apache-labs.com

UPDATE 12.1.8 1423 Aug 2021

DAC clock/data timing improvement.



Quartus project archive files for developers:

UPDATE 12.1.6 14 Aug 2021

Numerous bug fixes and timing improvements.
Thetis 2.8.11 build 21f will show the full firmware version number in Setup > General.



UPDATE 12.1_pre4 10 Feb 2020

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



Previous versions (the cryptic numbers at the end represent different Gigabit Ethernet PHY timing adjustments:








And even older versions here:

https://github.com/TAPR/OpenHPSDR-Firmware/tree/master/Protocol%202/Angelia%20(ANAN-100D)

Great, tnx working fine.

wa1oxt / / gary

Pure signal correcting, not stable. Back with the version Angelia_Protocol_2_v11.7_7-15-18.rbf
The only one that works 95% with my device and those of my friends.
No version 12 is working yet

Well guy's , after several day's I also found PS not stable with latest firmware.
Went back to PC2_ver 12.1_pre2_5550.rbf. firmware this morning and now PS is stable.

wa1oxt // garyradio

Hi Folks,

I am new to Anans. I acquired a 100D from the estate of a SK. I
would like to use P2 FW on it so that I can run Thetis SW. However I
am not sure whether my radio will successfully run P2 FW. I would
appreciate any guidance.

I loaded Metis_Angelia_Protocol_2_v12.1_pre4.rbf. I tried to use it
with Thetis. It worked for a short period but then Thetis lost
contact with the Anan. Subsequently Thetis would not maintain
communication with the Anan for anything but a very short period of
time.

At that point, I had to use HPSDRBootloader and a jumper to load other
FW.

Based on one of the entries above I tried
Angelia_Protocol_2_v11.7_7-15-18.rbf. It seems to be working though I
have only tried basic features.

However I found the text below at
OpenHPSDR-Firmware/Protocol 2/Angelia (ANAN-100D)/Release_notes_summary.txt
and it scares me.

"It should be noted that there is a potential concern for Angelia and
Hermes based radios (100D, 100B, 100, 10, 10E). With the Media Access
Controller (MAC) logic on the FPGA now running at Gigabit speeds this
causes the power consumption requirements of the FPGA to increase. On
these SDR cards, overheating of the "F3" positive temperature
coefficient (PTC) device which protects the 3.3V supply has been
reported, with resulting erratic operation due to current
limitations. It is not yet clear how significant this problem may
be. The development team has not, as of yet, looked into a possible
firmware based solution for this problem (e.g. creating 100BASE-T
versions of the Protocol 2 FPGA images). You may need to consider a
hardware modification to fully support Protocol 2 Gigabit operation."

MY QUESTIONS

Are there obvious indications of the current limiting issue mentioned
above?

Does running the MAC in the Angelia FPGA at a gigabit cause any
possible damage to the FPGA, some supply, or any other portion of the
Angelia?

I am not sure what is meant by the "hardware modification" mentioned
above. Has anyone done such a modification?

I guess the real question is whether it is safe for me to run P2 FW on
my 100D. If so, any recommendations on which version will be the
safest?

Best Wishes,
Andy Boughton
kc1dmm

Andy--see here: viewtopic.php?f=17&t=2777&p=8593

Hi Scott,

Thanks so much!

My symptoms seem similar to those of Bob, ko2f. I will order a 4A fuse though it sounds like the fuse may not be my problem. As I mentioned, I found a FW version that seems to work for me but I will also try the version that works for Bob since it is newer.

Best Wishes,
Andy
kc1dmm

This version has been working very well for me except for one issue I am noticing. When the LPF "6m/ByPass on RX" setting is selected (Under General,HPF/LPF) I get significantly more power output when transmitting on CW. For instance, on 40 meters when deselected I get around 100 watts, when selected I get over 125 Watts. This is measured using the watt meter on my KPA-500 amplifier (in bypass mode). and similar results using the watt meter on my MFJ Tuner. I have deselected and selected it multiple times and the power output increase seems to follow the check mark. Not sure why this setting has an affect on power output, but it does seem consistent with the setting. This is measured when transmitting into a dummy load.


John KD8K

Hi Scott and other Folks,

I have tried a few things. I found that upgrading my ethernet switch
and using a newer PC were the most helpful. With those
Metis_Angelia_Protocol_2_v12.1_pre4.rbf is working well for me.

I do have a couple small issues. I started seeing some DHCP issues
after I upgraded my "ethernet switch". The Anan 100D does not always
get a good address using DHCP. After assigning a static address the
Anan worked fine. My new "ethernet switch" is actually a wifi
extender, Netgear EAX80, connected to my old Verizon FIOS router by
wifi. The DHCP server is running on the Verizon router. My windows10
laptop and my windows8 laptop have no apparent issues using DHCP while
connected to the EAX80. I realize that this is a very limited
description but does anyone have any thoughts about why the Anan might
have DHCP issues in this configuration.

