The DIY'rs Cookbook

[johnjen]

Well I just finished replacing all of the wire in the analog section that carries the audio signal, with shortened and 'optimally' placed runs of silver wire.

This along with verifying that rectifier tubes are light sensitive, has dropped the s/n down to the -67dB level, which is a new low for the prototype amps we have built.
And I have added an accessory enhancement and continue to work on making the box/base purdy so I can take it to the upcoming Seattle head-fi meet along with the Purp-Amp.

JJ

 

[johnjen]

So another update…

The 6L6 amp responded well to the re-wire and the box is almost ready for the final finishing touches.

It now has s/n readings of -67dBv(L) & -66dBv(R) and we are shooting for -80dB or better
Here are 2 shots of the analog wiring.


and



Right now I'm working on tweaks for the Purp-Amp.
I've installed new tubes (a 'matched' pair of RCA Radiotron 71A's from the 1940's and a single GE 6BQ7 from the 1960's), added all the labeling, made an output impedance matching resistor network, and cleaned up some of the analog circuit wiring.
There is still some more circuitry fussing to do to lower the hum which is currently at -57dBv(R) and -68dBv(L).

And I'm listening to it as I type this and it is starting to blossom as the circuit gets used to playing music.

And most likely it will get re-purposed into it's Pre-Amp mode to drive the 6L6 amp, that is assuming I can dial it in to be quiet and musical so that they match up well together.

I'll take some pictures of it and post them tomorrow.

JJ

 

[johnjen]

So while the sun wasn't out here are some 'glamor' shots of the Purp-Amp.

1st the front panel


and a top view


and a rear view


and an overall view


Now it's time to fuss with the circuit a bit more to reduce the hum even more.

JJ

 

[johnjen]

And the sun did come out long enough to snag a few sun lit glamor shots


And a rear view


And a 'heroic' shot with some blue sky for effect…



Yeah the backgrounds are a bit 'rough' but hey it's all about the foreground…
hahahahahahahahaha…

JJ

 

[johnjen]

The Purp-Amp is starting to blossom as we continue to dial in the circuit.

The DHT magic is starting to shine thru and along with the simplicity of the analog circuit, there is a 'directness' to the SQ, along with impressive bass response, and a generous dose of MitM (Magic in the Mids).
And I figure these traits will only improve as more of the choke points are addressed and 'cleaned up'.

And I just received the replacement for the power supply input cap, in for the 6L6 amp, which I'm hoping should help clean up it's act even more.

And the local headfi meet is 1 week+ away so there will be a cut off for further circuit mods etc, as the bases need to be finished and other prep work needs to be ready for the meet.

JJ

 

[johnjen]

So I'm listening to the 6L6 amp now, after replacing the main input cap in the B+ supply.

It didn't seem to make any difference to the s/n #'s, but this amp has taken a decided step up in overall SQ.
Peaks and crescendos aren't as 'pinched', massed strings are heard as many violins playing together instead of as a single mass, etc.

And the Leading Edge Dynamic Impact has taken a 'mutha may I', step up as well.

My heads a bop'n and weave'n as I dance in my chair.
So the circuit is good to go to the meet, all I have to do now is finish the box, at least well enough to show it.

And I have a few last minute tweaks to try with the Purp-Amp as well, so my solder iron is gunna be flinging hot solder around here shortly.

JJ

 

[gefski]

So I'm listening to the 6L6 amp now, after replacing the main input cap in the B+ supply.

It didn't seem to make any difference to the s/n #'s, but this amp has taken a decided step up in overall SQ.
Peaks and crescendos aren't as 'pinched', massed strings are heard as many violins playing together instead of as a single mass, etc.

And the Leading Edge Dynamic Impact has taken a 'mutha may I', step up as well.

My heads a bop'n and weave'n as I dance in my chair.
So the circuit is good to go to the meet, all I have to do now is finish the box, at least well enough to show it.

And I have a few last minute tweaks to try with the Purp-Amp as well, so my solder iron is gunna be flinging hot solder around here shortly.

JJ

So you won't be setting off the library smoke alarms with solder smoke this time?

 

[johnjen]

No promises!!!!!!! hahahahahahahahahahaha

Besides sometimes we all need a little bit more excitement in our lives…
er, sumpt'n
hahahahahahahahahahahahahaahahahahahaha

JJ

 

[johnjen]

I switched back to the Purp-Amp and have rewired portions of the Power Supply with a new ground and other 'enhancements'.
I have switched to using a 6CG7 tube (like in the 6L6 amp) from the original 6BQ7, and stared using my 1930's Radiotronics globe UX-171 pair of 'exactly' matched 71A tubes.
Cleaned all the pins of these 55/80+yr old tubes (which helps with inner detail and spatial cues) and these new non-development tubes are still breaking in…

SDSG (SuperDuperSuperGlue) is back with a vengence, the extreme bottom end, which is a design goal, is most impressive and not just in the extension but in the 'texture', finesse, tonality, and 'richness' of even the subsonic frequencies.

And there is a sense of solidity and stability (related to the 3 C's, Cohesion Coherence Coupling) for the entire soundstage that has set new high water marks.
https://www.head-fi.org/threads/the-diyrs-cookbook.781268/page-3#post_11989064

In short, even thus far, these 2 amps are providing an intriguing and musically satisfying representation on my 800's.

