Minimalist valve tester design wanted.

[Graham G3ZVT]

I would like to build a valve tester with simple flying leads and croc clips for connections. What I envisage is something like this one built by Ron Soyland (Glasslinger on Youtube)



Not wanting to re-invent the wheel, so has anyone got a circuit for something similar?

[joebog1]

Try this:-

http://www.valvewizard.co.uk/valvetester.html

But best thing to do is search this site using "valve tester" as the search criteria.
It will take a little while to check out all the responses, but there is an absolute plethora of designs and information available ( like all the other subjects in this forum)

Joe

[joebog1]

Here is another

https://www.valveheaven.com/2015/03/...vetube-tester/

Joe

[Bazz4CQJ]

I'm sure that there have been two threads on this subject in the past year or so, but I can only find one https://www.vintage-radio.net/forum/...=153836&page=3.

I'm sure there was another thread in which someone came up with a really interesting and simple design which I think used just one valve. It had been published in a book or magazine many years ago. It was so simple that a few people said that they could not understand how it could work, though Wrangler expressed a very positive opinion on it (I stayed quiet ). The valve under test and the valve in the tester formed two arms of a bridge circuit.

I'm sure at least one person said they would definitely build it, but I don't recall him posting any follow up. That really interesting thread is the one I cannot find!!

B

[Ed_Dinning]

Hi Graham, first decide which valves you want to test as that could reduce the number of sockets etc needed.
If you need the tester for small signal valves and quick tests then small power supplies will do; if you are going to do prolonged testing "burn-ins" then much larger power supplies will be needed.

Ed

[Radio Wrangler]

Oooh heck.

I can remember someone posting a stripped-down fundamental circuit of a valve tester. It was quite elegant and did what it was supposed to. Can't remember any details. About a year ago?

The problem I see with extremely simple valve testers is that they are inevitably just indicators. Without multiple variable or switched supplies you can't set the valves up for the conditions in the data book, therefore anode current and Gm figures from databooks can't be expected.

Consequently, testing valves under fixed conditions needs custom databooks for those conditions.

Testing valves under databook conditions (existing databooks) forces either lots of anode and screen voltage tappings or else decently beefy variable power supplies and all the switching or metering to control same.

Maybe just replicating the sussex is the easiest way out?

David

[woodchips]

Wasn't that one from the BBC was it?

[Bazz4CQJ]

Originally Posted by Radio Wrangler

Oooh heck.

I can remember someone posting a stripped-down fundamental circuit of a valve tester. It was quite elegant and did what it was supposed to. Can't remember any details. About a year ago?

Wasn't there a stripped down circuit to show the principle and then a working circuit diagram for actual construction? I recall the debate being about "what constitutes a bridge circuit". Although simple, it claimed it could quickly and easily measure Gm (adjusting just one control to balance the bridge).
I'm so annoyed that I didn't copy the PDF of the pages which I think were from a 1930's book? Thought that I might build it from components in my junk box and sell my VCM163 .

I was totally sure that the discussion was actually the back end of the thread I cited in post#4, but if so, it's not there now.

B

[ms660]

I think I've posted this link before but here it is again:

http://www.r-type.org/articles/art-118.htm

Lawrence.

[Radio Wrangler]

Thanks, Lawrence, that's it!

Rather elegant indeed.

Looking at the simplified circuit, if you put a 1v sine in, the valve will swing its current by 1v x Gm mA, sinusoidally, and the anode will thus swing 1v x Gm x 1000 ohms. so the anode voltage swing will be Gm times that of the grid. Simple! but notice that the anode swing is in the opposite direction to the grid so looking down the potentiometer there will be a pace where cancellation occurs. There is (Gm +1) volts across the pot.... from -1v to +Gm volts (AC with -sign denoting opposite phase)

So you twiddle the pot to null the AC output, and the pot can have a scale to read off the Gm value.

Now to complicate it. The pot will load the anode resistor. So the elaborated circuit makes the anode resistor very small and plays a few games with the pot network and adds a buffer amp at the input andan AC amp on the output to bring the signal level up.

Note that you'd still need to add switched HT and switched g2 supplies to suit the range of valves you intend to test. An Ia current meter would be good. Grid bias can come from a pot and have a meter on it.

David

[Bazz4CQJ]

Yes, thanks Lawrence, that's the one I was thinking of.

Now...who was the guy who was definitely going to build and try it
Where is the thread hiding? Wasn't there also some discussion about the author, Mr G.A.French, for some reason?

B

[TowerRadio]

I confess,it was me who brought the subject up. I have to repair a few things first-a Cadet 3 for my son,look closely at the DAC90A my daughter bought me as a Christmas gift,sort out a Hitachi transcription unit and a Clarke and Smith school radio. I will build this tester as soon as I can.Les

[Bazz4CQJ]

Don't worry Les, my list "To Do" and "To Finish" list are of galactic proportions.

B

[PJL]

The one I wanted to build displays the full characteristics on a scope (X/Y). There is a design out there somewhere and it is surprisingly simple as it uses full wave rectified 50Hz for the scan and a shift register to step the grid voltage.