The Christiansen "DG" 300B Amplifier Build Thread

[Chromako]

  INo amount of shielding will do anything for 60 Hz hum induction. Unless we're talking shielding using µ-metal. EMI is not likely to downconvert to 60 Hz by chance. 
 
 

Nice that you mention that, with my obsession with EM shielding... I agree with you. Not 60Hz. That'd be a ridiculous coincidence. Unless you've got a poltergeist in the electrical grid.
 
Hmm... now that I think of it... have you tried doing an exorcism? 

 
Of course, this is all moot as he's sent it out already to get a second set of eyes, but it's fun to "armchair solder."
 
But yes, it worked before. It'll work again!

 

[holland]

In my experience, 60Hz is usually a grounding problem, and/or mains noise coming in somewhere.  The noise is probably on both channels, but different voltage levels.  You'd probably hear it very loud with IEMs.  I'm not convinced the noise was never there.  I think it always was, just not heard as loudly.
 
Assuming the disconnected transformer noise is a red herring....perhaps try grounding the primaries of the OPT and see if the noise is there still there.
 
Switching the wiring for output tubes might be a start, but I would look at the input transformer as well.  IIRC, the board is a single board for both channels with wires for the 300B.  Also, another IIRC, the DC heaters are floating, perhaps ground them if they aren't already.
 
I also believe there was prior discussion about pulling the driver tube and using a resistor divider to set the grid voltage for the 300B, can play with that again.
 
The cold spray is an interesting recommendation, I've not tried that one before.
 
If he sent it to Electra-Print, I'm sure Jack will find the problem.

 

[sceleratus]

   
If he sent it to Electra-Print, I'm sure Jack will find the problem.

Winner, winner, chicken dinner.

 

[sceleratus]

In no way do I think the design is bad.  Frank, Randy, and I have heard how nice it can sound and I believe it sounded better after the "pseudo balance" (Frank) inputs.   I greatly appreciate everyone's suggestions and support.
 
I am lacking the expertise to find all but an obvious problem.  
 
I check for shorts, loose wires, implementing ground isolation, purchased a 3rd party ground loop product,  disconnecting all inputs from the driver board, disconnecting the OPT's from the driver board.  I checked the capacitance values of the "off brand" caps.  They are exactly the value called out in the BOM.  I did not deviate from any component values in the BOM.   The build wiring diagram specifically shows XLR's for the inputs.  I followed the output wiring diagram and tied XLR Pin-1's together and to mains ground.  I have rolled everything possible to roll.  I have moved the OPT's an additional inch further out from the power transformer.
 
The box is still hear.  The dense foam packing materials arrive from U-Line today.  Jack told me not to change anything.  Just send it to him.  He has been very kind by exchanging many emails with suggestions.  Strike that "directions" as well as "what are you trying to do, knucklehead"  Yet he hung in there.  He said he's going to take a wall of gear and print it before and after.  I couldn't resist having that done.
 
Then when some one asks how quiet it is or what's the spec.  I will have it.
 
Tom,
You designed one hell of an amp.  Something went south, either organically or do to an error on my part.  So much of this baffles me as it hasn't produced the expected results from my testing and I am reticent to dive into the board.  My scope is maxed out at 400V and I don't know if I can probe with a resistor.
 
All I know the left side has escalated to "Bug Zapper' without the bugs.

 

[sceleratus]

  In my experience, 60Hz is usually a grounding problem, and/or mains noise coming in somewhere.  The noise is probably on both channels, but different voltage levels.  You'd probably hear it very loud with IEMs.  I'm not convinced the noise was never there.  I think it always was, just not heard as loudly.

 
 
That is a real possibility.
 
Quote:

  In my experience, 60Hz is usually a grounding problem, and/or mains noise coming in somewhere.  The noise is probably on both channels, but different voltage levels.  You'd probably hear it very loud with IEMs.  I'm not convinced the noise was never there.  I think it always was, just not heard as loudly.
 
Assuming the disconnected transformer noise is a red herring....perhaps try grounding the primaries of the OPT and see if the noise is there still there.
 
Switching the wiring for output tubes might be a start, but I would look at the input transformer as well.  IIRC, the board is a single board for both channels with wires for the 300B.  Also, another IIRC, the DC heaters are floating, perhaps ground them if they aren't already.
 
I also believe there was prior discussion about pulling the driver tube and using a resistor divider to set the grid voltage for the 300B, can play with that again.
 
The cold spray is an interesting recommendation, I've not tried that one before.
 
If he sent it to Electra-Print, I'm sure Jack will find the problem.

