Kraken SDR experience?

252 views
Skip to first unread message

fasleitung3

unread,
Oct 3, 2023, 5:22:35 AM10/3/23
to SARA
Hi guys,
Has anyone gained hands on experience with the Kraken SDR in
multichannel, phase coherent applications? I was wondering whether this
unit could seriously be considered for multi-baseline interferometry.
Thanks for any insights,
Wolfgang


Marcus D Leech

unread,
Oct 3, 2023, 7:58:06 AM10/3/23
to sara...@googlegroups.com
Hanish has used it for two channel interferometry.

Sent from my iPhone

> On Oct 3, 2023, at 5:22 AM, 'fasleitung3' via Society of Amateur Radio Astronomers <sara...@googlegroups.com> wrote:
>
> Hi guys,
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com
> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/d07c7a66589d2f8350b864e8e789e9040f20e63d.camel%40googlemail.com.

Jim Abshier

unread,
Oct 3, 2023, 2:08:40 PM10/3/23
to sara...@googlegroups.com

Wolfgang - I have been wondering about how they do the synchronization
of channels to get coherence. They obviously must do some sort of
correlation to align the data from the various channels, but I haven't
found any information on what exactly they do. For example do they do
sub sample alignment of the data. If they don't, there is a possible
reduction of correlation from sub-sample misalignment. People seem to be
really excited by this product, but my (admittedly inexpert) opinion is
that it is somewhat of a kluge.

Jim Abshier

On 10/3/23 5:22 AM, 'fasleitung3' via Society of Amateur Radio

B & MR Randall

unread,
Oct 3, 2023, 2:13:00 PM10/3/23
to sara...@googlegroups.com
I had heard some mention of all 5 RTL SDR's run on the same clock and there is som test signal used to allign data. Marcus probably knows more about this.
Bruce R.
--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/b4c05a76-cec7-97df-1527-36f93bacd551%40ameritech.net.

Marcus D. Leech

unread,
Oct 3, 2023, 2:24:21 PM10/3/23
to sara...@googlegroups.com
On 03/10/2023 14:12, B & MR Randall wrote:
> I had heard some mention of all 5 RTL SDR's run on the same clock and there is som test signal used to allign data. Marcus probably knows more about this.
> Bruce R.
Indeed, they have a directional coupler arrangement in each of the 5
antenna paths, and a noise source that is shared
  among all the receivers, and each receiver shares the same clock,
which drives both the ADCs and the PLL synthesizer.

The have a chunk of code that does an FFT correlation among all the
channels and computes the time delay, and sub-sample
  phase offsets.

They start out turning on the noise source, recording data from all 5
receivers, and then computing the correlations.  Once
  that is done, they insert the appropriate corrections in the live
data stream, and turn off the noise source.

Hamish Barker

unread,
Oct 3, 2023, 4:54:50 PM10/3/23
to sara...@googlegroups.com
I have dabbled, and run it for a year with 1 baseline (i.e. 2 antennae) at 1419MHz. I ran at that frequency as I have sawbird H1 LNAs but wanted to avoid the overwhelming H line emissions from everywhere in the sky.

in respect of phase coherency, I do get quite nice fringes of the sun even in 40+ degrees off axis in the sidelobes of my (two) dishes with the krakensdr. Operating at 1419Mhz so as to look at continuum radio sources, I do get much fainter fringes corresponding to various mapped peaks in an atlas of 1420 (off H line) galactic continuum sources. And this is only with 2x 85cm offset feed dishes with 30 degree circular cone feedhorns with a bit less than 300mm aperture feeding 150mm circular waveguides. With bigger dishes, I think those sources would show up much more clearly.  Stacking observations from multiple days according to Right ascension does improve the fringe visibility.  My baseline is about 10.7metres.

As I have said previously, I bought extra cables and LNA units to scale up to more baselines but have simply not gotten around to it.

