GainClone PS auditions

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
PS auditions

I finally got around to trying the 4-transformer low-flux PS Kuei proposed. The GC in question is 3875 inverted with just 2 resistors and no input cap. Initially a 160VA toroid was used per channel. The transformers i ordered are 100VA EI with dual secondaries which feed dual bridges. I compared the toroid to a single EI (not very fair powerwise) and to two EIs per channel with primaries and secondaries connected in series. Still haven't included dc blocking in the primaries.
The results in my system are a bit controversial. The toroid, as in past experience, produces the fastest and clearest mids, good dynamics but obvious congestion at higher volume and weak and inconsistent bass. The single Ei doesn't give the impression of being less powerful even at higher volumes (my transformer winder claims he's very conservative with the ratings); the sound is a bit slower with a nice deep bass and good depth layering and probably more listenable long term. The combination of two series connected transformers initially showed promise with its ability to efficiently move air. It has slightly overblown bass with good depth, a nice and steady soundstage but very distant mids and generally slow and uninvolving sound. This may have something to do with the low flux but previous experience shows this often to be the result of going over the top with the VA rating.
I also tried 4x12v/7ah batteries and while sounding great with acoustic music, on a wider selection the bass sounded disjointed and lagging from the rest, seriously screwing PRAT. I think the problem is not inherent in batteries but relates only to acid batteries, or even to particular makes, as i use NiMH in my MC stage with great results.
Incidentally my 'normal' amp is a way-over-the-top PX25 SET with 3 PS per channel.

peter
 
Re: PS auditions

analog_sa said:

I also tried 4x12v/7ah batteries and while sounding great with acoustic music, on a wider selection the bass sounded disjointed and lagging from the rest, seriously screwing PRAT. I think the problem is not inherent in batteries but relates only to acid batteries, or even to particular makes, as i use NiMH in my MC stage with great results.

What make batteries were you using ?

were they new or fairly new ?

how large were the bypass caps ?

and excuse my ingorance but what is PRAT ? my guess would be precise rhythm and timing but this abreviation is new to me.

cheers

mike
 
Hi Mike

Relevant questions indeed. The batteries are out of my UPS, fairly used but charge up to full capacity. Sorry, forgot the make, certainly not Yuasa or Panasonic. The bypass caps are simply the caps i use with the mains PS - 2x1000uF/chip BG. The character of the sound with batteries is slightly reminiscent of what you get if go overboard with the supply caps. Smaller batteries might sound better. Make no mistake though, the sound with batteries may be stunning on simple acoustical music, especially vocals. If girls with guitars is what you listen to mainly, batteries are great.
You're very close on PRAT - Pace Rhytm and Timing - a popular term in turntable-talk.

cheers
peter
 
I understand you are talking about this PS. Which diodes did you use? From what you mentioned, it seems like using transformers separetely may actually produce better sound than series connection?
 

Attachments

  • ps.jpg
    ps.jpg
    16.8 KB · Views: 1,897
Peter

That's the circuit which started me. I used MUR860 in a bridge as i didn't want to experiment with too many variables at the same time. Surely full rectification wil sound different to bridge - less switching noise from the diodes but more stress on the caps - may well be better. I might try it first in a preamp as it seems to make more sense with less current. My observations about the sound may only be relevant in the context of the particular transformers i use. Kuei seems to recommend this circuit with toroidal transformers and the results may well be different. All this is pretty inconclusive...sorry.

cheers
peter
 
The diodes limit the voltage across the caps to 1.4v. I still haven't tried this circuit. It will probably have a more pronounced effect on toroids than on EI and while some of this effect will no doubt be beneficial you'll have to cope with an extra cap colouration. it will be great if you give all this a try.

cheers
peter
 
I tried today PS suggested by Thorsten and pictured in post #4. I used 2 Plitron toroids rated 300VA and 2 X 35V. I compared it to single Plitron toroid rated 400VA and 2 x 18V. I used single bridge in new supply between primaries and MUR860 diodes everywhere.

It seems like a new supply has much more micro detail. Low noises and sounds, I normaly didn't notice, suddenly can be further detected, highs are also cleaner and soundstage seems to be bigger.

The old supply has more lower midrange but seems to be veiled somewhat. Please note that I didn't use fuse and switch in new supply, so that may affect things somehow. I also preferred when capacitors and bridge was used between primaries, as the highs were softer and possibly more detailed. I also fed the supplies from 10A Variac, as this brought further improvement.

I couldn't use secondaries of two separate transformers together with same diodes as posted, because of slight difference in voltage on both transformers (and this produced buzzing). I used one transformer for positive, the other for negative rails. I might unwind secondaries on one to match the voltages.