The other issue is that Thetis eventually gives warnings about out of
order UDP packets. Is this an issue? I have been ignoring it under
the assumption that UDP packets are allowed to be reordered.

By the way, I did replace the fuse with the 4.5A Littelfuse
1210L450SLWR. That did no damage but otherwise did not seem
to affect anything.

Best Wishes,
Andy Boughton
kc1dmm

Andy,

That's an interesting observation, and one I had not heard before, about the spotty DHCP performance.

The explanation may be that the Ethernet MAC (media access controller) is created in the FPGA, and it has the toughest job of any part of the FPGA code because it has to run faster than anything else in order to keep up with the Gigabit data streams. Thus it has proven to be difficult to obtain reliable timing closure in this functional area such that a given firmware version will run well on all serial numbers. The firmware you are using may therefore be not entirely reliable. However, going to a static IP is a good work-around.

All of the data that goes back and forth between the PC and the hardware unit is done via UDP. UDP packets that are out of order are not reordered and are lost. Lost packets are not re-transmitted. They are not desirable, but since UDP is not a lossless protocol like TCP/IP you have to expect a certain number of out of order or lost packets. UDP packets are only an issue if you are noticing a performance degradation: the littlest being audio glitches, the worst being Thetis crashing, that sort of thing.

73,

Scott

v12.1_pre4.rbf is stable for past 24 hrs on my 100D, no fuse mods. PS does appear to be broken.

Hello,

who writes the firmware for Angelia?
Do you have the sources of the latest firmware somewhere?

I want to try to build in the band filter control of OrionII, so that you have a separate RX filter for both receivers, like on the Orion MK2.

Maybe the manufacturer of the firmware could give me some hints.

best regards René

The firmware is developed on an open source basis, by a very, VERY small number of volunteer firmware developers. It does not come from Apache Labs. It is not "manufactured".

Read the three tacked topics here: viewforum.php?f=18

If you are looking for the latest Protocol 1 firmware source code it is in the Github repo referenced in that tacked topics linked above.

If you are looking for the latest Protocol 2 beta source code you will most likely have to contact Rick, N1GP, via this forum. You will find a great many posts by him in the various Protocol 2 beta firmware topics.

The firmware development environment is Quartus Prime Lite, which you can download for free from the Intel site. Each project is backed up as a ".qar" file which can be imported into Quartus.

Hi all,

I've had great success with P2 on my 100D except for the 6M band. Output power is very inconsistent - during TUNE my external power meter will fluctuate and wander around. SWR is MUCH higher than it should be. 2:1 in thetis / p2 instead of 1:1 in mRX PS / p1 as observed by an external meter.

There is also low level broadband noise being transmitted during silent periods, again only on 6M. Maybe spurious signals which could explain the SWR increase. It goes up and down with the VAC TX level adjustment even though there's no signal coming in on the VAC. Typical symptoms of RF feedback.

All other bands work normally.

Switched back to Angelia 5.7 and 6M operation is 100% back to normal. Stable, predictable power output, no noise, and low SWR on 6M.

It might be doing it wrong on 6M, but it is otherwise doing it right.

Thetis and Protocol 2 has put control of the filters where it belongs: solely in the hands of Thetis. All this nonsense about "manual" vs. "firmware" was crazy (refer to viewtopic.php?f=6&t=2764)

As to where the problem might be, if it's not just a problem with settings, firmware or software (Thetis), it's hard to say.

Thanks, I'm guessing this may be a thetis issue. A couple Q's -

Was firmware filtering removed in the P2 firmware?

Was there a specific point during thetis' development where firmware control was removed?

Was firmware filtering removed in the P2 firmware?

This was already answered above, again, filtering is solely controlled by Thetis.

Was there a specific point during thetis' development where firmware control was removed?

The initial release.

Thanks for the info. I scanned the source code and do not see anything wrong with the filtering code. I believe it is working correctly.

I did some more tests but this time with an FFT scope on the output, using the TUNE button at 50 watts into a dummy load. All processing disabled in DIGU mode.

My original belief that this was a 6M-only issue was wrong. It gets better as you move up in the bands but it's always there.

Attached are my 10M results side by side. 6M is worse.

Elevated noise floors are a symptom of the firmware not making timing on your hardware. Try a different version of the P2 firmware. The latest is not the best for everyone.

Scott

I tried every available firmware from 11.9 and up.