It does have the inherent RTLSDR limit of only 2.5MHz bandwidth due to the use of those receivers. On the plus side, it seems that the manufacturers have done quite a good job of filtering spikes and birdies etc from the frequency response and I don't see any of them.

For the low price, I think it is definitely worth trying multi channel, and I am sure that Wolfgang has the capability to make it happen, and it would be wonderful to see amateur, 10 baseline synthesis imaging actually happen! There is nothing with this number of coherent channels on the market anywhere near its price.

Cheers,
Hamish

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.

Jim Abshier

unread,
Oct 3, 2023, 5:43:02 PM10/3/23
to sara...@googlegroups.com
Hamish - Amateur synthesis imaging has already been done. This includes
images of the Cygnus A region using 6 baselines in one experiment and 17
baselines in another more recent experiment. A synthesis image was also
produced of the Taurus A region. This was done with a single 2-antenna
interferometer. The different baselines were achieved by moving the
antennas on different collections. A multichannel receiver was not
required. In fact the receiver was an analog phase switched
interferometer that used a computer only for data logging and fringe
data post processing.

Jim Abshier
> <mailto:sara...@googlegroups.com>> wrote:
>
> Hi guys,
> Has anyone gained hands on experience with the Kraken SDR in
> multichannel, phase coherent applications? I was wondering whether
> this
> unit could seriously be considered for multi-baseline interferometry.
> Thanks for any insights,
> Wolfgang
>
>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com
> <mailto:sara...@googlegroups.com>
> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com
> <mailto:sara-list-...@googlegroups.com>
> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to sara-list+...@googlegroups.com
> <mailto:sara-list%2Bunsu...@googlegroups.com>.
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com
> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sara-list+...@googlegroups.com
> <mailto:sara-list+...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sara-list/CAAdNG_K5u9JUGJ7QPU6%3DVqPeWR2doEbLgSxHWU87KDCqqG3yfQ%40mail.gmail.com
> <https://groups.google.com/d/msgid/sara-list/CAAdNG_K5u9JUGJ7QPU6%3DVqPeWR2doEbLgSxHWU87KDCqqG3yfQ%40mail.gmail.com?utm_medium=email&utm_source=footer>.

Marcus D. Leech

unread,
Oct 3, 2023, 5:46:49 PM10/3/23
to sara...@googlegroups.com
On 03/10/2023 17:42, Jim Abshier wrote:
> Hamish - Amateur synthesis imaging has already been done. This
> includes images of the Cygnus A region using 6 baselines in one
> experiment and 17 baselines in another more recent experiment. A
> synthesis image was also produced of the Taurus A region. This was
> done with a single 2-antenna interferometer. The different baselines
> were achieved by moving the antennas on different collections. A
> multichannel receiver was not required. In fact the receiver was an
> analog phase switched interferometer that used a computer only for
> data logging and fringe data post processing.
>
> Jim Abshier
Indeed, I remember your work, Jim.

Multiple antennas simple increase the rate at which you can "fill in"
the u,v plane and produce higher-quality images.

BUT, as Jim points out, if you have the patience, and mobile antennas,
and the space to move them around, image
  synthesis can be performed with only two antennas.

In fact, even with 5 antennas and 10 baselines, you'll likely still need
to move antennas around to get enough u,v
  coverage to produce nicer images.

Lester Veenstra

unread,
Oct 4, 2023, 9:33:38 AM10/4/23
to sara...@googlegroups.com
Anyone willing to share the: "chunk of code that does an FFT correlation among all the channels and computes the time delay, and sub-sample phase offsets. "?

Lester B Veenstra K1YCM MØYCM W8YCM 6Y6Y W8YCM/6Y 6Y8LV (Reformed USNSG CTM1)
les...@veenstras.com

452 Stable Ln
Keyser WV 26726 USA

GPS: 39.336826 N 78.982287 W (Google)
GPS: 39.33682 N 78.9823741 W (GPSDO)


Telephones:
Home: +1-304-289-6057
US cell +1-304-790-9192
Jamaica cell: +1-876-456-8898
--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/bbbea9fc-022a-0c15-64a9-df66895c074b%40gmail.com.

