[linrad] Re: PCM2900 sound chip in Linrad
Hi David, I have not yet found how to set up Linrad to compensate for one channel of a sound card having a delay. The PCM2900 chip seems to have this problem. A delay difference between I and Q will cause a phase error that grows linearly with frequency. In Linrad there is a calibration procedure that will compensate for an arbitrary variation with frequency of both phase and amplitude errors. The reason is that anti-alias filters (such as those used in the WSE units) would be extremely difficult to match between the channels. The calibration should take care of the delay so there is no feature in Linrad to treat it as a special case. 73 Leif / SM5BSZ # This message is sent to you because you are subscribed to the mailing list linrad@antennspecialisten.se. To unsubscribe, E-mail to: [EMAIL PROTECTED] To switch to the DIGEST mode, E-mail to [EMAIL PROTECTED] To switch to the INDEX mode, E-mail to [EMAIL PROTECTED] Send administrative queries to [EMAIL PROTECTED]
[linrad] Re: PCM2900 sound chip in Linrad
Leif, The PCM2900 has a design defect so that one channel is a clock tick later on the usb line to the computer. Sorry I didn't make it clear. I have played with small corrections in amplitude and phase. Actually in Winrad, I did the 1 clock shift and the fine adjustments, and never got the image at negative frequency to be less than 15 db lower than the main signal. David Underwood Leif Asbrink wrote: Hi David, I have not yet found how to set up Linrad to compensate for one channel of a sound card having a delay. The PCM2900 chip seems to have this problem. A delay difference between I and Q will cause a phase error that grows linearly with frequency. In Linrad there is a calibration procedure that will compensate for an arbitrary variation with frequency of both phase and amplitude errors. The reason is that anti-alias filters (such as those used in the WSE units) would be extremely difficult to match between the channels. The calibration should take care of the delay so there is no feature in Linrad to treat it as a special case. 73 Leif / SM5BSZ # This message is sent to you because you are subscribed to the mailing list linrad@antennspecialisten.se. To unsubscribe, E-mail to: [EMAIL PROTECTED] To switch to the DIGEST mode, E-mail to [EMAIL PROTECTED] To switch to the INDEX mode, E-mail to [EMAIL PROTECTED] Send administrative queries to [EMAIL PROTECTED] # This message is sent to you because you are subscribed to the mailing list linrad@antennspecialisten.se. To unsubscribe, E-mail to: [EMAIL PROTECTED] To switch to the DIGEST mode, E-mail to [EMAIL PROTECTED] To switch to the INDEX mode, E-mail to [EMAIL PROTECTED] Send administrative queries to [EMAIL PROTECTED]
[linrad] Re: PCM2900 sound chip in Linrad
Hi David,
The PCM2900 has a design defect so that one channel is a clock
tick later on the usb line to the computer.
Sorry I didn't make it clear. I have played with small corrections in
amplitude and phase.
Actually in Winrad, I did the 1 clock shift and the fine adjustments,
and never got the image at negative frequency to be less than 15 db
lower than the main signal.
Have you tried the Linrad calibration procedure? I think it should
take care of this problem - but it will not work at the
higher frequencies. At the Nyquist frequency one clock tick is
180 degrees:-(
Linrad should work and give good balance for the central region, but
at half the Nyquist frequency the phase error will be 90 degrees
and then the phase relation will be 0 or 180 degrees for upper and lower
sideband respectively and I do not know whether the Linrad routine can
handle that. If you care to give it a try I would be interested to
know over how many percent of the passband you can get good image
suppression.
In Linrad you can add a few lines in the code to shift the samples
one position for one channel. For Linux, the read is in
lsetad.c, the function thread_rx_adinput(void) and you should
add something to shift samples in the buffer rxin_isho.
if(ui.ad_device_no 256)
{
read(rx_audio_in, rxin_isho, ad_read_bytes);
}
In Windows, the read is in wsetad.c, the function
winthread_rx_adinput(PVOID arg) and you should also here manipulate
rxin_isho
if(ui.ad_device_no 256)
{
memcpy(rxin_isho,c1,ad_read_bytes);
rxin_isho is a short int array (16 bits) and it contains
ad_read_bytes/2 16 bit words. If your soundcard giver 24 bits,
two words in rxin_isho are used for each sample.
Save the last sample in a local variable, then shift all samples of one
channel one position, fetch the last sample of the previous read
from a global variable and finally copy the new last sample into
the global variable.
73
Leif / SM5BSZ
#
This message is sent to you because you are subscribed to
the mailing list linrad@antennspecialisten.se.
To unsubscribe, E-mail to: [EMAIL PROTECTED]
To switch to the DIGEST mode, E-mail to [EMAIL PROTECTED]
To switch to the INDEX mode, E-mail to [EMAIL PROTECTED]
Send administrative queries to [EMAIL PROTECTED]
