WinSDR-datastream-driver for Windows audio system
Rudolf Affolter
We have a
hum problem in our building. I am of retired age and find the analysis of this
problem very interesting. To analyze, I built some “PSN-ADC-Serial” and
“3-Channel-Amplifier/Filter” according to Larry’s documents on GitHub, changed
for a cutoff frequency of 150 Hz.
According to Larry’s documents, the replica worked really well. Many thanks to
Larry for these great templates. I know how much time and work it takes.
I use 3 channels from the PSN and amplifiers for 3D-Geophones-4.5Hz. I expanded the fourth channel of the PSN for a Piezo-IEPE-Sensor (4mA), which can be connected directly to my 3D sensor housings.
In parallel to data recording with WinSDR and the possibility of analysis with WinQuacke, a real-time spectrogram from 1Hz to 150 Hz proves to be very useful or even necessary.
That's the reason that I unfortunately can't use the "PSN-ADC24"!
The option via the PSN-to-WAVE utility with subsequent resampling and time adjustment to evaluate the spectral progression is possible but unsuitable because it is delayed and far too time-consuming.
My solution
so far is, that I use the analog outputs (also led out of the “PSN-ADC”).
I
connect 2 channels of choice with a USB sound adapter that I modified to a
cutoff frequency of less than 1 Hz, that is connected to Steinberg's WaveLabPro
program, running on the same Windows PC that I use for WinSDR. This makes it
possible to view the real-time spectrogram at the same time as the real-time
wave from WinSDR. Larry's system is really great for my hum problem analysis.
An even better solution would be, if there was a driver compatible with the Windows audio system either for WinSDR directly, or with the DLL provided by Larry. The selection of the channel pair (stereo), e.g. in the properties of the driver, should be possible.
As I would
like, this real-time spectral analysis would then be possible to view remotely,
as is already possible with WinSDR and the wave display.
Then I could also use the more compact and precise “PSN-ADC24”.
Is there perhaps such a driver already or code that would be suitable for Windows?
Could Larry, or perhaps someone else, who is familiar with this engineering, give me a tip?
Thanks a lot
Rudolf Affolter
Jonathan
--
You received this message because you are subscribed to the Google Groups "Public Seismic Network Mailing List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to psnlist+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/psnlist/f7ed1155-8c1d-40ea-884a-1fbee7ee6fa9n%40googlegroups.com.
Rudolf Affolter
Hi Jonathan,
the PSN digitisers work very well and I can use them very well, so I'm happy with them. The analysis with WinQuake is also very useful to me.
However, what I miss and what is important to me is the dynamic observation of the FFT spectrum over time.
With WinQuake I can only make an FFT over the duration of the PSN file. I am not aware of any software for dynamic observation of the FFT spectrum over time from the PSN files.
At the moment, I can only do this with software such as WaveLabPro from the audio sector. In real time or from audio recordings.
In
order to understand complex events as well as possible, this dynamic frequency
behaviour is important to me.
As a comparison, a video for example, is more meaningful and easier to handle, than just looking at a fiew individual images over time.
I hope I have succeeded in making my request easier to understand.
RudolfJonathan
To view this discussion on the web visit https://groups.google.com/d/msgid/psnlist/eebc1ba5-414f-4168-bb54-ad8354f395c7n%40googlegroups.com.
Larry Cochrane
Sorry for the delay. What I would do is run Earthworm and use the Swarm display utility. It has a realtime spectrogram
display. It will run under Windows or Linux. More information can be found here
https://volcanoes.usgs.gov/software/swarm/doc/swarm_v3.pdf.
The hard part would be getting Earthworm up and running with my ADC board. Here is some old information on how to run EW
on a Raspberry PI https://www.seismicnet.com/ew/RaspPi/index.html. To get the data into EW you would run my PSNAdSend
module. Besides that you would only need to run a few other modules the Wave Server, StatMgr and the StartStop. Once you
get the data into the Wave Server you can view the data with Swarm. Since Swam is a standalone program it does not have
to run on the Earthworm system. So you could run EW on a RaspPI and Swam on a PC laptop. All you would need is a TCP/IP
connection between the two.
Regards,
-Larry
> --
> You received this message because you are subscribed to the Google Groups "Public Seismic Network Mailing List" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to psnlist+u...@googlegroups.com
> <https://groups.google.com/d/msgid/psnlist/eebc1ba5-414f-4168-bb54-ad8354f395c7n%40googlegroups.com?utm_medium=email&utm_source=footer>.
