What causes IMD?

[gmikol]

I know this isn't specific to chip amps, but I just put together a quickie TDA2050 amp on a breadboard, and it sounds like #!%@!#%. I used RMAA to analyze it quickly, and it looks like the problem is IMD performance. It's about 0.25%. I know that putting it on a PCB will help a lot, but I want to work as many kinks out as possible before I do that. I'm wondering what things I sould be looking at as causes.

I really don't even know where to start. Any help is greatly appreciated.

Thanks--

Greg

[neutron7]

are you using it with split supply or single?

people here couild probably help more if they see the circuit diagram of what you have done.

anyways that is supposed to have much lower distortion than that, it says 0.02 into 8 ohms or so. it could be there is oscillation or if it is single rail, the output cap may be bad. (or a lot of other reasons, noone can tell without your circuit.

[gmikol]

Yeah, I guess you're right. I was more looking for a theoretical explanation than circuit debugging though. I don't have a schematic...I just built it up off the top of my head. I'll try to describe it, though.

I'm trying to build a test-amp for use with the Wallin Jig and Speaker Workshop. So this thing is battery powered (+/-9V). I'm using it in what I call a quasi-unity-gain configuration. The data sheet says that it's not recommended to use a gain <24dB for the TDA2050, so I'm using 110k and 6.8k feedback resistors (116.8/6.8=24.7dB), and I'm using a -24.7dB TL071 input stage with offset trimmed. This way, the voltage across any driver I'm testing is going to be compatible with my soundcard inputs. I've checked the outputs of my soundcard...virtually 0 DC, so the whole thing is DC-coupled for simplicity's sake. It's driving an 8Ohm resistor as load.

The circuit works...it just sounds bad. I'm thinking that the 60Hz tone of the IMD test is causing a wiggle on the supply rails which is getting modulated in (this might explain the bad test results?) due to the relatively high internal impedance of the battery source. When I go back to work on Monday, I'll grab a really stiff PSU we have there and see if that fixes things.

In the meantime, any other ideas as to what might be wrong (and how to fix it) are appreciated.

--Greg

[owen]

Try a single large cap between the rails - that'll show if its the 'supply' causing the problem...


Owen

[Gandalph]

Greetings from Norfolk

I just happened on this thread, and will probably be stating the obvious in my comments below.

Inter Modulation Distortion is normally caused by non linearity in an elemement in the signal path. The classic case of this is mosulation of an RF carrier by an audio signal, or the genreration of an intermediate frequency in a reciever.
Any non linearity in a circuit where two frequencies ar present will tend to generate additional frequencies which are the sum and difference of the two basic frequencies, and to a lesser degree 'N' times these frequencise, where N is an intiger.

This may help, or may just define the obvious !

Richard