RX phase & ampl. adjustment problem

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

RX phase & ampl. adjustment problem

magun
Hi

I am running quisk 3.6.1 under XP.

I am able to suppress the image and save the appropriate values for phase and amplitude adjustment. I did this for 1 frequency only. The value tuple is shown correctly in quisk_init.pkl However, when starting quisk again the image is strong as before, The amplitude and phase values, shown under 'RX Phase,' are the same as those found for maximum image suppression. Restaring quisk again the image is suppressed.

Supression non supreesion of the image alternates with subsequent starts of quisk.

73 de Andreas
Reply | Threaded
Open this post in threaded view
|

Re: RX phase & ampl. adjustment problem

ahlstromjc
Administrator
Hi,

I don't see this behavior in my tests.  First be aware that adjustments are saved for each band.  If you are on a different band the adjustment will be zero.

The adjustments are recorded each time the VFO frequency changes.  Please try starting Quisk at a time when the adjustment fails, and then either press the up/down band button or change band to and back to your band.  Either operation changes the VFO and should result in the correction being used.  That is, you
should be able to enable the correction by changing the VFO without restarting Quisk.

Of course this is a test and not a solution, but will give me more ideas of what might be happening.

Jim
N2ADR
Reply | Threaded
Open this post in threaded view
|

Re: RX phase & ampl. adjustment problem

magun
Hi

I did a the following tests:

- changed the frequency with 'Down/Up' and the band as well. The image does not disappear.

-  compared the .quisk_init.pkl files after subsequently started quisk (for the cases where the image is suppressed and not suppressed). No changes were found between the respective quisk_init.pkl files.

- the suppression of the TX image suppression works fine, i.e. it does not show the behaviour seen in the RX mode.

- when the image appears it can be cancelled with a different phase correction. The 2 values are -0.748 respectively  -0.148 for subsequent starts of quisk. The amplitude corections are nearly the same and close 2 zero (0.00436).

- rebooting XP and starting quisk always results in the suppression of the image

This is very strange. It seems as if information is stored in memory between the start of subsequent quisk executions.

In the quisk_conf.py is a parameter 'bandAmplPhase' that defines the dictionary for corrections? Ehere does it reside?

73 de Andreas

Reply | Threaded
Open this post in threaded view
|

Re: RX phase & ampl. adjustment problem

magun
In reply to this post by ahlstromjc
Hi

more tests with quisk 3.6.2:

previous tests werde on Laptop S20 (Atom270) under XP - the same effect was also seen under Ubuntu 10.04LTS

on desktop the effect disappeared under XP and Ubuntu 10.04LTS

repeated tests under XP after Laptop ran for a few hours: Effect disappeared, i.e. behaviour as expected.

switched laptop of and the effect reappeared.

I will do some further tests. I beleve that there is a hardware problem with the laptop.

73 de Andreas



73 de Andreas
Reply | Threaded
Open this post in threaded view
|

Re: RX phase & ampl. adjustment problem

ahlstromjc
Administrator
In reply to this post by magun
Hi,

I believe that Quisk is using the correction that was saved, but that the correction can become different somehow.  Some sound chips have a one sample delay between right and left that must be fixed in software.  See the channel_delay parameter in quisk_conf_defaults.py.  But that should have made the correction impossible.  Somehow the phase is shifting between runs.

Jim
N2ADR
Reply | Threaded
Open this post in threaded view
|

Re: RX phase & ampl. adjustment problem

magun
Hi

a sample delay can be excluded, but your remark regarding the sound chip gave me the idea that there might be a small change in the the filter characteristics of the analog input.

Anyhow the image suppression given by the TRX hardware is bearable (< -40 db).

Andreas
HBs9EHI