Group Buy: MOVED TO NEW THREAD

Felipe · « **Reply #20 on:** 7 Jun 2008, 07:50 pm »

Can you tell me difference between Sonicap Gen 1, Gen 2 or Platinum ??

I believe Wayne from Bolder uses Sonicap Gen II in his mods, so i assume that is the best one ?

whubbard · « **Reply #21 on:** 8 Jun 2008, 01:37 am »

Felipe,
The Platinum is the top of the line cap, considered by many to be they best cap in the world or to some, the second best after the V-Cap. The Gen-I and Gen-II are 'worse' caps that the Platinum. I'm not sure of the technical difference between the Gen-I and Gen-II, but according to SonicCraft:

Quote

The Gen I is basically a fairly balanced capacitor. However, some find our "balanced" slightly thin and bright compared to the more lush "colored" capacitors which would include the Gen II. While the Gen II has a nice Midrange bloom, Gen I has a slight advantage on the freq extremes (top and bottom). In many applications, larger values of Gen I bypassed with smaller values of Gen II yield a very favorable result.

The Gen-II also comes only in smaller values 0.001uF to 0.82uF while the Gen-I goes all the way from 0.001uF to 30uF. The Gen-I is also cheaper. Maybe it would be worth bypassing the 20uF with a small Gen-II?

If there is interest in bypassing with a .47uF 200V cap, which would add $11.2, $9.6, or $8 depending on the quantity. It is also a small cap that would easily fit along with the bigger Gen-I.

Hope that helps.

-West

kyrill · « **Reply #22 on:** 8 Jun 2008, 08:34 am »

I would be in for the smaller bypass cap

RonR · « **Reply #23 on:** 8 Jun 2008, 09:03 am »

Hi All,

At the moment I have Sonicap Gen I 20uF bypassed with Gen II 0.47uF (Not ideal I know, should be around 0.2uF) burning in at C7 on my GK-1. First impressions were quite good, but then things went a bit "Tizzy", I'm expecting them to come on song in a few more hours.

Thinking about it, I'm going to add a 1-2 Ohm resistor in series with the smaller cap in case there's a tendency to ring at these values.

BTW they are physically larger than the Solens.

Cheers,

Ron.

kyrill · « **Reply #24 on:** 8 Jun 2008, 09:16 am »

thx Ron for the work
we really wait for you

Yes they are quite large, but then..
when placed and put away it all will sing along singing songs..

whubbard · « **Reply #25 on:** 8 Jun 2008, 02:42 pm »

Ron,

Thank you very much for all you tests in choosing the best cap for C7 and C4!

You mentioned that bypassing with .47uF wasn't correct and that it should be .2uF, but I don't see why. Would you mind explaining how it could make a difference? (I'm just thinking that the circuit will either see 20.47uF or 20.2uF and I don't see why the difference will that big.)

-West

kyrill · « **Reply #26 on:** 8 Jun 2008, 09:09 pm »

common sense ( learned from Ginger alias Ian? or did i learn from Tinker alias Ben? I am not sure) that a good bypass is 1%-10% of the to be bypassed value. Preferably near the the 1%

Ian and Ben you have a GK-1 as well, and does this make you interested in this group buy?
K

PT914 · « **Reply #27 on:** 8 Jun 2008, 11:46 pm »

Hi Ron,

Since you are testing caps, you should try a 47 uF Solen. I just did a quick comparison on one channel replacing a 47 uF Solen with a 20 uF Aeon, which is close to a Solen according the Percy. True I did not burn it in but there is a difference. I also compared one channel to the other, which could be limited by channel imbalance but I also switched channels from source and the side with the 47 uF cap had the better sound stage and better highs. I am comparing channels with my small workshop system so I can't say much about bass.

Cheers,
Philip

AKSA · « **Reply #28 on:** 9 Jun 2008, 12:15 am »

Omigosh,

This could start a flurry of activity!!

First, let me confirm. You are all referring to C4, which is the bootstrap cap on the voltage reference of the CCS which supplies the output device on the SS section of the GK1. We might be referring to C7, which is the last cap decoupling the stage feed to the long tailed pair (T1/T2), but I don't think so...... correct me if this is wrong!

I'm really frustrated by this; I feel I missed something crucial during the design phase. Doggone!!

Now, let me suggest something. The CCS is merely a current provider, an energy stream if you like. It's single quality which detracts from the sonics is noise. C4 is used as a bootstrap cap between rail and the mid point of the resistive feed which keeps the LED alight. This means that any AC peturbation on the rail is reflected at this mid point by the cap, and then any differences across the cap are diminished by resistor to LED impedance ratio; it's a great system, one I've studied years ago and it works very well.

However, the quality of the cap is turning out to be a major issue here. Bearing in mind it's noise we are on about, and that the CCS active device is also a noise amplifier, perhaps we should choose a low noise transistor for this application?

I've used a BC557 transistor, which has a typical noise figure of 2dB and a max of 10dB at 0.2mA, a fair range. The working current in the GK1 is 12.6mA, MUCH higher, at which the noise should be much less, probably around the 2dB mark at the most, particularly as signal throughput is line level. IT IS POSSIBLE THAT A LOW NOISE TRANSISTOR MIGHT REDUCE THIS NOISE ALSO, THEREBY RAISING THE BAR ON RESOLUTION.

Something to consider, no?

