[sceleratus]

Quote:

 
An attenuator is basically a potentiometer with steps instead of linear travel. The number of sections needed for a volume control doesn't change just because it's labeled "attenuator" rather than "potentiometer". Hence, for a differential volume control, you'd need an attenuator with four decks.
 
I used to use an attenuator, but it drove me nuts. If you look at most attenuators, you'll see that to get a reasonable range on a 24-position rotary switch, the first couple of steps are *HUGE*. Like 10 dB or so. In my case, I ended up with a setup where step 3 was so low volume that I could barely hear the music, and step 4 was so loud I couldn't concentrate. 
 
In my preamp, I use a PGA2320. -95.5 dB to +31.5 dB in 0.5 dB steps. You won't find that in a rotary attenuator.
 
The absolute tolerance of the potentiometer may be 5 %, but really it's the tracking between channels that matter. This will improve with an attenuator. But if it isn't bothering you now, it clearly isn't an issue for you.
 
An attenuator will not fix the hum. Fix the ground loop before you start thinking about more modifications.
 
You had an amp that had no hum. Then you modified it and got an amp with hum. Something in the conversion went wrong and caused the hum. Trace it down and fix it. Then you can start talking about modifications.
 
~Tom

Precisely
No hum > Mod > Hum > mod fail
 
I looked into the attenuator because of the whole differential balanced interface thing and that the amp was designed for a pre-amp and that these decked stepped attenuators are pseudo pre-amps.....  That's how I got there.
 
Step one. get it working again. 

 

[sceleratus]

Volume update:
 
When I started this thread I said I would post the good, the bad, and the ugly.  Well….
 
Here's ugly.
After reading your suggestions and exchanging an email or two with Tom, I was getting the impression that what I wired up was wrong.  But I didn't know why.  Tom finally figured out what I was doing and asked "are you trying to use the volume pot with XLR"?
 
Well, yes. 
 
Here's a tip: RCA and XLR are fundamentally different.  Something about differential volume going in, or as Holland pointed out, move the pot to the other side of the IPT's.   I had no idea, I wasn't getting it, I now know the fix, but I can't say I understand it.  I'm not keen about munging with the board.
 
I "get" stepped attenuators.  Simply "stepping" through resistor values that are inline with the source.  Thinking…..(while typing) that's much different than a "T" configuration in a potentiometer.  Probably why my wiring didn't work.
 
If you have a 4 deck attenuator, it sunk in that each deck is for each side of the source connection.  A source to deck link-in and deck a to the board IPT.  2 x stereo = 4 decks.  What marries the decks?  No matter, if I go that way I'll figure it out.  Or some nice person can enlighten dopey.  (Find section in Morgan Jones)
 
I think I'm committed to balanced XLR.  The rub is the attenuators are quite large and with my centered volume control it might run into the driver board.  If so, I found a flexible shaft extension that might mitigate that.
 
 

 
 
The other choice that came up (my reading and Tom) was TI's PGA2320  volume control IC.  I see (NPI) that it has an RS232c interface, presumably for programming.  I also presume there is MMI software.  Parallels works with a USB Serial adapter on a Mac so I think I can run MMI.  However can I control it with a tactile knob?  Then there's  building a circuit.  Might be a good time to try to produce something with DipTrace.  Always wanted to etch a board.
 
So there you have it, Mr. Soldering Iron has run into, as Tom would say, Physics.
 
And lost.

 

[sceleratus]

I went to diyAudio and searched TI2320.
 
Sounds much too complicated.
Additional power supply and a microprocessor.
 
That leaves.
A) Stepped Attenuator
B) RCA.   I am so far along and such a buck naked rookie that I don't wan to carve up the driver board.
 
I had though about building a second Driver Board just for experimentation.  This might qualify.
 
I love this amp.  Hell it even sounds good when it's mis-wired.

 

[holland]

I use a relay attenuator with 100 steps for my nicer builds.  It's nice, but it takes up lots and lots of space.
 
You can wire the pot before the IPT, look at http://www.head-fi.org/t/664903/the-christiansen-dg-300b-amplifier-build-thread/300#post_9653109 for the schematic.  It adds extraneous steps, but it should work fine.  Easiest is definitely after the IPT, but if you don't want to hack the board you only have 3 options.  Back to RCA, 4-gang pot, or the above link.

 

[sceleratus]

Quote:

I use a relay attenuator with 100 steps for my nicer builds.  It's nice, but it takes up lots and lots of space.
 
You can wire the pot before the IPT, look at http://www.head-fi.org/t/664903/the-christiansen-dg-300b-amplifier-build-thread/300#post_9653109 for the schematic.  It adds extraneous steps, but it should work fine.  Easiest is definitely after the IPT, but if you don't want to hack the board you only have 3 options.  Back to RCA, 4-gang pot, or the above link.

Excellent.
Another choice.
 
I just ran the Driver BOM at Digikey.  It's $35 for the original parts,  I have a few changes.
Then there's the cost of the board and the Jensen transformers.  $140
 
But it might be nice to have, then I wouldn't be timid about changing things.  A learning tool.
If Tom came up with some things hed like to try on an HP amp (hint) I would be good to go.

 

[sceleratus]

Quote:

I use a relay attenuator with 100 steps for my nicer builds.  It's nice, but it takes up lots and lots of space.
 
