PS decoupling caps in Gainclone

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Member
Joined 2005
Paid Member
Saturday I've finally built my first Gainclone (PCBs from chipamp.com)

An externally hosted image should be here but it was not working when we last tested it.


and Sunday was tweaking day... :D

I've started with Panasonic FC bypassed by Evox-Rifa MMk5:

Nice, snappy sound but definitively harsh highs, terrible sibilants :xeye: and lack of basses! (is correct 'sibilant' in English for letters like 's'?)

mmh... :whazzat:

Swapped MMKs with Evox-Rifa PHE246 (polypropilene) from the Zobel:

An externally hosted image should be here but it was not working when we last tested it.


better tonal balance... less harshness and sibilants but not enogh!

Swapped MMKs with Evox-Rifa PHE246 also in PS decoupling:

A little better but not much :rolleyes:

Pulled out bypass caps and swapped FCs with FMs:

A great improvement! Sibilants quite under control, just a hint of residual harshness. :cool:

BUT

The sound was a bit on the cold side, voices a bit thinner than what I like and sound it's a bit tiring.

Well I've some Nichicon Muse KZ laying around so I swapped FMs with KZ:

An externally hosted image should be here but it was not working when we last tested it.


That's it! :cool: less harshness and sibilants under control, voices are bolder, great basses!

An externally hosted image should be here but it was not working when we last tested it.


You could listen to music in total comfort!

In effect that's quite different from what I've read quite everywhere... :confused: has anyone had similar experiences?
 
They are still available from some suppliers but some values are very hard to get. I found 4 x BG STD 100uf 50v caps on eBay which I now have in my LM3886 based amp. It was a straight swap over the stock caps and made a really nice difference. I haven't tried the Nichicon's but I think from what I have read that they have a similar sound to the BG's. I think I will build a diy amp from scratch soon and Im quite keen on the chipamp.com kits. How does yours compare to other amps you have tried?
 
Member
Joined 2005
Paid Member
If you refer to other Gainclones or DIY ones: nope ;)

This is my first DIY project. :D

Until now only modded my Speakers, Amps (Marantz, NAD, HK) and CD Players.

And components selection always been a hard task with a lot of try and change... :smash:

Just removed bypasses of main PS capacitors: nice improvement (another bit of harshness gone...)

Now I'll try with some others bypass caps... it could be that MMKs are not that good (but in my SACD they were a nice improvement over Vishay MKT1826...)

Answering you question:

well, it maybe sounds a little better than my modded marantz receiver (in this moment it is still the pre) but I have to swap back speakers to the Marantz to say for sure!
 
huy5005 said:
I would try with good 300VA power supply. I find that Panasonic FC is actually good with larger power supply ( > 300 VA). Of course, blackgate is better but too expensive for me. Just my 2 cents.

One or two 300VA?

The transformers that I've used are two 120VA R-Core = 240VA: quite similar, isn't it? :confused:

My knowledge of electronic is basic, so why a transformer with more power would make capacitors sound different?

Blackgate is too expensive also for me... ;)
 
ClaveFremen said:
Now I'll try with some others bypass caps... it could be that MMKs are not that good (but in my SACD they were a nice improvement over Vishay MKT1826...)

I've tryed some Epcos Polyester caps and no harshness but sound was somewhat compressed less soundstage, less highs, less life! :xeye:

Part of the residual harshness was due to MMKs, they have more air but here they're not so good. :mad:

But in effect no bypass to Nichicon KG Gold Tune seems better: less harshness, more air, bigger and wider soundstage. :cool:
 
gainphile said:
Do they actually make different sound and did you control the test environment (eg. loudness level)?

I have yet to play around with my caps as I have been agnostic about all these changes. But now really interested if more can be gained.

The test enviroment was flat (pure direct mode) no tone controls and no loudness, using my Marantz receiver as pre.

I'm sure that transformers influence sound, I've buyed R-Cores for that. ;)

What puzzles me is why a different tranformer would change a capacitor's sound... a capacitor that is on LM3886 supplies and quite far from transformer... :confused:

I'm also sure that capacitors do change how a device would sound. I've heard it in every modification I've made to my Speakers, Amps and CD Players.

In effect I've had quite consistent results using Elna Silmics in signal path (replacing stock caps).

Also had good results replacing stock big caps (2200uf, 3300uf) before voltage regulators in all my Amps using Panasonic FCs or Elna Cerafines after or using Rubycon ZL on opamps supplies.

