Remember to click G+1 if you like a post... It will make me happy :-)

4 March 2015

How to test a JFET

For the past couple of days I've been battling a problem with my latest project to build a Piezo Buffer. When we last spoke I'd got the prototype working on a breadboard and I was on the verge of experimenting. It went downhill pretty quickly from there...

Do you remember my impromptu experiment on Sunday where I tried to get an LED to light up without exploding? I only managed it once, but that was all the encouragement I needed to begin wondering whether I could add an LED to my Piezo Buffer circuit.

My biggest fear was that I would break my Piezo Buffer circuit in the process. With shaking fingers I made the required adjustments and what do you know, the bloody thing worked! Ha ha. I was bouncing off the walls with happiness!

That was until I started experimenting with different resistors and everything went dead on me. DAMN! What on earth had I done?

What happened next was hour after hour of debugging. Try as I might I couldn't find a fault. Eventually the penny dropped... perhaps one of my components was broken. I'd destroyed the JFET on earlier attempts. Surely this was the most likely one to have been broken. But how could I test it? Read on...

My moment of revelation came from watching Stan Gibilisco's fantastic video above where he explains how to test a JFET. I love Stan's accent - he certainly is an engaging speaker... and he knows his beans.

Having watched this video I picked up just enough information to set me off on my own experiment. At the end I looped back round and I can see now that Stan was right all along. ;-)

Watch the video for the "proper" explanation. I'm going to run through what I did just so you can see how to do this for yourself.

Before I go too much further let me tell you that as part of my research this week I've discovered that you can buy component testers. The one pictured above is available on ebay as a kit. I've ordered one. I don't know how good devices like this are, but what a great idea!

In theory, it can be used to test JFETs and all manner of other components. I'll report back on this some time in the future.

Let's stick with the long-handed way of testing JFETs for now...

You can see here that I have 5 JFETs. I've marked them 1 to 5 and stuck them in a breadboard for easy testing.

Behind is my multimeter. This is what I used to test the JFETs. In a nutshell, I measured the resistance between each of the pins in both directions and analysing the results I've been able to figure out which JFETs are good and which JFETS are bad. 

Above are my results from my tests. Let me explain how to read them...

First note the top left picture of a JFET with the pins marked D(rain), S(ource), G(ate). Pay attention to the "D" shaped body. We're looking at the flat side and the pins are ordered: D-S-G.

In the tables to the right you can see that I'm measuring the resistance between each pin. D-G means Drain to Gate; S-G means Source to Gate; and D-S means Drain to Source.

B-R means Black to Red; and R-B means Red to Black, and this is referring to the colour of the multimeter probes. So I'm measuring the resistance first one way and then the other. This is important as you will see soon.

Note: I'm testing N-JFETs (i.e. negative-JFETs). This same method should work with P-JFETs (i.e. positive-JFETs), but I think you'd expect the reverse of the values.

On the multimeter I had to flick between the 20k and 2000 ohm settings in order to get all of the measurements with these particular JFETs.

The results are interesting. Remember this bad JFET I showed you in a previous post? I blew it up without even trying very early on in the project. This is JFET number 1 in my test.

You'll see that it is measuring a resistance between all of its pins, any which way you measure it. BAD! This tells us that current is flowing. This shouldn't be the case. JFETs are specifically designed to not allow current to flow from gate to source or gate to drain unless there is DC flowing though source and gate. In the "off state" (i.e. with no DC) no current should flow.

See the results for JFETs 4 and 5. The Drain to Gate results are showing resistance only one way. Aha! 1 (or "infinity") means, no resistance and hence , no current! Sure enough, so too do the results for Source to Gate! These JFETs are working and I've proved it by plugging them into my Piezo Buffer circuit!

I'll quickly call out the Drain to Source results for the working JFETs. These measurements were jumping all over the place on the multimeter display. It seems impossible to get a fixed measurement on these with the working JFETs whereas the broken JFETS were easy to measure. Interesting.

And there you have it... a fairly painless way to test JFETs. What worries me now is how easy it has been for me to destroy 3 JFETs. I've only got 2 left. Maybe I should build my buffer quickly before I blow these too? :-O

No comments:

Post a Comment