12AT7 Testing Settings -- They Change the Voltages?!

Discussion in 'Tube Audio' started by thefragger, Jul 25, 2012.

  1. thefragger

    thefragger Certified Crazy. Subscriber

    Hey guys,

    So I bought a lot of ECC81 (12AT7) tubes for funsies, and now I want to test 'em. I crack open my book and, surprise surprise, that good ol' Eico data is all jumbled up and does not make any sense to me.

    I checked through all the different booklets and I'm seeing the same variation, but different numbers. :sigh:

    Can I just choose one set of grid and plate settings and go along testing away?

    Here's my problem:


    The Tone Lizard page on the Eico 666 has some great information on testing tubes and comparing the different versions of the manuals to one another (see here: http://tone-lizard.com/Eico666.html). Tone Lizard has found that the 666-04 settings seem to run the 12AT7 tubes more realistically.

    So, I'm assuming here that I know things, and what I think I know is that both sections of a 12AT7 tube are the same, which means that I shouldn't have to change the voltage settings when testing the two sections.

    Would I be within my right to, say, keep the voltage setting applied to the grid and plate, only changing the levers which assign the pins?

  2. battradio

    battradio AK Subscriber Subscriber

    My guess is that they some how , transposed 12dw7 ( 7247) , grid and plate to the 12aT7 setting , would use the setting that give the best reading .
    Last edited: Jul 25, 2012
  3. jsixis

    jsixis Nothing sounds better

    sounds like a mis print never seen that before
  4. thefragger

    thefragger Certified Crazy. Subscriber

    The thing is is that it's in every revision--the numbers bounce around, but triode 1 test is a different voltage/load than triode 2 every time! :no:
  5. battradio

    battradio AK Subscriber Subscriber

  6. Kegger

    Kegger Anything can be S "MODed" Super Mod Subscriber

    What I would do in that regard, is find me an good/known testing tube and test
    it at both of the suggestions just changing sliders for each half of the tube.

    Find out if one setting does a better job than the other at a more accurate read.

    Also see if there are settings for 6DT8 and 6AQ8 an compare how they are setup.


    Now in a pile of 12AT7's your bound to get plenty of bad/low/mismatched sections
    as it's just the nature of the 12AT7, many more "bad" tubes then you get with say
    a 12AU7 or 12AX7 which is one of the reasons I went to the 6AQ8 instead, they do
    tend to be better lasting/testing tubes per sample size..
  7. Kegger

    Kegger Anything can be S "MODed" Super Mod Subscriber

  8. thefragger

    thefragger Certified Crazy. Subscriber

    Yup, read the info there and the manual a bunch and it's been useful to know how the tester works so that I can run a quick mental check in my head whether or not the connections make sense (there was once an error with the heater pins) or I'll read that X tube is the same as Y tube but in a different bottle and develop my own settings for those rarer tubes.

    Anyways, here's ToneLizard's opinion:

  9. ConradH

    ConradH AK Subscriber Subscriber

    Sometimes my roll chart is right and sometimes the BAMA hosted reference is right. For the 12AT7 I have grid=85 and plate=90 for both sections, with the same switch settings as all the other 12A?7 types. V=2 and S=2.

    FWIW, there are slight differences in the two sides of the tube, at least with capacitance, so maybe Eico wasn't completely toasted when they did the chart. But I doubt it!

    My current question, also posted on DIY with no answers is:

    Has anyone published decent corrected settings for the 12A*7 series of tubes, settings that don't overload the plate?

    The manual says many contradictory things. It says dual triodes should be tested with voltage applied to both sections of the tube, and shows schematics for that mode. They boast of the speed of testing. Then the charts show to do every section separately! It would seem the lever settings should be 4511245 661111 (for the 12AX7 and the 667 model tester), though I'm not exactly clear on how the buttons get pushed and what leakage tests would be made.

