[cmscott6]

So, if this rumor is true:

http://www.partsconnexion.com/blackgate/future_productionOCT.htm

Does anyone have any last minute AKSA/GK-1/Aksonic tweaks to share?  Hugh?  Ginger?  Are there any nonpolar caps in the Aksonics we should think of swapping?

I guess I've swapped the obvious electrolytic caps in the GK-1, but who knows?

Granted, these rumors have been spread before, but PC is usually pretty reputable...

Any thoughts?

[stvnharr]

Why not just replace the little caps in amp and preamp with BG NX's.  The cost is very little, and you can take pride in having all those BG's in the units.  Maybe it'll even sound better.

[ginger]

The most obvious place to "bung" a blackgate is C7. This can be a BG Standard or a BG N BUT you may have problems getting the N to physically fit the PCB - I use a BG standard here myself.

You MAY also wish to fit a BG for C10 and C14. Again I use BG Standard here in my GK-1. I think I left C11 and C12 as original because I couldn't get BGs to physically fit once I'd changed C10 and C14 - That is space limitations dictated I could only change 2 of the 4 to BG and so C10 and C14 were the Best choices.

If it will fit a BG Standard for C6 would also be of some benefit - I can't recall if I did this or not..

The next place to use a BG also involves some minor schematic changes.
This mod gives some worthwhile top end extension and IS RECOMMENDED.
1) Fit the LED L1 in the C4 PCB position - make sure the anode of the LED goes to what was C4 +ve PCB pad.
2) Fit a BG Standard or NX (22uF/6.3V is OK) in the L1 PCB position
That is swap LED L1 and Cap C4 (SEE C4 notes below)
3) Change R13 from 2K2 to 470 Ohms
4) Change R14 from 2K2 to 10K

Notes on C4 in the modified circuit from my work book -
I ended up using 68uF/20V OSCON for C4 - just what I had in the "good caps" bin at the time.
I also tried 2 x 22uF/6.3V NX's in "Super E" (reverse connected parallel) for C4 which sounded even better (by a small margin) but was a pain ITA  to fit to the PCB (one of the NXs had to go under the board).
As another variant of the mod above which was elegant and almost as good sonically, use a wire link for R13 and don't bother refitting C4 in the L1 LED position - just leave it out.
ALL of these variations sounded better than the standard circuit.

Cheers,
Ginger

[jules]

Hi ginger,

the update on possible GK-1 mods is very interesting.

getting a bit OT here I guess and maybe one day BGs will really go out of production and surprise us all after the many cries of "wolf" but I'm considering the use of a Relcap at C1 and a 1/100th Relcap in parallel [and reverse to] the Auricap at C21. Both of these are probably going to have to go on the back of the board.

Aside from the obvious problems of access, board space and slightly tricky soldering are there any reasons why using the back of the board is not a good idea?

jules

[stvnharr]

Jules,
In my Gk-1, I used a BG NX at C1, as the input schematic is exactly the same is in 55N+.   I did use a Relcap .01 RTX bypassing on the Auricap.  You can do this piggyback style on top of the Auricap.  However, after reading around in various places, it seems that bypassing is most effective if the bypass cap is the same "brand" as the main cap.  So, maybe it would be better to bypass the Auricap with a .01 Auricap, rather than the .01 RTX.

[ginger]

Jules,
No reason except perhaps physical height why you can't use the back of the board for mods and add on components. I was trying to suggest easy mods which just involved replacing existing components so that it did'nt need much PCB rework skill level to try.
Your suggested mods should work fine - one caution I added a cap (1/100th value) across C21 att one stage to add some highs - it added highs OK, nasty, agressive,shrill, screaming highs. Needless to say it din't stay in.

Stvnharr's suggestion for C1 is also worth consideration. I use a 470nF Evox Rifa SMR Polyphenylene Sulphide (PPS) Cap for C1 with a 4700pF polystyrene across it. These Evox Rifa SMRs are a "best kept secret" in audio use. I really like them.

Cheers,
Ginger

[jules]

Ginger and stvnharr,

thanks for the suggestions. I will do some processing.

At C7, if the 47uF, 50V, N is too big @ 13x26mm how about a 6.3V NX Hi Q @ 6x7mm or is a standard polarised just as good?

I'm also planning on using a few well placed Riken Ohms and just to get back OT, partsconnexion says that these are going out of production too.

jules

[ginger]

Jules,
NO - C7 needs to be at least 15V rated - 25 or 30V would be better, REGARDLESS of if you are using a polarised or non-polarised capacitor. If 47uF/50V N is too big use 47uF/50V BG Standard.

My experience (from the 55N+ rather than the GK-1) was that the more I improved the circuit in terms of the capacitors the less I liked Rikken Ohms. Remember this is the preamp - you want it as accurate as possible, the sonics tweaks away from accuracy (which is what the Rikken Ohms do) should be confined to the power amp - well thats my philosophy anyway.

Cheers,
Ginger

[jules]

saved from my ignorance there ... thanks.

Final form is now looking like C1 .47uF 50V NX: C2 10uF 50V N; C7 47uF std  50V; C10 & C14 100uF 50V std.. RTXs at C1 or C21 are probably too costly (at ~$15 each) though I might have a go later. Rikens are out unless I do some 55N+ changes.

I'll also have a go at the C4/LED idea as it is a quite straightforward change in advance.

Thanks for sharing the wisdom of many [hundreds ?] of hours of testing. This is well beyond my knowlege so after this phase I'll be back to the more familiar zone of doing stuff with wood.

jules

[cmscott6]

Thanks much, Ginger and Jules, for the suggestions.  

