Noise, lots of it

[Deltat]

My senior design project is a subwoofer amplifier, using the Tripath TK2350 evaluation board. The input signal first goes through a MAX260 switched-cap filter IC then through a PGA2311 volume IC. The singal is then split and inverted then sent to both channels of the amplifier board bridged. Im having this issue where when the volume level of the PGA2311 is turned up i get a rumble, and its fairly prominant. Im pretty sure the noise is coming from the PGA2311. Could this be a grounding issue since i dont have a seperate ground for the digital and analog supplies on the chip? Or is it maybe that the input impedence into the PGA2311 isnt 600ohms or less? If i want to create an input impedence of 600ohms or less would i put a resister from the output of the filter to the input of the volume IC or would I place a resistor to ground on the input of the volume IC. Hopefully this wasnt too confusing. If it is i can send you a system schematic if that would help.

Heres a picture of my project if you are interested

[ACR]

Deltat, I know this is off topic, but that looks like a pretty nice housing - where did you get it from and how much did you pay (is it aluminium, mild steel?)

[macboy]

Or is it maybe that the input impedence into the PGA2311 isnt 600ohms or less?

That could be a problem, though I don't know if it is causing the "rumble" that you refer to. Still, it wouldn't hurt to correct this. From what I see in the MAX260 datasheet, the device is not capable of adaquately driving the PGA2311. The PGA2311 requires that the circuit driving it has an impedance of 600 ohms or less. Otherwise the signal will be distorted. And of course you can't load the outputs very much.

If i want to create an input impedence of 600ohms or less would i put a resister from the output of the filter to the input of the volume IC or would I place a resistor to ground on the input of the volume IC. Hopefully this wasnt too confusing.

You can't fix this by adding a resistor. You need to buffer the signal. The easiest way is with an opamp follower. This will drive the PGA2311 with a very low (effectively zero) impedance, and will load the filter with a high impedance. This will make both devices much happier. Don't forget that you need DC blocking caps for the MAX260.

Could you describe the "rumble"? Is it a constant hum, does its level follow the level of the signal? Is it like a low frequency distortion? Also, can you post a schematic? The forum allows you to attach gif, png, etc. files to your posts.

[sss]

Originally posted by macboy
You can't fix this by adding a resistor. You need to buffer the signal. The easiest way is with an opamp follower.

i'm with u on that

[Deltat]

The rumble sounds a lot like pink noise.

Kurt

[Deltat]

Ok let me be a bit more specific on why i think its pink noise. The noise is dependent on the volume level of the volume IC on the amp itself, it is unrelated to the amplitude of the input signal, it present with no input signal. As i adjust the crossover from 60Hz to 120Hz I can hear higher frequency noises as the filters cut-off frequency increases.

I tried to add my schematic but its too big. I copied the schematic from cadence into paint and it was 1M. I can email it to anyone who wishes to look it over.

I noticed during testing that if i had a signal from a function generator tied directly to an oscilloscope(no connection to the amp at all) and turn the amp on, the signal gets distorted. I disconnected my low voltage transformer from the power supply board and that went away, I cant figure out why.

Kurt

[SpittinLLama]

Start with just one block at a time to see where the problem occurs. So back up to just the first part and test, then add the next block and test. Then you can trace it down. Try it with out the Tripath amp so that there is no class D going on, maybe a known descent amp even a comercial one to see if it happens. Good luck.

-SL

[Deltat]

Macboy where would i want the caps for the DC blocking? I originally had a high pass RC filter on the output of the MAX260, but it wouldnt work, im guessing impedence issue.

This really sucks, the amp works and thats good enough to graduate. But id really love to work in the audio industry and have an amp that kinda sucks wont help that. I also dont want a $1000 worthless project.

Kurt

[pinkmouse]

Kurt

My suggestion, for what its worth. Put it away in a cupboard for a week, and forget about it. Go see a movie, drink beer, chase girls. All the things you haven't been doing whilst building this amp.

After a week, come back and open it up. The problem will probably be extremely obvious.