adding more solder fixed the problem.
I wasn’t earthed but was not likely carrying much static.
It doesn't take much; you don't need to see an arc to have enough voltage to damage the gate oxide.

If you're lucky, static sensitive devices will fail outright. If you're unlucky, they'll sort of work before eventually failing. I zapped a couple 2N7000. They sort of worked, but exhibited high leakage. Took me awhile to figure out that that was the case.

EDIT: I was grounding myself before touching the devices, but I was working in a room that was carpeted. I was sitting on the carpet (i.e. not walking around) and still built up enough static to damage them.
 
Last edited:

Thread Starter

JulesP

Joined Dec 7, 2018
383
If you had heated the joint properly, solder would have flowed through the hole. If you just melt solder on the lead, you might not have a good electrical connection.

I've had cases where I used insufficient solder paste when soldering surface mount components. Even after checking the joints with a magnifying glass, I couldn't find defects. The bad joints would fail after hours of use. Adding more solder fixed the problem.
It doesn't take much; you don't need to see an arc to have enough voltage to damage the gate oxide.

If you're lucky, static sensitive devices will fail outright. If you're unlucky, they'll sort of work before eventually failing. I zapped a couple 2N7000. They sort of worked, but exhibited high leakage. Took me awhile to figure out that that was the case.

EDIT: I was grounding myself before touching the devices, but I was working in a room that was carpeted. I was sitting on the carpet (i.e. not walking around) and still built up enough static to damage them.
Thankfully I’m working with a concrete floor. The solder hasn’t flowed all the way through as I was anxious of the amount of time that would be needed to do that but it’s well attached to the THT ring on the reverse side. I can add a bit more directly to the front side.
 

dl324

Joined Mar 30, 2015
16,839
Thankfully I’m working with a concrete floor.
You can build up static electricity by touching things with your hands.
The solder hasn’t flowed all the way through as I was anxious of the amount of time that would be needed to do that but it’s well attached to the THT ring on the reverse side. I can add a bit more directly to the front side.
You can heat the joint from the solder side and reflow the solder.

10 seconds is generally safe. I happened to have this datasheet in a viewer:
clipimage.jpg
 
The clothes you wear causes static buildup if you sit on fabric or varnished wood chairs and twist/shift while you work/handle components.
Use a grounded anti-static mat for your parts, they are commonly available.
I've been soldering for 42 years now, can't recall the last time I damaged a static sensitive part due to static discharge..
Most of my work these days concern static sensitive microcontrollers and MOSFET based outputs drives.
I often get a static zap as I exit a vehicle...twisting on the vinyl seat covers in dress pants builds up a charge real fast.
 

sparky 1

Joined Nov 3, 2018
756
The data sheet, should operate predictably in the recommended settings charted in the figures shown.
Not so bad range listed by symbol T. If the board is plenty warm by pre-soldering plenty of flux the soldering operation less than 8 seconds should be possible and with heat sink lowers the thermal conductivity, leads will not dissipate enough heat and the inside of the mosfet will not be affected very much if the iron clean tinned properly. Enough test runs right contact you will have an acceptable soldering rate and good solder joint.
 

Attachments

Last edited:
Top