Bizzaro fix for hum problem needs explanation! Pretty Please - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 30th March 2013, 10:20 PM   #1
diyAudio Member
 
Join Date: Jan 2012
Default Bizzaro fix for hum problem needs explanation! Pretty Please

Hi
In the process of trying to fix a major hum problem in a tube line stage, I stumbled on a weird temporary solution. I think this weird fix is trying to tell me what the underlying problem is, but being relatively inexperienced with electronics and tubes I havenít a clue.

Background and sad story of woe:
For the output stage of a DAC I choose to use John Broskieís BCF_9 broskie cathode follower board (to take the balanced output of the DAC and convert it to an unbalanced signal for my audio system).

John Broskie from GlassWare audio design has an excellent reputation for making decent circuits and PCBs but unless there is a missing or broken part he doesnít respond to emails for technical questions like what follows:

As you can see from the pics the build is nearly complete. I probably should have first tested the BCF_9 board before installing it but I thought I would just go for it.

Couple of days ago I checked the power supply voltages (H+ 12.7 VDC, B+ 300VDC, H- set to 1/4 B+), installed the tubes and was dismayed to have a major hum/buzz from the left channel. Also I would get a repetitive ticking sound.

Right channel is quite. Sound is overall very good from both channels but the hum/buzz/ticking from the left channel too loud to take.

I swapped tubes, tried a different set of 12BH7 tubes, checked all parts to make sure they were the recommended values (see circuit diagram), doubled checked with connections and resistor values with VOM, checked and reflowed every solder connection (twice). Still the left channel hums/buzzes and intermittently ticks (with or without an audio signal input). Completely disassembled and tested still hum in left channel

Pulled the tubes and the noise stops

Tried to trace the hum using an oscilloscope . I think the left channel noise signal is around 60Hz (compared to a signal generator) but not smooth (more saw tooth like and pretty distorted). But this is my first attempt using a scope.

What I did find is that the noise could be readily detected on left channel tube pins (V1 in diagram) 3 and 6.

However when I probed pin 2 or pin 1 the noise stopped. Note: I was using a 1x probe with ground attached to B-/ground. By chance I came upon the fact that the noise also stopped if I touched pin 2 or pin 1 with a test lead/probe that was not connected to anything.

So I tried several lengths of wire soldered directly to tube V1 pin2 (white wire in pics). To completely suppress the hum/buzz and ticking in the left channel the wire on pin 2 must be about 2-3 ft long and needs to be stretched out (cannot be wound into a smaller coil).

This seems antenna_ish to me with the wire influencing some reactive circuit. There must be a bad part or connection somewhere (V1 + input stage?). But how do I determine which component is defective and/or how to really fix it? Beside trying more voodoo?

This is driving me nuts and I would greatly appreciate your advice before I toss this cursed thing out
Attached Images
File Type: jpg BCF_9.jpg (296.9 KB, 242 views)
File Type: jpg photo1.jpg (418.5 KB, 207 views)
File Type: jpg photo2.jpg (640.2 KB, 207 views)
File Type: jpg photo3.jpg (637.6 KB, 197 views)
  Reply With Quote
Old 30th March 2013, 10:57 PM   #2
diyAudio Member
 
nigelwright7557's Avatar
 
Join Date: Apr 2008
Location: Carlisle, England
In my experience hum usually comes from poor ground wiring.
Its vital that power supply surging on the peak of the supply voltage causing pulses of current in the ground wire don't get into the audio signal chain.

In valves you have the extra dimension of heaters which can cause pick up if not carefully wired and routed.

