[JazzVinyl]

Listening to the Little Creatures LP....I love they way the lead electric guitar is intentionally de-emphasized on this entire album.

Instead, we have:

Powerful drum rhythms, crazy sharp deep bass notes, rolled, plucked, and abrupted - also cleanly belonging to the rhythm section....many vocal harmonies and juices added via intelligent horn arrangements...keyboard arrangements lifting spirits. Everything so very carefully orchestrated. Nary an unneeded effort was made, by any musician, present.

One wonders how many times you had to practice these songs, to play them, this 'oh so p~e~r~f~e~c~t~l~y'.

Also quite amazed at how incredibly wonderful these LP's sound, compared to their digital counterparts.

They spanked it, hard, y'all!

 

[L0rdGwyn]

Alright, phase 1 of the DarkVoice modifications is done. Took some troubleshooting, but everything is working as intended so far.

Here is what I did:

1) Removed CRCRC-CRC filter from the power supply - replaced with single 330uF resevoir cap and HV regulator
2) Removed plate resistor load from 6SN7 input tube - replaced with DN2540 - J310 gyrator load
3) Removed 6080 1K wirewound cathode resistors - replaced with Vishay 1.2K non-inductive wirewound resistors
4) Removed 6SN7 1K cathode resistors - replaced with KOA Spear 300ohm carbon film resistors

The goal was to use the original mounting hardware and wiring as much as possible, so the regulator and gyrator PCBs are both mounted using the existing standoffs for the old electrolytic capacitors. I used the original ground bus, but removed excess portions of it.

Here are the new tube bias points:
B+ 180VDC
6SN7: 80Va / 4.5mA / 1.4Vk
6080: 75Va / 85mA / 105Vk

This aligns very well with my simulations. Here are some pics.



And now the reward



From memory, there is a significant increase in soundstage and clarity as you might expect with an active loaded input tube, it sounds very good....granted, I only spent a short time with the stock DarkVoice, so it will be interesting to hear how it compares to the original and the La Figaro 339 when it is sent back to its owner.

Still more to do be done, phase 2 will involve fixing the heaters to hopefully get rid of the ridiculous hum issue. Phase 2 might also include increasing the output capacitance, will also be replacing the pot and power switch. Probably won't get to it until next weekend.

 

[L0rdGwyn]

I am very impressed by the sound coming out of this modded DarkVoice...

So I am going to throw together another pair of gyrator boards and swap them out with the CCS loads in my 841 amplifier tomorrow morning. The dimensions and standoff mounting holes on the boards aren't suitable for the 841 layout, but if there is an audible improvement, I will put together another gyrator board that fits those dimensions, in fact it is already drafted.

I sincerely hope it doesn't sound better, otherwise it is gong to send me down a rabbit hole replacing CCS loads for gyrators in all of my gear

 

[Zachik]

I sincerely hope it doesn't sound better, otherwise it is gong to send me down a rabbit hole replacing CCS loads for gyrators in all of my gear

"Perfect is the enemy of good"

 

[JazzVinyl]

Alright, phase 1 of the DarkVoice modifications is done. Took some troubleshooting, but everything is working as intended so far.

Fantastic job, LG! Wish I could have been there for the before and after listening...!!

 

[Tjj226 Angel]

Bah, 80s music.

Alright, phase 1 of the DarkVoice modifications is done. Took some troubleshooting, but everything is working as intended so far.

Here is what I did:

1) Removed CRCRC-CRC filter from the power supply - replaced with single 330uF resevoir cap and HV regulator
2) Removed plate resistor load from 6SN7 input tube - replaced with DN2540 - J310 gyrator load
3) Removed 6080 1K wirewound cathode resistors - replaced with Vishay 1.2K non-inductive wirewound resistors
4) Removed 6SN7 1K cathode resistors - replaced with KOA Spear 300ohm carbon film resistors

The goal was to use the original mounting hardware and wiring as much as possible, so the regulator and gyrator PCBs are both mounted using the existing standoffs for the old electrolytic capacitors. I used the original ground bus, but removed excess portions of it.

Here are the new tube bias points:
B+ 180VDC
6SN7: 80Va / 4.5mA / 1.4Vk
6080: 75Va / 85mA / 105Vk

This aligns very well with my simulations. Here are some pics.



And now the reward



From memory, there is a significant increase in soundstage and clarity as you might expect with an active loaded input tube, it sounds very good....granted, I only spent a short time with the stock DarkVoice, so it will be interesting to hear how it compares to the original and the La Figaro 339 when it is sent back to its owner.

Still more to do be done, phase 2 will involve fixing the heaters to hopefully get rid of the ridiculous hum issue. Phase 2 might also include increasing the output capacitance, will also be replacing the pot and power switch. Probably won't get to it until next weekend.

All you need to do to the heaters is ground them using 2 100 ohm resistors as a center tap.

If it is possible and you haven't already done so, I would swap out the sockets as well. I very rarely recommend screwing around with socket "upgrades" per se, but it's almost a requirement on the dark voice amps.

Also. PSSST. PSSSSSSSSSST.......find a solution to the output capacitor. Seriously, the WE421a and 5998 tubes have twice the transconductance of the 6AS7G tube. The 5998 tubes can drive LCD2s quite nicely if given half the chance.

The film caps in the dark voice are so chinsey that a high quality electrolytic in parallel with a high quality PIO cap, might actually sound a hell of a lot better.

