[stvnharr]

Recently I have made some changes and modifications to my GK 1.  These changes are as follows:
1. Incorporate the selector switch and shunt stepped attenuator from my passive pre.
2. Change signal wiring from cat 5 to same wire as I use for interconnects.
3. Modify the board with many of the suggestions in the old GK 1 Nirvana thread.
      These modifications are as follows:
            C1 stock to BG 22uf NIX
            C19 stock to RTX .01uf
            C21 bypass Auricap with RTX .01uf
            C4, 5, 7 stock to Oscon 47uf
            R11 and R22 to Rikens
4.  Install 2nd output, for subwoofer.  Output is direct from the attenuator and bypasses the tube section.

Results and comments:
1.  Attenuator works perfectly fine although shunt attenuators do have differing impedances at each step.  I have noticed nothing unusual in the operation.  Installing selector switch is reasonably simple.

2.  Signal wires are Vampire CCC wire, 2 strands for + and -, same as my interconnects.  I kept the wires for each channel the same length in each application.  Changing to this wire resulted in a noticeable sound difference as the stereo image became real crisper, cleaner, firmer, and clearer than before.  Not sure if the difference is from changing the kind of wire, going from single strand of cat 5 to 2+ and 2- of CCC, or just wire synergy as all signal carrying wire from sacd player into the 55 N+ is the same.  Having each channel wire the same length though is very important as this wire is just an extension of the interconnect.

3. Thus far I have heard no discernable difference in sound from changing the caps, no better/no worse, though these modifications were only recently done.  There could be changes later if there is any “cap break-in”.

4. Nice to have if adding a subwoofer in the future.

Future possibilities:
1.  Adding Vishay S102 resistors to some of the most used steps on the attenuator.  This is quite expensive at $11.50 per resistor, or $23 per step.  Guy Hammel, of Placette Audio, recommended this to me as a very important part of the attenuator.  It’s supposed to make things even cleaner and clearer sounding, though that’s hard to believe.

2. Tube rolling to 7308 tubes.  I doubt I will do this.  I’ve read that this is the one tube that can bring some improvement, in the form of a wider soundstage.  I already have a much wider soundstage with sacd.  And the cost of the tubes would buy a lot of music.

Finally, the initial comparison that I performed was to compare my passive preamp, which outshown the very expensive BAT VK 30SE, to the GK 1.  The passive and the GK 1 were fairly equal in sound quality although the GK 1 played louder, of course, due to the 6 db of gain.  Only then did I incorporate the stepped attenuator into the GK 1.  There were no ill effects from the changing impedances.  I also didn’t notice much of a sound change either.  The stock potentiometer is indeed quite good, though it does have a rather short shaft on the front making it hard to get a knob on after going thru the front panel.   The 6 db of gain means that I now operate the attenuator in different positions from when with the passive.
I rather like the music I hear in the system with the GK 1.

[jules]

congratulations on the test changes. I hope you'll give us a future report post BG break-in. I suppose it's going to be a bit harder to evaluate a pre-amp change in that they possibly only make up about 20% of the total amp influence.

It would be interesting to try some Vishays in just one step. If there's any part of a system that has to be easy to swap [and compare] from one option to the other it's a [shunt] stepped attenuator   .

Jules

[andyr]

Recently I have made some changes and modifications to my GK 1.  These changes are as follows: ...

Hi, Steven,

Can you tell me more about your signal wiring change?  You said you changed 24awg Cat 5 to 2 strands of the Vampire cast copper wire.  Which guage do you use ... 26.5, 21 or 20?

Re. caps: ... I also used BG 'NXs' at C2, 'FKs' at C4, C10, C11, C12 & C14, 'PKs' at C6 and 'Stds' at C15 & C16.  I also used a 1uF RTX at C21 instead of bypassing the Auricap with a 10nF RTX.

I hadn't picked up the change to C5.

I think you might have to wait a month or two to get the full benefit from the cap changes.

Remember, there is also the tweak suggested by Ginger ... to move C4  to be across L1 & change R13 to 470R and R14 to 4K7.