You can wire the pot before the IPT, look at http://www.head-fi.org/t/664903/the-christiansen-dg-300b-amplifier-build-thread/300#post_9653109 for the schematic.  It adds extraneous steps, but it should work fine.  Easiest is definitely after the IPT, but if you don't want to hack the board you only have 3 options.  Back to RCA, 4-gang pot, or the above link.

I didn't read your post close enough the first time.  I need to read things at least tree time due to some learning issues.
 
In the top diagram the pot is external to the board / IPT's?   I now see the +IN and -IN notation in the schematic.
The termination sequence of a schematic won't match the terminal wiring.. correct?  As +IN is in terminal block position #2.  Just stating what I think is obvious for peace of mind.
 
 

 
 
This seems like a low risk, high reward, nothing to lose choice.
 
The second choice is after the IPT's.  That I understand by the transformer icon.

 

[sceleratus]

Holland,
 
I have a 100K Alps pot and a 10K Alps pot.
I am using the 10K now.
 
Since your schematic is designed for 100K that's the one I should use?

 

[holland]

Quote:

I didn't read your post close enough the first time.  I need to read things at least tree time due to some learning issues.
 
In the top diagram the pot is external to the board / IPT's?   I now see the +IN and -IN notation in the schematic.
The termination sequence of a schematic won't match the terminal wiring.. correct?  As +IN is in terminal block position #2.  Just stating what I think is obvious for peace of mind.
 
 

 
 
This seems like a low risk, high reward, nothing to lose choice.
 
The second choice is after the IPT's.  That I understand by the transformer icon.

 
Yes, you can put the schematic before the IPTs.  The schematic is really for using an IPT or no IPT, but in your case it doesn't matter.  It will just feed into the IPT.
 
Either pot will work, but feel free to use the 100K first.  Whichever tracks better for you.  You should also be able to take the schematic from the 2 47.5K inline resistors and then the pot, but start verbatim to be sure.
 
Do wire the XLR without the pot first, to ensure you have no noise.  If you do have noise, then all of this is moot.  Fix the noise (if it's there without the pot), and then add the pot.

 

[sceleratus]

Quote:

 
Yes, you can put the schematic before the IPTs.  The schematic is really for using an IPT or no IPT, but in your case it doesn't matter.  It will just feed into the IPT.
 
Either pot will work, but feel free to use the 100K first.  Whichever tracks better for you.  You should also be able to take the schematic from the 2 47.5K inline resistors and then the pot, but start verbatim to be sure.
 
Do wire the XLR without the pot first, to ensure you have no noise.  If you do have noise, then all of this is moot.  Fix the noise (if it's there without the pot), and then add the pot.

This is worth trying because, as I said, it's low risk, but secondly, the smooth attenuation that a pot gives you is much nicer than steps.

 

[sceleratus]

Wow.
 
I'm reading M. Jones section on volume control, and I'm actually understanding it.  I need to understand SNR better.
Anyway, he hates carbon track logarithmic volume controls.  "They belong in the landfill."   "If quality is paramount, use a switched attenuator."
 
Page 553 -554

 

[sceleratus]

Holland,
 
Buddy..... Full Speed Ahead with your design.
Adding a 4 deck stepped attenuator was stupid money.  Between the 4 deck Goldpoint attenuator ($275) and flexshaft extension fittings ($100) and Misumi Automation custom "L" bracket $30
Crazy, even for me.
 
I will report the results.
 
Thanks

 

[tomchr]

I recommend Morgan Jones, "Valve Amplifiers", 4th ed to anyone wanting to learn about tube amps. But do keep in mind that Mr. Jones can be a bit opinionated at times. He also doesn't think highly of single-ended amplifiers - such as the DG300B. He wants power... That's not what SET is about.
 
~Tom

 

[sceleratus]

Just emailed Tom.
 
Eureka !
 
I wired home run from the XLR jack to the board.[size=medium]
No POT  (Hello Holland)[size=medium][/size]
[size=medium][/size]
I swear.  The snow is gone.[size=medium][/size]
A very tiny tiny tiny bit of hum.  Undetectable while listening.[/size]
Heck, it's a valve amp[size=medium]
[size=medium][/size]
Classical Violin.  Dead Bang Quiet ![size=medium][/size]
This amp has an absolutely incredible voice.[/size]
The best I've ever heard it.[size=medium]
[size=medium][/size]
[size=medium][/size]
I'm think'n a dummy knob is the way to go.[/size]
I'll use iPad Remote.
I can't complicate this, it sounds too damn good.

 

[sceleratus]

IMO
 
I will now declare "The Christensen Damn Good 300B"  a Headphone Amp.

 

[sceleratus]

Volume Update.
 
I'm going in the stepped attenuator direction.
Why, my inability to test and implement other solutions.  I can handle the attenuator.
Plenty of connect the dots and color inside the lines stuff.
 
Since iTunes has a 20 step volume adjustment, by my calculations, I will get 20 steps for each of the attenuators 23 steps.
 
I figured out how to move the giant 1.25" x 3" 4 deck stack.
I purchased a 6" flexible shaft extension.  (everyone should have one of these)
Then I ordered a custom "L" bracket from Misumi Automation.  The bracket will sit on top of the filament regulator stack.
The alignment was perfect.
 
Nice Photos: