The Dynaco Tube Audio Forum

Dedicated to the restoration and preservation of all original Dynaco tube audio equipment - Customer support for Dynaco VTA tube amp kits, all Tubes4hifi.com products and all Dynakitparts.com products


    runaway bias current?

    Share
    avatar
    evoroadster

    Posts : 45
    Join date : 2015-06-16
    Age : 67
    Location : Honolulu, Hawaii

    runaway bias current?

    Post by evoroadster on Sat Apr 15, 2017 12:30 am

    I just received a modified ST-70 that has the 10 ohm cathode resistor burned open and the corresponding test point for the El34 tube (V4) melted.  What would be the best way to proceed?  I need to get a replacement resistor and testpoint but is there anything I could check in the meantime?

    GP49

    Posts : 784
    Join date : 2009-04-30
    Location : East of the sun and west of the moon

    Re: runaway bias current?

    Post by GP49 on Sat Apr 15, 2017 2:54 am

    Check for the proper bias voltage on the grid of the affected tube...and you should probably check all the others, too. There should be a NEGATIVE voltage of roughly 40 volts or so on the grid, coming from the bias voltage power supply. If that NEGATIVE voltage is not present, the tube will conduct way too much current, and you see the result. This can also happen if the tube socket contact for the grid of the tube isn't making contact with the tube pin; you might measure the correct bias voltage but it wouldn't be getting to the tube, with the same dire result. Another possibility is that the coupling capacitor to the grid of that tube is shorted or leaky, and the positive voltage on the other side of the capacitor is swamping the negative bias voltage.

    Note that the "bias voltage" IS NOT the one-point-something volts measured at the test point, though that term is frequently used, incorrectly, for that measurement. Properly, the bias voltage is what is applied to the grid of the tube to put it in its proper operating range.
    avatar
    peterh

    Posts : 882
    Join date : 2012-12-25
    Location : gothenburg, sweden

    Re: runaway bias current?

    Post by peterh on Sat Apr 15, 2017 3:13 am

    evoroadster wrote:I just received a modified ST-70 that has the 10 ohm cathode resistor burned open and the corresponding test point for the El34 tube (V4) melted.  What would be the best way to proceed?  I need to get a replacement resistor and testpoint but is there anything I could check in the meantime?
    replace the tube ( i'd recommend to replace the pair with a matched pair, or
    even better get a quad and junk all 4 tubes)
    Then replace the 10ohm, check the other 10ohm , turn the bias pot to
    maximum bias ( most negative,CCW ?) and start adjusting bias.

    What makes the 10ohm burn ? A flash-over in the tube. Consider that tube dead.
    avatar
    Peter W.

    Posts : 546
    Join date : 2016-08-07
    Location : Melrose Park, PA

    Re: runaway bias current?

    Post by Peter W. on Sat Apr 15, 2017 7:13 am

    YIKES!!

    I wonder how many good tubes are trashed based on this sort of advice.

    Guy and gals, these days, tubes are expensive, ridiculously so, given their expected lives and numbers of apparent defects!

    Anyone seriously into a tube-based hobby these days should consider spending the functional equivalent of four quality small-signal tubes for a reasonable tube tester, even an emissions-type that does shorts and gas.

    The need for matched tubes in present and past Dynaco products is desirable, but not absolutely critical. And, given the dubious nature of so-called 'matching' these days with the exception of a very few excellent sources that do actual matching (for which one pays), I would be very careful of accepting such representations without re-checking. NOTE, however that unmatched tubes will have a shortened service-life proportionate to the out-of-match condition. AND! Note that matching per-se lasts no more than the first 50 - 100 hours of service in any case, even if done properly initially.

    So, to the OP, proceed with caution on the replacements. Purchase a few resistors, and make sure that each resistor for each tube in each channel is identical. Check each tube (even the small-signal tubes) before re-installing. Pay special attention to the rectifier, if so-equipped, and leave it on the tester for at least 10 minutes 'under load' to make sure it is stable. Before applying power *with* tubes, check everything on the AC side with no tubes in place. Then with all-but-the-rectifier, and only after all that checks out, with the rectifier. If you have an accurate fine-pitch ammeter, use it!

    Jim McShane

    Posts : 191
    Join date : 2011-10-19
    Location : South Suburban Chicago

    Re: runaway bias current?

    Post by Jim McShane on Sat Apr 15, 2017 3:15 pm

    Peter W. wrote:YIKES!!

    I wonder how many good tubes are trashed based on this sort of advice.

