Convenient USB based transistor tester - diyAudio
Go Back   Home > Forums > Design & Build > Equipment & Tools

Equipment & Tools From test equipment to hand tools

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 20th January 2014, 12:35 PM   #1
diyAudio Member
 
Join Date: Apr 2012
Location: Florida & France
Default Convenient USB based transistor tester

Here is a new thread, because I haven't seen one quite like this before.

I am proposing a project to design a BJT transistor tester that can do reliable matching and that can handle small signal all the way to power transistors. Also it should be able to do a non-destructive test for secondary breakdown on power transistors.

What I have in mind should not cost too much, be linked up to a computer, via USB preferably, and all the processing would be software based. It should allow to save all the test result data, do comparisons of multiple parts, such as combining curves into a single plot, and each part having its test data saved, could be recalled individually to be compared with others at any time.

I envision a tester with a sizable heatsink, on which both NPN and PNP could be tested at the same time, and in pairs (of each). For example a pair of NPNs could be tested to match them, or a pair of PNPs, or 2 pairs of each at the same time, so not only 2 NPNs and 2 PNPs could be tested for matching together, but also at the same time the pair of each sex could be tested for matching as a whole set of 4 parts.

I would put a good temperature sensor on the heatsink at each transistor location, so the test temperature could also be known, recorded along with it, and also the temperature of all present parts could be watched for equalization. Since all the parts would be on the same heatsink, they should eventually come to close to a temperature equilibrium so all parts would then be tested together at the same time at a similar enough temperature.

A "pre-test" delay could be afforded and monitored by software, giving the parts a certain somewhat minimal power to dissipate for a while, to bring them to a certain level, so they could be tested at a determined temperature.

I believe this should not be too difficult to achieve with a simple enough circuitry and all handled by software.

Both the secondary breakdown and matching could be done at a determined temperature.

I would not want a design that uses exotic parts or overly difficult or expensive parts. It must be easily buildable by any diyer, anywhere on the planet.

Being able to test power transistors with a different package, by sets of up to 4, and on the same heatsink, will require a little thinking, but I'm sure we can come up with something clever.

To keep it cheap and simple on the electronic side, I think making use of an arduino board might be a good idea, but there may be other choices usable. The arduino would allow easy USB without having to design all that interface circuitry, and it should have enough I/O ports to handle all of the tester's function. If an arduino nano isn't enough for example, there are larger models with more ports.

For the small signal parts, there could be various sockets to accommodate the various packages, and at least they should also be tested in pairs at a minimum, but more could be tested for matching at the same time.

Anything with a package mountable on a heatsink should be. Including the TO-126, TO-220, etc... Which might make for a trick heatsink design, but worth the effort.

I have come across a very interesting design, I think by a chinese guy, that does a good part of what I'm looking at doing. His is linked to a computer, windows only, and does plot curves, for matching, and saves the data, but only one part at a time, doesn't do any non-destructive second breakdown test, doesn't use a heatsink, and so it doesn't allow tests at pre-determined temperatures.

There is also that excellent (a bit older now) application note AN930 from motorola, that is only for a second breakdown tester, non-destructive, but that is only aimed at being used with an oscilloscope, so it doesn't allow recording of any data, doesn't allow for matching and doesn't work with a computer.

I believe this kind of tester would allow easy and reliable testing/matching, hopefully not cost too much, offer more features than available out there, and would be likely to suscitate quite some interest from many diyers (like me first).

I wouldn't be able to handle all the design by myself, so I'm hoping for a collaborative work.

And perhaps, a well designed device with all the trimmings could also become something that could be sold on the diyaudio store...
  Reply With Quote
Old 20th January 2014, 12:46 PM   #2
diyAudio Member
 
Join Date: Apr 2012
Location: Florida & France
Default motorola application note AN930

While I'm at it, here's the AN930 from motorola, so at least a good and important part of the design can be based on it.
Attached Files
File Type: pdf AN930.pdf (898.7 KB, 69 views)
  Reply With Quote
Old 20th January 2014, 01:45 PM   #3
diyAudio Member
 
Join Date: Apr 2012
Location: Florida & France
This guy (locky_z) here, did a nice job with his curve tracer:

DIY Curve Tracer for PC

It's in the spirit of what I have in mind more or less.

He's using a pc lpt port and it's windows only. Plus it's just a curve tracer, although it should be good enough for matching. It doesn't do power transistors in the way I describe and only does one part at a time. But this can also be an inspiration.

I would make something more universal, in regards to the computer being used, so it could work on linux/unix/mac as well and use a more universal connection: USB

I wouldn't bother with mosfets, jfets or whatever types of fets. Those should test quite well on his current design, although it doesn't work on anything else besides windows. This can keep the design simpler by restricting to bjts only.
  Reply With Quote
Old 20th January 2014, 02:26 PM   #4
diyAudio Member
 
AmpliFire's Avatar
 
Join Date: Oct 2011
Location: Tokyo, Japan
Good to hear you want to start something like that, you can count on me for crowd funding . I have Locky's curve tracer, and couldn't live without anymore. It is USB by the way.

