DIY Emission Tube Tester

[francisco]

Hi,

I'm currently building a analog tube tester based on knight 600 tester (http://bama.edebris.com/download/knight/600tube/600tube.pdf), but I've difficulties to determine  transformer's current, because I'm new in tube tester thing. Based on knight 600 schematic, there are two group in secondary, the upper one is for filament (.63V to 100V) and the  lower group (0-30-100-250V). I guess in filament group there are 3 sub group in series configuration,  <6.3V/ 3A, 7.5-19.6V / 1A, 25-110V /200mA, for lower group, If I set 200mA max for 250V, the capacity of transformer might be greater than 30VA. Normaly the capacity of transformer for emission tube tester is around 20-30VA CMIIW.

I'm hoping to learn from people in the forum to solve my problem.

Thank you in advance

[eleventeen]

This is not an easy thing to build, because of the fairly complex switching required. You are talking an emissions tester which does not place a power tube like a 6V6 or 6L6 under any kind of load that would be similar to how such a tube would be used in real world conditions. Such a tester only ties the elements together and checks to see that the cathode can squirt out some reasonable amount of electrons...these testers don't put 50 mils through a 6L6 and they typically only run the tube at about 250 plate volts. Very few 6L6/6V6 in the real world run at such a low plate voltage.


Nevertheless, these testers are not useless. If they show a tube to be bad, that tube is surely bad. I have one, a B&K 607, which is a solid state tester. Because it is solid state, in my imagination it should not require periodic calibration. I am not saying this is any kind of great tube tester or the best or anything like that. One very nice feature of that tester is its "lockout" feature which is helpful when you are switching from one type to another because it can prevent you from blowing up a 6 volt tube with a 12 volt heater voltage. (It probably would not burn up the heater; but it DOES cause the meter to read ZERO in many cases when I have forgotten to make one change in the switch settings and very very few tubes read ZERO, so this is in fact a clue that I have set something wrong. This takes a row of self-cancelling pushbuttons which would not be an easy part to find. I think I bought it for $65 plus shipping on ebay, it's in perfect condition. I couldn't build that for $65.



To build a tube tester, let's see...you need a meter; a transformer, a box to put it in, a bunch of switches, a bunch of tube sockets...would be nice to have a handle and a hinged cover on it because it's something you don;t really use every day.


I see these things going for about $100...do you really think you can build one for much less than that?


I also have a Hickock 600A which is a much better tester but it is a finicky sort of thing, uses an 83 mercury vapor rectifier that acts like it arcs when the tube is under test, and a #44 bulb as a bias fuse, and my tester blows this fuse now and again. Certainly it performs a much better test but it doesn't put any kind of 350 volts on a 6V6...and on a small signal tube like a 12AX7 you better have some tube samples to compare your readings to. I am glad I got it. I got it in a trade. 


But to get to your question, you are right, most low cost tube testers do not put much current through a given tube under test. They put "enough" volts on the plate and tie ALL the grids to the plate and check the action of the tube strapped as a diode. Which is not the way you would ever ever use such a tube. (Unless it WAS a diode)


I say: Buy one. I would consider building a super duper tester that ONLY tested 12AX7 tubes, for example, but a general "any tube" tester? I wouldn't try to build one. Just my opinion.


You know....every professional and semi-professional and amateur and semi-amateur and semi-semi-amateur TV repairman from the 50's and 60's had to have a tube tester. This is in large part how one repaired TVs in the 50s and 60s. There are tens of thousands of them out there and there are lots of people whose "dad" had a tube tester and who is now retired or passed. My local Craigslist shows 4 tube testers...but in any event, I believe if you wait it out on ebay long enough you can find a commercial tester equivalent for $50-$75 and get what you are trying to achieve by building one. And if you hate it, you can resell it on ebay for $50-$75 and try another. Or at a local hamfest.

[kagliostro]

I've do a search on e-bay and there are many tube tester disposable


Franco

[PRR]

In an emission tester, you only hold the button a few seconds. If you hold it, the tube will burn up before the PT HV winding is strained.

The heater winding must be good for a few minutes of the hungriest tube you plan to test. A 1A winding will carry a 1.6A KT88 for a few minutes.

I'd think a Champ PT would be plenty.

Agree that the many-many settings and switches would be better bought than built.

[Paul1453]

I have found that using one of my fixed bias amps to test tubes emissions is cheap and easy.

I use my HeathKit TC-1 for basic tube testing, and then use an amp to grade my stock of tubes emissions.

