Kindly suggest Hi Fi Sound chip


Hi Guys,

I have seen couple of post regarding this topic but got confused. Sorry i might be repeating the question which someone would have asked before.

Based on all your experience can someone suggest the so far best Hi-Mid (music) chip amplifier. The goal is produce good quality vocal and music.
I heard a lot about LM3875, 3886 1875 and TDA7294. Please suggest from your personal experience which of these or other chip would give me the best results in terms of clean sound quality(Hi-MiD).

Regards
Bhupender
 
Thanks Gootee and Arty. I wanted to buld the home audio system and i thing don't need very high power (best would be RMS - 40W - 80W). Have you guys ever build the surround sound setup, my main aim is build the amp and test the sound quality for home theater system (4:1). I know just amp is not enough, i need good speakers(woofer-mid-tweeters) . Need surround sound decoder kit as well.

If you guys have any link or suggestion to build the good quality setup then please share.Once again thank you very much for reply and suggestions.
 
LM3886 and TDA7293 are the two i can say sound good, the TDA7293 beeing the better of the two, at the price of beeing much more fragile.

The LM1875 is awful, i built an amp with that and threw it out the next day.
Very true on the lm1875t part,if you seen my thread
I think its the worst amp ever,my tda 2003 beats it in everithing including power and lower distorsion
 
Last edited:

Mihkus

Member
2012-07-22 8:01 pm
But lm 1876 sounds super neat compared to 1875 witch seems just to be 1 channel of the 1876...
You really need to put some effort in ur amp before you test it and throw away.
TDA2003 is not good at all, ive tested several HiFi NXP chips and i dont think they sound good.
National all the way.

I would build with LM3886T, why... T package needs 2x smaller heatsink :)
Not 500% sure buw thats how they say...
 

Junm

Member
2008-12-23 5:52 am
Dau
There are lot of factors to consider in good sounding chips..we have individual preferences, one might think a lm1875, 3886, etc ...sounds good but to the others are not. Try one that you have, then if your not satisfied try the others...you can breadboard or use point to point wiring for experimentation then finalize which one..
 
I can't imagine what kind of speakers one needs to tell a difference in sound between a properly designed (i mean not like 80 Hz high pass and not powered by a power supply filtered with 50 μF smoothing capacitors) lm1875 and a lm3886 or 1876 or 4780 or 3885 or whatever. And don't say "i have a pair of watt/puppy 6es" because then you wouldn't be looking at driving them with an amplifier that its connectors cost more than the IC itself. I'm not joking.
 
i here have 2 choices to suggest ... u can go for the LM3875 which have quite detailed midrange and treble .... or if u were on budget then u could also go for LM1875 ... which the sound is more flat compared with the LM3875 .... but works well on my speakers though ... BTW if u need more power u also can configure the LM1875 in BTL ... i tried that method ... if u would like to know just reply to me and i can kindly help u out ...
 
I can't imagine what kind of speakers one needs to tell a difference in sound between a properly designed (i mean not like 80 Hz high pass and not powered by a power supply filtered with 50 μF smoothing capacitors) lm1875 and a lm3886 or 1876 or 4780 or 3885 or whatever. And don't say "i have a pair of watt/puppy 6es" because then you wouldn't be looking at driving them with an amplifier that its connectors cost more than the IC itself. I'm not joking.

If you spend that much on speakers, then you might not have much money left over to spend on amplification!

Until now, I've preferred the mono versus the stereo versions:

1. Figuring a PCB layout to reach the back row of pins on the IC isn't obvious. My PCB process isn't good enough to pass traces between pins.

1. For the heatsink, you have two low-power heat sources instead of one hot spot.

2. You're trying to remove the heat from two amplifiers through a similar junction-to-case and case-to-heatsink thermal resistance.

But, like all things, it depends on your judgement...
 

