[L0rdGwyn]

Dumb but important question... how does one discharge his power caps? I typically just leave things unplugged for 24-48 hours and assume the caps will discharge naturally.

You can solder a high value bleeder resistor across the +/- leads of the first cap of the power supply, 220K to 1Meg, higher value will discharge more slowly, a lower value will load down the supply a bit more but not enough to drop the output voltage significantly.

 

[JKDJedi]

LOL! All my tube recommendations carry an implied disclosure: if something goes wrong, it's @JKDJedi 's fault.

 

[DeweyCH]

Now wait, there are problems getting the screws loose? Anyone want to bet me I won't have the whole thing disassembled in a half hour? I would do a video of course. I must have a screwdriver somewhere.

I'm sure you'll have it apart in minutes. I also suspect you're at the level where you can just glare at it hard enough that it disassembles itself.

 

[Paladin79]

I'm sure you'll have it apart in minutes. I also suspect you're at the level where you can just glare at it hard enough that it disassembles itself.

LOL I have a few tools and some experience. I have right angle mini ratchets for tight spaces,

 

[L0rdGwyn]

I finished fixing the heater situation in the DarkVoice this morning.

As I mentioned before, the heaters in the DarkVoice are not center-tapped, either at the transformer or virtually, which I suspected was a large contributor to the 60Hz noise issues numerous people are having with the amplifier.

Instead, what the designer did was ground one leg of the heaters at the 6SN7 socket here (it has been removed, but a connection was made between the 6SN7 heater and the front panel LED PCB ground connection).



So that connection was removed and a virtual center tap was made using two 100ohm resistors at the point where the heater windings hit the center PCB. The original wiring was removed and replaced with a twisted pair. I also mentioned that the tube heaters were being run over voltage, at around 7.15VAC. Just so happens I had the perfect voltage dropping resistors for the job on hand, two 130mOhm 5W wirewounds. One goes on each leg of the heaters since they are now balanced.



Now getting ~ 6.3VAC on the across the heaters (note that since the heaters are now balanced, each heater-to-ground voltage measurement will be 3.15VAC, the voltage from one heater to the other will be ~6.3VAC). Worth noting that the front panel LED is powered from the heater winding, the value of the resistor in series will need to be adjusted since the voltage has been halved, otherwise it won't light up.



So did balancing the heaters make a difference? Well, I have two 6SN7s on hand that were completely unusable in the DarkVoice due to heater-to-cathode leakage, the 60Hz noise was completely out of control. One was this black-plate Fivre 6SN7GT. I have had no problems with this tube in any other amplifier.



In terms of the hum level, the difference is massive, but there is still a low-level hum audible without music playing, took it from a 9 down to a 2. I suspect this is due to primary-to-secondary leakage in the mains transformer, which is not going to be fixed without completely replacing it, at which point you might as well just build a new amplifier. Regardless, I would still recommend this modification, center tap and balance the heaters and break the ground connection from the 6SN7 heater to the power LED ground. The hum could likely be further reduced by using a 500ohm trimpot with the wiper connected to ground instead of the two 100ohm resistors. The pot can then be adjusted to minimize the audible hum, it's called a humdinger.

 

[Deceneu808]

You could practice it on another (cheaper) tube first.

Went straight for the Tung-Sols. The second one started to pull the same trick on me. Took me around 10 minutes to do them both. They both work excellent now. Thank you !

 

[DeweyCH]

I finished fixing the heater situation in the DarkVoice this morning.

As I mentioned before, the heaters in the DarkVoice are not center-tapped, either at the transformer or virtually, which I suspected was a large contributor to the 60Hz noise issues numerous people are having with the amplifier.

Instead, what the designer did was ground one leg of the heaters at the 6SN7 socket here (it has been removed, but a connection was made between the 6SN7 heater and the front panel LED PCB ground connection).



So that connection was removed and a virtual center tap was made using two 100ohm resistors at the point where the heater windings hit the center PCB. The original wiring was removed and replaced with a twisted pair. I also mentioned that the tube heaters were being run over voltage, at around 7.15VAC. Just so happens I had the perfect voltage dropping resistors for the job on hand, two 130mOhm 5W wirewounds. One goes on each leg of the heaters since they are now balanced.



Now getting ~ 6.3VAC on the across the heaters (note that since the heaters are now balanced, each heater-to-ground voltage measurement will be 3.15VAC, the voltage from one heater to the other will be ~6.3VAC). Worth noting that the front panel LED is powered from the heater winding, the value of the resistor in series will need to be adjusted since the voltage has been halved, otherwise it won't light up.



