transistor tester

but what if i want to build a tester using microcontroller, all i have to do is plug the transistor into the appointed holes, the lcd would display everything that i want... ie, hfe, which one is the base, emmiter, collector, is it faulty etc.

is it possible to write out our own programming and programmed the chip? any recommendations of using programming language? will VB works? or do i need a higher language?

thanks^^

Yes, it is very possible. May not be economically practical however. If you want to... because you want to. By all means go for it. Much experience and knowledge can be gained from building such devices. Many years ago I built a transistor tester which provided variable drive signals with input and output signals monitored on a 4 channel oscilloscope. Supply and bias voltages were adjustable and individually monitored. It was a thing of beauty and a terrible waste of time and money. (I'd do it all over again though).

To each his own.

if so, how did you manage to build it?

That's a long story, If you are serious - e-mail me and I will e-mail you with the details. Much too intense for this thread.

Hello blunt Del the cat

You also earned a good answer mark, so I gave you one.

Kind Regards....

Hello Guest,

As advised above, it's not really worth building a transistor tester.

The cost of making one, with all the multi-type sockets, and switch combinations, means it is far cheaper to use a replacement, or several.

In the early days of electron tubes = Valves for UK folks, RCA, Mullard, Philips and a few others made "electron tube" = "valve" testers, which worked well.

As the years went by, the alterations to valve bases, and the increasing complexity of the electron tubes, made the once useful "tube testers" became expensive "doorstops", and eventually either scrapped, or an exhibit in a museum.

A transistor is best tested in the circuit it is installed, and by experience and careful deduction, the faulty part is quite easily established, then replaced, without much more than a couple of multimeters, an oscilloscope, and perhaps a frequency counter.

More interesting instruments are needed, in areas like RADAR, High power electronics, or TV Transmitter situations, of course

Kind Regards....

Sparky, you've got it wrong again!!

As the number and type of valve bases increased the valve testers became evermore so complex... The AVO valve tester is still in use today for a huge number of people involved with audio valve amplifiers etc...

Just take a look on ebay do a search for valve testers and be prepared to be amazed at how much these testers sell for, usually hundreds of £'s.

I could go on but I doubt if its worth it....

John.

Hello Electroman

I actually had a valve tester years ago.

I do remember the AVO valve testers, always wanted one, but in those days we had "Import Restrictions", and unless you had "Overseas Funds", the AVO units were so overpriced, they were unattainable.

The Philips valve testers were easier to purchase here, at that time.

I managed to get rid of mine when nuvistor triodes came out, as the makers (Philips) said they did not intend to supply any conductance tables or socket adaptors for new valves.

From memory, that was back in 1961 or so.

I had been frustrated with trying to test valves used in car radios, also the new type which operated with 12 Volt supply to plate and screen, if you remember those, because there were no tables or socket adaptors for those either.

While the unit I had could test valves such as the famous 80 rectifier, dating back to before 1935, transistors were obviously the coming thing, and Philips amongst others could see no point in making adaptor sockets or tables for anything later than the miniature 7pin and 9pin for ECC81 and similar valves.

So eventually all my valve manuals went the same way as the tester, given away or sold, along with several hundred valves.

As I now look at prices on the Internet, (Thank you for prompting me to look) perhaps I should have kept it all, as today all that hardware would be worth many tens of thousands of dollars which would be useful right now.

It's easy to be wise in hindsight, as a friend of mine also became, after selling his Model A Ford car for a few pounds, back in the late 1950's, and he thought he had gotten the bargain.

We had folks associated with Operation Deep Freeze, based at Christchurch, (Harewood Airport as it was called in those days), buying up old American cars, and sending them back to the US in otherwise empty Globemaster cargo bays, and making fortunes back in the US from the veteran and vintage cars easily found around Christchurch.

I found that I normally could test a valve in circuit, anyway, just using a couple of multimeters and some analytical thought, (thought which is always good to keep carefully used, to keep it "sharp").

Thanks for "honing" a few memories for me.

Kind Regards....

