Balance issue? Pot issue?

[Sp33ls]

Hey all,

I've been loving my Crackatwoa. However, there's been a quirk that has been bothering me.

It seems to come and go, which is what makes it more annoying. The other day I took a look underneath and performed resistance and voltage checks.

Everything was in spec, altho, a couple of voltages were a touch on the higher side:
Terminal    Voltage (DC)
Low Current C4S
IA        150V  (162)
OA 60-90V  (80)
KregA  3-6V  (5)
bRegA 150V    (162)
IB        150V  (162)
OB  60-90V  (70)
KregB  3-6V  (5)
bRegB  150V  (162)
High Current C4S
IA        190V  (215)
OA        150V  (163)
bA        0V  (0)
IB          0V  (0)
OB  90-110V  (102)
bB        150V  (162)
Headphone Jack
Tip          0V    (0)
Ring        0V    (0)


The issue that I'm having is that on my two-quiet attenuator, the 'fine' adjustment can give me a sense of improper balancing. I.e., when I select -4.5dB, the audio signal appears to be slightly stronger in my right ear.  When I select -7.5dB, the audio appears to be slightly stronger in my left ear. However, the rest of fine control steps seem to be... fine.  : 

The thing is, it doesn't seem to always be that way. Sometimes it seems to just operate completely as expected. Also, I noticed that after I used some deoxit on the selectors, it seemed to have temporarily fixed it. I'm a bit confused. After my initial build, I went through and touched up each solder point I could, attempting to reduce the chance of a cold solder point. Do you think that's the case here?

Thanks!

 

[Paul Birkeland]

IA        150V  (162)
bRegA 150V    (162)
IB        150V  (162)
bRegB  150V  (162)
OA        150V  (163)
bB        150V  (162)

These need to be resolved before making any assessments about the performance of the amplifier.  This voltage should be extremely close to 150V (like 149-151), and the fact that yours are so off suggests that the shunt regulators aren't working properly.  Since both are having the same issue, we can assume that the same problem is present on both channels.  Here are some things to look at:

1.  Are the TL431s and PN2907s in the proper places? 
2.  Are the 6AQ5s powered up and glowing? (I'm guessing so since you have 5V on Kreg)
3.  Are the white wires leaving each 6AQ5 going to the correct side of the center board?
4.  Are the triode strapping resistors on each 6AQ5 socket properly installed?  Be certain that the 33K resistors that would have been used on the volume/balance pots didn't get put on the 7 pin sockets by accident.

-PB

 

[Sp33ls]

Thanks for the quick reply, PB.

Here's the interesting part... my results from last year are posted in an older post:

Terminal    Voltage (DC)
Low Current C4S
IA        150V  (150)
OA 60-90V  (77.3)
KregA  3-6V  (5)
bRegA 150V    (150)
IB        150V  (150)
OB  60-90V  (73.6)
KregB  3-6V  (5.7)
bRegB  150V  (150)
High Current C4S
IA        190V  (197)
OA        150V  (150)
bA        0V  (0)
IB          0V  (0)
OB  90-110V  (98)
bB        150V  (150)
Headphone Jack
Tip          0V    (0)
Ring        0V    (0)

So, it seems that it's drifted since then..? Both tests above are with a Bendix 6080WB and Telefunken ECC82.

I ran the voltage tests again but this time with a Tung-Sol 7236. It resulted in the larger voltages (IA, bRegA, IB, etc) dropping by about 3V. Is that normal?

 

[Paul Birkeland]

Do you still have all the tubes we sent you?  Can you give those a shot if you do?

bRegA/B should be absolutely rock solid at 150V.

 

[Doc B]

PB is right that you should get those 150V regulated voltages working correctly. Regarding your original question, yes, keeping the attenuator contacts clean will often correct a slight imbalance as the added resistance of the oxidation on the contact can add with the value of the resistors. It's standard practice in studios to exercise all of the controls to wear off any oxidation they may have built up before a recording session.

 

[Sp33ls]

This is the results with the tubes you guys sent me.

Terminal    Voltage (DC)
Low Current C4S
IA        150V  (158.3) (10 mins later, 161V )
OA 60-90V  (74.1) (10 mins later, 75V)
KregA  3-6V  (5.3) (10 mins later, 5.5V)
bRegA 150V    (158.7) (10 mins later, 161V)
IB        150V  (158.1) (10 mins later, 161.8V)
OB  60-90V  (68.5) (10 mins later, 69.6V)
KregB  3-6V  (6.1) (10 mins later, 6.3V)
bRegB  150V  (158.3) (10 mins later, 161.6V)
High Current C4S
IA        190V  (208) (10 mins later, 214V)
OA        150V  (158) (10 mins later, 161V)
bA        0V  (0) (10 mins later, 0V)
IB          0V  (0) (10 mins later, 0V)
OB  90-110V  (97) (10 mins later, 99V)
bB        150V  (158) (10 mins later, 161V )
Headphone Jack
Tip          0V    (0)
Ring        0V    (0)

Judging by the differences over the course of 10-15 minutes, it appears that one of the bigger issues I'm experiencing is instability  :-\

Also, it appears I have the correct resistors in the 7-pin sockets, and I have the right-side white lead at KregA, and left side at KregB.