And now I gotta start making/modifying cables for the meet, finish up the boxes, configure a computer, and listen to the amps as they are still breaking in.

To paraphrase Blazzing Saddles,
'Work, Work, Work, hello boys, lets get together later'.
hahahahahahahahahahahahahahaha

JJ

 

[johnjen]

So after spending WAY to much time and effort getting a MacBook Air configured as a music player, I got it working well enough for the headfi meet yesterday.

It was slow as molasses since the internal SSD was dead and I was running the entire laptop on a dog slow USB stick.
But it worked well enough…

And now back to more experiments with tweaking and refining the circuits…
1st up, is dialing in the grid stopper circuits, with just the right amount of resistance, based upon listening tests.
This is one of those 'adjustments' to the circuit that often is made based upon what the conventional wisdom dictates rather than how this parameter affects the SQ of the circuits.
And since these resistors are in the direct signal path, the value and 'quality' of these parts are subject to scrutiny.

JJ

 

[johnjen]

So now the Purp-Amp is running with Grid Stoppers using 330Ω instead of 1KΩ.

This is a step up, or a return to previous SQ traits and suggests I could go even lower, say in the 200Ω range.
But 1st I'm gunna let these settle in and measure the s/n and sub 10Hz rumble.
And, oh yeah, just listen to some killer tunes.

JJ

 

[johnjen]

So I went a little 'overboard' in that I made several changes all at the same time on the Purp-Amp.

I changed the output Ω of the output xfmrs, added new compensation resistors to match the 32Ω rated impedance, (down from 150Ω), and swapped out the interstage caps to Miflex instead of the Hovland SuperCaps I was using.

It'll take a bit of time for the new caps to 'settle in', but they are opening up even as I type this.
And I have a few more tweaks to add as well.

And after I finish tweaking this amp, I have a few changes to make to the 6L6 amp, namely adding an input volume control.
At the last meet, it became abundantly clear I need to be able to control the gain of this speaker amp, especially when driving HP's.

And I'll get a chance to try out a few new (to me) sets of matched 6L6GC and rectifier tubes as well.

JJ

 

[johnjen]

So after making all these changes to the Purp-Amp, the SQ is starting to open up and has reached parity with the previous setup and has continued to 'blossom'.

And while measuring the outputs, one channel measured 0.000mvac which is the resolution of my Fluke 289 meter.
This equates to better than -120dBv s/n.
Now I just have to reduce the other channels s/n of 0.4mvac (≈ -68dBv) to near the same level.

And now that I know what can be achieved, I figure when I get both channels operating at near the same level of 'quietness', the SQ will again take another step up.
And I still have several more tweaks to experiment with, like the Ω value of the grid stoppers, and adding more primary capacitance to the power supply, as well as re-arranging the location of the input xfmrs, and who knows what other tweaks lurk in the wings…

Still, reaching -120dBv s/n is remarkable achievement for a tube amp regardless.

JJ

 

[johnjen]

So I tried adding another temco cap (100µf), just to see what would happen,
and nut'n honey, as in zip, nada, so it got yanked back out.

I also lowered the Ω of the grid leak resistors (down to 235Ω from 330Ω) and it also didn't help (in fact it made several SQ and operational factors worse) so back to 330Ω it went.
And I might go up the ladder to ≈400Ω or 500Ω just to see what sonic impact this will yield.

AND I'm gunna change the input xfmrs and their location to see what net changes this will bring.
I'm hoping the s/n gap between the 2 channels will close up and lower at the same time, we'll see.

I have also added a new (to me) dac to the stable.
It will be a meet and backup dac, that I already have a tweak (or 3) figured out to play around with.
It’s a Jolida Glass FX Tube DAC with a 2nd set of Telefunken 12AX7 tubes.

We may convert it to run different tubes as the 12AX7 is an OK tube but not exceptional in some regards.

And using a tube for the output stage in a dac opens the door for some interesting filter options.

But 1st I need to listen to it in stock form to get a handle on its sonic signature.
And I'll use it in my new headphone meet setup, and compare it to my reference desktop setup.
And THEN start tweaking it with caps, fuses, WAQy chips and different tubes.

JJ

 

[johnjen]

So I listened to the tube dac, and it was merely ok.
Nothing special to write home about…
And no matter if it was USB straight from my MacBook Air (limited to 44.1/48KB) or using SPDIF (set to 88.2) from my AOIP stack.
Yeah it was a bit better than Meh, but definitely in the mid-fi portion of the audio spectrum.

So I opened up the bottom to peer into the innerds…
And looky there, special 'tweako' 'factory' output caps, and
they were installed 'backwards' wrt each other.

So I guessed which one was in need of reversing it's 'polarity' (of these non-polar caps) and,
VOILA…! (I did have a 50/50 chance of getting it right the 1st time.)
MUCH better, in all sonic characteristics I consider important.

But, as usual, I couldn't just leave well enough alone, no, no…
So I added an old (1st gen) Synergistic Research fuse to the jolida dac, and
again, VOILA…!
only even more so.
I can just hear the choke points falling away… hahahahahahaha

So I'm on the right path…
And next up is swapping for a set of MONDO-Tweako yellow Hovland caps to replace the 'factory' special caps.
I'm figur'n this should lead to yet another,
VOILA…!

And of course there be tweaks a-plenty just waiting in the wings, and not just for this tube dac, but both of the tube amps as well.

Yep I'm hav'n sum fun now ya betcha!

JJ