I built a valve mixer and it hummed very badly.
Changing to DC heaters made a small difference.
However rerouting the ground wire from transformer to power supply capacitors and then to amplifier made a big difference.
  Reply With Quote
Old 31st March 2013, 01:26 AM   #3
DaveG is offline DaveG  United States
diyAudio Member
 
DaveG's Avatar
 
Join Date: Feb 2007
Location: Grease
I helped work on an amp with a similar problem. Heater AC bleeding into one channel.
Everything was swapped. No change.
Lifted various components. No change.
Ran wires off board to a separate stand alone tube socket. (completely bypassed the original socket with a new output tube and socket) No change.
Finally tracked it down to a problem with the tube socket, but the thing is, the socket and all soldering looked fine, even with an 8 power loop.
No amount of wiggling or poking of the individual socket pins made any difference when checking with a meter.
Eventually out of frustration unsoldered the tube socket and it finally stopped. Close inspection revealed nothing. Solder it back in and all was fine. Must have been an extremely small solder whisker somewhere that was hidden underneath the socket.
  Reply With Quote
Old 31st March 2013, 02:12 AM   #4
diyAudio Member
 
Chris Hornbeck's Avatar
 
Join Date: May 2011
Location: Little Rock
You're probably seeing a parasitic oscillation. Cathode followers oscillate just fine without some resistive degeneration at their outputs. I'd try just removing their cathode resistor bypass caps to see if that's enough.

All good fortune,
Chris
  Reply With Quote
Old 31st March 2013, 05:25 AM   #5
diyAudio Member
 
rfengineer2013's Avatar
 
Join Date: Feb 2013
Location: Colorado
Hi Sybednar,

C5, according to Broskie's documentation on the BCF-9, connects pin 9 of the tube to ground on his PCB design.

The 12BH7 you are using uses pin 9 for the heater center-tap. So if you put a capacitor in the C5 location, you are tying the center of the heater to ground through a capacitor.

Since you're using the B+/4 heater bias (if I read your post correctly), this might not matter... But I'd remove it anyway - it could only cause mischief. (The B+/4 heater bias should take care of the problem of heater hum coupling to the top-triode cathode - good that you used it!)

Also, was C10 part of the original design? I was looking at Broskie's manual for the BCF-9, and I don't see any capacitor from the grid of the top triode to ground in his documentation. It's unusual to put an AC bypass directly on the grid of the triode.

The cap is in a relatively high impedance part of the circuit, and could be causing the top triode to motorboat (if it's a "pop-pop" noise at low frequency)

I'd pull that cap out too, unless there's a good reason for it to be in there.

It's also possible that the C10 cap is picking up a bit of hum and injecting it onto the top triode grid - your gimmick antenna is probably injecting AC out-of-phase (and thus cancelling the noise). Your test probe touching the grid is providing a different path to ground, also cancelling the noise and stopping the motorboating.

All "educated guesses" - pull C10 and C5 and see if it takes care of your noise.

Best luck

~ Sam

EDIT - I just noticed that your board is a later version than the board in Brodskie's PDF from the TubeCad website... Might be a design change John put in - there is most certainly a space on your board for C10, whereas on the older version (2009) that was where the input wires went. I'd still pull those two caps - for sure C5 is not necessary (if it connects to pin 9 on the new board ) and C10 is some kind of bypass cap that IMHO is going to cause mischief. Good luck!
__________________
-- Just an average Man of Bronze --

Last edited by rfengineer2013; 31st March 2013 at 05:46 AM.
  Reply With Quote
Old 31st March 2013, 04:52 PM   #6
GoatGuy is offline GoatGuy  United States
diyAudio Member
 
Join Date: Dec 2012
Location: SF Bay Area
Specific question(s)

[1] if you use shorting-plugs on inputs, does problem go away?

The use of a long 18 gauge wire on pin 2 (the grid of the affected stage) indicates exactly what you surmise: induced pickup of EMF 60/Hz noise from the power supply.

Questions

[2] What are R7, R8 and R9 values?

GoatGuy
  Reply With Quote
Old 31st March 2013, 06:48 PM   #7
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
Sounds like parasitic oscillation, as Chris suggested. Add or improve grid stoppers. For CFs add a cathode stopper.

Add a grid stopper between R1 and pin 2. The existing resistor is not a grid stopper but part of a low pass filter.
  Reply With Quote
Old 31st March 2013, 07:21 PM   #8
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
Parasitic oscillations seem highly likely to me too.. I recommend removing C5 on both channels and reporting back.
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 31st March 2013, 08:10 PM   #9
diyAudio Member
 
rfengineer2013's Avatar
 
Join Date: Feb 2013
Location: Colorado
Quote:
Originally Posted by kevinkr View Post
Parasitic oscillations seem highly likely to me too.. I recommend removing C5 on both channels and reporting back.
Also remove C10.