    Guy and gals, these days, tubes are expensive, ridiculously so, given their expected lives and numbers of apparent defects!

    Anyone seriously into a tube-based hobby these days should consider spending the functional equivalent of four quality small-signal tubes for a reasonable tube tester, even an emissions-type that does shorts and gas.

    The need for matched tubes in present and past Dynaco products is desirable, but not absolutely critical. And, given the dubious nature of so-called 'matching' these days with the exception of a very few excellent sources that do actual matching (for which one pays), I would be very careful of accepting such representations without re-checking. NOTE, however that unmatched tubes will have a shortened service-life proportionate to the out-of-match condition. AND! Note that matching per-se lasts no more than the first 50 - 100 hours of service in any case, even if done properly initially.

    So, to the OP, proceed with caution on the replacements. Purchase a few resistors, and make sure that each resistor for each tube in each channel is identical. Check each tube (even the small-signal tubes) before re-installing. Pay special attention to the rectifier, if so-equipped, and leave it on the tester for at least 10 minutes 'under load' to make sure it is stable. Before applying power *with* tubes, check everything on the AC side with no tubes in place. Then with all-but-the-rectifier, and only after all that checks out, with the rectifier. If you have an accurate fine-pitch ammeter, use it!

    I must take issue with some of your statements:

    1. Unless each output tube has individually adjustable bias then matching of cathode current IS critical. The tubes must match well enough to allow both tubes to operate in similar ranges of current so one tube is not a "current hog" and/or the other is not doing much. I do agree that some matching is done better than others. I don't agree that the price the customer pays for quality matching is high at all - at least in my own case. And to say that matching only lasts 50-100 hours is misleading. Yes, the tubes may not be as tightly matched as when they were first installed, but they most often are still well within the range where they can operate successfully with each other even though they are a few more ma. apart than when first put into service.

    2. The value of an inexpensive emission tester is open to debate. Most emission testers can find a bad short or a very gassy tube - but that's about it. And they certainly can't be used to check matching.

    3. How are you going to check the other tubes in an amp if there is no rectifier in place? No rectifier = no B+ voltage - therefore the other tubes will only be getting heater power. That leaves a lot of unanswered questions, to say the least.

    I agree with much of what you are driving at - but I had to comment on the items above.
    avatar
    Peter W.

    Posts : 546
    Join date : 2016-08-07
    Location : Melrose Park, PA

    Re: runaway bias current?

    Post by Peter W. on Sat Apr 15, 2017 3:55 pm

    Jim McShane wrote:
    Peter W. wrote:YIKES!!

    I wonder how many good tubes are trashed based on this sort of advice.

    Guy and gals, these days, tubes are expensive, ridiculously so, given their expected lives and numbers of apparent defects!

    Anyone seriously into a tube-based hobby these days should consider spending the functional equivalent of four quality small-signal tubes for a reasonable tube tester, even an emissions-type that does shorts and gas.

    The need for matched tubes in present and past Dynaco products is desirable, but not absolutely critical. And, given the dubious nature of so-called 'matching' these days with the exception of a very few excellent sources that do actual matching (for which one pays), I would be very careful of accepting such representations without re-checking. NOTE, however that unmatched tubes will have a shortened service-life proportionate to the out-of-match condition. AND! Note that matching per-se lasts no more than the first 50 - 100 hours of service in any case, even if done properly initially.

    So, to the OP, proceed with caution on the replacements. Purchase a few resistors, and make sure that each resistor for each tube in each channel is identical. Check each tube (even the small-signal tubes) before re-installing. Pay special attention to the rectifier, if so-equipped, and leave it on the tester for at least 10 minutes 'under load' to make sure it is stable. Before applying power *with* tubes, check everything on the AC side with no tubes in place. Then with all-but-the-rectifier, and only after all that checks out, with the rectifier. If you have an accurate fine-pitch ammeter, use it!

    I must take issue with some of your statements:

    A reasonable request, certainly.

