DIY Tube Tester... Crazy idea???

[dbishopbliss]

I inherited a bunch of tubes a while ago. Most of them were not suitable for guitar so I sold them. But I have a large number  (40 maybe) of 12AX7s (AU and AT as well), ten or so 5AR4, and a number of pairs of power tubes including EL34, 6L6, 6V6.


I have put them in an amp and they seem to work but when I have looked at Tube Tester videos online there seems to be a way to measure how "good" they are. I have been looking around for a tube tester but it seems there is a lot involved with them. Tube charts, calibration data, etc. Lots of people have warnings about "don't buy one unless you know...".


Another thing I have gleaned from looking at some videos is that tube testers were designed to test hundreds of different types of tubes (seems that way). There are charts and you have to change settings for filaments as well as other things (they seem to use different names depending upon the manufacturer and model).


Got me wondering... what would it take to design a "simple" tube tester for the most popular guitar tubes? I'm thinking like, 12AX7, 12AT7, 12AU7, EL84, 6V6, 6L6, EL34, 6550.


Make a chassis with a socket for each tube type. Then, you just install the tube and maybe you can measure something with your mulit-meter that tells you how "good" it is.


I have no idea what I would be measuring, but would something like that be possible without costing more than $1K like all the calibrated fully functioning meters I'm seeing out there.

[acheld]

I've often thought about doing this . . .

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

So, no, not a crazy idea.

But I would like to see an actual schematic of a tube tester --would be an interesting project, though probably not cost effective.

[sluckey]

Go to BAMA (May want to bookmark this page). Select Hickok directory. Select tv-4bu. Everything you ever need to know about one of the best tube testers ever built.

[PRR]

Assuming you are not selling:

For tubes like you list:

Build a Champ. Use several sockets to handle the several pinouts. Add switchable cathode resistor to handle EL84 vs most others. Values come from Fender AA-Champ and Epiphone Valve Jr. Not critical.

Add probe-sockets so you can read tube voltages. Sounds like a lot but is really FAR simpler than a Real Tube Tester.

Pick a Standard Speaker you will "always" use with this test rig. Maybe you have a good but uninspired speaker. Or maybe your regular gig-speaker. Just so you can be consistent.

Put in known-good 12AX7 and 6V6. Play soft. Play loud! How does it sound? What voltages?

Put in unknown 12AX7. Then unknown 6V6, 6L6, etc. How does it sound? What voltages? If the gain and loudness are good with no crackle hiss or hum, the tubes are "GOOD" for your purpose. You don't need a uMho number, you don't need 28MHz response, if is plays OK.

12AT7 and 12AU7 can go in the 12AX7 socket. Plate voltages will be somewhat less. AT will be a bit less gain, 12AU7 will make you work to be loud. Like my mate's MX5 Miata with all-manual steering. It works good, but a parking-lot race will tire your shoulders. Not wrong, just less leverage.

6L6 will work in 6V6 hole, making 6V6-like power. If it does that, it will probably work at full 6L6 power.

EL34 can work in a 6V6/6L6 hole with a G3 jumper or 100K 2W resistor.

No you can NOT put a micrometer on a tube and, like a car tire or bearing, figure "67% life left". Most tube "wear" is unpredictable. If it matters, if you sell commercials for $100/minute, you read cathode current every day and change tubes at the least hint of fall-off. And even then, tubes just break with no warning.

[PRR]

...Select Hickok directory. Select tv-4bu. ....

Oddly, that link seems to be TV3-3BU.

The TB3 folder is another larger TB3 file.

[sluckey]

Oddly, that link seems to be TV3-3BU.

I noticed that too.

[dbishopbliss]

Assuming you are not selling:

I should have added that... I'm thinking about selling some of them. Most of these tubes are Mullard, Telefunken, RCA, etc which seem to command high dollar from online sellers. I think I would need to know how "good" they are if I want to sell for a decent price.


I also have a number of Dynaco labeled 5AR4 tubes that were made by Mullard.


A while ago I had the number of someone that tested tubes locally. He said he charged $20/tube. Seems like a lot to me so I passed. Maybe I should just buy a tester to test my tubes. Then I could start a "business" testing tubes for $10. Then I can sell the tester when I made enough to cover my losses. :-D

[shooter]

I've always "viewed" a tube tester as a GO/NOGO box, anything more needs testing "in circuit", hence PRR's suggestion of a Champ.
don't believe a tube tester will find a microphonic tube

[PRR]

If you are selling, you need a FANCY NAME BRAND tester.

Yes, you can test for microphonics. There were maybe a dozen (total!) such testers in the world by 1960. Though I gather the most usual test was to put it in a radio and turn it up.

[acheld]

I think I would need to know how "good" they are if I want to sell for a decent price.

   This suggests you are a good person.  I'd buy from you.

Do most ePay sellers who say they test actually test?   Maybe, maybe not.   And what does it mean?

Yes, I'm a confirmed cynic.

[HotBluePlates]

I should have added that... I'm thinking about selling some of them.

If you are selling, you need a FANCY NAME BRAND tester.

Do most ePay sellers who say they test actually test?   Maybe, maybe not.   And what does it mean?

This has come up on another forum.

The guys there who buy old-production tubes only buy from sellers who provide Gm and idle current measurements for the tubes they're selling.  The buyers understand what those numbers mean, and they have the gear to test it themselves and so they verify the performance of their new purchases.

There are other folks who might buy from sellers with less-detailed test results, but they're not gonna pay "old tube prices."  They're gonna pay "a little over new tube prices, before the pandemic."

If you're going the route of selling, I wouldn't bother with a home-brew tester, because the big-$$$ is not going to trust those results (even if they're 100% accurate).

I've always "viewed" a tube tester as a GO/NOGO box, anything more needs testing "in circuit"

The lab-grade setups are very much more than a GO/NO-GO test, and give everything a data sheet says & more.  There is some interpretation required to understand the results, and setup/test time is long.  But you can literally compare test results against data sheet conditions (and use data sheet conditions to establish the test parameters).

But nothing does gain/noise tests as well as using a real amp circuit.
If you measure in-circuit gain with a real amp circuit, you will likely also find out that tubes that measure very different on the tester don't perform that different in the amp (because cathode bias pulls the tubes towards an "average behavior" in a way that fixed bias doesn't).

... someone that tested tubes locally. He said he charged $20/tube. ...

And he probably hasn't made more than $20 or $40, because no one is gonna bother paying absurd prices like that.

... Maybe I should just buy a tester to test my tubes. Then I could start a "business" testing tubes for $10. Then I can sell the tester when I made enough to cover my losses. :-D

The testers people want to see mentioned in ads will run you $700-1k in good refurbished/calibrated condition.  The lower priced ones that don't claim a recent calibration might be like getting "a deal" on a used car...

[dbishopbliss]

The testers people want to see mentioned in ads will run you $700-1k in good refurbished/calibrated condition.  The lower priced ones that don't claim a recent calibration might be like getting "a deal" on a used car...

Thought exercise... Spend $900 on a calibrated tester. Test my tubes so I can get $100+ for my Mullard/Telefunken/RCA 12AX7s instead of $20 and then sell the tester. How much depreciation would there be on a slightly used calibrated tester? Would it go from $900 to $200 or could I get $800 back. If the latter then it could be worth doing.


Maybe I just need to build more amps to use them. :-D