Topics

Jumpy FFT Display with SDR# R1651

Ian DXer
 

Probably a video would be better at displaying this problem, but my description will have to suffice for the moment.

Setup:
Windows 7, 64 with Airspy HF+ & SDR# r1651
Tested Spectrum: 88-108MHz - not tested outside this spectrum.
Modulation: WFM
Bandwidth: 768kHz
Firmwear: Original (upgrade will happen sometime later)
Zoom Level: Minimum as usual

This is an usual problem & I'll do my best to describe it.
On the upper spectrum/amplitude display(FFT) (above the waterfall display) I note for maybe, 70% of the chosen frequencies
in the spectra of: (eg 88-108MHz) that at either end of the 768kHz BW (just outside the 768kHz bandwidth) that the amplitude of the noise floor are randomly jumping up by around 10dB
(jitter) Even within the 768kHz passband the noise floor (and maybe signals?) appears to be jumping up/down (jitter) by 1-3dB where they shouldn't.
At the edges of the passband jitter amplitude is random between 3-10dB.
Visually this is very distracting to the point of giving one a headache.
This fault was NOT evident on previous r1646 or any previously SDR# releases that I've used in the past.
I can partly negate the far end 10dB jumping by moving the zoom slider control up slightly so as not to see it, but still have the slight 1-2dB or 1-3dB jumping
of the display within the 768kHz rxer bandwidth.

I suspect this fault occurs more often in areas of the band where there are many strong signals, but I still find exceptions.
This is particularly a visual distraction at 768kHz bandwidth.

If I go back to using SDR# r1646 or earlier no problem. If I return to SDR# r1651 the jitter (for want of a better description) is there on most channels for
FM BCB.

Has anyone else noted this?
Or can others test for this with SDR# r1651 on FMBCB at 768kHz BW?

 

prog
 

On Thu, Feb 22, 2018 at 03:57 am, Ian DXer wrote:

If I go back to using SDR# r1646 or earlier no problem. If I return to SDR# r1651 the jitter (for want of a better description) is there on most channels for
FM BCB.
SETTINGS?

prog
 

And please keep these emails short and straight to the point.

Ian DXer
 

Hi Youssef,

Ok shorter msg & I'll try to be as specific as possible:

SETTINGS: 'DEFAULT' (ie/eg. just unzipped r1651 & tried for first time with just frequency changed)
Test Sample Freq: 105.1MHz - very weak signal near strong 150kW signal 104.9MHz. WFM. Also added time stamps...
See image below where I highlight roughly (in yellow the range of the jitter/random jumping of the signal levels.
In this sample outside of 768kHz passpand. noise jumping 15dB. Within passband noise/signals maybe jumping 3dB.
The jumping occurs perhaps 3 times a second - can't be exact.

IMPORTANT: Did some further experimenting, this time with FFT display resolution settings (given your recent work in this area)
& noted that the problem is only evident with FFT Display RESOLUTION settings of 32768 & lower.
The jumping of noise/signals is particularly distracting with 32768 & 16384 settings. The default program setting is of course 32768.

Is this information useful to you Youssef? I have no idea what other setting info I could provide that would be helpful, as it's all basically default settings.

Please let me know if you need further info.



 

prog
 

On Thu, Feb 22, 2018 at 01:14 pm, Ian DXer wrote:
Ok shorter msg & I'll try to be as specific as possible:
Good. Thanks. Check the latest SDR#.

PMM
 

I have tried to replicate this myself and failed myself, I have seen that behaviour but usually as an impedance error i.e antenna not 50ohm or local interference raising the noise floor, but i think the latter can be ruled out as it should not increase the genuine signal. I would check your antenna connections make sure the ground connect is good.

Inline images 1

On 22 February 2018 at 21:14, Ian DXer <baxterinoz@...> wrote:
Hi Youssef,

Ok shorter msg & I'll try to be as specific as possible:

SETTINGS: 'DEFAULT' (ie/eg. just unzipped r1651 & tried for first time with just frequency changed)
Test Sample Freq: 105.1MHz - very weak signal near strong 150kW signal 104.9MHz. WFM. Also added time stamps...
See image below where I highlight roughly (in yellow the range of the jitter/random jumping of the signal levels.
In this sample outside of 768kHz passpand. noise jumping 15dB. Within passband noise/signals maybe jumping 3dB.
The jumping occurs perhaps 3 times a second - can't be exact.

IMPORTANT: Did some further experimenting, this time with FFT display resolution settings (given your recent work in this area)
& noted that the problem is only evident with FFT Display RESOLUTION settings of 32768 & lower.
The jumping of noise/signals is particularly distracting with 32768 & 16384 settings. The default program setting is of course 32768.

Is this information useful to you Youssef? I have no idea what other setting info I could provide that would be helpful, as it's all basically default settings.

Please let me know if you need further info.



 

Attachments:


Ian DXer
 

Downloaded SDR# r1653 & tested briefly.
Yes problem now resolved Youssef - thanks so much, very impressed  :-)

Cheers.

Ian DXer
 

You're right, this kind of fault is similar to a faulty antenna connection & that was
one of the first things that came to mind (and checked), but the fault was specific to or noted only with recent software
changes to SDR# and not happening on all freqs. 
Youssef is like an Olympic sprinter at rapidly resolving these bugs. Fixed with release 1653 of SDR#.
Many thanks for your time PMM :-)