I also had a look on ebay after I posted that comment and there is a beautiful AVO valve tester for auction on ebay.co.uk with 3 days to run.

It looks superb and is already at £410... which surprised me!!

For those that are interested here's the link to the auction - which isn't mine of course!!

http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=250220386254&ssPageName=ADME:B:WNA:GB:1123

John.

Hello again, Electroman

I put a bid in at 650 pounds, on your behalf

A lovely piece of hardware, I remember lusting after same, but unable to obtain one.

Kind Regards, from far away....

I used to have a couple, in fact I think I still have 2 somewhere. But if it will ease your pain in not obtaining one, may I offer some thought. The ones I had were mutual conductance testers, testing valves (tubes) under loaded conditions. These were far more accurate than the emission testers. But in actual use were found wanting. For instance, a 3A3 might test fine with a plate voltage of 160 Volts but in circuit might not work properly with 25,000 volts applied. Also testing with a 60 hz signal a 6GH8 might test excellent but fail in circuit at 3.58 mHz. In-circuit testing under actual conditions still provides the most accurate diagnosis as to valve condition. (Maybe I should have grabbed my dust cloth and visited e-bay before I went and denigrated the things). Oh well.

I disagree with the guys who say a transistor tester is not useful. I have used one for many years. I didn't like the design of the one I was using, so I designed my own. Mine is not a curve tracer, but checks for shorts, leakage, and gain. It is about 1 inch by 2.5 inches by 4 inches. It has an edgewise panel meter, 2 rotary switches, and a socket. Originally it had a BNC output instead of a meter, and I used an external Voltmeter.

In the days of germanium transistors, leakage tests were essential (I still have some germanium in my stereo amplifier). The gain (beta) test is valuable still. I want to know the beta when designing a transistor circuit, or troubleshooting a low gain amplifier. Just knowing if a transistor is "good" when I take one out of my junk is very valuable.

S

P.S. I just got an open source program for making schematics (Dia), so if you want the schematic, let me know, and I'll draw it up.

Hello StandardsGuy

<"P.S. I just got an open source program for making schematics (Dia), so if you want the schematic, let me know, and I'll draw it up.">

Thanks for your generous offer.

If you could practice that schematic drawing utility, and draw the circuit, that may be helpful to some.

Kind Regards....

I just got an open source program for making schematics (Dia), so if you want the schematic, let me know, and I'll draw it up.

I always said that seeing one more schematic doesn't hurt. We, engineers, are a collective mind, we build based by our predecessors and, by adapting our projects to different situations, we innovate. I tip my hat off for inventors, but I consider that any engineer, by cumulating all their innovations, can by considered an inventor.

I will, curiously, look at your schematic, if you allow me. Thank you.

do you really have it?

can i have a look or have it? please?

thanks~

thanks for letting us see your diagram.. simple, neat yet sophisticated.. haha

but i was thinking of making one that works like a normal DMM.. something which display everything on the LCD panel that tells you the data.. fully programmed by the microcontroller... maybe... hah.. haven start thinking of how to really do it...

Not wishing to cast aspertions...

But the meter could do with some protection against large leakage currents and shorted collector - base etc... i.e. wrong transistor plugged in./...

John.

Hi Electroman,

You bring up a very good point. This is a simple design from a long time ago. I forgot to tell people how to use it. You start with the function switch set to short (I think the meter reads full scale with a short). If you get the full scale reading, then try the other type (PNP/NPN). If it still looks shorted, don't go any farther - throw the transistor away. If not, go to the next position. Silicon transistors will normally not read anything, but germaniums could. If the reading is on scale, go on to the next position. If you see any leakage in position 3, the transistor is bad, if not go on to the gain test (1000). If <=10%, then you can up-range with position 5 to get a better value.

I don't guarantee how accurate it is, but it works. Since, you brought this up, we could have a challenge to see who can come up with the best design using new technology (i.e microprocessor, auto-ranging, digital readout, etc), or just on improving the present one. What do you think? I will probably not compete, but am interested in other designs.

S

The values are reversed on the 84k and 840k resistors.