BTW - what is the part number for the C5/C10 (blue) caps? It looks like an Epcos logo, possibly a film capacitor, but the "M1091..." doesn't cross. Possibly there is a "Bxxxx" part number on the side?



C10 will be at about B+/2 potential - if the voltage rating is not sufficient then that C10 cap on V1 might be breaking down as it charges towards B+/2 and causing that sawtooth-like waveform you're seeing.

EDIT - Epcos has such a fun marking system... From your closeup photo, it looks like the 2nd line marking is " B7 -u1M(?)305(?)W(something) ".

If I'm reading it correctly, you've got a 0.1uF cap installed at C5 and C10 locations. Way too big, at least for C10 - take them both out.
__________________
-- Just an average Man of Bronze --

Last edited by rfengineer2013; 31st March 2013 at 08:40 PM.
  Reply With Quote
Old 1st April 2013, 01:40 AM   #10
diyAudio Member
 
Join Date: Jan 2012
Default Problem solved! Thanks

Thanks to everyone for their speedy and helpful suggestions. Based upon the phenotype of the noise and your suggestions I removed C5 and C10 (and the meter of additional wire from pin 2). I was elated to not hear anything. Left and Right channels without input are hum free and the motor boating I was hearing is also gone. Weird how happy this made me and the sound with audio input is very nice

The purpose of C5 and C10 according to Broskieís manual for the new version of the BCF_9 was to generate a low pass filter (<80kHz cutoff) which John recommended if using the BCF-9 as an output stage for a DAC.

Now that I have removed C5 and C10 will the lack of a low pass filter be a problem when I connect it to a DAC? Should I try different values of caps and resistor in the input section that would not create the motor boating and parasitic oscillation?

Again thanks keeping me from going nuts with this hum issue
Cheers
Sebastian

Below are a few additional responses and answers to some of the questions raised:



Quote:
Originally Posted by GoatGuy View Post
Specific question(s)

[1] if you use shorting-plugs on inputs, does problem go away?

Questions

[2] What are R7, R8 and R9 values?

GoatGuy
Values on R7, R8 and R9 are 1 Meg Ohm. I remember reading about using shorting plugs for trouble shooting hum problems but forgot to test this. What does this test reveal?


Quote:
Originally Posted by kevinkr View Post
Parasitic oscillations seem highly likely to me too.. I recommend removing C5 on both channels and reporting back.


Quote:
Originally Posted by rfengineer2013 View Post
Also remove C10.
how right you two were. Thanks

Quote:
Originally Posted by rfengineer2013 View Post
EDIT - Epcos has such a fun marking system... From your closeup photo, it looks like the 2nd line marking is " B7 -u1M(?)305(?)W(something) ".

If I'm reading it correctly, you've got a 0.1uF cap installed at C5 and C10 locations. Way too big, at least for C10 - take them both out.

These caps came with the board from Broskie and the manual recommended C5 and C10 values of 100pF-0.01 uf. Yesterday I spent some time trying to determine what their value but the Epcos website and docs werenít much help. The label on the top of the caps (same for both C5 and C10) is B7-u?1M305WU. For future reference can you tell me what their value is?
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Bizzaro fix for hum problem needs explanation! Pretty Please sybednar Tubes / Valves 0 30th March 2013 10:17 PM
tube amp hum........normal or suggested fix? sybednar Tubes / Valves 30 4th January 2013 02:23 AM
Unable to fix hum on my F5 anilva Pass Labs 43 24th May 2011 07:06 PM
Explanation for this problem? Rob11966 Tubes / Valves 4 22nd September 2007 04:39 AM
Tivoli Model1 PSU hum fix / upgrade? jives11 Solid State 1 24th May 2004 08:30 AM


New To Site? Need Help?

All times are GMT. The time now is 02:38 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2