This is not just another gainclone

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
we could try active power supply regulation

Post 1183


" The +/- 27 volt suppy is regulated to +/- 20 volts by two LT1084 regulators with 47uF Black Gates at the regulators inputs and 100uF Black Gate at the regulators outputs. Two additional 100uF Black Gate caps are mounted next to each LM1875 for high frequency decoupling."

Your right, it is a good idea........ Glad to see the French backing the US on this one.
 
regulated supply sound?

Hi Fred,

I wonder if the sound you are hearing (very analytical) is as much because of the regulators you employed as the other parts. I have a few friends who have replaced their SET tube amps in favor of a gainclone. In fact, one friend sold his 45 amp upon hearing the LM1875 amp I built for him. He tells me that he swaps the SS amp out for his SET using #10 tubes depending on what he's listening to and his mood at the time. Heck, he had had more $$ invested in just the OPTs of the SET than I had in the whole SS amp.

That being said, I ended up building him 2 power supplies, one that went with the first amp and another so he could take a crack at building a 'clone. Acording to him, they sounded very different (on the same amp module, he has not had success with his attempt) even though they used the same Nichicon caps. The voltages were different and I used no regulators on either. The salvaged 200VA Sony transformer with 2*20 volt taps sounds better/more tube-like in his opinion.

Food for thought?

Tom
 
Re: we could try active power supply regulation

Fred Dieckmann said:
Post 1183


" The +/- 27 volt suppy is regulated to +/- 20 volts by two LT1084 regulators with 47uF Black Gates at the regulators inputs and 100uF Black Gate at the regulators outputs. Two additional 100uF Black Gate caps are mounted next to each LM1875 for high frequency decoupling."

Your right, it is a good idea........ Glad to see the French backing the US on this one.
when the idea is good, we back you ;)





I assembled my 2nd channel this evening
with even more cable mess than the 1st channel :D

and I had more hum than music :bawling:
so I rewired everything, but with a category 5e cable, for everything except output, so the chip only has a big cat 5e cable, and a speaker cable connected to it

and now it only has a very very little hum




but now, the 1st channel has more hum than before :cannotbe:
 
Re: regulated supply sound?

45guy said:


That being said, I ended up building him 2 power supplies, one that went with the first amp and another so he could take a crack at building a 'clone. Acording to him, they sounded very different (on the same amp module, he has not had success with his attempt) even though they used the same Nichicon caps. The voltages were different and I used no regulators on either. The salvaged 200VA Sony transformer with 2*20 volt taps sounds better/more tube-like in his opinion.

Food for thought?

Tom

That's what I'm repeating all the time. To make this amp sound really good, the power supply is of greatest importance. I found only one type of diodes that sound the best with this amp, and trying different transformers also bring mixed results. The cable from PS to the amp is also important. When everything is chosen properly, the amp produces real magic.
 
Re: Re: we could try active power supply regulation

Hi!
Bricolo said:

and I had more hum than music :bawling:
so I rewired everything, but with a category 5e cable, for everything except output, so the chip only has a big cat 5e cable, and a speaker cable connected to it

and now it only has a very very little hum

but now, the 1st channel has more hum than before :cannotbe:

After my GC experiences from the last days I have to say that I always had hum when I left out the bypassed resistor to ground at the non-inverting input. Resistor value can be calculated like this:
R_bias = (R1*Rfeedback)/(R1+Rfeedback), where R1 = the resistor from input to inverting input IC.
I took a 9k53 Welwy RC55 resistor, which closely matches my optimal value (R1 = 10k, Rfeedback = 221k), bypassed with an Epcos 0.1 uF MKP (will try Wima FKP soon).

Try it, especially if both IC's are connected to same star ground.

Also maybe try to put the connection from the 1000uF caps to ground a little bit away from the signal connections, and connect case to ground (for example with 100 ohm resistor, maybe bypassed).

Hope this helps :nod: ,

bye, Arndt
 
Re: Re: Re: we could try active power supply regulation

Cradle22 said:
Hi!


After my GC experiences from the last days I have to say that I always had hum when I left out the bypassed resistor to ground at the non-inverting input. Resistor value can be calculated like this:
R_bias = (R1*Rfeedback)/(R1+Rfeedback), where R1 = the resistor from input to inverting input IC.
I took a 9k53 Welwy RC55 resistor, which closely matches my optimal value (R1 = 10k, Rfeedback = 221k), bypassed with an Epcos 0.1 uF MKP (will try Wima FKP soon).

Try it, especially if both IC's are connected to same star ground.

Also maybe try to put the connection from the 1000uF caps to ground a little bit away from the signal connections, and connect case to ground (for example with 100 ohm resistor, maybe bypassed).

Hope this helps :nod: ,

bye, Arndt



what case?
http://www.diyaudio.com/forums/showthread.php?postid=143091#post143091

:D
 
Hi Peter and all,

Peter, I would wholeheartedly agree with you about the power supply and associated parts. Minimalist designs using quality parts are my thing. I come from a tube amp background and have never really played with an actively regulated B+ supply. Simple filters and chokes and caps have worked best for me, with a brief exploration of tube type voltage regulators in a linestage. It did not sound good to my ears. A much simpler supply did wonders though. Maybe that is some part of the "tube sound" that many love...along with the distortion products and other tube attributes. And now along come these chips and as much as I love my SET amps, I love these little chips even more due to the simplicity, cost, and, of most importance, sound.

Has anyone alse tried regulating the supply? I would be interested to hear if the results are similar. My theory is that the as the chip needs more current, instead of drawing from a bank of storage caps, the required current comes from an oversized (?) transformer. I don't know how this might be possible with regard to low frequencies since we are operating at 60Hz, but that may show my ignorance on the topic. I do know that a more than a few tube builders advocate small caps and many chokes.

