[hpsdr] Diversity - OpenHPSDR mRX PS v3.3.6

[Barry N1EU]

Thanks Warren.  Yes, I think you fixed part of it, for sure.  But the
issue still lingers in some form, thanks to Windows.

73, Barry N1EU

On Tue, Dec 8, 2015 at 11:59 AM, Warren C. Pratt <warren at wpratt.com> wrote:
> ***** High Performance Software Defined Radio Discussion List *****
>
> Hi Barry,
>
> There was a bug sometime ago where RX1 and RX2 were not always getting set
> to the same DSP buffer size ... that would of course cause this kind of
> effect due to the differences in delay.  However, I recall that we fixed
> that, AND, that would not explain the timing difference changing after a
> transmission.
>
> That all said, there is nothing in the software that would explicitly
> guarantee that RX1 and RX2 would produce their output at exactly the same
> time.  Their processing is independent and, to an extent, it depends upon
> when Windows wakes up their processing threads.  I think that may be what
> you're hearing.
>
> This is of course very different for "pure Diversity" where the buffers of
> samples are mixed on a single thread and the result then passed to the RX1
> receiver channel in the DSP.
>
> 73,
> Warren  NR0V
>
>
> On 12/8/2015 7:49 AM, Barry N1EU wrote:
>>
>> ***** High Performance Software Defined Radio Discussion List *****
>>
>> I use the Stereo Diversity function, not pure Diversity, and I often
>> observe a sudden anomalous introduction of a phase shift between the
>> receivers which is usually (not always) subsequently corrected after a
>> xmsn.  I've been observing this in all versions of OpenHPSDR mRX PS
>> since v3.2.23 added the Stereo Diversity "SD" feature.
>>
>> Barry N1EU
>>
>> On Tue, Dec 8, 2015 at 10:42 AM, Joe Martin <k5so at k5so.com> wrote:
>>>
>>> ***** High Performance Software Defined Radio Discussion List *****
>>>
>>> Hi Jim,
>>>
>>> Very interesting report.  In my initial investigation I see that with
>>> either HPSDR PowerSDR mRX PS v3.3.6 or HPSDR PowerSDR mRX PS v3.2.27
>>> (probably others as well) a random phase shift is introduced between the two
>>> IQ streams if the RX2 button is activated any time after diversity mode has
>>> been opened/closed/reopened.
>>>
>>> As long as the RX2 button is not cycled (diversity mode off) the phase
>>> relationship between the two diversity IQ streams is maintained regardless
>>> of whether PowerSDR is shut down and restarted and/or the hardware is
>>> powered off (for at least 15 seconds!) and restarted.  The deep null
>>> settings are identical in these cases.
>>>
>>> Contrary to your stated process for bringing up diversity mode, you
>>> should NOT need to cycle the RX2 front panel button at all.  In fact, it
>>> seems that cycling the RX2 button introduces a random phase shift.  I will
>>> investigate this further but the issue does not appear to be associated with
>>> the new HPSDR PowerSDR version exclusively but rather seems to be associated
>>> with the use of the RX2 button on the front panel in between running
>>> sessions in diversity mode.
>>>
>>> To bring up diversity mode you need only power up HPSDR PowerSDR then
>>> click on the Diversity tab on the front panel, no use of the RX2 button
>>> should be required whatsoever; indeed, it appears that using RX2 before
>>> enabling diversity mode introduces a random phase shift between the two IQ
>>> streams, the source of which, admittedly, we need to understand.
>>>
>>> Hope this observation helps you in the short term.
>>>
>>> 73, Joe K5SO
>>>
>>> On Dec 7, 2015, at 8:38 PM, Jim Barber wrote:
>>>
>>>> ***** High Performance Software Defined Radio Discussion List *****
>>>>
>>>> A bug confirmation, I don't remember the original poster.
>>>>
>>>> PowerSDR mRX PS v3.3.6
>>>> Anan 100D
>>>> Windows 10, AMD 1090T (6-core, 3.4 gHz)
>>>>
>>>> To operate diversity here at least it's necessary to first:
>>>> (1) Set RX2 to ADC2 if not already
>>>> (2) Turn RX2 on, then off
>>>> (3) Enable Diversity
>>>>
>>>> It also seems that the the phase angle between receivers changes between
>>>> program runs. IOW, if I null a noise source then close PowerSDR and reload
>>>> it, the best null may be at a different phase angle.
>>>>
>>>> I enjoy diversity a lot, since most of my operation these days is
>>>> evenings on 160M. Without it, 160M is basically useless between power noise,
>>>> plasma TV's and the like.
>>>>
>>>> Thanks and 73,
>>>> Jim N7CXI
>>>>
>>> _______________________________________________
>>> HPSDR Discussion List
>>> To post msg: hpsdr at openhpsdr.org
>>> Subscription help:
>>> http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
>>> HPSDR web page: http://openhpsdr.org
>>> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/
>>
>> _______________________________________________
>> HPSDR Discussion List
>> To post msg: hpsdr at openhpsdr.org
>> Subscription help:
>> http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
>> HPSDR web page: http://openhpsdr.org
>> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/
>
>
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at openhpsdr.org
> Subscription help:
> http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
> HPSDR web page: http://openhpsdr.org
> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/