I think it's a worthwile upgrade to a regular supply (although I didn't consider it originally;)), especially if you can find cheap toroids. Mine were taken from older amp I didn't use anymore.Makes me wonder now, what batteries can do?;)
 

Attachments

  • ps.jpg
    ps.jpg
    67.5 KB · Views: 1,418
Might try 6.8uf, Analog...

Roncla over on Amp Chip DIY posted :

" low value caps around 6.8 uf on the power rails with a 1 uf between the power pins with just battery power alone."

And when the charger is connected:

"use 1000 uf at the P/S (batteries and running chargers) on the posts , both caps going to either power posts and the ground with a 1uf between the chip pins."

What's the sonic difference?

"The second seams to give a slightly brighter sound and has the advantage of plug and play without having to switch the chargers on and off, i just leave it on 24/7. "
 
tiroth said:
It's not surpising to me that battery supplies don't sound good on GC. The 1000uF caps should be the tipoff.

these are very vague comments

why is it not suprising ?

why should the caps be a tipoff ?

have you tries batteries ?

If not, how do you know what they sound like ?

there are some people on this forum, who have tried both methods, that regard them as the ideal supply for chip amps, although I know that this is not a universal view.

I think to give batteries a fair trial it will be necessary to:

use 4 x 12V per channel. Otherwise the earthing is compromised and there will be some cross talk via the supplies.

experiment with the best value/make of bypass cap.

mike
 
It's not surpising to me that battery supplies don't sound good on GC. The 1000uF caps should be the tipoff.

What Tiroth probably means is that a 7A/h battery is the equivalent of a much larger cap. My limited experiments with batteries lead me to believe that acid batteries do not sound to my liking - they seem to destroy PRAT big time. Of course the midrange is glorious and the highs very different to what we normally expect from SS - strangely untiring. The bass is deep but lags a bit after the rest. Stereo is astonishingly wide - seems a common trait of all batteries. I fully believe that a type of battery (probably NicD) exists which will only improve upon mains supply.
 
Exactly. mikelm, I used SLA in my recent active crossover and was pleased, but it seems that they address a nonexistent problem with GC-type amp. People are not exactly coming out of the woodwork complaining about mains hum or lack of bass, the two biggest issues that I would see SLA addressing. And in fact people have made comments like "capacitance up, quality down" and "too much bass at +-25V."

analog_sa, how do you feel about battery supplies for line-level applications? Do you think the difference with your phono stage is due to battery type, or do you think that battery supplies are just not as good for power amps versus line-level?
 
tiroth said:
Exactly. mikelm, I used SLA in my recent active crossover and was pleased, but it seems that they address a nonexistent problem with GC-type amp. People are not exactly coming out of the woodwork complaining about mains hum or lack of bass, the two biggest issues that I would see SLA addressing. And in fact people have made comments like "capacitance up, quality down" and "too much bass at +-25V."

This is quite a big area to discuss.

The benefits I noticed from batteries had nothing to do with hum or bass. Much more to do with dead quite backgrounds, very refined low level detail, and overall a nice natural smooth sound. I achieved similar results using a multiple choke PSU. In other words it has to do with low noise or perhaps a different kind of noise. I have to say here that I used the batteries in conjunction with 10,000uf T network caps which have a extremely effective HF filter inherent in the design.

Unless people are using big high efficiency bass speakers ( 15-18 inch 98-100db/w/m ) I find it amusing that the power supply is blamed for slow bass.

Russ Andrews, a hi fi innovator in the UK in the seventies & eighties was of the imformed opinion that getting 'fast' sounding bass by having small PSU caps is simply the sound of a PSU running out of energy. This may be desirable if one is trying to get the impression of 'fast bass' from small inefficient speakers, which almost by definition are not capable of producing a truly realistic fast bass sound. Hence the comments 'too much bass' or 'overblown bass' perhaps.

I do agree however that some small caps may give better mid & treble reprodution.

mike
 
I did some more comparisons today.

I had 225VA Plitron with electrostatic shield between primary and secondary windings and 400VA Plitron without shielding. Both had double 22V AC secondaries. Each transformer was tried with a stereo amp.

Smaller transformer sounded smoother and less edgy, but the bigger was a much better choice, providing more depth, ambience and 3 dimentionality.

I also tried PS build with 2 EI, 250VA, older type transformers (ea. transformer supplying one rail). This PS didn't even come close to what toroids offered.

It also seemed that a 400VA toroid with 18V AC (instead of 22V) was less edgy sounding, but it's hard really tell OTOH, because one supply was well broken in and the other was brand new.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.