[jules]

thanks Peter,

I won't be going overboard with the expense here but these parts justify putting in an order where a short  list of vital but cheap resistors I've been wanting for some time, did not. Why are there so many different values to resistors and why don't I ever have the one I want?

I might give the paralleling a go at the same time.

jules

[AKSA]

Hi Peter,

Many thanks for a terrific contribution to AKSA art - this Sonicap with teflon bypass sounds like the cat's meow......

Jules,

Not ever having the correct resistor in stock is a variant of Murphy's law, that hapless fellow immortalised in folklore just before he committed suicide on being informed by his fiance of an impending son and heir to the family's misfortunes.....

I have thousands of dollars of resistors.  Quite frequently, out of all proportion to numbers in stock, I do not have the one I need.  In a determined and vengeful attempt to meet the requirement, I actually bought many, many way out values (464K, 422R, 44K2, 215R, 487R etc), but still the problem persists.....

Cheers,

Hugh

[TimS]

Jules,
I bypassed the 2.2uf Sonicap I used for C1 with a Relcap 0.022uf 200v TFT (dimensions 0.23" D x 0.75" L), roughly $USD 20 each at Parts Connexion.  I chose that one because I wanted to mount the bypass on the foil side, under the C1 Sonicap.  The TFT was small enough to fit between the GK-1 Analog PCB and the top plate of my chassis.

This cap is still breaking in, but (after 100+ hours) it is sounding like it might be a keeper.  My intention is to listen to music now... if the AKSAphile community finds another cap for C1 that is just wonderful, I may give it a try someday, but--in my system--a 2.2uf Sonciap ($7.60 each), possibly bypassed with a really nice smaller cap ($$$), sounds very, very nice to me.

Take care,
Peter   

Hi Peter

Going back to your tweaking, you found that initially changing C1 to a 22 uF electrolytic (Blackgate) improved the sound considerably, but you settled on the 2.2 uF film capacitor as the best option in the end. 
Do you think that the major improvements in changing C1 are down to the fact that you have increased the capacitance from 470 nF to 2.2 uF (and 22 uF) or mainly due to changing C1 to a film capacitor? 
The reason I ask is that I have a couple of spare (and rather large) 470 nF RTX film capacitors for C1 that I am thinking about trying but wonder if I should also be going for an increase in capacitance as well?

cheers

Tim

[PSP]

Hi Tim,
Because I am building Orions and SL recommends that the preamp be good down to 5 Hz (I'm guessing that this would minimize phase error at 20Hz), I asked Hugh what I might do to extend the GK-1 bass performance a bit.  He suggested that C1 be changed to 2.2uf (good film cap) and that C2 be set to 33uf (good electrolytic--I used BG).

After reporting nice results with this setup, Jens and Hugh asked me to evaluate the 22uf BG NX at C1; after a lot of adventure, I liked the film cap better (I used a 2.2uf Sonicap Gen I).  I have since bypassed the 2.2uf Sonicap at C1 with a 0.022uf teflon TFT (this cap was small enough to fit under the board) and I like that even better.

My subjective impression is that increasing C1 and C2 noticably extended bass depth and resolution; in fact with my current speakers (GR-Research Paradox 3  http://www.gr-research.com/discontinued/paradox_3.htm ) the bass might now be a bit too heavy.  As I mentioned in an earlier post, I expected deeper bass (based on Hugh's recommendation) but I was very surprised to hear improvement across the whole spectrum when I made these changes.  I don't really know (because I did not do the entire experiment), but I think that increasing C1 and C2 extended bass response while using a better film cap at C1 was probably the dominant contributor to the generally improved across the board sonics I heard.  I have no way of knowing (without doing more soldering) whether the new value of C2 contributed to the overall improvements I heard... maybe Hugh's thoughts here would be helpful.

So, if I was not building the Orions, I would definitely try a good 0.47uf film cap at C1 and an RTX should fit the bill nicely (although squeezing it in might be a trick).  I would make a careful comparison of hum and noise before and after you make the switch, just in case the larger cap and/or longer leads pick up enough noise to outweigh any sonic inprovements.

Let us know what you think, OK?
Peter

[pea]

Peter, Hugh,

This tweak sounds great and I'm about to do this to my GK-1. However, after receiving the parts (C1 - 2.2uf Sonicap Gen I Film Cap, C2 - 33uf Black Gate N Electrlytic Cap and a Relcap 0.022uf Teflon TFT Cap), I'm wondering how you mounted the Sonicap?

It's pretty big. Do you mont it on end on the componet side or from the foil side of the board laid on it's side?

Thanks for the tips guys!

Bruce

[jules]

Bruce,

they are big aren't they 

Hugh answers this on page two of this very thread in answer to a similar Q that I had asked.

jules

[pea]

Hi Jules,

Thanks for your note, however I think the reply Hugh gave on page 2 of this thread is about switching C1 on an AKSA 55N+ amp module and not the GK-1 preamp.

