Self made Diode Tester or Fluke Scope Meter Component Tester...

Only because you didn't waste the 10 seconds to type it into google.

Can

The concept of a curve tracer is trivial. You described exactly how to do it above. The devil is in the details...variable voltages, current limits, current step generators, and on and on and on.

There are a zillion curve tracer pages. Google is your friend.

I went thru the exercise of trying to build a curve tracer by using a uC to generate fixed voltages and currents, then the A/D to read the result and plot the points. Works in concept, but the device temperature changes between measurements and the curves came out all wonky. Ain't no substitute for a good ole Tektronix 576. Does everything you'd ever want, including helping you drive in the snow if you stow it in your trunk...assuming you can lift it that high without getting a hernia.

Anyone knowing how a curve tracers curves should look like for a Power N-FET transistor??

regards Geir

Archie?

There are better curve-trace adapter designs out there... IF you really are intent on doing serious testing. ...Jim Thompson

Den 02.01.2011 22:28, skrev Jim Thompson:

I want to have a quick and dirty way to check if components are okay or not. However back in those days (70-80s I only worked with bipolar transistors, hence I wondered how the curves would look like with a FET, or do I risk to fry the gate..?

Geir

Jim Thompson wrote in news: snipped-for-privacy@4ax.com:

if you can find a Tektronix 7CT1N,7B53A,and 7000 series mainframe,you'll have a nice small-signal device curve tracer,up to 300 Vcc.I had one on my bench at Tek. There's also the 5CT1N and 5000 series Tek mainframe,but the 5K stuff was el-cheapo compared to the 7K.

I saw a really nasty 7603 at the Ocala Hamfest recently. The guy didn't know how much he wanted for it, and got really pissed when I offered him $25. There was no sign of any calibration stickers, or that he had even been plugged in for decades. The plugins were really beat up, and nothing to brag about.

There's a thing called the Huntron Tracker. I never had one, but they appear to be much like what's already been described. They're SERIOUSLY OVERPRICED, but a little research might enlighten you about what kinds of tests are useful. If I understand the manual to your scope, it can do a similar function up to 3V.

The Tektronix 5CT1/7CT1 are nice plugin curve tracers. Two drawbacks. They're expensive cause there are not many for sale. The mainframe is huge. Good idea if you already have the mainframe. If you don't have the mainframe, not so much. I still have one sitting on my bench, mainly 'cuz I'm afraid I'd hurt myself trying to get it up into the attic. Tek 576 is the granddaddy of curve tracers. Can do 10A or 1500V, thankfully, not at the same time or I'd probably be on fire by now. I use mine more for blowing shorts out of NiCd batteries than for testing transistors.

For bipolars, your ohm-meter will tell you if they look like two diodes. If so, they're probably good. Yes, there are many failure modes that don't show up that way, but as a percentage of cases you'll encounter, one or both junctions are either blown open or shorted. Of course, everybody wants to test in-circuit. Can sometimes do that if you know the circuit. Typically, you don't and have to take 'em out anyway to be sure.

FETS are a bigger problem. Other than checking for shorts, I've not found any useful way to test 'em with less than three test leads and out of the circuit. Curve tracer can help there...assuming you can even determine which lead is which element of the FET.

My TV quit, no sweep. Turns out that virtually every circuit in the thing had some kind of interlock with the sweep, so nothing was running. Was a fet in the horizontal oscillator. Tested the thing in circuit until I was blue in the face. Finally ran out of options, gave up and pulled it out. A tiny bug had committed suicide between the leads. Bug Juice has some interesting nonlinear properties. Didn't conduct enough to see at low voltages, but at higher voltages, the pins were shorted. Not sure a curve tracer would have helped in-circuit.

A friend came over once with some RF power transistors out of his 450 MHz. transmitter. They all looked the same on the curve tracer at normal working currents. A better indicator of a bad transistor turned out to be some small Collector-Emitter leakage that you could read on an ohm-meter.

Bottom line is that there's no easy answer.

Any comment on this tester? Adequate in an hobbyist's shack...?

Geir

I couldn't get the page to load. I then found the following comments dated Nov. 25, 2010:

"Except that they stopped making it about a year ago and show no sign of manufacturing anymore. I've emailed them about it and requested a response on the DIYAudio forum, but it seems to be a no go! On the DIYAudio forum someone told me about the Elektor 'project' and it seemed quite attractive until I received a whole bunch of reports from people who said that they had problems with it."

The page that included the above comment also had this link, which does load:

I'm not sure what to make of these things, but there it is.

I have a small circuit I built decades ago that I built into a 35mm plastic film container (with a BJT socket mounted) that had two LEDs, one red and one green. No micro, just a simple oscillator and a few parts. I'd put the BJT into the socket and if it is NPN the red LED lights, if it is PNP the green LED lights, and if it doesn't work for some reason then neither or both will light up. Never failed me. But it doesn't do curve tracing or tell you any quantitative things

-- just go/no go for the device and what type it is. I could provide the schematic, if that's what you'd like.

Jon

Jim Yanik Inscribed thus:

I've got an old 7000 mainframe in the garage, no plugins though. It must have been in there for at least ten years or more. Bet it goes bang when I plug it in...

Jon Kirwan Inscribed thus:

That sounds like it should be a toolbox item ! :-)

I found something similar to it on the web. It's on this PDF file's page 3:

Jon