Marcus D Leech

unread,
Oct 4, 2023, 9:37:20 AM10/4/23
to sara...@googlegroups.com

Lester Veenstra

unread,
Oct 4, 2023, 10:01:02 AM10/4/23
to sara...@googlegroups.com

Excellent!

  Thank you

 

Lester B Veenstra  K1YCM  MØYCM  W8YCM   6Y6Y W8YCM/6Y 6Y8LV (Reformed USNSG CTM1)

les...@veenstras.com

 

452 Stable Ln

Keyser WV 26726 USA

 

GPS: 39.336826 N  78.982287 W (Google)

GPS: 39.33682 N  78.9823741 W (GPSDO)

 

 

Telephones:

Home:            +1-304-289-6057

US cell          +1-304-790-9192

Jamaica cell:    +1-876-456-8898

 

image001.png

fasleitung3

unread,
Oct 5, 2023, 3:20:19 AM10/5/23
to SARA
Thanks, Guys for your insights and comments.
I also looked at the forum of Kraken in order to better understand what
exactily is going on with the phase calibration using the noise source.
It still remains somewhat of a mystery to me. Nevertheless, it seems
that this calibration procedure is somehow triggered also when using
their gnu-radio block. What irritates me a bit was a comment in the
forum where someone had the issue that the phase calibration changed at
regular intervals due to an automatically triggered re-calibration. It
seems to me that one really has to have one of the units at hand to
fully understand the capabilities and limitations. Obviously the team
at Kraken concentrates on their own applications like direction finding
and is paying less attention to people with their own application.
Thanks again,
Wolfgang

Anthony

unread,
Oct 5, 2023, 7:39:17 AM10/5/23
to Society of Amateur Radio Astronomers
That is correct, the support KrakenSDR provided to me, helped with my decision to move forward with USRP-B210, thought it would have been nice to use the KrakenSDR to add a third, 3-meter dish, or many smaller dishes.

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.

Jonathan Pettingale

unread,
Oct 5, 2023, 3:27:25 PM10/5/23
to Society of Amateur Radio Astronomers
Wolfgang.

There is little documentation that comes with the software, I have enclosed what there is.

Jonathan

HDAQ_firmware_ver1.0.20201130.pdf
gr-krakensdr_natively_on_Windows.pdf

Marcus D. Leech

unread,
Oct 5, 2023, 5:45:20 PM10/5/23
to sara...@googlegroups.com
On 05/10/2023 03:20, 'fasleitung3' via Society of Amateur Radio
Astronomers wrote:
> Thanks, Guys for your insights and comments.
> I also looked at the forum of Kraken in order to better understand what
> exactily is going on with the phase calibration using the noise source.
> It still remains somewhat of a mystery to me. Nevertheless, it seems
> that this calibration procedure is somehow triggered also when using
> their gnu-radio block. What irritates me a bit was a comment in the
> forum where someone had the issue that the phase calibration changed at
> regular intervals due to an automatically triggered re-calibration. It
> seems to me that one really has to have one of the units at hand to
> fully understand the capabilities and limitations. Obviously the team
> at Kraken concentrates on their own applications like direction finding
> and is paying less attention to people with their own application.
> Thanks again,
> Wolfgang
>
I hadn't heard about the "periodic re-cal".   If that is a thing,
turning it off shouldn't be that hard.

The main thing that I ran into, over a decade ago (!!!!!) in having two
RTL-SDRs, with shared clocks, and
  with the modified driver that turns off "dither" is that you get
phase-coherence, but there's a residual
  constant frequency offset between the devices, and you have to
characterize it and insert a rotator
  to compensate --said offset is very small indeed, and results in
"fringing" for a constant source, at a rate
  that is a fraction of Hz.  So, quite small, but at similar
time-scales to actual astronomical fringes.  It is fixed, though,
  and I'm guessing that Kraken compensate for this in their "DAQ"
code--they would have to for direction-finding.