Am I reading that correctly?

Thanks -
Bruce

[AKSA]

Hi Bruce, Jules, Peter, et al!!

There is little doubt in my mind that in any series signal application like C1 a film cap beats an electrolytic, even a BG.  However, there's a downside;  the film caps of 2.2uF are normally rated to at least 50V, way more than we need here (it has about 0.2 volts across it!!), and the size of the cap increases with the square of voltage, which makes 'em big!  This size also makes them large from an RF injection standpoint, and some attempt needs to be made to keep them away from pesky EM fields.

If, on the power amps for example, a very large cap for C1 is mounted under the pcb close to the inductor, positive feedback results since the inductor radiates its output back into C1, causing oscillation.  On the GK1 the problem is not so great since the voltage output is not high and there is no inductor;  however, space is short and for layout reasons these larger C1 caps can't be mounted either on the component or the foil side.

A single lead from the input board (or RCA panel input on the EL) carries input to the SS section of the GK1, so this can be broken midstream and C1 neatly inserted.  Naturally this takes C1 offboard, but there is no penalty for this.  The C1 mounting point on the pcb should then be bridged with a tiny wire link.

This should solve the problem, Bruce!

Cheers,

Hugh

[pea]

Hi Hugh,

That's the ticket sir!

Thank you for the advice,
Bruce

[stvnharr]

Hello,
This thread has mostly been devoted to the input cap possibilities in the GK-1.  My post here is to be about the other two "significant" caps in the GK-1, but they are still quite a simple "tweak".
I'll just make a long story short and say that I recently, like a couple weeks ago, changed both C-19 and C-21 in my GK-1: C-19 from .01 RTX to .01 Sonic Cap Platinum teflon cap, and C-21 from 1uf Auricap bypassed with .01 RTX to .47 Dynamicap E series metallized polypropylene.
I've changed a lot of caps in my GK-1 thru time, and never have really been able to discern any real difference.  Maybe this was because the caps were all new at installation, or something else.  There may have been subtle changes after a time, but nothing that I can honestly really note.

This time was different though.  The Sonic Cap and DynamiCap were well broken in from a year or two's service in one of my N+ amps.  Shortly after installation I thought I was hearing a more precise stereo image, more detail, and less "polypro" sound from the fibreglass reinforced polypropylene drivers in my speakers.  I initally ascribed the difference in sound to things like moving the speakers a little, or moving my listening chair a bit to get into the "equilateral triangle" position with the speakers. 
But then last weekend I put on a disc that I had always found problematic.  I mostly listen to classical sacd's, and the worst thing that can happen is to spend $30 on a disc and have crummy sound in part of the disc, in this case, muddy, cloudy, veiled piano sound.  But last Saturday night I only heard crystal clear piano sound on this disc, in the previously offending portions of the disc.  Other parts of this same disc were also much much cleaner and clearer sounding.

Well, after mulling it all over, and trying a few things, I can only conclude that the new caps, being the only change in anything really, have brought this improvement in sound, as that was the only thing that was really changed.

Many have extolled the virtues of teflon caps.  The bypass caps didn't really do anything in my N+.  But many others have written of "sound transformation" simply by installing teflon caps at appropriate places.  I used the Sonic Cap as I had it, and they are half the price of the V-Cap teflons.  Both are made at the same place, though I'm sure to different specifications.
I simply used the DynamiCap 'cause I had it, and Michael Elliot, the former Counterpoint guy now with a new company, extolled their virtues in a cap comparison including RTX's and Auricaps.  But most caps have their group of "supporters".

I dunno though.  First time for everything I guess.  Be interesting if any one else, other than Peter, has had experience with teflon caps in the GK-1.

[DSK]

...C-19 from .01 RTX to .01 Sonic Cap Platinum teflon cap, and C-21 from 1uf Auricap bypassed with .01 RTX to .47 Dynamicap E series metallized polypropylene...

Stvnharr, thanks for sharing your caps swapping experience with us.

Hugh, just wondering, aside from any sonic differences between different caps, what is the likely effect of changing the value of C21 from 1uf to .47uf?

[AKSA]

Steve,

My thanks too!

Darren,

The substitution with a smaller output cap on the GK1 would raise the corner frequency of the bass.  This is because a smaller cap would have more series impedance, reducing the amplitude of the bass frequencies more than the treble or midrange.  6dB down (half voltage) figures for feeding a 40K Zin amplifier such as the AKSA/Lifeforce would be:

1uF:  4Hz
470nF:  8.5Hz

Since 8.5Hz is low enough - the lowest bass guitar note is only 43Hz - I would say no problem at all!

Cheers,

Hugh

[andyr]

Steve,

My thanks too!

