1. Time for some upgrades in server hardware and software to enhance security and take AK to the next level. Please contribute what you can to sales@audiokarma.org at PayPal.com - Thanks from the AK Team
    Dismiss Notice

Tracking Down Hum and Buzz

Discussion in 'Tube Audio' started by Paully, Nov 7, 2018.

  1. Paully

    Paully AK Subscriber Subscriber

    Messages:
    272
    Location:
    Alabama
    Touching the outer shells together does nothing one way or the other. Accidentally touching the middle pin conductor to the outer shell of the other definitely did raise buzz.
     

     

    Please register to disable this ad.

  2. maxhifi

    maxhifi AK Subscriber Subscriber

    Messages:
    2,974
    Location:
    Edmonton Alberta Canada
    Okay, now we are getting somewhere. This should have killed the buzz, not raised it.

    What's happening is power supply noise is getting into the signal ground. Did you build this kit yourself? Has it ever worked right?

    Once you post photos, I can give some more ideas.
     
  3. Paully

    Paully AK Subscriber Subscriber

    Messages:
    272
    Location:
    Alabama
    I did build this myself and it did work perfectly before. I will post some photos soon, just can't get to it right now.
     
  4. maxhifi

    maxhifi AK Subscriber Subscriber

    Messages:
    2,974
    Location:
    Edmonton Alberta Canada
    Cool.
     
  5. Paully

    Paully AK Subscriber Subscriber

    Messages:
    272
    Location:
    Alabama
    I do have two top photos, getting into the amplifier is of course a little more involved!

    [​IMG]

    [​IMG]
     
  6. maxhifi

    maxhifi AK Subscriber Subscriber

    Messages:
    2,974
    Location:
    Edmonton Alberta Canada
    What type of input jacks are those, are they isolated from the chassis, or are they grounded to the chassis?

    If you measure the resistance between the two input jack shells with a DMM, what value do you get?

    How about from the input jack shell, to the metal top of the main filter capacitors? If you get infinite ohms there, scratch through the coating on top of the cap with the tip of the probe a little so it touches bare metal.
     

     

    Please register to disable this ad.

  7. Paully

    Paully AK Subscriber Subscriber

    Messages:
    272
    Location:
    Alabama



    I don't know the specifics of the Jacks, they are from Dynakits and were old stock jacks and they should have been installed in exactly the way the schematic indicates. I put a link a while back. Resistance between the shells of the two input jacks measures 0.0 ohms. Ringing out the shell to the chassis shows they are connected and therefore grounded. Resistance from the shell to one of the large electrolytic capacitors doesn't give consistent readings. Thought one time was 11 ohms but then 30 Kohm, 70 kohms, 90 kohms and the readings start to fall while leaving the meter leads in place. If put the red lead on the shell I get no reading. If I swap the leads then I get a reading.
     
  8. maxhifi

    maxhifi AK Subscriber Subscriber

    Messages:
    2,974
    Location:
    Edmonton Alberta Canada
    Might be prudent to flip it over, and measure all the grounds. If this worked properly before, you're probably correct about a cold solder joint. The negative terminal of the big filter caps should be 0.0 ohms to ground too. Also try to measure continuity of the shield of your patch cords, just to check.
     
  9. Paully

    Paully AK Subscriber Subscriber

    Messages:
    272
    Location:
    Alabama
    I hate to be dense (I often can't help it) but what does "measure continuity of the shield of your patch cords" mean and how do I do that?
     
  10. maxhifi

    maxhifi AK Subscriber Subscriber

    Messages:
    2,974
    Location:
    Edmonton Alberta Canada
    Set the meter to ohms, and measure between the shells of the cable on each end, and then the tips. The reading should be reasonably close to zero ohms. Do it for both the left and right cords.

    You already said you tried a second set of patch cords, but it doesn't hurt to double check things, the way touching a shield made it buzz made me think one of them is acting up.
     
  11. Paully

    Paully AK Subscriber Subscriber

    Messages:
    272
    Location:
    Alabama
    Got it, I didn't recognize the term patch cords. Good advice and I will check the cords. Should be able to get to all of this late this afternoon or tomorrow morning. Thanks again!
     

     

    Please register to disable this ad.

  12. maxhifi

    maxhifi AK Subscriber Subscriber

    Messages:
    2,974
    Location:
    Edmonton Alberta Canada
    Oh, patch cords are what people called "interconnects" before companies decided to charge big money for them and call them something pretentious :) That's me being inconsistent with terminology.
     
    UncleBingo and Paully like this.
  13. Paully

    Paully AK Subscriber Subscriber

    Messages:
    272
    Location:
    Alabama
    Honestly, I like patch cords a lot better.
     
  14. maxhifi

    maxhifi AK Subscriber Subscriber

    Messages:
    2,974
    Location:
    Edmonton Alberta Canada
    https://en.wikipedia.org/wiki/Patch_cable

    "A patch cable, patch cord or patch lead is an electrical or optical cable used to connect ("patch in") one electronic or optical device to another for signal routing. Devices of different types (e.g., a switch connected to a computer, or a switch to a router) are connected with patch cords:

    "The term "patch" came from early use in telephony and radio studios, where extra equipment kept on standby could be temporarily substituted for failed devices.[citation needed] This reconnection was done via patch cords and patch panels, like the jack fields of cord-type telephone switchboards. Furthermore, patching could also create temporary atypical connections between devices for unusual needs. Analog music synthesizers typically use patch cables to interconnect functional sections, such as oscillators, filters, etc."
     