Not including FETS would be a BIG miss. Matching is a key use/objective of such devices for us audio folks, and FETs are as much used and matched as BJTs, think of differential input stages for example. So please don't discard that.

What I would be looking for in any case is a real high voltage / high current tester to really test power output devices, say up to 100 volts and 10 amps. Yep, that will be a very hefty transformer.

Mounting and measuring multiple DUTS at the same time needs some thinking about the casing touching the heatsinks (TO-3); and I agree with you that comparison (matching) should be done under strictly same body temperature, hence solid heat sinking and fixation. Lucky's curve tracer is missing this and I have considered several times to install sort of quick-clamp heatsink, as the DUTs heat up during test. Though, why measure the temperature if all devices are then measured on the same heat sink (same temperature), if you cannot change the heatsink temperature anyway?
  Reply With Quote
Old 20th January 2014, 03:01 PM   #5
diyAudio Member
 
Join Date: Apr 2012
Location: Florida & France
Quote:
Originally Posted by AmpliFire View Post
Good to hear you want to start something like that, you can count on me for crowd funding . I have Locky's curve tracer, and couldn't live without anymore.
Glad to "hear" that!

What he's done is very nice and helpful, but I need more and I want it more universal as far as platform. I do not use windows.

Quote:
Originally Posted by AmpliFire View Post
It is USB by the way.
I forgot about that, I was thinking about his first incarnation.

Quote:
Originally Posted by AmpliFire View Post
Not including FETS would be a BIG miss. Matching is a key use/objective of such devices for us audio folks, and FETs are as much used and matched as BJTs, think of differential input stages for example. So please don't discard that.
Well, I think this can be done if we make it "modular", so those who don't need to work with fets don't need to build that part of it. I'm quite sure we can come up with some clever way of doing this.

Quote:
Originally Posted by AmpliFire View Post
What I would be looking for in any case is a real high voltage / high current tester to really test power output devices, say up to 100 volts and 10 amps.
Absolutely!! My thoughts exactly. We should be able to test those modern types with voltages over 350V.

Quote:
Originally Posted by AmpliFire View Post
Yep, that will be a very hefty transformer.
Maybe not! I would go for switching psu for that. Much smaller and likely cheaper if we can find a way to use commercially available units, if feasible.

But then again, there are ways to make good switching psus without spending too much. I used ICs before that are specialized for this, the SG3524, which make it possible to make them with a fairly low part count. The one thing I don't like about switching psus are those "wierd" transformers. They are small and light, but I'm not sure they can be found easily...

Quote:
Originally Posted by AmpliFire View Post
Mounting and measuring multiple DUTS at the same time needs some thinking about the casing touching the heatsinks (TO-3);
Yes, the issue doesn't come up when we're only working with same sex types, but we must isolate when mixing sexes unfortunately.

BUT!!! I may be wrong, but there may still be a way, if someone can confirm this: I think we might be able to put all the collectors' potential together, so we could forget about the isolation. This isn't possible in an amp, although some amps are designed with all collectors together, even in a complementary topo, but those are either at ground or the amp's output...

It would really be a plus if we avoid insulation, not only it would reduce the thermal resistance, it would also make it much easier and more convenient to mount and unmount parts without insulators or worse, grease!

The one big thing that would remain to be careful about if no insulation is needed, is the potentially very high voltage present during testing on the heatsink. So either: do not touch during testing! or put some type of mechanical shield on that sink in case someone still touches it during testing.

Quote:
Originally Posted by AmpliFire View Post
and I agree with you that comparison (matching) should be done under strictly same body temperature, hence solid heat sinking and fixation. Lucky's curve tracer is missing this and I have considered several times to install sort of quick-clamp heatsink, as the DUTs heat up during test. Though, why measure the temperature if all devices are then measured on the same heat sink (same temperature), if you cannot change the heatsink temperature anyway?
It's a must! Those tests are temperature dependent and to avoid doing a half way job, the temperature must be taken into account.

My reasons for temperature sensors on each part are:

1) tests can be done at various pre-determined temperatures
2) the tests results can be recorded for each temperature

It is quite possible to obtain those pre-determined various test temperatures, as I listed in my first post in the wanted features, as we can have the devices dissipate some power before testing, up to the pre-determined wanted temperature, wait for it to stabilize and then quickly do the tests while all the parts at at that wanted temperature and before it changes.

I know all this seems a little bit of a tall order, but nothing that can't be done and it doesn't require very advanced engineering to achieve it, and it's a really useful feature that I really want in the tester. Plus since we make a software operated tester, we only need the basic electronics and do most of it with software.

This is going to be a very cool project I think. And I will be very happy with such a device when we make it work.
  Reply With Quote
Old 20th January 2014, 03:21 PM   #6
diyAudio Member
 
Join Date: Apr 2012
Location: Florida & France
The features wanted are coming together.

We can make it modular so the testing of parts other than bjts can be done on a separate optional module attached to a main module, so those like me who do not use any kind of fets can opt not to build the option.

Perhaps the bjt testing can itself be a an optional module like the fet testing one, and the requirement would be that at least one module be attached to the main one to function properly.