Peter East

unread,
Oct 5, 2023, 6:16:37 PM10/5/23
to Society of Amateur Radio Astronomers
You are right Marcus - I met this problem with my quad RTL.
Although all the RTL devices run from the same clock, there are slight differences in the RTL sample times, I measured between 15 to 150ns. This means that the relative phases change with frequency (phase= frequency x time) and so you get a constant drift.
I measured the phase slope at the center frequency and corrected the I/Q values digitally by continuously rotating the vectors. The alternative is to do what Kraken do and re-calibrate on a regular basis.

Marcus D. Leech

unread,
Oct 5, 2023, 6:19:01 PM10/5/23
to sara...@googlegroups.com
On 05/10/2023 18:16, Peter East wrote:
You are right Marcus - I met this problem with my quad RTL.
Although all the RTL devices run from the same clock, there are slight differences in the RTL sample times, I measured between 15 to 150ns. This means that the relative phases change with frequency (phase= frequency x time) and so you get a constant drift.
I measured the phase slope at the center frequency and corrected the I/Q values digitally by continuously rotating the vectors. The alternative is to do what Kraken do and re-calibrate on a regular basis.
That alternative is a poor alternative for long-term observing (like, even a few minutes at a time).


--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.

fasleitung3

unread,
Oct 6, 2023, 9:51:04 AM10/6/23
to sara...@googlegroups.com
Thanks Jonathan,
That explains how they do the phase calibration.
Wolfgang
--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.

fasleitung3

unread,
Oct 6, 2023, 10:11:25 AM10/6/23
to sara...@googlegroups.com
Hi, I just happened to run across a website which describes a setup very similar to the Kraken SDR:
It is in German but maybe Google translate can help. Here are the key points:

There is a continous phase drift between two RTL-SDRs even when clocked from the same source.
The reason for this is a hidden 17th bit of the fractional-N PLL which is set arbitrarily on each retune.
In order to determine that a noise signal is injected and the correlation is calculated.
Not sure how the compensation is done in this example.
Wolfgang

Am Donnerstag, den 05.10.2023, 12:27 -0700 schrieb Jonathan Pettingale:

Hamish Barker

unread,
Oct 7, 2023, 2:11:20 AM10/7/23
to sara...@googlegroups.com
I found following the kraken sdr install process from the GitHub page and man pages straightforward. I know my way around Unix type systems but am not more than an occasional programmer, usually limited to getting something ported. So if Wolfgang or someone in his group is a decent programmer, I would expect no problems for them.

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.

Peter East

unread,
Oct 9, 2023, 12:55:58 PM10/9/23
to Society of Amateur Radio Astronomers
Question for Marcus, is this the dither that the -N command stops?

Marcus D. Leech

unread,
Oct 9, 2023, 1:03:42 PM10/9/23
to sara...@googlegroups.com
On 09/10/2023 12:55, Peter East wrote:
Question for Marcus, is this the dither that the -N command stops?
The -N command on what, exactly?

The librtlsdr library, unfortunately, has had MANY "forks" over the years, some of those "forks" include an API to
  turn off dither when you're tuning.  Which "fork" your particular OS includes is a bit of a crap-shoot.

Once you have a librtlsdr that has that API, then access to it has to be propagated "upwards" in whatever application
  stack you're using--for Gnu Radio that would be the gr-osmosdr and SoapySDR  device abstraction libraries.

This is one of the reasons that my approach, when I get around to it, would be to simply pick up the pre-packaged
  system image for rPi 3/4 that the KrakenSDR project has produced.  Their software takes care of all that, and from
  an application (Gnu Radio or whatever) perspective, you just "see" a radio with 5 channels and a simple tuning
  interface, all delivered over 1Gbit ethernet.  The rPi+KrakenSDR is an "integrated network-radio" in that case, and
  someone *else* has taken on the pain of making sure they have the correct "fork" of librtlsdr.


--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages