How do i test FETS without removing?

[NeilP]

Well after less than a week of use my 18 FET Lyen edition controller has died..

have ordered a new 18 FET from Max Kogan...sent priority shipping via Fedex...shipping more than controller but I want to be back on the road ASAP

Anyway
Lyen suggested testing the fets for short circuit, while still on the board, going from leg to leg, but not sure quite what I am expecting to see.

Well when I say that ..I know a short will obviuously give near on zero resistance..but as we are testing FETS here, then they should conduct in one direction any way...and sticking voltage across the pins from multi meter is going to open/close the FETs is it not?

So how do i do this? I imagine the best I will be able to achieve is which bank of FETS the problem FET is in...if it has not taken other out with it...along with drivers for them as well.

So what polarity for the continuity tester leads? voltage on this tester at the test leads is 2.9 volt...so that is going to open the FETS anyway isn't it?


Edit:

I'll leave this here as the description is more fitting...at least for me to search for in future...but found a similar thread

http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=26927&p=395404#p395404

also recent.

will continue with my results of tests I have done over there to help the other chap as well

 

[Electroglide]

I think if you search, you will find that others have posted how to test fets quickly. Some really quick checks can be in the form of looking to see if the body diodes are present, by using a diode checker from phase wire to Bat+ and from Bat- to phase wire. This doesn't proved they are good, but it sometimes pinpoints which phase to look at. It will howver not help if the fet is blown open in a parallel bank. If you have a shorted fet, it is hard to tell which fet in a parallel bunch has actually shorted. You may be able to tell by looking at the resistance seen into the gate to source of the mosfet too. If you have values in the 100 ohm range, then it is damaged or the gate driver is damaged. Normally I replace all fets in a blown parallel bank because you never know if one of the others have taken a hit. Other ways are to check for differences in impedance when measuring the phase to phase and phase to battery wires. If phase to phase measures are different, then you have an issue someone on the bridge. The phase to battery wire measurements are less of an indication unless you have another good unit to compare against. Good luck with it.

 

[NeilP]

I did a search ...but as with al search engines...not a lot of joy...Found one thread as I mention above, where I have posted my results...I found that thread just by scanning the forum though. nothing useful came up using the search function

Would you mind having a look at my post on the other thread and seeing the results I posted there..save duplicating the conversation, and it may hlep the other chap too

viewtopic.php?f=2&t=26927&p=395404#p395404

Cheers

 

[grindz145]

FETs are hard to fully test in circuit, but you could apply a gate charge and a load and measure the voltage drop, that should be pretty easy to do.

 

[number1cruncher]

Neil,

Here is the link to the way I tested my FETS. http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=26949


I have not had the chance to finish repairing the controller, but will try to get to it this week and post back my results.

Basically, I set my DMM to ohms and probed between each phase to Vcc, phase to GND and each FET drain to source. The drain and source pins are on the right side when the FET is facing you(edited to add: although I tested from the bottom of the board, which puts the pins on the left :wink: ). From my understanding of what Jeremy Harris and Fechter have posted in the past, the value on each of these probes should be around 10K Ohms.


Good Luck!