If done right, other modules could be designed and extend the features to test other types of parts, and perhaps more kinds of tests could be performed later if a updated existing module was designed to add more functions.

One thing to look at, is how to test the power transistors in a way that has all their collectors together, so no insulation would be needed on the heatsink(s).

It should remain as convenient as possible, so the power transistors should be easily mounted, only plugging into a socket and with screws to keep them tightly in place.

It's easy to have safeties built-in regarding overheating, since all parts locations on the heatsink would have a sensor.

Being able to test even the highest voltage parts would be great, like those with Vce of 350 or 400V. Obviously this will require a hefty psu for testing at high currents and being able to reach the high secondary breakdown voltages on such parts. To avoid a huge transformer, a switching power supply seems in order.

The parts' saved test data could be used at a later time for comparisons, so parts matching doesn't have to be done right then and there on the spot with the parts at hand, and this means parts could possibly be matched from remote, for example someone with this tester could test some parts on hand, and share this with someone else, also with the tester, who could be far away, and other parts could be tested and matched...

There is nothing like this on the market that I know of, and I think we can easily get addicted to such a thing.
  Reply With Quote
Old 20th January 2014, 03:36 PM   #7
Krisfr is offline Krisfr  United States
diyAudio Member
 
Krisfr's Avatar
 
Join Date: Nov 2007
Location: near the House of the Mouse
And just think, I bought one of these for 24 dollars...

http://www.ebay.com/itm/New-Transistor-Tester-Capacitor-ESR-Inductance-Resistor-LCR-Meter-NPN-PNP-MOSFET-/251334183807?pt=LH_DefaultDomain_0&hash=item3a84af 4b7f

And it runs on a 9 volt battery...

That schematic looks like a great place to start, I will help in any way I can...
Attached Images
File Type: jpg $3.JPG (102.3 KB, 234 views)
__________________
Thanks
Never give a Systems Analyst a screwdriver or hammer

Last edited by Krisfr; 20th January 2014 at 03:42 PM.
  Reply With Quote
Old 20th January 2014, 04:06 PM   #8
diyAudio Member
 
Join Date: Apr 2012
Location: Florida & France
Quote:
Originally Posted by Krisfr View Post
And just think, I bought one of these for 24 dollars...

http://www.ebay.com/itm/New-Transistor-Tester-Capacitor-ESR-Inductance-Resistor-LCR-Meter-NPN-PNP-MOSFET-/251334183807?pt=LH_DefaultDomain_0&hash=item3a84af 4b7f

And it runs on a 9 volt battery...

That schematic looks like a great place to start, I will help in any way I can...
Great!

I have something similar to this also, it was less than $20 when I got it on ebay.

It's very convenient and it does a lot of things, but it won't do all those things I'm looking for, so that's why I'm digging into this.

I want it linked to a computer, and besides locky_z's excellent tester, I haven't seen anything out there that does what I'm looking for.

I'm sure a good collaborative work can work well and we can achieve this.

I'm afraid some commercial entity will later grab on to this idea and put something out on the market, but that may take a while.
  Reply With Quote
Old 20th January 2014, 04:41 PM   #9
diyAudio Member
 
Join Date: Apr 2012
Location: Florida & France
Here is what I'm thinking we could do:

- A main root module that can receive one or more optional module attachment
this main module could very well be for example an arduino, since it already has the usb port built-in, is very cheap and has communication protocol already built-in as well. I'm not yet familiar with arduino, so I'm not sure how a separate software could use the arduino built-in communication protocol for its own purpose and properly interface with the arduino software loaded into it.

- a switching psu with enough umph to provide all the various needed voltages for the modules and the test voltages and currents.

- a module for bjt testing, with heatsinks and all that.

- a module for fet testing. the mosfets would also need a heatsink if they are to be tested at high power levels.


A good method for attaching modules to each other has to be devised so more than one optional module can be attached to the same root module. Avoiding bad contacts is not an option!

I have a couple of arduino nano boards and also an uno, but I haven't yet done much besides basic stuff on them, mostly tutorials to learn the abcs of it, but nothing yet to know enough to make full use of it.

If someone has a better idea than using arduino as a base, please share your thoughts.
  Reply With Quote
Old 20th January 2014, 06:02 PM   #10
Krisfr is offline Krisfr  United States
diyAudio Member
 
Krisfr's Avatar
 
Join Date: Nov 2007
Location: near the House of the Mouse
The "secret Sauce " will be the SOFT ware. NO hard ware is good with out the Software. There are ways to lock the firmware up. Be reasonable and you will win.
__________________
Thanks
Never give a Systems Analyst a screwdriver or hammer
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Hickok 1890M transistor tester trobbins Equipment & Tools 0 23rd June 2013 02:20 PM
useful transistor tester siliconray Equipment & Tools 0 25th November 2011 01:48 AM
Transistor tester wgpet74 Swap Meet 1 6th October 2011 10:24 AM
Transistor Tester Current Levels Stocker Parts 87 22nd December 2010 03:30 AM
transistor tester janey Solid State 0 4th January 2002 02:44 PM


New To Site? Need Help?

All times are GMT. The time now is 11:55 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2