Darren,

The substitution with a smaller output cap on the GK1 would raise the corner frequency of the bass.  This is because a smaller cap would have more series impedance, reducing the amplitude of the bass frequencies more than the treble or midrange.  6dB down (half voltage) figures for feeding a 40K Zin amplifier such as the AKSA/Lifeforce would be:

1uF:  4Hz
470nF:  8.5Hz

Since 8.5Hz is low enough - the lowest bass guitar note is only 43Hz - I would say no problem at all!

Cheers,

Hugh

Hi Hugh,

Then it seems to me that is not a good substitution (1uF --> 470nF)??   

As I understand it, increasing the input coupling cap certainly produces more bass and increasing "the other one" which was referred to in the original post - is that C22?? - has also been noted to do this.  So why would you want to go the other way by decreasing C21?

If you want a teflon cap in that position, why not simply substitute the same value - or even bigger!!??  ("Nothing succeeds like excess"        )

Regards,

Andy

[AKSA]

Hi Andy,

Thank you for your post!

You would want a smaller value cap here for two reasons.  One is cost, the other is physical size.

There are arguments that a very large cap at C1 will reduce phase shift to vanishingly low values at low frequencies, and this is indeed valid, but I was asked the effect.  I gave some computations to show the effect was not very significant from a measured LF point of view, and while I can broadly agree with your points, I tried to answer the questions objectively.

Until this is implemented and road-tested, then the sonic impact will not be known.  However, based on past experiences and the figures, it seems to me that the effect is marginal at best, and you will notice I neither recommended it nor condemned it.  I try to keep an open mind - there is no other way to learn - and these issues are left to the individual to decide.  My 'house' sound, FWIW, is just that;  the considered choice of one man, and nothing is more dangerous than one man's opinion.   

Thus, for you, your opinion should prevail.  No one else's opinion will do!   


Cheers,

Hugh

[stvnharr]

My GK-1 schematic shows C-21 value of .47uf, thus I figured it was okay to use the cap I had.  FWIW, the 1uf DynamiCap is same price as smaller value, or nearly the same.  I only used it as I had it and was curious as to the difference.  A teflon cap of either .47 or 1uf is very very expensive, any brand you like and especially when bought in pairs.
Again, it's all just my experience here, and the increase in sound clarity, etc. can only be from either changing the two caps, or moving my listening chair a little bit, as that's all that changed.  I didn't think moving the chair would change the dull piano sound, only that something in the circuit could do that.

[DSK]

...I didn't think moving the chair would change the dull piano sound, only that something in the circuit could do that.

Easy to check, just briefly move the chair back and remove any doubt.

[andyr]

My GK-1 schematic shows C-21 value of .47uf, thus I figured it was okay to use the cap I had.  FWIW, the 1uf DynamiCap is same price as smaller value, or nearly the same.  I only used it as I had it and was curious as to the difference.  A teflon cap of either .47 or 1uf is very very expensive, any brand you like and especially when bought in pairs.
Again, it's all just my experience here, and the increase in sound clarity, etc. can only be from either changing the two caps, or moving my listening chair a little bit, as that's all that changed.  I didn't think moving the chair would change the dull piano sound, only that something in the circuit could do that.

Sorry, Steven,

Your previous post said "... and C-21 from 1uf Auricap bypassed with .01 RTX to .47 Dynamicap E series metallized polypropylene"

so I assumed you must've had what the schematic listed.  I didn't realise you'd upped C21 from 0.47uF, initially.

Regards,

Andy

[stvnharr]

Darren,
The dullish piano sound I mentioned is completely gone now, no matter the chair position.  I had not listened to the recording in a year, so hope my memory is serving me well, and I think it is.
Memory can play tricks.  I spent a full week of, "gee, things are sounding really good" time to see if the newer sound was real, or just imagined.  Even now I remain slightly sceptical of myself, as cap changes never produced much for me before, and I didn't really expect these 2 changes would be any different.  But, the perceived difference in clarity and precision in the stereo image is quite apparent to me.

Andy,
My GK-1 schematic, which is an early version, and not entirely correct, shows .47uf for C-21.  The Auricap came stock and is listed as such in the component list.

In addition: neither of the caps I installed were on the market when the GK-1 was designed, so therefore could not have been tried and tested at that time.  I think the REL teflon cap was tried, but I could be incorrect.  I have read in Arthur Salvatore's spiel on the V-Cap that the V-Cap is a noticeable improvement on the REL teflon cap.  I think the V-Cap and Sonic Platinum are similar, but others may dispute that.
As for the DynamiCap, well, everyone seems to have their favorite metallized polyprolylene cap.

[stvnharr]

Here's something new on the subject of capacitors for eveyone to mull over -

http://www.vhaudio.com/21capacitorshootout.pdf

It's just one person's view and opinion of course.  But it seems reasonable.

[andyr]

Here's something new on the subject of capacitors for eveyone to mull over -

http://www.vhaudio.com/21capacitorshootout.pdf

It's just one person's view and opinion of course.  But it seems reasonable.

Great article!! 

Regards,

Andy