SparkSDR and WSJT-X vers. 2.6.0-rc4

[Juergen, DL8LE]

When trying to work D60AE on 10m in FT8 I have made the following experience:

SparkSDR in DigiU, Hermes Lite II Firmware Version 73, Firmware Patch 2, Board ID 5, WSJT-X connected via an virtual audio cable to SparkSDR:

1. Noise floor at around -150, many signals in the waterfall in SparkSDR, only very thin traces in the waterfall of WSJT-X, no decodes

2. Setup another virtual receiver in SparkSDR with FT8 and not DigiU, huge number of decodes in SparkSDR and - all of a sudden - similar but lower number of decodes in WSJT-X and waterfall showing the signals

3. SparkSDR shows SNR figures which are typically min. 10 db better than those ind WSJT-X, it decodes signals with SNR - 13 which are not decoded any longer in WSJT-X.

Questions:

1. Any possible explanation why VAC provides enough signal from SparkSDR after switching from DigiU to FT8 in SparkSDR (and keeps the level after switching back to DigiU again)?

2. What is the meaning of the 3 to 4 digit number in the line of the decoded signal (first one is the flag of the country, second one the signal level, third one is ?, fourth one is the call etc.)? I could not find it in the help.

73

Juergen, DL8LE 

[ahop...@googlemail.com]

Hi Juergen,

the number is the frequency offset of the signal, the history display does label it but I removed the labels on the operating view to gain space.

I'm not sure what could cause the vac issue.  One thing to check is that the virtual audio cables are set to 48khz stereo in the system settings.  Some people have found that using the same cable for in and out can also cause problems.

73 Alan M0NNB

[Juergen Sturhahn]

Hello Alan,

I have found the explanation for the numbers already ... it's not the frequency offset, those are the distance to the station with the call displayed in that line (realized this when I clicked on the history button). Sorry for this question. Also interesting somehow when looking at the SNR-figures.

Regarding the VAC: I am using the standard 48khz, only one virtual cable from SparkSDR to the audio input of WSJT-X (don't use the same cable for in- and output). The same virtual cable, however, is not only used in the first but also in the second virtual receiver. Therefore there is still the question what is triggering the different behaviour of the HL II (it looks as if the noisefloor has been reduced significantly) when switching the 2. virtual receiver from DigiL to  FT8. Just for info: I have 2 virtual receivers running on this HL II, one in DigiU for operating with WSJT-X (sometimes JTDX), the second one in DigiL so that I can easily use RTTY (AFSK) together with WinWarbler (part of DXLab). Might be this explanation is more precise than my first mail. As switching the second one to FT8 helps to solve the problem it is already a work around but it would be helpful to understand this topic.

Another question: is there any way to get the IQ-data of the HL II from SparkSDR transferred to WSJT-X (or JTDX)? I know that it is possible to feed the IQ-data of each slice of a Flexradio 6400/6600 via DAX to a different device but I have never used it when having the chance to operate the Flex at DA0UP. But when I see the huge difference in SNR figures of FT8 and DigiU with WSJT-X then I believe it is worth to think how to avoid those losses.

73

Juergen, DL8LE

73

Juergen

--
You received this message because you are subscribed to a topic in the Google Groups "SparkSDR" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/sparksdr/SaenY4K0PDY/unsubscribe.
To unsubscribe from this group and all its topics, send an email to sparksdr+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sparksdr/5a2c3db5-649f-4f0f-a334-8edfcc18bf95n%40googlegroups.com.

[ahop...@googlemail.com]

Hi Juergen,

are you using the same vac for two virtual receivers at the same time? If so all sorts of odd things could happen, at best the audio from both will be added together.  

I would love there to be a way to feed iq to WSJTX etc directly without using vac, I've failed for about 7 years to tempt anyone to do this (other than multipsk but no one has ever mentioned  using it).

73 Alan M0NNB

[Brian VK4BAP]

Hi Alan,

I hadn't seen that you have now added RawIQ to the modes. Is there any documentation? I'm happy to use VAC - just set (like DigiU) Audio Output to cable A input, Audio Input to cable B output???

I found using a single cable to be unreliable. Maybe just my low powered PC.

Meanwhile I will have a play with multipsk.

Thanks for the good work.

73 Brian.

[ahop...@googlemail.com]

Hi Brian,

the raw iq mode is really an internal mode used in an upcoming feature where one copy of spark can share radios connected to another, it appears in the menu as an unexpected side effect:) That being said I could make it emit iq over stereo for local use if anyone wants that, I'm not sure what comes out locally at the moment.

73 Alan

[Brian VK4BAP]

Yes Alan,

that would be useful to have iq output as two channels. SDRUno/SDRConsole does something like that. I've used it with CWSkimmer.

73 Brian.