Do Chip Amps (LM3876T) Require a Lot of Burn-In Time?

[Steve M]

Just built my first kit chip amp using the LM3876T (with the mute function bypassed), its a very nice sounding amplifier!

See pic & comments at: http://forum.audiogon.com/cgi-bin/fr...146370182&view

My question is how many hours of burn-in is required before the chip really frees up and sounds at its best? I ask this because I have a friend's identical LM 3876T chip amp here which seems to sound better, its more holistic and open sounding ( his is probably 5-6 years old and well run-in).

Thanks in advance...

Steve M.

[AndrewT]

Originally posted by Steve M
how many hours of burn-in is required before the chip really frees up and sounds at its best?

none.
Some of the caps may require time to reform the dielectric and maybe some better types could be swapped in. Film generally sound better but are more expensive.
There may be some PSU noise getting past the PSRR of the chipamp.
It may be verging on being unstable (insufficient phase margin).

[Steve M]

* "There may be some PSU noise getting past the PSRR of the chipamp. It may be verging on being unstable (insufficient phase margin)." *

Thanks Andrew, please explain further what you mean and perhaps what to look out for and how to resolve these issues ...and what do you mean by insufficient phase margin??

If I had to describe more fully what the difference is between my new amp and my friend's older one is that, his seems to give better separation of the performers on the stage; has better front-to-back imaging and possibly sounds softer and more natural ), it generally sounds more open hence the holistic comment above.

Other than a different chassis layout we are also using different potentiometers on the input, his is a cheap RS Components 100K log single gang Bourns pot ($9.28) while mine is a more expensive stereo 100K log Alps Blue (@$50).


Regards,

Steve M.

[cabbagerat]

Are you sure you used a suitable value for Rm (see the datasheet)? The 3876 exhibits really strange non-linear distortion when partially muted. For +-25V supplies Rm must be 22k or less.

If I had to describe more fully what the difference is between my new amp and my friend's older one is that, his seems to give better separation of the performers on the stage; has better front-to-back imaging and possibly sounds softer and more natural ), it generally sounds more open hence the holistic comment above.

How did you do volume matching during your listening test? Even 1dB of mismatch can lead to one amp sounding "better" than the other.

[AndrewT]

Hi,
read up on amplifier and opamp stability.
Start with Walt Jung's web site and try following some of his "audio" articles. Then download the app notes of Ti, National etc and try Walt's opamp handbook, it gets a bit deep but everything (well almost) is in there.

You will some equipment to fully check out your chipamp.

[Steve M]

Thanks Cab, we are both not using the mute function. The kit amps we've both got suggests a simple wire jumper to bypass the mute circuit.

There is a slight anomaly between his board and mine near the mute circuit. There is a 22uF electrolytic cap in this position which on his board appears to be connected a*se-about(wrong way around)! The negative of the cap on his board is in the positive position on my board (my board is clearly marked + where positive should be). The reversed position of this cap is the ONLY difference between his board and mine.

With level matching and A-B testing, just using my long experience and 'golden ears' to make such judgments ...

Regards,

Steve.:

[markiemrboo]

Originally posted by Steve M
Thanks Cab, we are both not using the mute function. The kit amps we've both got suggests a simple wire jumper to bypass the mute circuit.

There is a slight anomaly between his board and mine near the mute circuit. There is a 22uF electrolytic cap in this position which on his board appears to be connected a*se-about(wrong way around)! The negative of the cap on his board is in the positive position on my board (my board is clearly marked + where positive should be). The reversed position of this cap is the ONLY difference between his board and mine.

With level matching and A-B testing, just using my long experience and 'golden ears' to make such judgments ...

Regards,

Steve.:

Do you two also have identical speakers in an identical size / shape room? If not, take your amp to his set up. I'm sure it will sound the same then.

[Steve M]

Mark: I've got his amp here in my room and have been directly A-Bing the two ...all components and cables kept the same.

The only things I can think of causing the difference in sound are:
- burn in time for the LM3876 chip (or maybe for the whole amp?)
- slightly different chassis layout
- the reversed (+ve) position of the 22uF cap near the mute
- the cheap Bourns pot might be outperforming the Alps Blue

Who knows... ????

Regards,

Steve.

[markiemrboo]

I see.

It was previously mentioned that if the chip isn't properly out of mute then it will exhibit some strange distortion. I am not sure if having the mute cap the wrong way would prevent it from going in to mute properly.

Try putting the mute cap the right way around on his. You should probably do regardless, if you're sure it's the wrong way around!

[cabbagerat]

Originally posted by Steve M
Mark: I've got his amp here in my room and have been directly A-Bing the two ...all components and cables kept the same.

The only things I can think of causing the difference in sound are:
- burn in time for the LM3876 chip (or maybe for the whole amp?)
- slightly different chassis layout
- the reversed (+ve) position of the 22uF cap near the mute
- the cheap Bourns pot might be outperforming the Alps Blue

- The levels are not properly matched. Research has shown that even small differences in level can make two identical amps sound very different. The levels will need not be obviously different.

Try level matching within 1dB with a good meter, and switching to a double blind testing protocol.

Originally posted by markiemrboo
I see.
It was previously mentioned that if the chip isn't properly out of mute then it will exhibit some strange distortion. I am not sure if having the mute cap the wrong way would prevent it from going in to mute properly.

It could be. That mute cap should be in parallel with a resistor of between 22k and 10k. The function of the cap is to reduce pop on turnon and turnoff, so a dead electrolytic shouldn't effect the normal sound of the amp, unless it were shorting.