Pages: 1 [2] 
|
 |
|
|
Author
|
Topic: Restseitenbandunterdrückung - schlecht? (Read 4521 times)
|
|
DD4WH
positron
alter Hase
 Offline
Posts: 462
Ich liebe dieses Forum!
|
 |
Re:Restseitenbandunterdrückung - schlecht?
« Reply #15 on: 19. July 2018, 11:15:08 »
|
|
Hi Danilo,
I did not find the old thread/threads where we were discussing about that. There we had evidence of a TUNE signal that was considerably worse than an SSB signal.
However, I dont remember whether that problem was ever solved or not.
73 Frank
|
|
 Logged
|
-----------------------------------------
Teensy Convolution SDR
https://github.com/DD4WH/Teensy-ConvolutionSDR
|
|
|
|
|
S53DZ
schon länger dabei
Offline
Posts: 58
Ich liebe dieses Forum!
 
|
|
Re:Restseitenbandunterdrückung - schlecht?
« Reply #17 on: 19. July 2018, 12:18:20 »
|
|
When there is such a large difference at work as 40 dB than there should be more evident cause for that but it seems it is well obscured.
As a first try this is what pops out of my thoughts:
1. Possible different phasing of a synthetic sine wave (I/Q) compared to the ADC generated signal.
2. Unwanted coherence between TUNE side band signal and carrier signal.
I have seen cases where they produced such an internal signal, put it to a free DAC and back in through the normal signal path to sustain the same amplitude / phase (delay) characteristics.
EDIT:
But in both cases I think there should be different IQ calibration values when calibrating with a mic / line_in signal compared to a TUNE signal.
73 Bojan
|
| « Last Edit: 19. July 2018, 12:25:53 by S53DZ » |
Logged
|
|
|
|
|
|
DB4PLE
positron
Urgestein
Offline
Posts: 1278
|
|
Re:Restseitenbandunterdrückung - schlecht?
« Reply #19 on: 19. July 2018, 16:43:38 »
|
|
Hi Bojan,
When there is such a large difference at work as 40 dB than there should be more evident cause for that but it seems it is well obscured.
As a first try this is what pops out of my thoughts:
1. Possible different phasing of a synthetic sine wave (I/Q) compared to the ADC generated signal.
2. Unwanted coherence between TUNE side band signal and carrier signal.
I have seen cases where they produced such an internal signal, put it to a free DAC and back in through the normal signal path to sustain the same amplitude / phase (delay) characteristics.
EDIT:
But in both cases I think there should be different IQ calibration values when calibrating with a mic / line_in signal compared to a TUNE signal.
73 Bojan
|
|
The SSB tune signal is not IQ but a simple audio sine wave. CW in UHSDR is btw directly generated IQ using the same sine/cosine generator code as for the tune signal (for SSB tune we simply use only sine, for CW we use both sine and cosine).
So in SSB the signal processing is very much identical to the one for audio. I am not sure if we did that last time this popped up, but we could have a look at the generated tx IQ signal recorded via USB and check if we see a difference there between tune and the feeding of a sine wave of same characteristics (as digital audio) via USB.
Anyway, this has indeed not the highest priority but uses our resources from time to time, so maybe we should at least figure out the cause so that this discussion can be put into the FAQ with some conclusions.
73
Danilo
|
|
Logged
|
|
|
|
Vlad
Neuling
Offline
Posts: 1
Ich liebe dieses Forum!
|
|
Re:Restseitenbandunterdrückung - schlecht?
« Reply #20 on: 19. July 2018, 19:05:26 »
|
|
Und in CW planen Sie, Ordnung zu bringen? Das ist auch sehr relevant.
73!
Vlad
|
|
Logged
|
|
|
|
S53DZ
schon länger dabei
Offline
Posts: 58
Ich liebe dieses Forum!
|
|
Re:Restseitenbandunterdrückung - schlecht?
« Reply #21 on: 21. July 2018, 21:44:20 »
|
|
Hi,
I did a few more measurements with the same I/Q calibration numbers as before.
The results show almost no difference when using TUNE compared to USB.
The setup:
OVI_UI FW 2.9.42, mcHF RF, 2W.
With the USB mode I have used line-in input (not mic!) and an external NF White noise generator, TX filter SOPRANO.
TuneSSB:
wanted/unwanted = 55.1 dB
wanted/carrier = 33.7 dB
USB, line_in, white noise:
wanted/unwanted = 53.0 dB
wanted/carrier = 19.6 dB
The difference in wanted side band to carrier ratio is due to lower modulation level when using line-in whereas the carrier leakage stays constant.
So, I think this outcome completes the issue of having different results when on TUNE.
73 Bojan
|
|
|
|
|
Pages: 1 [2] 
|
|
|
|
|
|
|
sideband_meas.zip
