I am going to make a MOSFET tester (really only concerned in Logic Level versions).
Something along the lines of this.
An Arduino Pro Mini with a SSD1306 OLED display will show voltages, amperage and Rds(on).
Open to suggestions on what to add to the tester.
Even if you use Logic Level versions, you still need to know the gate voltage at which a certain current can be turned on. I think that you need a PWM with RC-filter for the gate.
Measuring a few millivolts (drain-source) will not be easy. Are you adding a OpAmp ?
Making good contact with a temperature sensor is also not easy.
Your design should use kelvin contacts wherever needed. How are you going to determine the die temperature as the threshold changes with temperature. Do you have the analog skill set this will require? When we measured the Gate threshold voltage of a MOSFET, we shorted the Gate to Drain then set up our ID at 250μA. At this point we could monitor the voltage difference between Gate-Source. Please note the VGS(TH) has a negative temperature coefficient. After you build this how are you going to assure calibration?
Thanks for the comments.
This is what I have settled on.
Will add an Arduino Pro Mini and a SSD1306 OLED display.
The display will show Vgs from the 10 turn pot, Vds to see if the MOSFET is fully ON and Rds(on) as calculated from the current flowing divided by Vds.
I like @Koepel 's idea using PWM instead of the 10-turn pot. Using Timer1, could get 12-bit resolution at 3.9 kHz PWM frequency, therefore near 1mV resolution. Very little ripple if you use 10K/1μF, fast enough response (cutoff freq at about 16Hz). Test here.
Another advantage is being able to automatically program a series of voltage levels and generate a response curve.
This is how I did that:
but for a mosfet with a very high input resisitance you wouldnt need an op amp circuit.
The advantage is it becomes hands-free. Press "go" and it measures all your parameters against a changing Vin
Have you any experience with the rail to rail single supply (3V to 15V) op amp LMC6482 ?
No Larry, I was going to get one but its very similar to other rrss op amps I've tested.
I'll point this out as a general note - "rail to rail" op amps can not maintain rail to rail outputs if supplying current.
So should you remove the R5 & D1? Of course if D1 is not plugged in for testing it wont change the "gate" voltage; and if youre testing the LED the IV curve will still be valid.
I ordered 50 of the LMC6482, it will be interesting to see how they work.
Supposed to be improved over the TLC272s that I am use to.
Yes, LED D1 is installed only when testing LEDs, not when testing MOSFETs.
I like that the LMC6482 can be used up to 15V.
Supposed to be able to drive RL = 600 Ω with a output voltage of 14.1V
Will be fun to see what traces I can get with my digital storage scope. 
Will breadboard everything before making the PCB.
Have you see “‘Ardutester “ worth a google
And others projects are worth a look for inspiration
Got the LMC6482s.
These are great, with Vdd at 5V, we get full 5 volts out !
Some results on the solderless breadboard (have to make up a hard wired version breadboard).
This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.
.
