HamVNA

[rodwall1]

Hi Steve,

I sent a problem report to Alex the author of HamVNA.

Alex indicated that HamVNA is only able to read either S11 or S21 data from a S2p touchstone file but not both at the same time. This indicates that is is not able to display both lots of data.

This may have come from the fact that there is only one detector with the Hermes hardware. So HamVNA may only generates a reflection S11 or transmission S21 touchstone file.

I've never generated a touchstone file from HamVNA so I am only guessing the above.

Regards,

Roderick, vk3yc.

[Steve Haynal]

Hi Roderick,

Thanks for investigating this. Yes, there is only one detector in the Hermes-Lite. I'd like to have an option for a second coherent receiver but that is farther out. I think you are right that what HamVNA does is the best you can offer for a Hermes-like architecture with a single detector. Also, you'll have more limited frequency range than other VNA-specific hardware. Take a look at QVNA and the quisk VNA to see more of what you can expect with a VNA on similar hardware. I think the main uses will be antenna analysis and some lower HF frequency filter analysis.

Maybe you could persuade the author of myVNA to add support for Hermes/Hermes-Lite, at least for what can be supported?

73,

Steve

KF7O

[rodwall1]

Hi Steve.

On talking to Alex he indicated that the reason HamVNA only generates either S11 (reflection) or S21 (transmission) files. Is because Hermes hardware only has one detector. So you get S11 or S21 in separate files. Not together in one file.
I think you will find that myVNA will work with one detector and is able to combine each scan into one touchstone s-parameter file. I have not tried this as my N2PK VNA has two detectors. I still have to turn the DUT around to scan S22 and S12. I will investigate and find out if it can. The advantage in doing this is shown in the YouTube video below. You can then use software like RFsim99.

https://m.youtube.com/watch?v=C8wUUngmxew

Roderick, vk3yc.

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[rodwall1]

Hi Steve,

Have you received my email indicating that Dave the author of myVNA. Needs to know what the Hermes/Hermes Lite interface is so that he can write a interface for his great myVNA software.

Dave also indicated that he will also  modify myVNA to do separate scans and to combine them into one touchstone s-parameter file. Required because Hermes Lite only has one detector.

Regards,

Roderick Wall,  vk3yc

[Steve Haynal]

Hi Roderick,

Sorry for the delayed response - there is a lot going on here at the moment. That is great that Dave is willing to add support to myVNA. I have yet to dig into the VNA portion of the RTL so don't yet have a good understanding of how the Hermes people implemented the VNA interface. I will in a few weeks. As a first cut, I don't plan to do anything different than what has been done in the Hermes. Given this, and that there are many Hermes out there, Dave might want to start by supporting the Hermes. Perhaps people on the openhpsdr list can help, or the author of HamVNA. Links to the Hermes protocol can be found at https://github.com/softerhardware/Hermes-Lite/wiki/Protocol-Coverage.

I'll be happy to help in a few weeks once I learn enough to offer proper help!

73,

Steve

KF7O

On Tuesday, June 9, 2015 at 6:48:11 AM UTC-7, Rod Wall wrote:

Hi Steve,

Have you received my email indicating that Dave the author of myVNA. Needs to know what the Hermes/Hermes Lite interface is so that he can write a interface for his great myVNA software.

[Steve Haynal]

Also, Alan Hopper on this list has written code to interface with a Hermes/Hermes-Lite. He may be willing to share this code or help with setting up the ethernet-based interface code.

73,

Steve

KF7O

[John Laur]

Also, the HPSDR protocol itself is about to change so it may not be
worth writing support for the legacy protocol until after this
transition, though not terribly much would change for a VNA
application.

If Dave happens to be able to obtain access to an Angelia board he
could begin writing against the new protocol now. Angelia does have
dual phase locked ADCs so I believe it could be used to make
simultaneous 2 port measurements. I do not know if any of the current
software has done that yet. It would be a great feature to have, and
an excellent first HPSDR board to support.

As far as the current "VNA Protocol" goes the only difference is that
a bit is set in the mic data at the point at which the receiver
changes IF frequency. As the VNA sweeps the receiver frequency through
the full range, it repeats the process of sending a tune command
receiving a sufficient number of samples. Because of network delay the
software does not know at what point in the stream of received samples
the receiver changed frequency. Before "VNA Mode" was created the VNA
software just had to discard a few ms of samples after the tune
command was sent before it could trust the samples. Now that the flag
is set it knows instantly when the change has occurred and can
therefore sweep much faster. I do not think the VNA mode contains any
special logic to lock the transmitter and receiver together in phase
though; I think basically everything has to be inferred in software by
counting samples etc.

I hope that made sense. I also do not know if it is absolutely 100%
correct. I am considerably out of my element discussing it, but I do
have a bit of information. This was interesting to me because I wanted
to try to move and zoom the panadapter/waterfall in PowerSDR to match
the users tuning, but I did not know the point at which the receiver
data changed frequency. I decided to try to delay this until after the
protocol was extended to provide the change indication in the standard
protocol.

I think (hope) the new protocol may handle this a little more
elegantly. In my mind ideally the receiver should send the actual IF
frequency, phase word, and bandwitdh at the exact point in the stream
when this changes. Of course you still have to wait for some number
of samples to pass through the filter chain in the FPGA before you
have 'clean' data from the new freuqency but it is better than what
we've got now.