So did balancing the heaters make a difference? Well, I have two 6SN7s on hand that were completely unusable in the DarkVoice due to heater-to-cathode leakage, the 60Hz noise was completely out of control. One was this black-plate Fivre 6SN7GT. I have had no problems with this tube in any other amplifier.



In terms of the hum level, the difference is massive, but there is still a low-level hum audible without music playing, took it from a 9 down to a 2. I suspect this is due to primary-to-secondary leakage in the mains transformer, which is not going to be fixed without completely replacing it, at which point you might as well just build a new amplifier. Regardless, I would still recommend this modification, center tap and balance the heaters and break the ground connection from the 6SN7 heater to the power LED ground. The hum could likely be further reduced by using a 500ohm trimpot with the wiper connected to ground instead of the two 100ohm resistors. The pot can then be adjusted to minimize the audible hum, it's called a humdinger.

Oh good something else for me to do! (I am totally doing this BTW - thanks for sharing!)

 

[therremans]

I finished fixing the heater situation in the DarkVoice this morning.

As I mentioned before, the heaters in the DarkVoice are not center-tapped, either at the transformer or virtually, which I suspected was a large contributor to the 60Hz noise issues numerous people are having with the amplifier.

Instead, what the designer did was ground one leg of the heaters at the 6SN7 socket here (it has been removed, but a connection was made between the 6SN7 heater and the front panel LED PCB ground connection).



So that connection was removed and a virtual center tap was made using two 100ohm resistors at the point where the heater windings hit the center PCB. The original wiring was removed and replaced with a twisted pair. I also mentioned that the tube heaters were being run over voltage, at around 7.15VAC. Just so happens I had the perfect voltage dropping resistors for the job on hand, two 130mOhm 5W wirewounds. One goes on each leg of the heaters since they are now balanced.



Now getting ~ 6.3VAC on the across the heaters (note that since the heaters are now balanced, each heater-to-ground voltage measurement will be 3.15VAC, the voltage from one heater to the other will be ~6.3VAC). Worth noting that the front panel LED is powered from the heater winding, the value of the resistor in series will need to be adjusted since the voltage has been halved, otherwise it won't light up.



So did balancing the heaters make a difference? Well, I have two 6SN7s on hand that were completely unusable in the DarkVoice due to heater-to-cathode leakage, the 60Hz noise was completely out of control. One was this black-plate Fivre 6SN7GT. I have had no problems with this tube in any other amplifier.



In terms of the hum level, the difference is massive, but there is still a low-level hum audible without music playing, took it from a 9 down to a 2. I suspect this is due to primary-to-secondary leakage in the mains transformer, which is not going to be fixed without completely replacing it, at which point you might as well just build a new amplifier. Regardless, I would still recommend this modification, center tap and balance the heaters and break the ground connection from the 6SN7 heater to the power LED ground. The hum could likely be further reduced by using a 500ohm trimpot with the wiper connected to ground instead of the two 100ohm resistors. The pot can then be adjusted to minimize the audible hum, it's called a humdinger.

The mods look great. Would it be possible that the potentiometer upgrade (probably an ALPS), will fix the remaining hum? It did great things to the noise floor. I look forward to trying these mods out myself when you’re finished. Thank you.

 

[DeweyCH]

I notice a lack of Fitz mod. I suppose because (I think I've read) you consider the two caps to be detrimental to sound outside of the hum, and this is a solution that doesn't have negative side effects?

 

[Paladin79]

I finished fixing the heater situation in the DarkVoice this morning.

As I mentioned before, the heaters in the DarkVoice are not center-tapped, either at the transformer or virtually, which I suspected was a large contributor to the 60Hz noise issues numerous people are having with the amplifier.

Instead, what the designer did was ground one leg of the heaters at the 6SN7 socket here (it has been removed, but a connection was made between the 6SN7 heater and the front panel LED PCB ground connection).



So that connection was removed and a virtual center tap was made using two 100ohm resistors at the point where the heater windings hit the center PCB. The original wiring was removed and replaced with a twisted pair. I also mentioned that the tube heaters were being run over voltage, at around 7.15VAC. Just so happens I had the perfect voltage dropping resistors for the job on hand, two 130mOhm 5W wirewounds. One goes on each leg of the heaters since they are now balanced.