 

[L0rdGwyn]

Bah, 80s music.


All you need to do to the heaters is ground them using 2 100 ohm resistors as a center tap.

If it is possible and you haven't already done so, I would swap out the sockets as well. I very rarely recommend screwing around with socket "upgrades" per se, but it's almost a requirement on the dark voice amps.

Also. PSSST. PSSSSSSSSSST.......find a solution to the output capacitor. Seriously, the WE421a and 5998 tubes have twice the transconductance of the 6AS7G tube. The 5998 tubes can drive LCD2s quite nicely if given half the chance.

The film caps in the dark voice are so chinsey that a high quality electrolytic in parallel with a high quality PIO cap, might actually sound a hell of a lot better.

Yes, a virtual center tap will be made. The heater windings are soldered to the center PCB before they hit the tube heaters and one end appears to be grounded through a PCB trace, so it will need to be cut or the heater windings removed from the PCB and run directly to the tubes. The wiring between the tubes is not twisted, like, at all. Also the heaters run hot, around 15-20% above nominal, to may need some dropping resistors as well. So yeah, not just a virtual center tap, unfortunately but nothing too tedious. I have heard about the "hum problem" of the DarkVoice for years, it is very likely a simple heater grounding issue. I won't be replacing the sockets, I'll leave that for other people if they want to pursue it. I will be increasing the output capacitance with some radial TKD polypropylene film caps at 33uF a pop.

 

[L0rdGwyn]

For the sake of experimentation, I put together another set of gyrator boards to replace the 841 CCS loads in my headphone amplifier. DN2540 was replaced with the IXTP08N100D2, increased the reference voltage resistor to put 380VDC on the plates.



Better? Worse? The same? I do think it is better, improvement in clarity and small increase the the staging, expectation bias be damned. So looks like I will be ordering some gyrator boards that fit my cascode CCS PCB footprint.

Another change I made was replacing the current set resistor on the Rod Coleman regulators with NOS Mills MRA10. A small improvement in imaging and holography.

 

[L0rdGwyn]

I am doing a big vinyl ultrasonic cleaning session. In between, listening to the modded DarkVoice. Threw on a 6J5 to 6SN7 adapter and some GEC L63.

Sounds really good, but not a massive difference from the 6SN7 driver. I've found that once the driver has an active load, it is something of an equalizer, differences between different makes of tube become more subtle.



I will eventually need to do something for all of my OTL power tubes, I might build this circuit from scratch for myself some day but in a chassis with enough space for giant output capacitors and either a choke on the 6080 cathode or a FET current sink.

 

[bcowen]

I am doing a big vinyl ultrasonic cleaning session. In between, listening to the modded DarkVoice. Threw on a 6J5 to 6SN7 adapter and some GEC L63.

Sounds really good, but not a massive difference from the 6SN7 driver. I've found that once the driver has an active load, it is something of an equalizer, differences between different makes of tube become more subtle.



I will eventually need to do something for all of my OTL power tubes, I might build this circuit from scratch for myself some day but in a chassis with enough space for giant output capacitors and either a choke on the 6080 cathode or a FET current sink.

Nice! Really nice.

But OMG!!!! What *is* the deal with that power cord? Looks like the freebie that came in the box with my AM transistor radio (that runs on batteries).

 

[L0rdGwyn]

Nice! Really nice.

But OMG!!!! What *is* the deal with that power cord? Looks like the freebie that came in the box with my AM transistor radio (that runs on batteries).

I think that's the one that came with the amp LOL it isn't the sexiest. I'm not huge on power cords though, at least not yet, that is like item #876 on the list of things to experiment with. Same with interconnects, there is just to much to do

 

[JazzVinyl]

I am doing a big vinyl ultrasonic cleaning session. In between, listening to the modded DarkVoice. Threw on a 6J5 to 6SN7 adapter and some GEC L63.

Sounds really good, but not a massive difference from the 6SN7 driver. I've found that once the driver has an active load, it is something of an equalizer, differences between different makes of tube become more subtle.

Good! Will keep one from falling into the 'gotta hear it" rabbit hole.

You had the Crackatowa at one time, LG? How does LGDV336 compare to that one?

 

[L0rdGwyn]

Good! Will keep one from falling into the 'gotta hear it" rabbit hole.

You had the Crackatowa at one time, LG? How does LGDV336 compare to that one?

Oh boy, it's been a while, but I think this is better. The Crackatwoa is a Crack + Speedball with a tube shunt-regulated power supply. This has a better power supply regulator and I think a gyrator load is better suited for direct-coupling over the CCS load used in the Speedball as it regulates the driver tube plate voltage, and thus the output tube grid bias. The Crackatwoa also uses LED bias on the driver, I prefer an unbypassed cathode resistor when practical. One advantage that goes to the Crackatwoa is it uses current sink on the 6080 output tube, I can't say how big of a difference that will make vs. a resistive load, something I will explore if I rebuild this circuit from scratch.

 

[L0rdGwyn]

I made a change to my ultrasonic cleaning process. Since I am using this as a singular cleaning step (i.e., no pre-clean or rinse), I was finding some sediment build-up on the record at the end of a clean, even while running the filter in between batches. The solution? Running the filter continuously while cleaning. MASSIVE difference, this is the way. I also marked the arm level with a permanent marker so I can quickly get it into position.

 

[L0rdGwyn]

Well the vinyl is clean, so there is only one thing left to do.