Any Theories on 6U8A Hum?

Discussion in 'Tube Audio' started by OBMG74, Feb 18, 2018.

  1. OBMG74

    OBMG74 AK Subscriber Subscriber

    Messages:
    155
    Location:
    Rockford, IL
    I think that we are mixing apples and oranges here or are at least attempting 2 different discussions. Please allow me to clarify:
    The 6U8's (didn't seem to matter what brand) developed a hum over time. As posted earlier in the thread, several knowledgeable members suggested that the dc voltage on the heaters needed to be raised to prevent the 6U8A deterioration over time, which I did.
    The fact that the NOS JAN's hummed immediately and were microphonic was strictly due to the amp not liking them. (some time ago, I had some GE's do the same thing) I believe that some time ago when I searched AK for 6U8's that people liked in this amp, JAN's were not recommended, RCA's got a few kudos, so I have stuck with them and like them.
    I ordered the JAN's before I started this thread to see if they would last longer but I found out as mentioned earlier that it was recommended that the heater circuit be modified, regardless of the brand of tube used.
    Basically, I threw the JAN's in there as a test in case something went wrong after I modified the circuit (and used a dim bulb tester, of course!)
    As I said, when I installed the NOS RCA's the amp performs wonderfully, I am confident that the hum that used to appear over time will not re-appear.
    Hope that I have answered your question.
     

     

    Please register to disable this ad.

  2. 6DZ7

    6DZ7 Super Member

    Messages:
    2,009
    Ok, it just seems like you had already made the decision on the affectiveness. I would judge the affectiveness by using tubes that seemed to be problematic in longevity to see if they had longer life, rather than using a "good" tube and having to wait a long time or ever to actually know if the additional DC did anything toward your goal.
     
  3. OBMG74

    OBMG74 AK Subscriber Subscriber

    Messages:
    155
    Location:
    Rockford, IL
    Yes, I see your point and am actually doing that. The tubes that I liked the sound of and which developed a hum over time were RCA's.
    I have put in new RCA's after the circuit change and in that way will be able to judge the new circuit's effectiveness on tube longevity.
     
  4. Retrovert

    Retrovert AK Subscriber Subscriber

    Messages:
    3,295
    Location:
    New York City, NY
    The cumulative damage is the key point. Voltage stress gradually degrades the insulation causing leakage which increases until the tube is damaged or destroyed. Elevated heaters is the only solution and it completely solves the problem.
     
    Pio1980 likes this.
  5. billyz

    billyz Active Member

    Messages:
    240
    Location:
    Spokane
    I have found it necessary to elevate the heater on 7199’s as well . I usually elevate them by 25v or so . I also find a 0.1 uf film cap from the voltage divider junction to ground eliminates the last bit of buzz .
     
  6. Retrovert

    Retrovert AK Subscriber Subscriber

    Messages:
    3,295
    Location:
    New York City, NY
    The lack of elevated heaters on the 7199 explains why the Dynaco ST-70 eats them like candy.

    25 V, however, is insufficient as the alumina insulation is really only rated for 90 V at that thickness. Yes, the datasheet specifies more, but at those ratings the RCA papers report dramatically reduced lifespan. That higher number is, I believe, a fiction created to allow the 7199 to be used as disposable items in a tube TV, its intended application, which was itself a disposable item. But what we know is that the continual stress eventually causes insulation failure, hum, noise, and eventually heater-to-cathode shorts.

    In addition to the stressing of the insulation is metal migration which occurs at any potential difference above about 5 V.
     

     

    Please register to disable this ad.

  7. 6DZ7

    6DZ7 Super Member

    Messages:
    2,009
    7199 was developed and released by RCA and promoted for Hi-FI, not TV. A year later Sylvania developed an "improved" version they called the 7687, also intended and promoted for HI-FI.

    The 6GH8A, was developed for TV then adapted for HI-FI and now is being inappropriately conflated with the 7199 because of its sub-ability.
     
  8. gadget73

    gadget73 junk junkie Subscriber

    Messages:
    35,878
    Location:
    Southern NJ
    The trick, and I've mentioned this previously, is that the driver tubes generally share a heater string with the rest of the tubes. Get the driver H-K within 5 volts and you're probably some 50-90 volts different on the rest of the tubes. Short of adding a separate heater supply for those tubes, about the best you can do is find a happy medium that keeps everything reasonably happy. Not perfect, but better than nothing.
     