Jules, I've bypassed the Auricaps on my GK-1 with 1/100th value polystyrene (generic industrial surplus   ) caps, and I soldered them on the back of the board.  I thought the top end was better afterwards...

So nobody has tweaked their Aksonics?  

Chris

[stvnharr]

Jules,
I did some checking on exactly what I did to my Gk-1, and it is as follows:
C4, 6, & 7 - Oscon 47uf
C1, 5, 8 BG NiX .47uf
C19 - .01 uf Rtx
C21 bypass - .01 Rtx
I seem to remember changing a couple resistors likely R11 &R22 to either Holco or Riken.

The C10 & 14 change to BG std. is one I missed, but may look to do.  And I suppose C9 &13 to little BG might be a consideration.
Also, may change the C21 bypass cap and use the Rtx somewhere else.

I didn't notice much sound change/difference with the above changes.  It's just the diy thing about doing it just because you can do it.
I suspect that changing tubes might yield some change.  But the tubes needed to affect change are a lot more expensive than a few BG caps.

[jules]

thanks again stvnharr,

I have a parts order on it's way to me at the moment so my inital arrangement will be pretty much as specified in my last post although I did order an RTX for C21 on the basis of a suggestion some time back from Hugh that this was slightly superior to the Auricap.

Ginger made the point here recently that in order to test different components he made changes on one chanel only, used "mono" and with BGs allowed a break-in period. That's very rigorous and quite remarkable when you consider the number of possible combinations.

I'm slightly wary of overdoing the BGs as I recollect an earlier discussion about AKSA 55s suggesting that a blanket replacement of caps with BGs lead to a loss of the characterisitic warmth, leaving a more ordinary SS sound.

I suppose that if BGs really did disappear it would be possible to work something out with a combination of the many other brands available but that would mean a lot more testing of the abovementioned numerous combinations.

jules

[stvnharr]

I have thought of the one channel only method for trying to discern mod changes.  I have used this for some other things.  I find it very trying to discern anything, even with mono programs or various other ways of listening to only one channel at a time.  The powers of perception can get severly strained!!!  But perhaps this can work for some.
It's basically quite hard to truly perceive any mod changes as few have a reference unit and the time to only change one thing at a time in another, and then wait a bit for everything to happen.

The C21 Rtx change is one I have thought about.  Let us know how it goes, when it happens, etc.

[TimS]

The next place to use a BG also involves some minor schematic changes.
This mod gives some worthwhile top end extension and IS RECOMMENDED.
1) Fit the LED L1 in the C4 PCB position - make sure the anode of the LED goes to what was C4 +ve PCB pad.
2) Fit a BG Standard or NX (22uF/6.3V is OK) in the L1 PCB position
That is swap LED L1 and Cap C4 (SEE C4 notes below)
3) Change R13 from 2K2 to 470 Ohms
4) Change R14 from 2K2 to 10K ....

As another variant of the mod above which was elegant and almost as good sonically, use a wire link for R13 and don't bother refitting C4 in the L1 LED position - just leave it out
 ...

Hi Ginger

Thanks for your posts on the GK-1 so far.  Just a couple of questions on swapping LED1 and C4:  
Does the swap require the C4 value to change to 22uf/6.3V or can it remain at the original value of 47uf/35V?  If C4 remains at 47uf do the R13 and R14 resistors still change to the new values of 470R and 10K?
Secondly, the other variant using the wire link in place of R13 - does R14 remain at 2K2 or change to 10K?

cheers

Tim

[ginger]

Latest news is that a contract HAS now been arranged to continue the manufacture Blackgate Capacitors. Prices MAY rise and we MAY find that there is a reduced product range BUT Blackgates will continue to be available for at least several years.
 
Tim,
R13 and R14 do need to change to the new values I listed. C4 does not need to change - you can use the old cap BUT using a Blackgate NX would give better results. The 22uF/6.3V was suggested as this is the value used else where for Nivarna Plus mods to the 100w and 55W amps. The value is not that critical.

Cheers,
Ginger

[jules]

Thanks Ginger, good to know they'll still be around in some form,

I've had my amp out of action for a couple of weeks while making some changes to my speakers and it was very noticeable when I started things up again that the BGs need a bit of a run to get them going again even after a break like this. Initially the sound had elements of a tinny portable transistor radio but it recovered pretty well after about an hour. I seem to remember that lost81 [now lost] reckoned that larger BGs took longer to settle in. All in all I suppose this isn't a very satisfactory thing for normal commercial requirements though it doesn't trouble me.

Black Gates come several forms [or have so far] and I'm wondering if anyone can tell me what the functional difference is between the compact polarized PK and the Standard Polarized forms, apart from size of course?

edit: Could I push my luck a bit here and also ask about the qualities of the non-polarized N types of BG, relative to the polarized types?

jules

[ginger]

Jules,
You can ask BUT unfortunately I don't know the full answer. You can wade thru' the gobblygook here
http://www.partsconnexion.com/resources.php
There is a bunch of stuff on Blackgate Capacitors, jokingly (not intentionally I'm sure), called Blackgate Technical Reports. There is some useful stuff in there BUT be cautious, there is also a LOT of pure marketing B.S., unsupported opinion, and some stuff which is just plain WRONG.
There are a couple of the technical reports which describe the difference between polarised and non-polarised caps there. Look for stuff on the PK and FK series.

Basically it has to do with the DC voltage at which the caps start to look like polarised componenets. Below this "standoff" voltage they work like non-polarised components (and therefore supposedly work better).
I have run polarised PKs and FKs for the Vbe multiplier bypass cap in the 55N+ for example (where there is about 1.2 Volts DC across the cap) and they sounded great. We ended up using an NX non-polarised which we though sounded even better.
 
Cheers,
Ginger