Diode Tester for Current Leakeage. / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

Diode Tester for Current Leakeage.

Move Thread

LAN_403

si2030

Thu Jul 17 2008, 04:48AM

Registered Member #1571 Joined: Wed Jul 02 2008, 03:26AM
Location: Bendigo Victoria Australia
Posts: 44

Hey there,

This is the first project that I am placing here and to be honest I am a babe in the woods with respect to electronics. I am getting better but there has been no bolt of lightning to make it all understandable.

My interest is ultra high voltage. So from "whats a capacitor" to having to understand the basics of V= IR etc and just talking to people and doing a LOT of reading I have started to understand.

This project is part of a much larger and time consuming project.

It was pointed out that I really needed to make sure that my diodes in each string for a Cockcroft Walton Multiplier must be similar. If one leaked more than the others then I might end up with a firey outcome and lots of money down the drain. That was the start of this project.

OK, the first thing was to have a chat with my good friend and electronics genuis Andrew Pullin. In a day he knocked up this design for me to test the leakage of diodes.

One of the things I found using the breadboard was there was too much fluctuation. So Andrew spent another hour putting the components into a tin. This RF shielded the system and as long as temperature was stable I could sort out the diodes according to their leakage.

I ended up with diodes being in different piles

The circuit consisted of the following:

1. A calibration section. I needed to know I had approximately 1 nano Amp going through the system so I could set my meter. To do this I used a voltage divider to obtain 100mV via 220k/1k8 divider. This was then fed into the system which has a 100Megohm resistor to provide another resistive divider network. The signal out of this divider network was DC and around 1nano Amp.

2. This was fed into a buffer using an LF353 opamp.

3. This was then fed via a couple of reisistors and capacitor decoupling to an op amp inverting amplifier using another LF353 (the other half). it was then fed into a resistor, a pot and then, as a negative value, into the negative end of my meter which has Full Scale Deflection of 50mAmps.

I can use the pot to adjust the 1 nano amp deflection to a desired swing and if the diode leakage where to be say 2 nano amps then the deflection will be twice the 1 nano amp deflection.

The important thing is not so much the value of the leakage but leakeage in relation to the other diodes.

Once I had this I could pick my diodes for each string and be fairly confident I didn't have a dodgy diode in the string.

For power to the diode, I used a Cockcroft Walton Multiplier from an old xray machine that Andrew had junked. this supplied approximately just under 15Kvolts out for 24 volts in. Once set I had to run through all the diodes in one go so nothing changed including ambient temperature. Andrew has said you might want to find a power supply from an old photocopier.

I also used one of my large capacitors at the output of my small Cockcroft Walton Mulitplier.

Here is the schematic of the system. This works in LTSpice and I would be happy to email the file to those that need it. I couldnt find a place to save the file on this site and I dont have a website myself unfortunately.

1216269127 1571 FT0 Diode Tester Schematic

Anyway thats it. Hope this helps others if they want to test diodes. Hey maybe you can improve.

Again A big thanks goes to Andrew for this as with out him I would still be struggling with the most basic principles!!

Cheers

Simon.

Steve Conner

Thu Jul 17 2008, 09:59AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

We take files with a .zip extension, so just zip your stuff. Also, with avalanche rated diodes, there's arguably no point in sorting them, as they distribute the voltage evenly by avalanching.

si2030

Fri Jul 18 2008, 03:52AM

Registered Member #1571 Joined: Wed Jul 02 2008, 03:26AM
Location: Bendigo Victoria Australia
Posts: 44

Sorry, I should have considered the option of a zip file.

If you want to see this work, then you will first have to have LTSpice which you can download for free here... LTSpice

By the way there is a Yahoo group for LTSpice that is really very good.

Once you have installed LTSpice then download the zip file, save it to your computer and then unzip.

]diode_tester.zip[/file]

It should run OK.

As for avalanche diodes, I have to be honest and say I only just realised these were an option recently. Unfortunately I have no way of knowing whether the diodes I have are Avalanche as I have a scant datasheet ]2clg1020015.2clg1020020.pdf[/file] and I purchased them from China. The company Tjsemi it appears is massive and while I got samples in this case they really deal in very large quantities and I think wont return my emails unless I am a now a company...

Simon

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob