apache-labs.com

This is for the 16-bit Hermes platforms only, which are the ANAN-10 and ANAN-100. This is NOT for the 14-bit Hermes platforms (10E, 100B).

UPDATE 10.8_pre20 25 Mar 2020

This update changes the timing such that it compiled to work with a cooler FPGA temperature range. RX1 and RX2 must be set to the same sample rate.



Quartus project archive file for developers:

Read the entire topic for other releases between pre6 and pre20, however most of those did not pan out and were not re-posted here


UPDATE 10.8_pre6 8 Mar 2020

Experimenting with timings (mostly ethernet phy) trying to get PS more stable on all bands.



UPDATE 10.8_pre5 12 Feb 2020

More bug fixes per the discussion below.



UPDATE 10.8_pre4 10 Feb 2020

Additional bug fixes including transverter operation, PTT fixes and timing adjustments.



Previous updates here:

https://github.com/TAPR/OpenHPSDR-Firmware/tree/master/Protocol%202/Hermes%20(ANAN-10%20and%20100)

w-u-2-o wrote:This is for the 16-bit Hermes platforms only, which are the ANAN-10 and ANAN-100. This is NOT for the 14-bit Hermes platforms (10E, 100B).

UPDATE 10.8_pre4 10 Feb 2020

Additional bug fixes including transverter operation, PTT fixes and timing adjustments.

Metis_Hermes_Protocol_2_v10.8_pre4.rbf

Previous updates here:

https://github.com/TAPR/OpenHPSDR-Firmware/tree/master/Protocol%202/Hermes%20(ANAN-10%20and%20100)

Thank you all for the firmware update I ‘m having two issues, version zero shows up in Thetis under options and also pure signal still in and out.

Jorge

Hi Jorge,

Could you try this updated FW. It fixes the FW version and has some timing changes
for PS:

https://www.dropbox.com/s/5ssofoegz0qcov6/Hermes_Protocol_2_v10.8_pre5.rbf?dl=0

Tnx

-Rick / N1GP

n1gp wrote:Hi Jorge,

Could you try this updated FW. It fixes the FW version and has some timing changes
for PS:

https://www.dropbox.com/s/5ssofoegz0qcov6/Hermes_Protocol_2_v10.8_pre5.rbf?dl=0

Tnx

-Rick / N1GP

Hi Rick,

The FW is now showing up correctly, PS is working much better but we still have some in and out.

Thanks for help.

Jorge kd4jm

Thanks for the new Version.
For me the firmware in Thetis 2.6.9 and 2.7.0 is correctly reported as 10.8.
No further tests have been carried out so far.

DH1KLM wrote:Thanks for the new Version.
For me the firmware in Thetis 2.6.9 and 2.7.0 is correctly reported as 10.8.
No further tests have been carried out so far.

Hi Sigi,
Have you tested to see if your pure signal is working correctly with this new version 10.8, thanks.

Jorge kd4jm

No further tests done so far. Hope for a little spare time on weekend.

Hi everybody,

Just flashed my ANAN 10 (16-bit version) to "10.8 pre" and tested under Thetis v2.6.9_c3.
In my case Pure Signal did not much improve (if any) in "Auto Mode".
Unfortunately it's practically still unusable except I'm using "Single Cal" (as done before)

Any advise to get things improved at my end please let me know.
I am more than willing to help and assist with any kind of tests if even possible.

Anyway - it is a great step forward and it is more than appreciated that the firmware is further under construction….after a loooong time.
Many thanks to all who are and who were involved!
Please don't give up…...think the installed base is huge.

Once again thanks and 73,

Andy

Sri - need to add I tested 10.8 pre4 (will do pre 5 tomorrow)

Tnx, Andy

Link to 10.8 pre5 added to the first post.

Quick feedback after an 1 hour test of firmware 10.8 pre 5:

Pure Signal is way more stabil - comes even back after auto disabling.
In about 70% of all tx periods it worked good and stayed alive - sometimes it switches off.....still a bit instabil.
This version (10.8 pre 5) does actually display the firmware version 10.8 correctly - the former 10.8 pre 4 didn't….blank field.

Using an ANAN 10 (16-bit) here with external PA. Feedback level switches the ANAN" att to around 16 dB (build-in att).
The same PA works next to perfect on my ANAN 7000.
Single cal works all the time as usual.

Many thanks for getting the firmware to this latest level - big improvement! Not perfect but already a lot better than before.

73,

Andy

Hello everybody,
I have been using the Hermes Board with PureSignal for many years.
Under protocol 1 P.S runs also 100% !

Now I use the following configuration:
HW: Hermes Board
SW: Thetis v.2.6.9 (c3)
FW: 10.8 pre 5

Pure signal is not yet stable. I guess 50% TX is okay and 50% is not okay.
I noticed that if P.S is not OK "GetPk" and "SetPk" are not the same.
"GetPk" is then half of "SetPk".
(-> see Calibration Infomation)

Thank you very much for the further development of the FW for Hermes !