    1. Unless each output tube has individually adjustable bias then matching of cathode current IS critical. The tubes must match well enough to allow both tubes to operate in similar ranges of current so one tube is not a "current hog" and/or the other is not doing much. I do agree that some matching is done better than others. I don't agree that the price the customer pays for quality matching is high at all - at least in my own case. And to say that matching only lasts 50-100 hours is misleading. Yes, the tubes may not be as tightly matched as when they were first installed, but they most often are still well within the range where they can operate successfully with each other even though they are a few more ma. apart than when first put into service.

    a) Many of the amps discussed here (VTA, for instance) do allow for individual bias.
    b) Critical - as you suggest to prevent current hogs. And I did note that service-life is affected to the degree of mismatch. But at the same time, unless one is certain of the matching from the seller, *OR* one has the means to check themselves, one is largely at the mercy of the seller.  You are one of the few good sellers out there as it happens. BUT - my point is that it is not a good idea to toss tubes simply because one fails or that they have been stressed due to some sort of issue. Basic tests for emissions, shorts and gas will allow basic screening - which is infinitely better than a trash can.


    2. The value of an inexpensive emission tester is open to debate. Most emission testers can find a bad short or a very gassy tube - but that's about it. And they certainly can't be used to check matching.

    I have a 1950s Simpson that will discern temperature-dependent shorts, and gassy tubes. Point being that it is enough to discern whether a flashover was actually that and whether it is fatal or not. I also have a Hickok 539B, which is my final arbiter in these things, and can match with additional instrumentation - but it is surprising how effective the Simpson is at determining fatal faults. Again, the point is to separate the wheat from the chaff, when the wheat is in the $50/stalk level. Not to match, or even begin to do so. That I paid $10 for the Simpson and another $10 for up-to-date charts is a matter of pure luck, but equivalent testers with paperwork are typically less than $100 in good, clean condition.  

    3. How are you going to check the other tubes in an amp if there is no rectifier in place? No rectifier = no B+ voltage - therefore the other tubes will only be getting heater power. That leaves a lot of unanswered questions, to say the least.

    This tells me if there is anything going on that is, for lack of a better term, mechanical vs. electronic. The bias supply is not tube-dependent. Nor is the filament supply, nor the primary AC voltage. All these things can be verified prior to applying B+. Remember, per the OP, this amp has been traumatized and arrived with a burnt test point and burnt resistor. I like to *exclude* what I can before investigating further. By running it without a rectifier, I eliminate an entire class of faults. There are lot of cold tests as well that I would do with an unknown amp - but I did not want to get too deeply into the minutia.

    I agree with much of what you are driving at - but I had to comment on the items above.

    Please do - you help me clarify my thoughts by asking questions about things I take for granted
    avatar
    peterh

    Posts : 882
    Join date : 2012-12-25
    Location : gothenburg, sweden

    Re: runaway bias current?

    Post by peterh on Sat Apr 15, 2017 5:17 pm

    Tubes are expendable articles, not very far from lightbulbs. They can be damaged in many
    different ways but most damages will make the tube unreliable or behaving bad.

    A tube that has "flashed" ( had an internal discharge) might have damages of grid wires that will
    cause flashing to occur again. Tubes that has been abnormally hot ( redplated) very likley have
    been distorted internally that tube changed parameters. Or debris might have been deposited
    on the g1 that will cause it to draw grid current when hot, thus causing an avalache that results
    in overcurrent or even redplating again in a circuit that is sane.

    Tubes are cheap ( in relation to other parts, and in relation to the time spent in servicing). Thus
    replacing a suspect tube with a new ( tested and matched by a reputable distributor) will save
    time and money in the long run, and most importent, make the listening experience free from
    annoying failures.

    Don't try to save money by reusing suspect tubes, it's like trying grandma's old medicine when
    feeling sick.
    avatar
    evoroadster

    Posts : 45
    Join date : 2015-06-16
    Age : 67
    Location : Honolulu, Hawaii

    Re: runaway bias current?

    Post by evoroadster on Sat Apr 15, 2017 7:57 pm

    Here is some more info: The amp has individual bias for each output tube and it is using a dynamull driver board. It was working prior to being sent to me so I assumed all I needed was to install the tubes, set bias and go. Stupid me. Originally I was going to power it up with no tubes, then add small signal tubes followed by output tubes and then set/check bias. At turn-on I was looking for red-plating or flash over but smoke got there first. The 10 ohm resistor burned open and melted the panel mount plastic bias test point. I did give the amp a good looking over and had to tighten some output xformer mounting screws but everything else looked good. I picked up some 10 ohm resistors and a replacement test point and will install them tonight. The tubes sent with the amp had been used but I do have a new, unused set of El34 tubes. The main thing I am lacking is a schematic
    avatar
    evoroadster

    Posts : 45
    Join date : 2015-06-16
    Age : 67
    Location : Honolulu, Hawaii

    Re: runaway bias current?