Hugh

andyr · « **Reply #29 on:** 9 Jun 2008, 12:27 am »

Quote from: AKSA on 9 Jun 2008, 12:15 am

I've used a BC557 transistor, which has a typical noise figure of 2dB and a max of 10dB at 0.2mA, a fair range. The working current in the GK1 is 12.6mA, MUCH higher, at which the noise should be much less, probably around the 2dB mark at the most, particularly as signal throughput is line level. IT IS POSSIBLE THAT A LOW NOISE TRANSISTOR MIGHT REDUCE THIS NOISE ALSO, THEREBY RAISING THE BAR ON RESOLUTION.

Something to consider, no?

Hugh

Absolutely, Hugh!

But maybe you need to experiment with some different transistors and let us know? Not having to have "boat anchors" hanging of the side of the main GK-1 PCB but achieving the same sonic improvement would certainly be a good thing!

Regards,

Andy

PT914 · « **Reply #30 on:** 9 Jun 2008, 12:30 am »

Hi Hugh,

The Gk-1R I am playing with only had a 47 uF Solen at the "C4" positions. I still have Ginger's mod where the LED is switched with C4. C7 is a BG 47 uF. So I just replace the right channel at C4 with a 20 uF Aeon. I'm surprised that Ron heard an improvement with the 20 uF cap. There is quite a difference between a 20 uF and 47 uF cap at C4.

Philip

AKSA · « **Reply #31 on:** 9 Jun 2008, 12:42 am »

Phil,

Thanks for explaining; this makes quite a difference. A 47uF in the 'Ginger Mod' position will be vastly more effective than a 20uF; but won't make much difference using the pre-existing topology.

The Ginger Mod relies on a large capacitance here since the series resistor powering the base of the CCS is necessarily small, probably around 1K. I would go for a 100uF cap, but it need not be large physically because it's VERY low voltage, only working at 2VW.

The [pre-] existing topology with the bootstrapped decoupling regime has much more voltage across it, around 7V. This means that the energy content of a cap in this position is 49/4 times more - that's a useful 12 times, so the cap need not have such a high capacitance, around half would still give it an advantage. Add to that the multiplying effect of the 2K2/LED impedance ratio (that's typically 2.2k to 75R, a ratio of almost 30 times) and you have a very effective voltage reference for the CCS.

I'm surprised that the Ginger Mod returns a better noise figure, to be honest, but if it's definitely quieter, then it's irrefutable.

So, to sum up; If you use the Ginger Mod, the voltage across the cap is MUCH smaller, and the capacitor MUST be at least 47uF and preferably 100uF, or even more if you can stretch it. When film caps likes Solens are rated to 250VW, it seems overkill, and better to use it in the bootstrap (cf. existing) topology.

Thanks Phil, appreciated,

Hugh

whubbard · « **Reply #32 on:** 9 Jun 2008, 01:58 am »

Well I guess here is the big question:
Has anyone compared the ginger mod to replacing both C4 and C7 with a good cap?

I'm really pleased at all the discussion as I think its going to help us all obtain amazing GK-1s!
Maybe we will even approach the Swift!

-West

kyrill · « **Reply #33 on:** 9 Jun 2008, 08:05 am »

haha
so we need a low noise transistor AND 2 big filmcaps, no? aa

curly joe · « **Reply #34 on:** 9 Jun 2008, 08:31 am »

Hi all, just got back from a weeks holiday and stumbled across this thread. I'd definitely be interested in joining a GB when a decision is made on which cap to go with. Thought I'd better chime in to give an idea of numbers.

Cheers
CJ

RonR · « **Reply #35 on:** 9 Jun 2008, 12:06 pm »

Hi Phil,

Just before I started the tests, I reversed the "Ginger" mod in my GK-1, returning it to Standard configuration. I suppose that explains my results.

Unfortunately I only have 1 pair of Sonicaps, which are at C7 at the moment. TBH I don't hear much difference between it and the Solens. I'll wait another day for break-in, then swap them into the C4 position.

Hugh, I have some Fairchild BC560s, which are low-noise versions of the BC557: noise is 1.2dB typical, 2dB Maximum. Would these noise figures be low enough to substitute, or are we looking for something better?

Cheers,

Ron.

kyrill · « **Reply #36 on:** 9 Jun 2008, 09:32 pm »

Dear Ron
Dont look to much to db
soundwise what sound better is much more than linear db

oh i see, it maybe much more than 2db difference

AKSA · « **Reply #37 on:** 9 Jun 2008, 11:27 pm »

Hi Ron,

Yes, BC560s should be the cat's meow, do try 'em out!!

Cheers,

Hugh

PT914 · « **Reply #38 on:** 10 Jun 2008, 06:11 am »

Hi Hugh,

You were right! With the Ginger mod, 47 uF is better than 20 uF at C4 and 67 uF is even better than 47 uF at C4. 100 uF may be the ultimate end but I'm running out of room. As the cap goes higher the improvement in the highs are even better and the improvement reaches to the bass. I have probably reached the limits of these little bookshelf Paradigm speakers. I've spent the last hour listening to these little speakers and am just amazed how good they sound with these mods compared to the Orions.

Cheers,
Philip

aurelius · « **Reply #39 on:** 10 Jun 2008, 09:57 am »

Hi Philip,

Quote

I've spent the last hour listening to these little speakers and am just amazed how good they sound with these mods compared to the Orions.

Interesting statement... I can read it one of about 4 ways:

1) The Paradigms sound better than the Orions. period.
2) The Paradigms with the modded GK-1 sounds better than the Orions without the mods.
3) You can hear the effect of mods more readily with the Paradigms compared with the Orions.
4) Dollar for dollar and pound for pound, the paradigms perform well against (but are not an improvement on) the Orions.

Very interested in which you meant?