73, DL5AI
Arno

hello,
after update to 10.8 pre. 5 firmware Pure signal is a little bit more stable as before but sometimes restarting self.
working arount 60 percent of the time.
a tip for all who have no change to version Firmware 10.7 .... please reset your database !
best regards and thank you !
Tom

Thanks so much for your hard work on this !! I am running ANAN 100, just installed 10.8 pre5. The keyboard space bar control of the toggled function (PTT/MOX, VOX etc) is not working. Is this to be expected with this version of firmware?

Running Thetis v2.6.9 (1/24/20)

Thanks,

Jeff
N8RWS

Whoops... I did not have the keyboard shortcuts enabled... sorry

73

Jeff

I've been experimenting with timings (mostly ethernet phy) trying to get PS more stable on all
bands. This one is working pretty well from 160-6 mtrs on my ANAN-10.

Please give it a try and report back when you have a chance:

https://www.dropbox.com/s/1ea2d62xlye4f1k/Hermes_Protocol_2_v10.8_pre6.rbf?dl=0

-Rick / N1GP

n1gp wrote:I've been experimenting with timings (mostly ethernet phy) trying to get PS more stable on all
bands. This one is working pretty well from 160-6 mtrs on my ANAN-10.

Please give it a try and report back when you have a chance:

https://www.dropbox.com/s/1ea2d62xlye4f1k/Hermes_Protocol_2_v10.8_pre6.rbf?dl=0

-Rick / N1GP

Hi Rick,

Thanks for the update, but I'm still experiencing in/out on PS on all bands.

Jorge kd4jm

Hi Jorge,

Tnx for giving it a try.

Here's another one to try. This time I've changed the clock feeding the DAC.
It's no longer 90 degrees out of phase with the ADC clock. Still working for me
but I do notice that PS is locked in better:

https://www.dropbox.com/s/082gk2ov37egglj/Hermes_Protocol_2_v10.8_pre7.rbf?dl=0

-Rick / N1GP

n1gp wrote:Hi Jorge,

Tnx for giving it a try.

Here's another one to try. This time I've changed the clock feeding the DAC.
It's no longer 90 degrees out of phase with the ADC clock. Still working for me
but I do notice that PS is locked in better:

https://www.dropbox.com/s/082gk2ov37egglj/Hermes_Protocol_2_v10.8_pre7.rbf?dl=0

-Rick / N1GP

n1gp wrote:Hi Jorge,

Tnx for giving it a try.

Here's another one to try. This time I've changed the clock feeding the DAC.
It's no longer 90 degrees out of phase with the ADC clock. Still working for me
but I do notice that PS is locked in better:

https://www.dropbox.com/s/082gk2ov37egglj/Hermes_Protocol_2_v10.8_pre7.rbf?dl=0

-Rick / N1GP

Hi Rick,
Thanks for working on this pre-7update, PS is working about the same in/out, what I do notices now I’m getting some SEQ errors.
I am running Thetis 2.7.0 (a2) I did not notices any SEQ errors on pre-6, Hopefully will get more folks to try this new release and get some feedback, I really appreciate all your efforts.

Jorge kd4jm

Hello
and ... wow ... Pure Signal working much much better as before.
thats the right way now !
in the moment not really time for more testing here ... but thats a short report.
many thanks for your work !
regards and best wishes Tom

@DLDXer

Which version is working well for you?

Hi
I´m using Thetis 2.6.9 C3 and Firmware 10.8 pre7 now and Pure Signal is much more workable as before.
I mean its mostly working right here thinging around 75 percent of the transmitting time.
sometimes it does not want to start, but then starts again when the next PTT is pressed. So I'm sure it works better than before.
But it is still not as 100% as with PowerSDR and protocol one.
73s Tom

Thanks for all the work !!

I am running Thetis v2.6.9 ANAN 100

With 10.8 pre6 PS will lock in and stay locked for several seconds then the "feedback" will stop and soon after that "correction" will stop. After I unkey and re key the cycle repeats.

With 10.8 pre7 PS locks in but there is an elevated splatter 40 kHz wide (see image). PS remains locked for several seconds then it unlocks and at this point I hear distortion in monitor and very high splatter 40kHz wide.

Thanks again for the work.

73
Jeff
N8RWS

This firmware image changes the DAC clock to 45 degrees of the ADC.

The last image changed from 90 degrees to zero, and seemed to make a difference towards PS
being more stable.

Just getting some data points, tnx for trying and reporting back!

https://www.dropbox.com/s/oxylrph2ggi4yks/Hermes_Protocol_2_v10.8_pre8.rbf?dl=0

This one narrows a gap in the mux timing between tx & rx during PS. All feedback much appreciated!

https://www.dropbox.com/s/wz3d8ond15oea2z/Hermes_Protocol_2_v10.8_pre9.rbf?dl=0

-Rick / N1GP

n1gp wrote:This firmware image changes the DAC clock to 45 degrees of the ADC.