  9. Retrovert

    Retrovert AK Subscriber Subscriber

    Messages:
    3,295
    Location:
    New York City, NY
    Heater transformers are inexpensive and small, and that way all the tubes are within limits.

    Costs a whole lot less than unobtainium tubes. Just sayin'.
     
  10. OBMG74

    OBMG74 AK Subscriber Subscriber

    Messages:
    155
    Location:
    Rockford, IL
    So I now have 39VDC to the hum pot wipers on the ASR-433 due to the circuit modification.
    I am going to see how this works out on tube longevity.
    Or, would it be worth the trouble to install a separate filament transformer for the 6U8A's, remove the current circuit modification, and increase the DC voltage to the 6U8A heaters only?
     
  11. peterh

    peterh AK Subscriber Subscriber

    Messages:
    1,792
    Location:
    Gothenburg,Sweden
    It's not possible to optimize the filament offset in this amp as the penthode section has filament in common
    with the triode that is used as cathodyne. The penthode has it's cathode only a few volts above ground, the
    triode has ( according to the schematic i have ) 70V above ground. Thus 39V seems to be a very reasonable
    compromise in this case.
     
    Retrovert likes this.

     

    Please register to disable this ad.

  12. OBMG74

    OBMG74 AK Subscriber Subscriber

    Messages:
    155
    Location:
    Rockford, IL
    Ok, Peter- thank you for the explanation.
    I am going to leave well enough alone and just enjoy listening.
     
  13. gadget73

    gadget73 junk junkie Subscriber

    Messages:
    35,878
    Location:
    Southern NJ
    yes but ....

    so the best you can really do is compromise. I believe all of the commonly used triode/pentode tubes have the same situation with a single heater connection for both elements.
     
  14. Pio1980

    Pio1980 AK Member Subscriber

    Messages:
    25,137
    Location:
    Angel Station, Alabama
    Sort of like anti-skate, it's an approximate compromise, and some is better than nothing.
     
  15. Retrovert

    Retrovert AK Subscriber Subscriber

    Messages:
    3,295
    Location:
    New York City, NY
    Yes, you are absolutely correct.

    I'd forgotten about the shared cathode in this instance. As a general rule one should optimize for individual tubes but with cathode sharing one's hands are tied.

    So 39 V is the best one can do under the circumstances.
     
  16. Pio1980

    Pio1980 AK Member Subscriber

    Messages:
    25,137
    Location:
    Angel Station, Alabama
    I checked the schematic for my Silvertone/ Warwick console amp, they tie the heater chain to the output tube cathode bias.
     

     

    Please register to disable this ad.

  17. Retrovert

    Retrovert AK Subscriber Subscriber

    Messages:
    3,295
    Location:
    New York City, NY
    Since I don't have the schematic in front of me, some minor speculation which needs validation.

    If the elevation voltage is pulled from the cathode bias resistor the voltage slightly fluctuates, even with a capacitor. Removing the degeneration is the purpose of the capacitor, but it cannot be fully removed.

    Any fluctuation in heater voltage modulates the heater winding and thus the cathode. True, this may be minor, but it can still inject noise. Any foreign signal in this context is noise.
     
  18. Pio1980

    Pio1980 AK Member Subscriber

    Messages:
    25,137
    Location:
    Angel Station, Alabama
    There was some bypass I replaced substantially more low esr capacitance with film augmentation to each channel during the upgrade.
     
  19. gadget73

    gadget73 junk junkie Subscriber

    Messages:
    35,878
    Location:
    Southern NJ
    Many of them do tie the cathodes to the heater string. With 6bq5 thats typically somewhere about +13v. Not all phase inverters run the cathode at a lot of voltage either, some are basically at ground potential so there isn't much need to elevate it. Split load / cathodyne / concertina where it has output from both plate and cathode to the output tubes are where you have issues. Also designs with direct coupled stages tend to have at least one cathode that is a bunch of volts above ground.

    Deriving it from B+ is probably the better way, but the cathode voltage was "good enough" and it cost zero extra parts to use.
     
  20. Pio1980

    Pio1980 AK Member Subscriber

    Messages:
    25,137
    Location:
    Angel Station, Alabama
    One of the advantages of the paraphase front end (Magnavox) is the cathodes being nearly 0 volt bias. Elevating the heater chain via the output cathode bias is free and more than adequate.
    Mine is a cathodyne phase splitter.
     

Share This Page