Tubes that test equally well on the TC-1, can be dramatically different in emissions in the amp.

Pairing the tubes in a PP output circuit based on my emission testing seems to work quite well.

Just for putting numbers on this example, lets say I test all my 6L6s emissions.

I get tubes with say 80, 75, 60, 58, 48, 45, 30, and 28 mA conduction in my fixed bias amp.

Pairing the 75 with the 30 produces noticeably different response then if I pair the 75 with the 80.

But the funny thing is that pairing the 30 with the 28, doesn't sound all that different from the 75 & 80 pair.
In this case the weaker pair sounds nearly as good as the stronger pair in tone and response, they just don't put out as much power.   

[francisco]

Thank you for your response

Actualy this is my 4th tube tester project, I made my first from tube tester kit, I bought this kit from France via ebay. I added some rotary switch to make this tester more practical for testing many tube, plus I added some variable PSU for filament supply and another B+ to test full penthode.  It's a digital tube tester with PIC microcontroller to display the result, B+, volt  bias, filament, grid voltage. The tester can test Plate current (Ip), transconductance (S), Internal resistance, and amplification.

In my 2nd project, I made the tester from French magazine articel, Radiofil No.43 by Mr. Fabert, I replaced the card type section with rotary switch, and the 7 segment display with 2x40 LCD display. The tester can preform as my 1st project, but this one with transistor base tester, my 1st is with IC.

My 3rd project, I made from Mr. Steve Bench web, RAT tube tester, but recently I scrapped this tester because the tester performance the same as my 1st and 2nd project. It can tests tranconductance, as I remember.

But for emission tester, this is my first project. The reason I have to build this emission because I wanted to sell some of my old tubes, like 26, GZ32, AZ1, EF12, etc and my customers want to perform an emission test, not tranconductance test. I do aware to build a tube tester its a hard job, with many switch configuration, gathering components, plus box, I bought an aluminum box from China, it's around $10. For this knight case I've got 99% components, most of them belongs to my uncle, the last one is transformer, I wanted to wind toroidal transformer myself, I've heard the latest model is OA shape, like in jantzen inductor.  I'd to buy knight 600 tester, but the cheapest knight 600 tester is around $125.

Thank you guys for the suggestions

[francisco]

This is my tube tester projects

[kagliostro]

As I can see you are skilled on that

A friend of mine has build 2 versions of the Metrix U61B, the first with rotary switch (modified as to have also a signal input)

and as he wasn't satisfied of the reliability of the rotary switch he has build, a second version with matrix

There are only few sockets because he uses adapters



















so if you find that you have problems with the rotary switch or simply with space, as you can see there is an optional solution

Franco

[eleventeen]

Your tube testers look GREAT! Obviously not your first project! Excellent work.

[HotBluePlates]

... I have to build this emission because I wanted to sell some of my old tubes, like 26, GZ32, AZ1, EF12, etc and my customers want to perform an emission test, not tranconductance test. ...

You are probably already aware that an emission test can be damaging to the tube.

If your customers want to verify cathode emission as a means of predicting the remaining life of the tube, perhaps you could simply use a Hickok-style Life Test.  Transconductance (Gm) is tested as usual.  Then, then filament voltage is reduced to the next-lower common filament voltage (e.g., from 6.3vac to 5vac), and Gm is tested again.  If the new Gm reading stays same/similar as the full-filament-voltage reading, the cathode is assumed to have strong emission and plenty of remaining life. If the new Gm reading is severely lowered (more than 50%, or starting at a normal reading then falling dramatically), the tube cathode is judged to have less reserve emission capability and probably less remaining life.

The problem with such life tests is that there is some subjective judgment required; you get a feel for how new and worn out tubes present differing results on your tester, and judge them accordingly.  However, the same argument can be made for any test of a tube other than a test of performance in the actual circuit in which the tube will be used.

If the Hickok-style life test is acceptable, it requires only that you have a means of varying the filament voltage (which could also then be a way to rejuvenate some tubes).

[jjasilli]

Along that line of thought:  http://www.diyaudio.com/forums/tubes-valves/262033-testing-emission-without-risk.html

[slowboy2012]

Ciao. Complimenti, ottimi lavori. Vorrei costruire anch'io un provavalvole e dotarlo di un display lcd 2x40 dove visualizzare tensioni e correnti. Mi puoi aiutare dandomi consigli o schemi per realizzare il voltmetro-amperometro multi-ingresso? Grazie in anticipo.