Now getting ~ 6.3VAC on the across the heaters (note that since the heaters are now balanced, each heater-to-ground voltage measurement will be 3.15VAC, the voltage from one heater to the other will be ~6.3VAC). Worth noting that the front panel LED is powered from the heater winding, the value of the resistor in series will need to be adjusted since the voltage has been halved, otherwise it won't light up.



So did balancing the heaters make a difference? Well, I have two 6SN7s on hand that were completely unusable in the DarkVoice due to heater-to-cathode leakage, the 60Hz noise was completely out of control. One was this black-plate Fivre 6SN7GT. I have had no problems with this tube in any other amplifier.



In terms of the hum level, the difference is massive, but there is still a low-level hum audible without music playing, took it from a 9 down to a 2. I suspect this is due to primary-to-secondary leakage in the mains transformer, which is not going to be fixed without completely replacing it, at which point you might as well just build a new amplifier. Regardless, I would still recommend this modification, center tap and balance the heaters and break the ground connection from the 6SN7 heater to the power LED ground. The hum could likely be further reduced by using a 500ohm trimpot with the wiper connected to ground instead of the two 100ohm resistors. The pot can then be adjusted to minimize the audible hum, it's called a humdinger.

I already have other transformers picked out and ordered, I was not kidding when I said mine may have few original parts. Maybe the name plate if I can work that in.

 

[jonathan c]

To pursue the thoughts engendered by o@therremans, if tubophilia is bad enough, what about the excitement of a shipment (from Langrex) of several dozen 9-pin protectors? Utmost care for those 7308s, 7316s, 7730s, and so much more...

 

[L0rdGwyn]

The mods look great. Would it be possible that the potentiometer upgrade (probably an ALPS), will fix the remaining hum? It did great things to the noise floor. I look forward to trying these mods out myself when you’re finished. Thank you.

Changing the pot is unlikely the affect the hum, unless some weirdness is going on where the stock pot is prone to noise pickup. Not surprising the noise floor dropped though, I'm sure the original pot is quite noisy, but again, not likely to affect the 60Hz noise.

I notice a lack of Fitz mod. I suppose because (I think I've read) you consider the two caps to be detrimental to sound outside of the hum, and this is a solution that doesn't have negative side effects?

Correct. Of course it is going to depend on what is more important to you if the hum cannot be completely eliminated by center tapping the heaters (like I said, a humdinger gives you the best chance to minimize it). IMO, bypassing the cathode resistors is a situation where the cure is worse than the disease, I would probably leave them unbypassed and only use tubes with low heater-to-cathode leakage, I place a high priority on not putting electrolytics in the signal path as they negatively affect the sound. Center-tapping the heaters and using twisted heater leads gives you the best chance of lowering the audible hum without resorting to a bypass cap.

Get a 500ohm trimpot, connect the two terminals to the heater winding pads on the PCB, connect the wiper to ground. Place the pot at its midpoint, put in a tube that is prone to the hum issue and listen with headphones and no music playing, adjust the pot until the hum is minimized. If it is still too much, you can always add the bypass caps back on.

I'll get a 500ohm trimpot and see if it can be further reduced over a pair of 100ohm resistors.

 

[L0rdGwyn]

Replacing the caps on the PCB without removing it is proving more annoying than anticipated, looks like I'll have to take it out to get the job done, which means I am done for today

 

[DeweyCH]

I know this is somewhat underwhelming vs. the Magnum Opus on display above but I'm kinda proud of how I (after several failed attempts) set up this attenuator to actually be able to have things soldered to:

It's clean, it's low-profile, and it's got actual pins to solder resistors and wires to. I also took a couple of pins I clipped off of something else to connect the 2 ground pins on each channel for the shunt mod. Last time I ran long, ugly-ass lines of solder between them and it was less than good

EDIT: I know the first picture looks like I scorched the hell out of it. It's just the lighting. Doesn't look at all abnormal in person.

 

[therremans]

I know this is somewhat underwhelming vs. the Magnum Opus on display above but I'm kinda proud of how I (after several failed attempts) set up this attenuator to actually be able to have things soldered to:







It's clean, it's low-profile, and it's got actual pins to solder resistors and wires to. I also took a couple of pins I clipped off of something else to connect the 2 ground pins on each channel for the shunt mod. Last time I ran long, ugly-ass lines of solder between them and it was less than good

EDIT: I know the first picture looks like I scorched the hell out of it. It's just the lighting. Doesn't look at all abnormal in person.

Looks better, but shouldn’t this pot not be shunted?