73, John K5IT

[John Williams]

Roderick,

Looking at Dave's web site I see that they have previously had boards
available for the bridge circa 2004. Can you ask if we can resurrect
some of that to supply this group with solid bridge designs? Looks like
they had multiple iterations of this to arrive at an acceptable design.
I can do the gerber work if necessary.

Perhaps we can arrange a built 1.22 unit for him. I will contribute...

John - W9JSW

[Steve Haynal]

There are also some links for bridges in the Hermes VNA documentation. It would be great if you or someone else creates a frontend card that is a VNA bridge.

73,

Steve

KF7O

[rodwall1]

Hi John,

Dave's myVNA was written for Paul's N2PK VNA. The following link has three links, Part 1 PDF and Part 2 PDF. Part 2 PDF has the details for the bridge. Link 9 also has the build  information for the bridge including the PC board layout.

http://www.n2pk.com/VNA/VNAarch.html

I purchased my PC boards for my N2PK VNA and bridge from:

http://www.makarov.ca/vna.htm

Paul's home page:

http://www.n2pk.com/

The following link shows the bridge being attached to the N2PK VNA.

http://www.n2pk.com/VNA/Pix/VNApix.html

Regards,

Roderick Wall, vk3yc.

On Jun 17, 2015 9:15 AM, John Williams <jswi...@gmail.com> wrote:
>
> Roderick,
>
> Looking at Dave's web site I see that they have previously had boards
> available for the bridge circa 2004. Can you ask if we can resurrect
> some of that to supply this group with solid bridge designs? Looks like
> they had multiple iterations of this to arrive at an acceptable design.
> I can do the gerber work if necessary.
>
> Perhaps we can arrange a built 1.22 unit for him. I will contribute...
>
> John - W9JSW
>
> On 6/9/2015 12:10 AM, rodwall1 wrote:
> >
> > Hi Steve,
> >
> > Have you received my email indicating that Dave the author of myVNA.
> > Needs to know what the Hermes/Hermes Lite interface is so that he can
> > write a interface for his great myVNA software.
> >
> > Dave also indicated that he will also  modify myVNA to do separate
> > scans and to combine them into one touchstone s-parameter file.
> > Required because Hermes Lite only has one detector.
> >
> > Regards,
> >
> > Roderick Wall,  vk3yc
> >
>

[John Williams]

Roderick,

Do you know anything about the Dual TC-1 Bridge that is described on Sam
Wetterlin's web pages. It seems that the HPSDR folks have a design on
oshpark for that at a very reasonable $7 USD for 3 boards. To use the
bridge below, I would have to use ExpressPCB at a cost of around $90
USD, or have to purchase a program to convert the .PCB format into
gerbers for Oshpark at a cost of $49 USD. Is there any substantive
difference in usage for our rudimentary needs?

John - W9JSW

[rodwall1]

Hi John,

Thanks for your comments. I have sent them onto Dave as they may help him to write the Hermes interface for myVNA.

Thank for your comments,

Roderick Wall, vk3yc.

> > 73,
> >
> > Steve
> > KF7O
> >
> >

> > On Tuesday, June 9, 2015 at 8:05:56 AM UTC-7, Steve Haynal wrote:
> >>
> >> Hi Roderick,
> >>
> >> Sorry for the delayed response - there is a lot going on here at the
> >> moment. That is great that Dave is willing to add support to myVNA. I have
> >> yet to dig into the VNA portion of the RTL so don't yet have a good
> >> understanding of how the Hermes people implemented the VNA interface. I will
> >> in a few weeks. As a first cut, I don't plan to do anything different than
> >> what has been done in the Hermes. Given this, and that there are many Hermes
> >> out there, Dave might want to start by supporting the Hermes. Perhaps people
> >> on the openhpsdr list can help, or the author of HamVNA. Links to the Hermes
> >> protocol can be found at
> >> https://github.com/softerhardware/Hermes-Lite/wiki/Protocol-Coverage.
> >>
> >> I'll be happy to help in a few weeks once I learn enough to offer proper
> >> help!
> >>

> >> 73,
> >>
> >> Steve
> >> KF7O

[Steve Haynal]

Hi List,

Attached is a VNA plot of my qrp-labs 20M filter made with a Hermes-Lite and the Ham VNA software. One interesting thing to note is that just before 25 MHz and at -50 dB I run out of dynamic range and the readings are not valid from 25 MHz and up. I'm tweaking the settings and my test fixture to hopefully get in the expected 60-65dB dynamic range. Even still, with a 12-bit ADC, we will have to switch in the LNA with some gain at various points to extend the dynamic range.

Rod - I will send notes on what I learned in the next few days to help with myVNA porting.

73,

Steve

KF7O

[rodwall1]

Hi Steve,

OK on dynamic range etc.

And great on the information for Dave's myVNA interface.

Roderick vk3yc.

[Steve Haynal]

Hi Rod,

I updated the Protocol Coverage/Software Development wiki page with a section on VNA functionality. This page includes links to all the protocol documents that Dave will need. I found the source code to the OpenHPSDR VNA application (also linked on the wiki) to be the most helpful in understanding what is going on with VNA functionality. Dave may be able to just reuse the ethernet interface from that code.

I hope to finish up and release the new firmware this weekend. It will support the two existing Hermes VNA applications.

73,

Steve

KF7O

[rodwall1]

Hi Steve,

Thanks for the VNA operation information. I have forwarded this onto Dave. OK on the firmware support for VNA operation.

I will let you all know the status as progress is made.

Roderick, vk3yc.

--