I ran a 20 Hz signal line out (at unity) and back in to a 002. Here's what happened:
www.bplr.com/Temp/Picture%203.png
it went out a square and came back a trapezoid. can somebody tell me why?
D/A/D Conversion weirdness
- bplr
- takin' a dinner break
- Posts: 179
- Joined: Thu Jan 01, 2004 1:49 pm
- Location: Seattle, WA
- Contact:
D/A/D Conversion weirdness
Bipolar Production
I might be remembering this wrong, but it looks like what happens when you send a signal through a board that needs to be re-capped. A 1 kHz square wave will tilt when it goes through a circuit without the benefit of a properly charging and discharging capacitor.
Now, having said that, the fact that it's at 20 Hz says to me that it's not that big a deal. I think you are going to have a loss of bass response whenever you go through a conversion from D to A back to D. Especially if you are using 44.1 or 48kHz. If you get the same slope at 1k then it's time to worry. Even a well-calibrated mid-level tape deck will show that amount of loss at 20 Hz.
Now, having said that, the fact that it's at 20 Hz says to me that it's not that big a deal. I think you are going to have a loss of bass response whenever you go through a conversion from D to A back to D. Especially if you are using 44.1 or 48kHz. If you get the same slope at 1k then it's time to worry. Even a well-calibrated mid-level tape deck will show that amount of loss at 20 Hz.
not to worry, just keep tracking....
- Disasteradio
- pushin' record
- Posts: 278
- Joined: Tue Sep 16, 2003 8:39 pm
- Location: Wellington, New Zealand
- Contact:
- bplr
- takin' a dinner break
- Posts: 179
- Joined: Thu Jan 01, 2004 1:49 pm
- Location: Seattle, WA
- Contact:
i sent the signal straight out and directly back in through the 002 ins and outs. no hpf involved.Disasteradio wrote:Surely that's the wave under a bit of HPF for DAC/ADC? the standard spec for most converters is 20Hz up right? you'd expect some nonlinearity around that filtering point for sure...
Bipolar Production
Ah, but in effect there is a HPF filter from the fact that the 002 ins (and probably outs) are capacitor-coupled, to block DC. Try this: make a new test signal that's like your square wave, only after it jumps from zero to the high value, it just stays high. Now run that through the i/o. You'll get to watch those capacitors discharge, like those tilted tops you're seeing in the red waveform, only that ramp will go all the way to zero... at least, that's my prediction...bplr wrote:i sent the signal straight out and directly back in through the 002 ins and outs. no hpf involved.
Leigh
Who is online
Users browsing this forum: No registered users and 45 guests