Film-Tech Forum ARCHIVE: Diode testing, can I test without removing??

This is topic Diode testing, can I test without removing?? in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=008676

Posted by Frank Dubrois (Member # 3042) on 04-29-2007, 02:23 PM:

I have asked that question about 100 times to various techs, and I seem to get both yes and no answers. I am looking for someone who has tested diodes (Christie SLC 3000/4000) without removing them. We don't have a spare rack to swap out, so when a diode goes out, we are left to removing/replacing each diode until we find the bad one. This is time consuming simply because of the enduring pain I go through on my knees, on tile floor, scratching up my hands and arms digging into the lamphouse taking out each individual diode. It excited me to hear that you can actually test a diode without taking it out. IS THAT TRUE? If so, HOW do I test it? Even if I could narrow down the bad diode to one of two, it would save a huge amount of time. Anyone who has/does test diodes without removing them PLEASE let me know how it's done before we lose another diode. Now, before I'm told this by everyone, I KNOW all the diodes should be replaced when one goes, but I am looking for a quick way to test and replace the one bad diode to get the show up and going as quickly as possible. Thanks for any help!

Posted by Gordon McLeod (Member # 33) on 04-29-2007, 02:29 PM:

You must disconect the lead at least so the meter only sees that one device not the rest of the stack or the filter

Posted by Steve Guttag (Member # 268) on 04-29-2007, 04:40 PM:

You will see that each diode has a lead going up to connect with its mate on the termination board. For the smaller rectifiers, there is a cable coming straight from the diode to the stud on the termination board...on the big ones the diode lead goes to a copper strap first.

Either way you need take the lead off so it isn't connected to its mate (or the termination board) but you don't have to physically remove the diode...then test each diode using its lead and the stack it is connected to...pretty simple really.

Steve

Posted by John Walsh (Member # 168) on 04-30-2007, 08:28 AM:

Practically speaking, you have to isolate at least one of the diode's leads.

Impractically, there are only two things I can think of. Thirty years ago, I repaired PC boards and we had a piece of test equipment from General Instrument called a Tracker. It had two sharp-point probes which would be placed "across" the lead of a device, usually a DIP-style IC. You would hold one probe at the top of the lead where it enters the IC, and the other at the bottom. The IC lead is now just a shunt and the Tracker would read the current. It was a very finicky piece of test equipment, but after a while you could tell if the IC was sinking too much current - without having to unsolder the IC. This kind of test equipment would not really work with a xenon DC power supply because it was only meant for steady-state DC (not pulsating DC) and couldn't handle the high current. BUT a very sensitive meter might be able to show a difference between a good and bad diode while the supply was running.

The other idea is to use an oscilloscope to see each diode conducting. The trick though is to find out which of the 'humps' shown goes with what diode.

Nether of these ideas are good, and would probably be more trouble than it's worth to set up. But maybe someone could add a practical spin to them.

Posted by Tim Reed (Member # 137) on 04-30-2007, 08:55 AM:

Idea for ultimate console: metering in series with each diode, feeding display on panel! [Razz]

Posted by Dave Macaulay (Member # 813) on 04-30-2007, 09:57 AM:

I think the only reliable test is with the diode isolated.
I used something called a "Huntron Tracker" years ago, it had a little CRT and showed the characteristics of a semiconductor junction you connected it to. We used it for in-circuit troubleshooting transistors and ICs, maybe it would show a connected diode but I'm not sure. You can get them on Ebay for under $100.00.

Posted by Steve Guttag (Member # 268) on 04-30-2007, 01:05 PM:

Tim...look at MiTs rectifier...LEDs on each diode to determine its status...in fact, they are not diodes at all...SCRs...thus no contactor either.

Steve