Odd distortion problem

[soundNERD]

I built an amp using an OPA544 and am experiencing a strange problem with it.

The bass is fine at all volumes, but the treble is highly ditorted at low volume, and by the time it gets better, the amp is clipping.

What's going on here and how do I fix it?

[Netlist]

A little schematic would be a great help.

/Hugo

[soundNERD]

Sorry, I built it using the schematic on the first page of the LM3875 datasheet.

http://www.national.com/ds/LM/LM3875.pdf

[Netlist]

Perhaps you need a few more components like on page 5 of the pdf: Single supply application circuit.

/Hugo

[soundNERD]

I'm using a split supply, though, so I'd have to use the split supply circuit.

[carlmart]

Originally posted by soundNERD
Sorry, I built it using the schematic on the first page of the LM3875 datasheet.

Perhaps you shouldn't use that circuit. Why don't you try the one on page 6 of the OPA544 datasheet?

The gain bandwidth of this chip is smaller than the 3875, so perhaps you should use less gain. All OPA amp chips seem to be like this. You can't use the same 3875/3886 circuit on the 549 either.

Why don't you try a smaller resistor instead of 20K? Or put a 20K multiturn trimpot there and go down slowly. If it works you can then replace it with a fixed resistor.

You need to have a blocking cap in the feedback to do this, or at least watch for DC changes. In fact DC should go down too.


Carlos

[Mad_K]

I have experienced the same problem with the opa541, and the problem is indeed the limited gain of the opa chips. The open loop gain is infact even lower than the graphs in the datasheet when loaded with 4-8 ohm speakers...

In my experience the opa chips should not be used with more than 3,3x gain (10dB).

[digi01]

Originally posted by carlmart


Perhaps you shouldn't use that circuit. Why don't you try the one on page 6 of the OPA544 datasheet?

The gain bandwidth of this chip is smaller than the 3875, so perhaps you should use less gain. All OPA amp chips seem to be like this. You can't use the same 3875/3886 circuit on the 549 either.

Carlos

sure,the opa544 is only a huge current op amp,it is not a power amp.less gain should be fine
here is my opa548 amp with input buffer,the 548 work on unity gain stage.
http://www.diyaudio.com/forums/showt...834#post329834

digi

[soundNERD]

Thats exactly why then.

I used a 51k feedback resistor.

It must be way too much.

So I should probably change it to something less than 10k?

[soundNERD]

Thinking about this issue, does it apply to the 549 also? Because I have built a bridged amp using two of them and 470k feedback resistors. I drive a sub with it so I wouldn't pick up HF distortion.

Since I'm not even sending a signal to distort through it, should I even bother trying to fix it? Especially because I lose too much gain without a large value resistor.