 
I think I'm just going to box it.  Why?   I'm sending it to Jack regardless and if I mung up something else during testing will just make it more difficult to find.  I'm sure he'll find it.  Then I will report back and end the speculation (Best case scenario)
 
I don't care if I have to order another Driver Board and built it from scratch.  Hell, depending on the outcome, I might do just that, so I have a stock board and board for modding. It's gonna get fixed.  Then I'm going to listen to music for quite some time.

 

[sceleratus]

Question for educational purpose.
 
I want to revisit the screen captures below.  It's odd, to me, and I'd like to understand what might be happening.
 
I put a 114Hz tone on the left input, turned the attenuator to medium, probed the left output with the scope and captured the wave pictured 10.0V at 5.00ms.  Right and Left XLR outputs had an  R50 25W "Load Plugs" across pins 2-3.
 
I moved the probe to the right side, keeping the signal on the left input.  I got nothing until I turned the volume all the way up and captured the small wave in the other picture.  Is there a reason for this?
 


 
 
[size=medium]Next question:[/size]
Why can I capture a 60Hz sine wave (below) with nothing connected to the inputs, yet when I put 114Hz  freq (Image above) on the input, the 60Hz is not visible?  The 114Hz is perfect.  Is it because the 144Hz tone is at 10.0V and the 60Hz noise is at 200mV?  Also the phase shifts about every second.  I setup my camera to capture this..  Thanks

 
​

 

[tomchr]

How are you grounding your scope probe? The 60 Hz pictures you post look like the scope either isn't triggering properly or you forgot to ground the scope probe.
 
~Tom

 

[sceleratus]

J6 P3-4

 

[tomchr]

  Also, another IIRC, the DC heaters are floating, perhaps ground them if they aren't already.
 
I also believe there was prior discussion about pulling the driver tube and using a resistor divider to set the grid voltage for the 300B, can play with that again.

 
The heater supplies for the 300B are grounded. The heater for the small-signal tubes are biased by a resistive divider to about -100 V. Do NOT change this as the Vhk spec of the little tube will be violated. The heater is at AC ground through a 100 nF cap.
 
I suggest pulling the 300B and measuring the voltage of the grid pin. That'll tell you if the hum gets in before or after the 300B.
 
My money is on:
1) Bad connection
2) Missing ground
or 3) ground loop.
 
I'd say 90 % chance of 1). 9.9 % chance of 2), and 0.1 % chance of 3).
 
~Tom

 

[tomchr]

  J6 P3-4

 
What happens if you use J4 (or J8) pin 1 of the driver board? You can also use 300B pin 1.
 
~Tom

 

[sceleratus]

   
What happens if you use J4 (or J8) pin 1 of the driver board? You can also use 300B pin 1.
 
~Tom

I will set up in a few minutes and let you know.
When I use the scope I terminate  a lead to J6 P3 ground and clip the probe to that.
 
With a DMM I use the probes an touch the terminal block screws.

 

[sceleratus]

   
The heater supplies for the 300B are grounded. The heater for the small-signal tubes are biased by a resistive divider to about -100 V. Do NOT change this as the Vhk spec of the little tube will be violated. The heater is at AC ground through a 100 nF cap.
 
I suggest pulling the 300B and measuring the voltage of the grid pin. That'll tell you if the hum gets in before or after the 300B.
 
My money is on:
1) Bad connection
2) Missing ground
or 3) ground loop.
 
I'd say 90 % chance of 1). 9.9 % chance of 2), and 0.1 % chance of 3).
 
~Tom

What about TU1 and TU3, pull those as well?
I can't tell you how many times I have tugged on leads tied down to the blocks.
Can't tell you about solder connections.  
The candidate for most sensitive are loops from the board to TU2 and TU4.
Followed by upgrades / reworks of C4A C9A   R6, R7, R20, R21.   R9,R23.   R10 R24
 

 

[sceleratus]

Without the valves there is a quite hum. Both sides are equal.  Nothing like the loud buzz/hum with them in.
 
I will take measurements

 

[tomchr]

  Without the valves there is a quite hum. Both sides are equal.  Nothing like the loud buzz/hum with them in.
 

 
There's hum in the headphones even with all the valves unplugged? That makes no sense at all. I suppose it could be coupling from the power transformer to the output transformers. Regardless, I'm not sure it's relevant as you won't be operating the amp with the OPTs disconnected.
 
~Tom

 

[sceleratus]

J4 / J8 Pin1 works fine as a ground with the DMM
With both sets of tubes pulled...
[size=medium]All voltage measurements for TU1, TU3, TU2, TU4  conform with the correct values in the startup procedure.[/size]
[size=medium]OPT input voltage is 400V[/size]