  15. dcgillespie

    dcgillespie Fisher SA-100 Clone Subscriber

    Messages:
    9,709
    Location:
    Ball Ground, GA
    Max is going exactly where I am thinking as well. I am completely unfamiliar with the Bottlehead, so I'm at a disadvantage there. But I would suggest two things:

    1. Make sure you are using quality interconnects. In a pinch, I recently went to Home Depot and purchased a 6 foot set of interconnects for use between my Fisher preamp and power amp -- only to have it hum notably. Swapped the interconnects out with a different set -- no hum. What I found was that cheap cables today do not even remotely wrap or shield the inner conductor but rather, simply run a bare ground lead next to the inner conductor, and count on the very low output impedance of today's modern SS devices to reduce any hum pickup to minimal levels. So, make sure you're using good, well shielded interconnects.

    2. Taking Max's suggestion even further, using a known good set of interconnects, plug one end of the interconnects into the Dynaco, and at the other end, bring them together face to face, so that the outer ground shield of one plug is grounding out the top of the other plug, and visa versa. In this way, the top of both inputs will be grounded out by the ground lead of the other channel.

    If the Dynaco is quiet in this test, then the problem must lie within the Bottlehead, and particularly with the grounding circuit between the output jacks of the two channels. On the other hand, if the buzz and hum are present, then the problem may be the inherent issue with the physical lay out of the Dynaco due to the power transformer being mounted in between the two amplifier input jacks. This allows the input and ground circuits for each channel to be on opposite sides of the magnetic field generated by the power transformer, which then generates a ground current through the interconnects when their ground leads are connected together at the source end. This is often a problematic way to do things, but normally, the ST-35 is not cited for this kind of problem -- but that is also based on it being used in conjunction with a Dynaco PAS preamp. Most particularly however, note that to minimize such concerns, Dynaco did not connect the ground terminals of the output jacks in the PAS directly to ground, but to ground through 10Ω .5 watt resistors. These resistors act to allow an appropriate ground reference to be made, but also minimize any ground current that might flow from the ground loop conditions described. You might look at using these resistors in your bottle head if the suggested test indicates that the noise is originating from the Dynaco.

    As Max said, this problem is all but eliminated in the ST-70, since the input jacks are spaced right next to each other in the middle of the power transformer's center line. This kind of configuration allows for little if any ground currents to be developed between the channels, all but eliminating the problem. However, just to be sure, Dynaco included the 10Ω resistors inside the ST-70 as I recall as well.

    I hope this helps -- let us know!

    Dave
     
  16. Paully

    Paully AK Subscriber Subscriber

    Messages:
    272
    Location:
    Alabama
    I appreciate everyone's input. I have used the Dynaco with the Bottlehead preamp for a good while prior without incident and they sounded wonderful together. So I wouldn't have thought it would be a design problem inherent in the Bee Pree. And the preamp doesn't buzz or hum with other another amp. Doesn't mean something hasn't gone wrong but I just don't understand how it could be the preamp by design.

    I am using Cardas cables and Blue Jean that again don't have a problem in any other context, and I thought they were decent quality. I will of course go do the test touching the ends of the cables. What is so baffling to me is that shorting out the inputs and having no noise should prove it isn't the amp. The preamp has no problems in any other context and never had a problem with the Dynaco which should prove it isn't the culprit. I am using two sets of pretty good cables. Shouldn't be them. Nothing is left! I am going to follow everyone's advice and go searching for a cold ground solder joint and do that test you suggest with the interconnects. I hope to get to it tomorrow but it may have to be Monday. Thanks to both of you for trying to help me, it is appreciated.
     
    Last edited: Nov 8, 2018

     

    Please register to disable this ad.

  17. trobbins

    trobbins Active Member

    Messages:
    471
    Location:
    Melb, Australia
    It's interesting that the ST35 doesn't take the COM lead from the speaker terminal strip back to #2 on the pcb for that channel, rather than what would appear to be a noisier point on the CT connected lug of C8.
     
  18. knockbill

    knockbill Addicted Member

    Messages:
    7,782
    Location:
    SE PA
    Are you using the same cables in the quiet amp/preamp connection, with the amp/preamp that is noisy? that should take the cables out of the equation...
     
  19. Paully

    Paully AK Subscriber Subscriber

    Messages:
    272
    Location:
    Alabama
    Yes
     
  20. knockbill

    knockbill Addicted Member

    Messages:
    7,782
    Location:
    SE PA
    OK,,, If the jacks are well secured to the amps/pres, and both either grounded to the chassis or insulated, that will give ya an even playing field... I rebuilt a pair of Heath pre amps a while ago, that had a hum... The manual said to make a pair of RCA patch cords with the shield only connected to jacks on one end of the cable, grounded end plugged into amp, to prevent ground loops... It worked on couple different amps... Just a thought, may not be your problem tho...
     
    Paully likes this.

Share This Page