Regards,

Andy

[AKSA]

Very interesting, Steven,

Thanks for publishing these changes, I'm paying attention!  When next we meet for multiple coffees, can I ask some questions?  

Cheers,

Hugh

[stvnharr]

I've been gone for a long weekend, so will catch up now.

Wiring-the CCC wire I use is 25.5ga I believe.  It's what I've gotten from Chimera and use throughout everything.  I think Michael Percy sells some 26.5ga., and it's a lot cheaper.

Hugh, not sure when I'm coming to Melb again, but will let you know.

[stvnharr]

Update:
Backordered BG NX's arrived from pcx, and have now been installed at C5 and C8.  That finishes the component changes.

Resistors are now on their way from Michael Percy for the attenuator.  Will compare S102, naked S102, and CaddockM132 to hear if any make a difference in the ground side of the shunt attenuator form the Dale RN60's. With luck they will arrive before Easter.

Steve

[jules]

I'm all ears ........

jules

[SamL]

Update:
Backordered BG NX's arrived from pcx, and have now been installed at C5 and C8.  That finishes the component changes.

Resistors are now on their way from Michael Percy for the attenuator.  Will compare S102, naked S102, and CaddockM132 to hear if any make a difference in the ground side of the shunt attenuator form the Dale RN60's. With luck they will arrive before Easter.

Steve

Steve,

Where did you get the naked S102 from and how much are they?

Sam

[stvnharr]

Sam,
The Vishays, both S102 and "naked" VSH's are available from Michael Percy, www.percyaudio.com.  The VSH's are 3.50 USD and the S102's are 11.50 USD and up, depending of R value.  The VSH's are same for all values.

Steve

[stvnharr]

I now have the resistors and have them installed in the attenuator.  There are two Vishay S102’s, and a single Vishay VSH and a single Caddock MP132.  They are installed in the four listening positions that I use the most.   If I had wished to make a real proper     devaluation, I would have had the same valuation for each of the three different resistors, but I didn’t do that as I didn’t want the extra resistors.  Thus it’s hard to really do a proper evaluation.  Also, I didn’t want to solder and unsolder any more than necessary on the attenuator boards.  
Thus far in my listening evaluations, I can hear NO difference between any of the resistors, including the stock Vishay RN60’s.  All I can hear is the volume difference in each step, but no sonic difference at all.  I have listened on the output direct from the attenuator (subwoofer output), and the normal output after the tube output stage.
If I do end up hearing any sonic difference, at a later point in time, I will duly report it here.

Thus I think I am finished with any modifications and such to the GK-1.  I have thoroughly cleaned and relacquered the board, and will content myself with listening to music.  It’s possible that curiosity may get the best of me at some time in the future, and I will try a pair of 7308 tubes, but I think that will be awhile.

Thus, of all the recent changes I have made to the GK-1, the only one that made a noticeable and immediate difference was the changing of the signal wires.  That really gave a tight focus to the stereo image.  The GK-1, as originally designed, is extremely well thought out.  Of course all the changes I have made have not made the GK-1 any worse either.  The pursuit of sonic excellence is always a worthy quest.

.

[jules]

Steve,

I noticed that you seem to have tried replacing resistors on the ground side of your attenuator [rather than the signal side]. I might not be understanding what you have written correctly:-

"Resistors are now on their way from Michael Percy for the attenuator. Will compare S102, naked S102, and CaddockM132 to hear if any make a difference in the ground side of the shunt attenuator form the Dale RN60's"

but shouldn't the changes be in the signal path [or am I mis-understanding the operation of a shunt attenuator]?

I'd be most relieved to find that changing resistors in a shunt [or ladder] attenuator makes no difference as I'm not all that keen on removing and re-soldering the 96 odd resistors that make up a ladder  . On top of that, it can be hard to get a complete enough range of values in exotic types, leading to problems with even steps.

Jules

[stvnharr]

Jules,
I have a Goldpoint 20k shunt attenuator, not a ladder.  The series resistor is a 20k Vishay S102.  The ground resistors are the ones that are different.  At the RMAF, Guy Hammel suggested to me that I change some of the ground resistors to S102's as they are all in a complete circuit.  That's what I did.