Should I check some specific terminals for the power supply?

Also to mention -- I noticed this morning that after having it up and running for a good long while, it seems to correct itself [or atleast to my ears] with the balance issue on the -4.5dB step. I've noticed this behavior before, too. Once again, it leads me to think that my amp is just not stable 

 

[Paul Birkeland]

On the middle PC board, there is a 2.49K resistor on each side.  Can you measure the DC voltage at each end of that resistor?  Can you post some photos of your build too?

 

[Paul Birkeland]

That is either indicative of a symmetrical miswire (switching the Kreg wires, for example), a pair of flaky solder joints, or 431 regulators that aren't operating properly.  The 431 will work itself until it explodes to make 2.5V at the reference pin if it is provided with adequate voltage to do so (5V is plenty). I am accustomed to seeing damaged 431s (they can be damaged by other issues in a build over taxing them) showing a lower reference voltage (usually 1.8-2V), so your high voltage there is perplexing. You could still try replacing them, as well as looking at the jumpers and solder joints on the bottom of the center PC board.

On your attenuators, it looks like there may be some partially soldered or not soldered resistors on the fine switch.

 

[Sp33ls]

That is either indicative of a symmetrical miswire (switching the Kreg wires, for example), a pair of flaky solder joints, or 431 regulators that aren't operating properly.  The 431 will work itself until it explodes to make 2.5V at the reference pin if it is provided with adequate voltage to do so (5V is plenty). I am accustomed to seeing damaged 431s (they can be damaged by other issues in a build over taxing them) showing a lower reference voltage (usually 1.8-2V), so your high voltage there is perplexing. You could still try replacing them, as well as looking at the jumpers and solder joints on the bottom of the center PC board.

On your attenuators, it looks like there may be some partially soldered or not soldered resistors on the fine switch.

Waaaait a minute. I decided to just go ahead and grab another multimeter (a cheaper one actually), and threw a new battery in it. Now, I'm getting 2.48V across those resistors.  ???

Terminal    Voltage (DC)
Low Current C4S
IA        150V  (150)
OA 60-90V  (70)
KregA  3-6V  (5.1)
bRegA 150V    (150)
IB        150V  (149.2)
OB  60-90V  (64.6)
KregB  3-6V  (5.9)
bRegB  150V  (149.2)
High Current C4S (right-side)
IA        190V  (197.5)
OA        150V  (149.2)
bA        0V  (0)
IB          0V  (0)
OB  90-110V  (92)
bB        150V  (149.2)
High Current C4S (left-side)
IA        190V  (197.5)
OA        150V  (150)
bA        0V  (0)
IB          0V  (0)
OB  90-110V  (95.1)
bB        150V  (150)
Headphone Jack
Tip          0V    (0)
Ring        0V    (0)

So, what I see now is more appropriate numbers. Also, is it common for one side to be 149.2V and the other side to be 150V? And the OB terminals to vary?
[Edit: swapping out the 6005 tube over the board did not make an effect on OA or OB]

 

[Paul Birkeland]

Yeah, that looks better.  Good call on trying a different meter.

The 150V voltage will depend on the actual value of the 2.49K resistor and 147K resistors.  The "book perfect" voltage there would be 149.5, but a difference of +/- 1 volt or so is pretty normal.

Did you look at the junction of the resistors on the fine attenuator?  If one was a little loose and not all the way soldered in the bundle, it could make contact or not make contact depending on the vibration of actually switching the attenuator.

 

[Sp33ls]

Wellp, so far so good. I went through and touched up the junctions again to make sure they weren't cold, and added a bit more solder. I tinkered with the pot a tad, too.

I'll continue taking her for a spin and see if jumps up again, but hopefully not! 

I very much appreciate the help PB and Doc! You guys are awesome.

Just a couple of quick questions... besides the required pin adapter, is there any modifications required to run 6SN7 tubes? Any downsides to running one?

 

[Paul Birkeland]

There seem to be quiet ones and noisy ones.  YMMV with those adapters!

 

[Sp33ls]

Gotcha. I have one made in the USA by Garage1217. It appears to be high quality construction. I briefly tried it with a 6sn7, but the tube itself appeared to be microphonic. I might give it another go in the future.

Also, for painting / finishing the top of the power transformer, should we just unscrew the four top screws? I don't want to go poking into power supplies [even with the power unplugged] in areas I shouldn't be

Thanks again, PB.

 

[Paul Birkeland]

Also, for painting / finishing the top of the power transformer, should we just unscrew the four top screws?

Well, if it's already mounted, things get a little tough.  What I have done in the past is to pull all the tubes out and take painter's tape (blue or green) and tape off the chassis plate around the bottom of the power transformer really well, then apply something like VHT Roll Bar spray paint over the stack and transformer cover.

It's also a good idea to clean the steel cover before painting.  A paper towel with mineral spirits or lacquer thinner will cut through the goo pretty quickly.