The last image changed from 90 degrees to zero, and seemed to make a difference towards PS
being more stable.

Just getting some data points, tnx for trying and reporting back!

https://www.dropbox.com/s/oxylrph2ggi4yks/Hermes_Protocol_2_v10.8_pre8.rbf?dl=0

This one narrows a gap in the mux timing between tx & rx during PS. All feedback much appreciated!

https://www.dropbox.com/s/wz3d8ond15oea2z/Hermes_Protocol_2_v10.8_pre9.rbf?dl=0

-Rick / N1GP

Hi Rick
I’m running Thetis 2.7.0 a2 Anan 100: FW 10.8pre9 Big improvement PS working now 90% of the time.
Thanks for all your work it means a lot to us for taking our feedback.

Jorge kd4jm

Hi all,

Just tested "pre 6" on my ANAN 10 (16 bit version):

Pure signal wasn't really much better here - but "get signal" and "set signal" showed the correct level now at least half of the time.
I observed a total loss of signal recognition several times - so "correcting" and "corrected" disappeared totally.
Once pushing the PTT again it changes very often and Pure signal came back on......but even when activating the 2-tone test (while PS is on) it happens often that there is still no correction…..so I had to restart 2-tone test several times to get it back on.

Summarizing: No big changes on my ANAN 10.....no stable recognition of the feedback signal.
Output was about 1 KW (incl. the amp) and att set automatically to 19 dB....a good value right in the mid range I believe.

Any recommendation what I could try or do any better?

At this point: "Many, many thanks for all the work on that new F/W release! Please keep on going!

73,

Andy

Andy: try pre6 through pre9 and report which works best for you.

Hi all - hi Scott,

Realized a bit too late that there are more versions to test. So I went back for another hour testing in my lab.
In direct comparison I would choose pre 9 (tested pre4 - pre9)…..

Pre 9 is may be slightly more stable - the cleaning process is better (and way better than pre4). Worst was pre 8.

But here is the problem: Open the Linearity menu - then watch the GetPk and SetPk numbers.
In all FWs the GetPk number gets lost from time to time - sometimes it's not even recognizing any value from the beginning.
It looks like no feedback cable is connected…..but doing several transmissions it finds the feedback again and works again….
If I should guess what's the problem I would say that the FW is not continuously monitoring the feedback - it jumps off and on.
Once the feedback signal gets lost Pure Signal stops correcting - no wonder.

I tried to analyze 2 things:
1) How often does the signal gets lost - (correction stops)
2) How good is the cleaning process - how about resulting IMDs - yes there are differences

There are differences in both.
Regarding 2) I believe pre9 is pretty good - if not excellent
Regarding 1) there is still a major problem. If the signal gets lost the GetPk value stops at about 0,14......working values are 100% higher at about 0,29.

Don't know if this helps - regarding that 1)-issue I do NOT see any major difference comparing all pre versions - but they are all better than the older versions before. Think there is still this one bug - may be the monitoring function of the feedback signal needs just a higher priority?

In many cases I observed the following scenario:

Feedback gets lost - release Ptt - push Ptt again, PS is back on.....but in some rare cases this doesn't help either.
Enabling PS-A sometimes needs the 2-tone test…..it appears to me that the 2-tone test pushes harder and the correction process kicks in.

Sorry to say but it's still kind of unstable operation in PS-A.....may be the real reason for that strange behavior has not been touched so far.

Finally, I am more than happy to continue with further tests - so please don NOT stop working on this.
It might only be peanuts once you adjusted that one hidden screw….

I used again my ANAN-10 (16-bit Blue box) on my i7 NUC / 32 GB.

Thanks again for all your efforts!! Let me know if you need me to test anything….

Andy

Hello Scott
Hello Rick
Hello everybody

I use the following configuration:
HW: Hermes Board
SW: Thetis v2.7.0. (A2)
FW: v10.8 pre9

I can confirm what Andy (DL6EAT) described.
I don't need to repeat it completely.
Pure signal runs better, but not yet stable!

As I reported earlier on ^^ "Thu Feb. 13, 2020 7:12 pm", I also see a problem with "GetPk" and "Set PK".
However, I do not know the exact function - so it should only be an indication.

I am looking forward to a new FW version for Hermes and would be happy to test and report everything.

I cannot say it often enough: Thank you very much for the work you have done!

73, DL5AI
Arno

Tnx guys for your testing a great feedback.

pre9 works flawlessly on my ANAN-10. I can't make PS misbehave!

This next test image I added a reset to the rx0 & rx1 fifos on a transition of PTT with the
thought that rx0 & rx1 may get out of sync on PTT transitions.

Purely experimental, but looking forward to see if this may make PS more stable:

https://www.dropbox.com/s/9b9b2kdnh1wmyhh/Hermes_Protocol_2_v10.8_pre10.rbf?dl=0

-Rick / N1GP