[jules]

Steve,

thanks for the clarification. I guess it's obvious that the series resistor has to be of high quality and I think it was worth having a go at the ground resistors although in one sense it doesn't look as though it would make any difference since they are simply bleeding signal off to ground. On the other hand, electrons do flow from negative to positive and as Guy says, it is part of a complete circuit so maybe they could have had some influence   . You've taken it past guesswork and done the test so it seems your conclusion is that it's not worth doing.

Have you tried different resistors in the series position?

Jules

[stvnharr]

I have only used the Vishay S 102.

[SamL]

A slight off topic. Can anyone tell me the GK1M's gain? I remember something like 2 but not sure if it is gain of 2 or 2db.

TIA,
Sam

[AKSA]

Hi Sam,

The gain is (68 + 47)/47 * 0.94 = 2.3 which is around 7dB.  However, many are reducing this gain by adding a 10K resistor between the SS section and the top of the attentuator pot (22K), and this is reducing maximum gain from 2.3 down to just under 1.6, which is almost exactly 4dB.

Cheers,

Hugh

[SamL]

Thanks for that Hugh  
Don't mind me asking a few more questions. What is the ideal input quiescent current for aksa n+ power amp? How much quiescent current is on GK1 output?
I am looking for preamp that will (eventually) partner with GK1 in a 5.1 setup. Would like the gain and output of the preamp to be as close as possible.

TIA,
Sam

[stvnharr]

Well, here's my latest and last post on my GK-1.
Earlier this week I transferred everything to a new case of aluminum and wood.  The case is a small enclosure with wood sides, similar to the IAG chassis design.  Tubes are exposed out the top.  When I get access to a digital camera I'll post some pics in the gallery.

As noted in my post in Felipe's thread, there is NO hum now with my
GK-1.  In the old Jaycar case there was always a small amount of hum, but never audible very far away from the speaker.  I think it may have been sloppy wire routing, but really don't know, and doesn't matter now.

I'll list all the modifications from the stock unit, as there are several, including some that resulted and were inspired from last month's disaster.

Capacitors:
C-1 - .47uf BG NX
C-2 - .47uf BG NX
C-4,6,7 - 47uf OsCon SP
C-5,8 - .47uf BG NX
C-10,11,12,14 - Rubycon ZL 100uf 50v
C-15,16 - Rubycon ZL 470uf 35v
C-19 - Rel RTX .01uf
C-21 - Rel RTX .01uf bypassing the Auricap 1uf

Resistors:
R11, 22 to Riken's

Tubes:
Amperex 7308, 1960's NOS

Attenuator:
Goldpoint 20K shunt
Vishay S-102 as shunt resistor, and a few fancy resistors in some grnds.

Wiring:
All signal wiring is 25.5 ga CCC wire, braided or twisted, and enclosed in fibreglass cloth sleeving.

Subwoofer output added: direct output from attenuator to rca.

Case: 330mm X 300mm X 60mm
1mm aluminum top, bottom, front, and back, 25mm jarrah wood sides.


Final Thoughts:  While I might prefer to have dual mono attenuators in place of the stereo shunt attenuator, I can think of nothing else to really do.   I considered changing C21 to several expensive options.  All the teflon caps are very expensive, Cardas's hybrid being cheapest at about 30USD per.  Full service RTX's at about 15USD per were considered, but I had no other use for the auricap and little RTX, so decided to just keep them in place.
The sound is now DEAD QUIET, and so good, that it's at the point of DON"T CHANGE A THING.
Case looks nice, even if fine sanded aluminum does look a little dullish.  Tubes seem to stay much cooler exposed.  One of these days I'll get some pics taken and posted.

[stvnharr]

It is with much sadness that update my report with news that Hum has now returned from the blackness of dead silence.  When something is too good to believe, as dead silence seemed to be, well, beware.  The hum is small in amount and one has to put one's ear up close to the speaker to hear it, but it's there.  But it does not detract on bit from the music and it's enjoyment, and that's the most important thing here.
And there's lot's of musical enjoyment to be had with the GK1 controller in the music system.