Low output on audio slicekit

Technical discussions related to any XMOS development kit or reference design. Eg XK-1A, sliceKIT, etc.
User avatar
pstnotpd
XCore Addict
Posts: 161
Joined: Sun Jun 12, 2011 11:47 am

Low output on audio slicekit

Post by pstnotpd »

Hi all,

I've been testing the audio slicekit demos.
Although functionally some demo's seem to work, the audio output to headphones is very low and mic input does not appear to be picked up at all. I only got results on connecting the audio output of my laptop to the inputs.

The long delay and long reverb give horrible screeches.

Is this normal? I cannot find any reference to the expected input/output levels.


User avatar
lilltroll
XCore Expert
Posts: 956
Joined: Fri Dec 11, 2009 3:53 am
Location: Sweden, Eskilstuna

Post by lilltroll »

Hi

I havn't looked at the schematics for the audio-slice, but if it uses the same CODEC/output as the USB Audio reference design, the output is a line-out intended for > 1 kOhm (Small headphone is often 32 ohms), meaning that you would get a distorted and weak sound using headphones.
The input is a line-in without a microphone preamplifier and without a PHANTOM or a electret J-FET voltage to bias the microphone, meaning that connecting a microphone would give a very weak signal and if you use an electret microphone it would be extreamly distorted.

A laptop has an output that can drive 16 or 32 ohms, and can supply a microphone bias voltage on the input including analog pregain.
User avatar
pstnotpd
XCore Addict
Posts: 161
Joined: Sun Jun 12, 2011 11:47 am

Post by pstnotpd »

lilltroll wrote:but if it uses the same CODEC/output as the USB Audio reference design, ....
Just compared the schematics. It indeed looks like we're dealing with the came Cirrus Logic CS4270 codec and the same input/output circuitry. The multichannel reference design uses a different codec and does apparently feature a dedicated headphone amplifier, which the slice doesn't have.

The quick start guide suggests headphones and a mic to be used. I'd urge xmos to change this as it doesn't seem to work as one might expect.

Unless I'm wrong of course ;o)

But I wonder if this also explains the "banshee shriek" I get when using the long delay and reverb examples.......