I've always thinked that Sony, for example, in my SACD Player wasted money using Silmics before and after regulators and on analog supplies and Nichicons Fine Gold on digital supplies...

Tryed to swap Nichicons with Oscon SCs there with bad results. :whazzat:

After my Gainclone experience I'm starting to think that audio caps has to be considered also in PSs. :scratch2:
 
huy5005 said:
Try with 20hz tone, measure the supply rail voltage if you have voltage fluctuation, then your need larger cap or larger power supply or regulated power supply which provides enough current for the amp.

Interesting! :cool:

Nice Idea the 20Hz Tone but 10000uf +100uf isn't enough? :scratch: (see photos from first post)

Next test session I would give it a try! :smash:

raypalmer said:
Has anyone found any significant improvement jumping from the 1500 Panny FC caps to a higher value panny fc? 3300 etc....

Aren't you referring to values of Peter Daniel's kits'reservoir/smoothing caps?

Mine is from BrianGT so reservoir/smoothing caps are 10000uF! ;)

Or you refer to 100uF PS decoupling caps?
 
mikesnowdon said:
I think I will build a diy amp from scratch soon and Im quite keen on the chipamp.com kits. How does yours compare to other amps you have tried?

ClaveFremen said:

Answering you question:

well, it maybe sounds a little better than my modded marantz receiver (in this moment it is still the pre) but I have to swap back speakers to the Marantz to say for sure!

Well, yesterday I branched again my Marantz receiver to the front speakers...

The difference is HUGE: the Marantz is bit more polite with highs (a bit of harshness less) but soundstage is a lot smaller, particularly in depth, basses are less controlled, channel separation is in another league (thanks dual-mono!!) and there is quite less dynamic.

Conclusions:

- My Gainclone is a big improvement over the power amp section of my Marantz receiver.
- Small values bypass of good electrolythics isn't always the best thing after sliced bread...in this case they're a detriment of sound quality.
 
mikesnowdon said:
Hi Dario.

Were the different caps you tried all the same value's?

Yes.

100uF 50V for the Elcos and 0,1uf for film bypasses.

I've ordered some Wimas MKP2 0,1uF from Germany. I think they will arrive at the end of next week.

I want to try some different bypass caps... MMKs and Epcos were not good there.

If someone had good results with others 0,5mm led spacing bypass caps not too costly I would try them.
 
Do you have voltage fluctuation at the supply rail?

For the cap, i tried with so many different caps. The larger cap will make the beauty of mid and high disappear. In fact, Panasonic FC 1500 cap is a good choice when you have a large transformer (larger cap doesn't help the bass though for my case). I think the bass issue is the problem of amp and speaker matching. if you want, you can try Carolsm snubbed power supply which i found that it is excellent with small transformer and picky speakers. Don't forget to check Peter's threads 'cause he has done so many experiments perhaps,, drop him a line ;)
 
huy5005 said:
Do you have voltage fluctuation at the supply rail?

For the cap, i tried with so many different caps. The larger cap will make the beauty of mid and high disappear. In fact, Panasonic FC 1500 cap is a good choice when you have a large transformer (larger cap doesn't help the bass though for my case). I think the bass issue is the problem of amp and speaker matching. if you want, you can try Carolsm snubbed power supply which i found that it is excellent with small transformer and picky speakers. Don't forget to check Peter's threads 'cause he has done so many experiments perhaps,, drop him a line ;)

I've measured voltage fluctuations with a 20Hz sine tone and voltage varies from 33.16 and 33.19 Volts on V+, hence with 0.03 Volts variations (V- the same).

To me it seems OK, isn't it?

Your Gainclone is a kit from Peter Daniels or similar?

I asked you because mine is from BrianGT and includes Snubberized CarlosFM PS, Zobel Network, 10000uF smothing caps and 100uF Decoupling caps on chip rails.
 
huy5005 said:
With the 50% volume of sine 20hz from my sound card, my voltage drop down 1.50 volts V+ to gnd.

Seems to me that we're talkink about different things ;)

I've played the 20Hz tone and measured how V+ to gnd varies reproducing the tone.

With no music or tones V+ to gnd measures (V+ to gnd) 34.99 Volts, so it's a voltage drop of 1.80 Volts

I suspect you were talking about the second measure, isn't it? :D
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.