Sus

Member
2013-02-24 7:38 pm
My opinion is weak because I'm a noobie. I'm in the same boat you are in. This audio swamp is indeed deep. But it is starting to come together for me,finally. I know nothing about how a good sound system should sound. I said good not excellent. Are discretes better? I don't know. So while I'm dragging my feet my first purchase will be from Peter at Audio Sector. His reputation is impeachable. Every body likes him and when you buy a kit from him you will know it has good parts and not fakes or counterfeit.

I may buy his LM 470 kit. $84.00.. Maybe I will buy the LM3875 precision kit. That seems to be the Gold standard For either one you need a transformer. . So now you have a reference for what a good amp should sound like. Without a reference you can't learn or compare. I assume you have decent speakers. If not, that the next swamp on the right. I don't think I will get a whole not of don't do this here. If I do I will listen.

My AR 2's are very old and need to be rebuilt or I need to get new speakers. As far as I can determine one of the best things going is MJ Kings new compact BLH design Very inexpensive with a Fostex speaker if you have the tools to build the cabinet. . Being a new design there are not many that have one. Maybe only Dr. Boar who did some heavy duty tests on it. I plan to build one in the next month or so. Back log you know slows one down. I may also rebuild the AR's and then that will be good comp.

After this learning experience I should be able to walk on my own with out help.

Good luck.
 

picbuck

Member
2008-05-31 12:50 pm
The amp itself is a chip and 4 or 5 resistors, maybe a capacitor. Do you really need a kit for that?

<< best Hi-Mid (music) chip amplifier. The goal is produce good quality vocal and music >>

Possibly the real secret is conservative design. Don't believe that a 50 watt chip is really good for 50 watts. No matter what the engineers say, the marketing department is always in there bending the numbers.

The schematics in the data sheets are always good, because they always come from the engineers. But look at the graphs in the data sheet and figure on a wattage output 25% to 50% less than what's advertised. When you don't know what a graph means just ignore it, because using the data sheet schematic you can't go far wrong.

Then use a bigger power supply (in watts/VA) than you ever dreamed you could possibly need. This means don't stint on the power transformer. The value of toroids is questionable, in my opinion, but you gotta have a big enough (in watts/VA) transformer.

Of course...all this poring over graphs and specs can be a confusing and frustrating process. But the knowledge stays with you for next time.
 
Last edited:
Then use a bigger power supply (in watts/VA) than you ever dreamed you could possibly need. This means don't stint on the power transformer. The value of toroids is questionable, in my opinion, but you gotta have a big enough (in watts/VA) transformer.

the only thing he will manage with a huge transformer and big smoothing capacitance is trip the mains fuse every time he turns on the amplifier.

these chipamps have pretty good CMRR so you don't gain anything from a huge power supply
 

picbuck

Member
2008-05-31 12:50 pm
As the saying goes, watts is watts. A 70 watt power supply will no more blow a fuse or circuit breaker than would a 70 watt light bulb. Nor would a 100 watt supply, which is no more load on a circuit than two 50 watt light bulbs burning at the same time.

More to the point of this thread, I guess "a bigger power supply (in watts/VA) than you ever dreamed you could possibly need" was a pretty silly way to describe a power supply, at least in this case.

If you're building a 50 watt amp, then use a 70 watt power supply, bare minimum. That's just to give an eyeball kind of number, but there's no such thing as too big a power supply.

The primary reason for this is demonstrated by the presence of the amplifier's heat sink. Some one-third to one-half of the power an amplifier draws is simply dissipated as waste heat. But even though it's wasted, this power still has to come from somewhere, and that somewhere the power supply. This means that a 50 watt power supply, for instance, cannot provide 50 watts of music power, it can only supply some 30 watts. The rest is wasted as heat. This is simply the present state of the art in audio amplification.

A second important reason is peaks, which happen constantly in music. All commercial recordings are heavily [criminally?] compressed, but there are still heavy peak loads put on the power supply. If the power supply is not adequate then voltage drops, and the result is distortion.

I sometimes wonder whether people saying this or that chip is no good are simply using an insufficient power supply.