Hermes Protocol 2 firmware v10.3 released

FIRMWARE TOPICS ONLY--non-firmware topics will be MOVED
Forum rules
Until such time as the New Protocol firmware goes into general release, all discussion will be concentrated here.
K5SO
Posts: 12
Joined: Mon Apr 17, 2017 2:20 pm

Hermes Protocol 2 firmware v10.3 released

Postby K5SO » Thu Apr 27, 2017 3:31 am

All,

Hermes_Protocol_2_v10.3 firmware is available at

https://github.com/TAPR/OpenHPSDR-Firmware

This version of protocol 2 firmware restricts the ethernet connection speed to 100Mbps because of two issues that are unique to the Hermes board:

1) on TAPR produced Hermes boards the FPGA exceeds available power that is available from the SMPS source when running at gigabit rates
2) on both TAPR produced and Apache Labs produced Hermes boards the resettable fuses that power the FPGA have voltage drops across them that result in marginal (i.e., below FPGA spec) voltage to the FPGA resulting in issues when running at gigabit data rates to the ethernet connections.

This firmware runs fine on my standalone Hermes board, including the PureSignal features and firmware peak detect functions for FWD_PWR (AIN1) and REV_PWR (AIN2), at sampling rates of 192 ksps and lower.

Recognizing that some Hermes owners have modified their Hermes boards to be able to use external 5v power supplies instead of the on-board SMPS and have replaced their resettable fuses with higher current varieties, I am working on a 1000T version of the firmware so that those users may be able to run at the 1 gigbit data rates and will post it to the repository when/if I am successful in getting it operational. We appreciate your patience as we work through the various and sometimes unique Protocol 2 issues on this platform. This initial Protocol 2 firmware release should suffice for those wishing to test Protocol 2 firmware on their Hermes-based radios, I think.

73, Joe K5SO
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby Tony EI7BMB » Thu Apr 27, 2017 8:52 am

Thanks Joe. Do you (or Simon perhaps) happen to know if the lower data rate will work with SDR console ?
User avatar
Simon
Posts: 40
Joined: Sun Apr 09, 2017 5:19 pm
Location: Mawnan Smith
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby Simon » Thu Apr 27, 2017 9:23 am

Tony EI7BMB wrote:Thanks Joe. Do you (or Simon perhaps) happen to know if the lower data rate will work with SDR console ?


I don't see any reason why not. I have not downloaded this firmware as I'm happy with the version I'm currently using.
sdr-radio.com
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby Tony EI7BMB » Thu Apr 27, 2017 10:07 am

Thanks Simon
User avatar
W1AEX
Posts: 425
Joined: Sun Apr 09, 2017 6:17 pm
Location: Connecticut, USA
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby W1AEX » Thu Apr 27, 2017 4:46 pm

Thank you Joe!

I installed v10.3 and THETIS v.2.3.11 came up fine with my ANAN-100. After making a few TX profiles for SSB and AM I transmitted with all of them into the dummy load and they all sound fine. I also ran Pure Signal and it is correcting reliably and behaving perfectly. I am running it with RX1 and RX2 set for a 192K sampling rate as you suggested it should it be at this time and so far everything has been running smoothly for the past hour.

At this time I'm listening to WCBS in NY while I am munching lunch and also watching the WB screen. Very nice!

So far so good!

73,

Rob W1AEX

Image
Last edited by W1AEX on Thu Apr 27, 2017 5:00 pm, edited 1 time in total.
"One thing I am certain of is that there is too much certainty in the world."
Walter
Posts: 7
Joined: Sun Apr 09, 2017 8:52 pm

Re: Hermes Protocol 2 firmware v10.3 released

Postby Walter » Thu Apr 27, 2017 4:58 pm

Hello all,
Do I need to use Bootloader to install this firmware in myAnan 100. Also will I have to use bootloader for future protocol 2 firmware? Or once the protocol 2 firmware is installed can I use programmer from then on.Thank you for your time ,Walter kj6gea
User avatar
W1AEX
Posts: 425
Joined: Sun Apr 09, 2017 6:17 pm
Location: Connecticut, USA
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby W1AEX » Thu Apr 27, 2017 5:19 pm

Hi Walter,

I have read that you can use HPSDR Programmer to load the Protocol 2 firmware but that HPSDR Bootloader is necessary to go back to the Protocol 1 firmware. I would definitely make sure that HPSDR Bootloader is working for you before playing around with Protocol 2 since THETIS might not be ready for everyday use!

