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.
OK Peter I undestand now what you meant since you want to get on with other projects.Time is also a valuable commodity which we must maximize.
However even though I can believe you have made or "voiced" a great sounding amp I still think the priority should have been to select the best -sounding op-amp first (there aren't that many contenders) and build on from there.
 
Well I have only tried OPA 549 and Lm3875.The other main contenders would probably be TDA 7294 and LM 3886 to keep it simple.
It could be my system or set-up but when I repeated an A/B
this weekend I found the 3875 a bit too "ringy" on instruments such as xylophone and piano while on horns were too "blarey".
I found the 549 more neutral and correct. As you know it's very hard to describe tonality differences in words.

By the way the BG super-E configuration has improved considerably now to the point where I am about to take back my comments that it doesn't make an essential difference. In fact now it has come to the stage where it seems to make a fundamental difference to musical enjoyment. What I mean by fundamental difference is that the component in question does not achieve just a more pleasing balance in the sound but it manages to convey the impression that you are actually hearing a better e.g. remastered type of recording where you can hear not only more detail but more of the performance too.I hope this makes sense.
 
Peter Daniel said:
What values BG are you using?

680uf 35V BGNx. For the Super E you parallel two for each rail which makes it 1360uf per rail on each chip.You connect the parallel caps in opposite "polarity" to each other.I have some 1 uf BGn 's which I plan to use as bypass in super -e config.to above.I found that bypassing my previous gc 1000uf panasonic caps with 2,2uf film polycarbonate and 0,1uf sounded noticeably better and had more bass believe it or not. We shall see if the Bgs will sound better bypassed or not.I will wait a couple more days to see whether the breaking in process has evened out.
 
I built my first gainclone last night using 3875. It worked fine driving a single test driver but when I hook it to my Ariel the current draw on the variac goes way up to about 2 amps.

Is this likely an instability oscillation?

would I reduce the feedback resistor to get more feedback and stabalize this?
 
Fritz said:
Yes it draws the current at idle or with an input signal. Its almost as if I was shorting something when I hook the speaker up, th ecurrent steps right up and there is audible hum from the speaker.

Hmmmm. Not sure if it's an oscillation problem or if something's wired up wrong. The hum is a bit unusual if it's an oscillation problem, but then if you're using the small value filter caps, oscillation can drain them way down.

Which schematic did you build it from and what were the part values?

Whats a zobel?

It's basically an impedance compensation network tied to the output to help isolate the amp from reactive loads.

Let's see if we can track down the actual problem first.

se
 
Ok,

another gainclone is working :) , and the sound is just find. I made it with lm3875 (samples ;) ) and some generic parts.
I use a 500VA 2x25V toroid.

Now here is my question:

I mounted the amps on a tiny heatsink but it doesn't get warm at all (aspecially compared to my 1 channel aleph5).
So is this normal ?

The music is coming out of my speakers just fine (about 20 mV offset so that's not to bad either)

Ralph
 
Covering all the bases.

Fritz,

I will assume that you have constructed the amp using point-to-point method. Even so, make sure you have read the sections of the data sheet for the LM3875 that address supply bypassing, lead inductance, and especially layout, ground loops, and stability. The LM3886 and LM1875 data sheets also have very good information albeit similar, however, the sheets provide other valuble information for these IC amplifiers.

When I was experimenting with the LM1875 I initially had a ~375 kHz oscillation with only power applied. After reading the documentation and careful mearsurement it was determined that the signal ground path shared too much of the power ground path. Answer--re-route the board and run all grounds seperately.

You should not have to use a Zobel with any of these IC's. I have found them very robust even when driving more difficult 4-Ohm loads.

Later,
 
Re: Free Lunch!

jam said:
If a Zobel is required you owe me lunch.;)

Now now, don't go jumping the gun. :)

I didn't say a Zobel WASN'T required. I'm just trying to root out the actual problem so that the appropriate solution can be applied. At this point, the information is a bit too ambiguous.

It also could be that the ic needs to be bypassed with local caps at the B+ and B- pins.

Yes. Could be. Could ne not. Which is why I'm holding off on recommending a specific solution.

Then again if it is a wiring problem I owe you lunch. Deal?

Ok, if it's a wiring problem, you owe me lunch. Heads I win, tails you loose. :)

se
 
Getting Hot Too ?.

Fritz said:
Yes it draws the current at idle or with an input signal. Its almost as if I was shorting something when I hook the speaker up, th ecurrent steps right up and there is audible hum from the speaker.
Whats a zobel?

"Hmmmm. Not sure if it's an oscillation problem or if something's wired up wrong. The hum is a bit unusual if it's an oscillation problem, but then if you're using the small value filter caps, oscillation can drain them way down."

This is a classic symptom of ultrasonic oscillating condition, and due to inadequate supply bypassing, or more likely incorrect earthing arrangements.

Steve, you owe us all a lunch break.

Eric.
 
Re: Diodes

vladimir said:


What I like the best is ultra fast recovery BYV28-200 and in the single bridge (4 diodes only for whole amp).
Bigger BYV29-200 are not as good as 28-200.


I tried today a single bridge using MUR860. However, two separate bridges, using same diodes, sounded way better in my setup.

Also I did some tests comparing Holco and Riken in feedback position (220k) and my choice is Riken now. Holco, indeed sounds a bit dry and Riken offers more depth and highs. The difference however, wasn't that big.
 
rwagter said:
I mounted the amps on a tiny heatsink but it doesn't get warm at all (aspecially compared to my 1 channel aleph5).
So is this normal ?
Ralph


It's a class AB amp, so that's normal. My prototype a few weeeks ago didn't warm up either and I had it mounted on a small heatsink too. After a couple of hours it got nicely warm but nothing to be concerned about.
Some (like Peter Daniël) use the case or a part of the amplifiers housing as a heatsink and it seems to work fine.
I am planning to do that too, first I want to finish a couple of speakers.

Also check out this site of your fellow Dutchman

http://www.platenspeler.com/

Look for 'Geenkloon'. He uses small Conrad Alu cases without any extra heatsinking.

Marc
 
Re: Diodes

vladimir said:
Hi I'm playng with gain-cloning for one year now.

Ordinary bridge doesn't sound very good for me.

I'm using 2x1000uF Fa (Fc) Panasonic. I've tried also some ordinary 'noname' caps as Hitano or Jamicon: tends to be forward sounding (maybe a bit agresive). Coupling caps are 1uF polycarbon (better than polypro wima or polyester wima).

I'v tried doble schottky bridge: using SR390, SB5100, and paralleled
MBR20100. The sound have less presence, this may be an advantage with forward sounding speakers.

What I like the best is ultra fast recovery BYV28-200 and in the single bridge (4 diodes only for whole amp).
Bigger BYV29-200 are not as good as 28-200.

I can also confirm that diodes have the very long break-in time (~1000 hours).

v.
The FA Panasonic are the same as FC with longer lifetime or just different series?
Giorgio
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.