Angel Rodriguez
To view this discussion on the web visit https://groups.google.com/d/msgid/psnlist/CAOY0kB3Kx4MhWT%3DGBh2xwQKzXKDQOF7uQh011bGTiLkhPtOyeg%40mail.gmail.com.
Rudolf Affolter
Hi Jonathan,
I know some very good programs in Windows that are suitable for analyzing and, above all, always provide a very good overview of at least two channels for comparing two situations in real time. I also have experience with some of them.
Unfortunately, none of them can connect a seismic data processing system.
A common denominator would be the Windows audio system.
I'm looking for a way how I could get or build this interface!
Best regards
Rudolf
Rudolf Affolter
Hi Angel,
I have already built enough digitizers that
are suitable for me, based on Larry's documents, thank you very much.
Best regards
Rudolf
Rudolf Affolter
Hello Larry,
Thank you very much for the information I didn't know. I've taken a look at Earthworm and Swarm and noticed that it doesn't offer the options I'm used to in the audio sector. It is important to me that I can always see at least two situations at the same time and in real time in a clear and quickly changeable display. Audio stereo offers this from the ground up, which is probably why it is often used for measurement systems such as oscilloscopes and others.
I think your multi-channel PSN-ADC’s together with WinSDR and the very precise evaluation with WinQuake are brilliant and I like to use them.
An absolute highlight is, that WinSDR can be started on a remote PC and act as a client connected to the base PC (with the PSN ADC) and can even be set independently on the remote PC with your own channel settings, such as other filter properties etc.
It seems to me that WinSDR can make its data available multiple times and even via TCP/IP. Wouldn't it somehow be possible to intercept the data stream you've already prepared and then make it accessible to the audio system at the same time (pure data with counter)?
Is there a DLL module from WinSDR for this use, possibly also for use on the remote client?
Best regards
Rudolf
Larry Cochrane
>
> It seems to me that WinSDR can make its data available multiple times and even via TCP/IP. Wouldn't it somehow be
> possible to intercept the data stream you've already prepared and then make it accessible to the audio system at the
> same time (pure data with counter)?
data in the TCP/IP data packets and send it to the audio system. I personally do not have any software to do this.
>
> Is there a DLL module from WinSDR for this use, possibly also for use on the remote client?
-Larry
Rudolf Affolter
Hi Larry,
I see the opportunity to read the port data stream to the WinSDR-client and extract the desired data. To do this, however, I would need to know the structure of the data stream. Finding this out with Wireshark would be a bit tedious.
I could also write a minimal handler that behaves as a client. To do this, however, I would need to know the necessary information for the handshake with the WinSDR-server.
Please, could you recommend one or the other variant, or perhaps a better one, and give me the relevant information?
Rudolf
Larry Cochrane
WinSDR will send MINISEED packets to the client. So all you need to do is find a MINISEED encoder/decoder library that
will work with your project. Each MINISEED packet is 512 bytes long and contains about 200 samples, if I remember
correctly. You feed the data into the decoder and out will come the raw ADC data. One problem you are going to run into,
if you want to use the Windows sound subsystem, is sample rate. Since you are dealing with very low sample rates
(<=200sps) from the ADC card compared to audio you will need to up sample the ADC data to 8kHz somehow so you can feed
the data into Windows.
Good luck...
-Larry
> You received this message because you are subscribed to the Google Groups "Public Seismic Network Mailing List" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to psnlist+u...@googlegroups.com
> <https://groups.google.com/d/msgid/psnlist/a06d706c-5239-4c8f-b39d-54ba4160f3d3n%40googlegroups.com?utm_medium=email&utm_source=footer>.
Rudolf Affolter
thank you. Did you need the libmseed-main library on GitHub?
Rudolf
pastor...@gmail.com
Rudolf Affolter
Hi Jeff,
all of my built PSN-ADC systems are in use.
My questions were intended to find an integrated solution to see FFTs and spectrograms from the remote systems in real time also. I had already built an additional solution, only usable locally.
As Larry suggested, implementing could be very elaborate. After thinking about it for a while, I see it that way too. Real-time resampling alone would be a project by itself.
I will now expand the current local solution in streaming the separately digitized audio data from remote to the Windows audio system, in parallel to the PSN-ADC data to WinSDR.
Rudolf
Jeff Mauer
Best regards
I posted to the wrong thread