73,

Rob W1AEX
"One thing I am certain of is that there is too much certainty in the world."
Walter
Posts: 7
Joined: Sun Apr 09, 2017 8:52 pm

Re: Hermes Protocol 2 firmware v10.3 released

Postby Walter » Thu Apr 27, 2017 5:29 pm

Thank you Rob for your quick post :D I will wait until the bugs are worked out then move forward. This radio already exceeds my expectations...73's walter
PS. I really enjoy your videos and your doucumentations , you and Scott have made this radio extremely fun.
w9mdb
Posts: 446
Joined: Sun Apr 09, 2017 5:53 pm

Re: Hermes Protocol 2 firmware v10.3 released

Postby w9mdb » Thu Apr 27, 2017 8:23 pm

I loaded Thetis 2.3.11 and this release of V2.
I use WSJT-X pretty much exclusively.
Receive seems to be going well.

But ...no transmit.

TX Meter is set to Fwd Pwr.
When I press TUN on Thetis the Tx Meter says a constant 42W regardless of the Drive setting. But...my wattmeter does show the power adjusting based on the Drive setting. And the display shows an adjusting signal also.
When I try and transmit from WSJT-X I also see 42W but no power comes out at all. So it appears the audio stream is not being passed correctly.

This, of course, all works fine on PowerSDR and the previous firmware.

de Mike W9MDB
Mike W9MDB
User avatar
W1AEX
Posts: 425
Joined: Sun Apr 09, 2017 6:17 pm
Location: Connecticut, USA
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby W1AEX » Thu Apr 27, 2017 9:07 pm

Mike,

I noticed the same FWD PWR behavior with the TX meter when I began to go through the band-by-band PA Settings. After I finished up setting all the bands to 100W output with my external meter the TX meter was sticky at 108W even with the rig idling at no output. I think I saw on the GitHub site a note by Joe that they are experimenting with peak detection for FWD PWR and REV PWR so it's probably a work in progress with each hardware version they are dealing with. That's my best guess anyway!

73,

Rob W1AEX
"One thing I am certain of is that there is too much certainty in the world."
User avatar
Mike_N1JEZ
Posts: 37
Joined: Sun Apr 09, 2017 6:05 pm

Re: Hermes Protocol 2 firmware v10.3 released

Postby Mike_N1JEZ » Fri Apr 28, 2017 10:52 am

I just loaded firmware 10.3 on my bare Hermes board using KV0S's Web based programmer. It went smoothly and Thetis appears to run fine. My PiHPSDR controller also seems OK. Now for more exhaustive testing.

Mike
73,
Mike, N1JEZ
"A closed mouth gathers no feet"
User avatar
DL1OF
Posts: 28
Joined: Tue Apr 11, 2017 12:38 pm

Re: Hermes Protocol 2 firmware v10.3 released

Postby DL1OF » Sun Apr 30, 2017 11:27 am

Hi Group,

does that issue mean for those, who aren't able to run their Hermes based Transceivers with additional 5Volt supply will never be able to make use of a 1 GBit protocol?
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby Tony EI7BMB » Sun Apr 30, 2017 12:30 pm

I was wondering that myself. If its not possible then perhaps "stitched receivers" could be enabled in Thetis as a work around
User avatar
w-u-2-o
Posts: 5525
Joined: Fri Mar 10, 2017 1:47 pm

Re: Hermes Protocol 2 firmware v10.3 released

Postby w-u-2-o » Sun Apr 30, 2017 1:17 pm

Because the new protocol is designed to run 384, 768 and 1536KHz sample rates, the stitched receiver option was removed from Thetis.