    Post by evoroadster on Sat Apr 15, 2017 9:53 pm

    Replaced resistor and TP.  Turned bias pots full CCW.  Brought up on dim-bulb tester with no tubes.  All well.  Added small signal tubes...all well.  Before I try the new output tubes I will wait to see if anyone has any other checks I should make.
    avatar
    peterh

    Posts : 882
    Join date : 2012-12-25
    Location : gothenburg, sweden

    Re: runaway bias current?

    Post by peterh on Sun Apr 16, 2017 4:44 am

    evoroadster wrote:Replaced resistor and TP.  Turned bias pots full CCW.  Brought up on dim-bulb tester with no tubes.  All well.  Added small signal tubes...all well.  Before I try the new output tubes I will wait to see if anyone has any other checks I should make.

    If you have the el34 tubes sorted ( known which ones were placed where) then
    i suggest that you take the pair from the opposit side ( the one that did not
    have a blowd 10ohm resistor) and install on the side where you have replaced the 10
    ohm resistor , leave the "opposit side" empty for now, and power up. See if
    you can adjust and keep bias steady , and that the amp works with the single channel.
    Then order a new pair instead of the ones that blew the resistor.
    If you don't have the tubes sorted out, order a matched quad before continuing.

    Whoops, i did not read all, it looks that you have new tubes. Wise. Go on,
    install and enjoy the music.


    Last edited by peterh on Sun Apr 16, 2017 4:46 am; edited 1 time in total (Reason for editing : Failure (my) to read all text.)
    avatar
    peterh

    Posts : 882
    Join date : 2012-12-25
    Location : gothenburg, sweden

    Re: runaway bias current?

    Post by peterh on Sun Apr 16, 2017 8:26 am

    evoroadster wrote:Replaced resistor and TP.  Turned bias pots full CCW.  Brought up on dim-bulb tester with no tubes.  All well.  Added small signal tubes...all well.  Before I try the new output tubes I will wait to see if anyone has any other checks I should make.

    There is one thing to do before installing the el34;

    Identify pin5 ( g1) on the output tube sockets.
    When powered up with preamp tubes and rectifier installed, measure the voltage
    between pin 5 on each output tube socket and ground.

    You should see a negative voltage -10-30V that can be adjusted with the bias
    adjustment potentiometer. Adjust all to be the highest negative voltage and
    leave them there.

    A leaking coupling cap would cause this voltage to be smaller or even positive, if
    so that cap have to be replaced.

    If all 4 sockets shows an simular amount of negative then power off and install
    the el34. With speakers or other load power up, wait a few minutes and then
    start to adjust the bias pots while observing the voltage across the 10ohm
    resistors. You will have to adjust all 4 in a circular manner as they will
    inflict each other due to loading of B+. Make sure no tubes get's over 1V ( 100mA)
    in the process as this might overheat a tube.
    When all 4 tubes measure equal, keep it online and repeat after 30m.

    Do check bias after a day or so, then it should be stable , check after a month .


    avatar
    Mandopicker

    Posts : 12
    Join date : 2016-07-29
    Age : 47
    Location : Danville, CA

    Re: runaway bias current?

    Post by Mandopicker on Sun Apr 16, 2017 2:56 pm

    "A leaking coupling cap would cause this voltage to be smaller or even positive, if
    so that cap have to be replaced." peterH

    8 out of 10x this was the cause of bias runaway in my restorations.

    I do not mind changing the caps in this case for best function...certainly when it happens to the old ones.
    avatar
    evoroadster

    Posts : 45
    Join date : 2015-06-16
    Age : 67
    Location : Honolulu, Hawaii

    Re: runaway bias current?

    Post by evoroadster on Sun Apr 16, 2017 3:07 pm

    I had set the bias pots to -40v with only the small signal tubes installed.  This morning I pretensioned the octal plugs, installed a new set of output tubes and set bias all around to 0.400VDC.  It now seems to be fine.  I will run it for 30 minutes, reset bias to be equal all around and run it for a few hours on the bench on a dummy load.

    This amp is about 2 years old; it was a new build with new parts with the exception of the 2 output transformers which are original Dynacos with cloth-covered wires.


    New info: previous owner now tells me the quad I first used had an "iffy" tube at one point but seemed to come around after a bit which is why it wasn't mentioned. Well, stuff happens. I know it wasn't intentional.

    Sponsored content

    Re: runaway bias current?

    Post by Sponsored content


      Current date/time is Mon Dec 11, 2017 5:40 am