Perhaps someone can shed more light on this for me.

Thanks,
Tom-who is now sourcing some R cores!
 
45guy said:


Tom-who is now sourcing some R cores!

Let me know if you find something. I'm in a process of developing monoblocks and would also like to use some R core transformers. Here's one link I found: http://www.custommag.com/CMIHOME.HTG/rcore1.htm

I'm not sure if active PS will sound good. I was only changing different rectifiers and the difference was tremendous. So active regulator might bring changes that are not really desired . Anyway, if 1000u cap is enough, what to regulate for?
 
Peter,

I plan on being a competitor of yours and probably should not reveal my potential sources, but I started this audio obsession on the Bottlehead BBS and the sense of community and willingness by all to help each other is what kept me hooked...and learning. It undoubtedly exists here on this site also.

In that spirit, you might check out the following if you haven't already done so-
http://www.kitamura-kiden.co.jp/english/products_e.html
http://www.tortran.com/

Tom
 
I'm not sure if active PS will sound good

And you never will making idle speculation without trying it. The soft recovery diode stuff is very old news. I don't think anybody has even mentioned the really good diodes to use or experimented with snubber circuits or RC filters for the supply. It is very interesting that everyone is only concerned about bass response and ripple with 1000uf caps. I would be more concerned about a 1000uF cap being an adequate high frequency decoupling cap. Also Black Gate caps often don't work well in parallel with other caps. I guess everybody has all this stuff figured out already....... I think I will keep experimenting though.
 
Fred

do you mean to say that IF and or RF will be introduced by the 1000 uF value or that the 1000 uF value will fail to attenuate the higher frequencies. Also, is there anything unique about the BG caps that makes this more of an issue with other caps? Oh and since you brought it up, which do you think would be the best type of diode/bridge to rectify the PS?
 
Re: I'm not sure if active PS will sound good

Fred Dieckmann said:
And you never will making idle speculation without trying it. The soft recovery diode stuff is very old news. I don't think anybody has even mentioned the really good diodes to use or experimented with snubber circuits or RC filters for the supply..

From what I remember, at least pinkmouse was experimenting with snubbers and he didn't like it. I tried a filter on AC line and didn't like it either. I even added additional 1000u BG at the bridge and it sounded worse. I used AC conditioner with the amp and it also made things worse. This amp seems to be in favour of making things simple.

One other thing I found out, is that when you parallel two fuses it improves the sonics a lot.;)
 
1000 uF value will fail to attenuate the higher frequencies
and possibly contribute to amplifier instability depending on the exact capacitor type. The National Semi app notes show 0.1 and 100uF local decoupling for the LM1875. I once built an audio circuit that sounded much diiferent with caps of the same brand, value, and series but had different form factors. One cap type was taller and smaller in diameter than the other type. The caps were both Panasonic HFQ types.


"Is there anything unique about the BG caps that makes this more of an issue with other caps?"

Yes and there are some other caps as well that could have problems. Putting a film cap in parallel with a Black Gate was described to me by one designer as "The Kiss of Death" for sonics. I have had the experience certain electrolytics sounding bad with Black Gates and paralleled Black Gates sounding fine. This is not a given and your milage may vary. I have often heard 0.01 film caps screw up the sound on supply bypassing. It pays to do substantial experimentation on capacitor bypassing especially with the Black Gates.
 
This amp seems to be in favour of making things simple.

I like snubbers on some high speed soft recovery diodes. ALL amplifiers are sensistive to all this stuff. Most line filters sound bad. Putting 0.1 or 0.22uf metalized film caps across fuse holders often helps the sonics. Jeff Rowland was doing this 10 years ago as well as capacitor coupling the AC input to the power transformer.

Like I said...... you guys think you have all this stuff figured out.
 
Fred,

I hope you have not taken my questions the wrong way. I certainly don't think I have it all figured out. I feel that if you ever stop learning, you might as well just roll over and die. I also try not to take the viewpoint that there is only one way to accomplish the same goal, especially if the end result is something that can be judged subjectively.

I am actually interested in what you came up with and if anyone else has employed something similar. Voltage regulation is rarely seen in tube designs, hence my questions. I know I'll be looking through my parts bins for some LT108x tomorrow. There's more than one way to skin a cat!

I have not noticed the lack of bass that many claim, especially when using the LM1875. I also did not find my version to be too hard sounding at all, unless pushed to volumes that are too loud for my ears. I would say that it is full and tube-like without getting at all flabby in the low end, like so many of my tube experiments. That being the case, I still can't explain why it can sound the way it does with so little capacitance to draw from. Part of the mystery of this chip, I guess.

I know Peter has tried MANY variations and has stated that if he took the time to audition every possible combination of topologies and parts, his product would never make it to market. Time is my worst enemy too...anyone with kids knows what I mean.

Keep up the good work,
Tom
 
Fred,
We know we are not trying to reinvent the wheel, just trying to see for ourselves how different types of wheels ride on the wagon. Nobody said anything about having it all figured out. As you were expressing before concerning BG caps, its the combination of parts that can make the difference. Are you trying to move this thread in a different direction? If you have an idea, out with it, Man. :) :) :) You seem to favor a regulated supply. What are these really good diodes. Are they something we should try? Give up the goods! :nod:
 
Re: This amp seems to be in favour of making things simple.

Fred Dieckmann said:


Like I said...... you guys think you have all this stuff figured out.

I didn't say that and I don't think anybody will ever have it figured out completely. New discoveries happening every day.;)

BTW, I don't even have it figured out, what I'll be doing tomorrow.;)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.