    Can anybody clarify this stuff?

    edit- curious how Tonelizard got the currents- was it by calc or actual measurement?
    Last edited: Jul 25, 2012
  10. Kegger

    Kegger Anything can be S "MODed" Super Mod Subscriber

    I don't know what you mean by, "settings that don't overload the plate"..
  11. ConradH

    ConradH AK Subscriber Subscriber

    There's a dissipation rating on the plate, different even in the same family of 12A?7 tubes, and it appears some of the test settings exceed it. It's just P=IE, so it depends on the voltage. I probably need to do actual measurements of a tube in the tester.
  12. thefragger

    thefragger Certified Crazy. Subscriber

  13. Kegger

    Kegger Anything can be S "MODed" Super Mod Subscriber

    I know what plate dissipation is and how to calculate it there. ;)
    But an "normal" setting on a tube tester comes nowhere near that.

    The one posted above with 20ma would but that is a major exception.
    But still it's only 1.8w on a 2.5w rated tube, the other 2 are quite low.
  14. thefragger

    thefragger Certified Crazy. Subscriber

    The example above states that at the voltages the tester is using (low, ~100V) that the current should be 3.7mA according to the RCA manual; the author states that at 100V the tester is putting 20mA at the plate (potentially damaging it?).

    That's what I got out of that paragraph, at least... :scratch2:
  15. ConradH

    ConradH AK Subscriber Subscriber

    I definitely need to check the real numbers! BTW thefragger, that's a really nice 667. I also notice that your line control is at the bottom right whereas mine is at the upper right.
  16. Kegger

    Kegger Anything can be S "MODed" Super Mod Subscriber

    No doubt the 20ma is pushing the tube pretty good, still under rated plate dissipation.
    (but yah 20ma on a small signal tube like this makes no sense, not a good way to go)

    I would use the more realistic real world ones they have posted there as well, 4-5ma.
    (on a 12AT7 3-6ma is pretty typical of what they'll do in audio circuits under use)

    About the only typical small signal tube that's gunna see 10ma or more in action is an
    6DJ8, so most of your usual 12axx I wouldn't run more then 5ma through them, to get
    a real feel for how good that tube is, but what is posted there isn't "past" the ratings.
  17. thefragger

    thefragger Certified Crazy. Subscriber

    Ha, thanks! Photogrphy's another hobby of mine, so that might be the trick ;)

    From reading around mine is a newer version from when I did research on it. :thmbsp:

    There are also intermediary sockets that have little breakout tabs (plugs into tester, tube plugs into it) and that way you can directly read any of the pins!

  18. ConradH

    ConradH AK Subscriber Subscriber

    Here's a short commercial adapter I've had for ages, plus two I made up for biasing 6L6 and similar guitar amp tubes. I'm lazy so having two allows me to just use two meters for both balance and setting. The 1 ohm resistors are on the ground side. The loop is to the plate and allows me to get my (428?) HP current probe around it. I had at least one miniature adapter but I can't seem to find it right now :-(

    Attached Files:

  19. ConradH

    ConradH AK Subscriber Subscriber

    Last night I made an adapter to monitor plate current and such for the miniature tubes with the hope of developing some sensible settings. I have quite a few NOS tubes of some types, but not many 12AX7As. Lots of 12AU7As for some reason. The range of performance is far greater than I would have thought! Basically one chooses a current and sets the grid to get it. Then it's just a matter of choosing a sensitivity for the meter reading. IMO, there's nothing precise about this and without a "perfect" reference tube or a large group for a statistical average, it's hit and miss. I can imagine that Eico never had every tube in the chart in their possession, and that they changed recommended settings as more info became available. Add the errors of a manually kept list, and all bets are off. Soon I'll post what I think are reasonable settings for common audio tubes, but anybody who thinks a tube tester "verdict" is the end-all be-all determination of tube condition is probably deluding themselves!

    edit- not quite right- see my later spreadsheet in this thread
    Last edited: Jul 29, 2012
  20. Kegger

    Kegger Anything can be S "MODed" Super Mod Subscriber

    The best way to use a tube tester is get use to your particular.

    If you test enough of the same tube you can a get a good idea
    of what a tube "should" test as, combine that with running that
    tube in a piece of equipment you know how it should run that is
    even better to "test" the tube after checking it to verify results.

    Testers for the most part if your unfamiliar with how it responds
    are just a no/go deal just to tell you it's still Alive or Dead there.

    I've used my testers enough with the good amount of each tube
    to see where the tester wants to run on certain tubes, then ran
    in piece of gear taking voltage/current readings to verify results.
    (even built an small stereo single ended tester amp to run many)

Share This Page