TDA2050 better than LM3875 and transformer VA rating

Hi All,

I seem to recall reading somewhere that TDA2050 gave a more refined sound than LM3875 in gainclone implementations. Any views on this?

Also, I have 2x300 VA transforners (20V secondary) which I intend using for my LM3875 implementation. Is it corect to say that teh higher the VA rating of the transformer the better? I have seen that others are typically using only 120-225 VA transformers.

Ryan
 
Ryan

Within certain limits higher VA rating may be beneficial for the sound. I have, however, tried a very large (1000VA) transformer once and the sound was remarkably flat and unexciting. How much you can blame on the core size, wire quality, core type etc is not very clear as these experiments tend to get very expensive. I am still determined to try high quality, very thin J-laminations with distributed gap which a local winder can get. Won't be cheap :)

cheers
peter
 
Dr.H said:
Hi All,
Also, I have 2x300 VA transforners (20V secondary) which I intend using for my LM3875 implementation. Is it corect to say that teh higher the VA rating of the transformer the better? I have seen that others are typically using only 120-225 VA transformers.

Ryan

160~200 VA is good for one channel.
I use one single 384VA toroid for the two channels.
Your 20V, 300VA transformer will do fine for two channels.
 
Dr.H said:
Hi All,

I seem to recall reading somewhere that TDA2050 gave a more refined sound than LM3875 in gainclone implementations. Any views on this?

I've tried both, and I can tell you: Forget the TDA2050!
It simply stinks! Giving a very coloured sound, and is hard to get stable. Definately not suited for a minimal GC.
 
The TDA2050 does sound much better than the LMs but needs battery or regulated supply because of it's low PSRR and hence should not be implemented with a minimal style power supply a la Gaincard.

Other TDA chips e.g. TDA7294 have more complex circuitry (such as mute) and probably may not sound as good.
 
Alex S said:
The TDA2050 does sound much better than the LMs but needs battery or regulated supply because of it's low PSRR and hence should not be implemented with a minimal style power supply a la Gaincard.

I'm glad you said that. I was begining to doubt my own experience. someone asked me to make an amp that would run from a 12V battery and because of the low voltage I chose tda2050. two per channel balance working. everyone that heard that amp really liked the sound and it was that, and thorsten, that got me back into the chip amp scene.

cheers

mike
 
TDA2050 GOOD

Hi All,

I seem to recall reading somewhere that TDA2050 gave a more refined sound than LM3875 in gainclone implementations. Any views on this?

Also, I have 2x300 VA transforners (20V secondary) which I intend using for my LM3875 implementation. Is it corect to say that teh higher the VA rating of the transformer the better? I have seen that others are typically using only 120-225 VA transformers.

Ryan

I ve friend that use TDA2050 for years beside almost all chipamp,
he said the clue is to find the right resistor by using variable resistor first