11.9 - Clean RF, glitchy audio
12.0 - No RX or TX
12.1 Pre1 5070 - Clean RF, clean audio
12.1 Pre1 5171 - Spurious RF, glitchy audio
12.1 Pre1 5670 - Clean RF, clean audio
12.1 Pre1 5770 - Clean RF, clean audio
12.1 Pre2 - Spurious RF, clean audio
12.1 Pre2 5550 - Spurious RF, clean audio
12.1 Pre4 - Spurious RF, clean audio

So guys/gals, even if your radio seems to be working fine and you're getting good signal reports, you might want to check on a external scope that you aren't emitting out of band RF spurs. Puresignal seems to work fine on the "clean RF, clean audio" releases I listed above, and does not work on the spurious RF versions. So if your PS isn't working, you may be emitting spurious RF....

Scott, I'm willing to be a guinea pig if needed for any potential future releases.

~Randy K6RP

Glad you found a version that works for you, Randy. We all have to thank Rick, N1GP for taking the time and effort to create them and post them here.

Hi all, I am new to the world of Anan but have had fun over the last couple of week playing with a "new to me" 100de.
It came to me with Protocol 1 v5.8 firmware installed. I used if for a week with the last version of PowerSDR mRx to get a feel for the software and radio. No problems other than user errors caused by me not knowing my way around the software.
Wanting to try Thetis I upgraded to Protocol 2 with Firmware 12.1 pre1-5770, to my surprise everything seemed to work but I kept getting "seq=>2 (ddc0)" errors. So after looking around I was under the impression to try different firmware so I jumped up to the latest 12.1_pre4. This got rid of the seq errors but now I had "adc1 overload" all the time but only on 20m. I was also finding my Auto ATU was constantly returning on any band 20m and above. I thought it was an rf issue and spent some time going down that avenue. In the end I though I would go back to the initial protocol v1 and see if everything went away. It did, ATU stopped retuning every time I keyed up.
The new Thetis 2.8.7 has now been released so today have done more tests with this . Using Protocol 1 everything works fine so I thought I would try P2 again with 12.1_pre4. Same issue with constant ATU tuning but the "adc1 overload" didn't appear.
Having read a lot on this forum I decided to try P2 11.7_7-15-18 and everything appears to be fine. I am running 384000 sample rate and the only thing I can say is a had the "seq" error once while using FT8 via VAC.
Can anyone explain why my ATU behaved like it did on the other firmwares?
Also using Puresignal I see feedback in green but sometimes in red, never did in mRx. is this anything to worry about? where can I read more about the correcting and feedback warnings?


Tnx
Paul

If the other versions of firmware were not making timing closure on your hardware then it is possible that the RF waveform emitted during tuning could have been non-sinusoidal and thus confusing your tuner.

Before I answer the PureSignal question, are you using an external coupler with an external amplifier?

Thanks for getting back to me. No not using a coupler or amp with the radio. I did find a bit more info last night that explained the red feedback was due to not being enough drive for it to sense properly, but it didn't explain how to increase this.

As for the tuning issue, it did tune fine with the tune button. it was when transmitting on SSB it went nuts.

Thanks again Paul

Hello All,

Just some of my experiences with Angelia Firmware on my Anan 100D.

I have only just updated to Protocol 2, I found Protocol 1 reasonably stable and as I don't use multiple receivers, other than Rx 2, there seemed little point in updating. But as a trial being locked down with Covid, I decided to give Thetis a try, latest version.

I downloaded the firmware revisions from 11.6 through to the 12.1 to try and see. The first problem I had was a failure of the web based firmware programme which after a few updates bricked my radio, it shot off to 100 plus seconds to clear the firmware and just ran on and on.

Bootloader came to the rescue, much more reliable in operation, and the radio comes apart easily and not difficult to fit the jumper.

The tests, well a bit disappointing really, not sure what happened in the firmware development but it all seems to have gone astray after version 11.6, which is the only nearly reliable version on my Anan box.

12 was hopeless, no transmit facility at all, the different timing versions of 12.1 had their issues as well. Stuck with 11.6, and that has a habit of stopping when the box is cold, recycling the power brings it back. I appreciate this is not a very scientific report but it does highlight the fact that something happened after V 11.6 firmware. So possibly back to reliability and protocol 1 for me.

Thanks, Mike

The good news if that if you can't get P2 to run on your hardware you can now use Thetis with P1 starting with Thetis version 2.8.

w-u-2-o wrote:The good news if that if you can't get P2 to run on your hardware you can now use Thetis with P1 starting with Thetis version 2.8.

Interesting... Thanks

To keep up with news like that (the entire forum, really), you should look at the "Active Topics" link periodically:

search.php?search_id=active_topics

Also accessible via the "hamburger" menu at the top left of every page on the forum.

Hi,
is the Metis_Angelia_Protocol_2_v12.1_pre4.qar source code available somewhere?

thank you
73

Hello,

It's been a while since I've touched this code, but this is the latest I have circa pre4:
https://www.dropbox.com/s/9ubn7hxjq4v4glv/Angelia_Protocol_2_v12.1_pre4.qar?dl=0

73s

Rick / N1GP