It makes no difference anyway, since two stitched receivers at 192KHz equates to the bandwidth of a single receiver at 384KHz, etc.

As Joe alludes to above, I suspect that ultimately two versions of the new protocol firmware will be necessary for Hermes based radios, one running at 100Mbit/s, the other at 1Gbit/s. Obviously the 100Mbit/s version will have all of the usual bandwidth restrictions, e.g. PureSignal at 192KHz only, etc.

73,

Scott
w9mdb
Posts: 446
Joined: Sun Apr 09, 2017 5:53 pm

Re: Hermes Protocol 2 firmware v10.3 released

Postby w9mdb » Sun Apr 30, 2017 1:19 pm

What is the higher bandwidth good for?
Trying to figure out if I should check my ANAN 100 for upgrading the power supply.

de Mike W9MDB
Mike W9MDB
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby Tony EI7BMB » Sun Apr 30, 2017 1:33 pm

I understand why it was removed Scott but if I run at 192 as at present with Power sdr will I not lose some functionality with Thetis? For example on 6m band I can see over 500KHZ with "limit stichted Receivers" unchecked, when its checked i can see only 160khz or so.
User avatar
w-u-2-o
Posts: 5525
Joined: Fri Mar 10, 2017 1:47 pm

Re: Hermes Protocol 2 firmware v10.3 released

Postby w-u-2-o » Sun Apr 30, 2017 4:21 pm

Tony EI7BMB wrote:I understand why it was removed Scott but if I run at 192 as at present with Power sdr will I not lose some functionality with Thetis? For example on 6m band I can see over 500KHZ with "limit stichted Receivers" unchecked, when its checked i can see only 160khz or so.
You should be able to see all 192KHz with the zoom all the way out.

Although it remains to be seen, you shouldn't lose any functionality. Just run RX1 at 384 or 768KHz and don't use PureSignal. Same as stitched receivers, essentially.

73,

Scott

P.S. Mike--more bandwidth can be handy for monitoring bands that are fairly large in extent, like 40, 15, 10 and 6M. It's nice to be able to see the entire band, or at least an entire sub-band such as the phone portion, all at once. This allows quicker scanning of the entire band, whether it's for rag chew, DX, contesting or whatever. I normally operate with an entire phone sub-band up on the display and CTUN turned on. Then I work my way from one end to the other.
User avatar
W1AEX
Posts: 425
Joined: Sun Apr 09, 2017 6:17 pm
Location: Connecticut, USA
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby W1AEX » Sun Apr 30, 2017 9:02 pm

With two different kinds of 16-bit Hermes boards out there it gets a little complicated. The original Hermes designed and produced by TAPR and the later Hermes designed for and produced by Apache Labs are not exactly the same.

If you have the original Hermes by TAPR you would need to jumper it to use an external 5v supply rather than the onboard SMPS supply that is on the Hermes board. Then, as I understand it, you would need to replace the 3 resettable fuses with higher rated devices.

If you have the Apache Labs ANAN version of Hermes Abhi has stated several times that the onboard SMPS supply can handle the supply demands of the 1G protocol and if you are running the ANAN-100 you already have a separate 5v supply mounted on the base plate. Again, Abhi has indicated that it can handle the current demands of the 1G protocol. That leaves replacement of the 3 resettable fuses with F1 and F2 mounted on the top of the board and F3 mounted on the bottom of the board. I'm uncertain if all 3 would need to be replaced or if only F3 is involved but it looks like a fairly simple job to lift them off and replace them and it appears that Mouser (and probably Digikey) carry them. It would be interesting to know what rating of resettable fuses were installed on the Angelia and Orion boards as they appear to be fine.

At any rate, I appreciate the 100 mbps version that Joe released to allow us to get in the game! I'm unclear about whether or not it will be all right to load the 1000T version as long as you use the 48/96/192 sampling rates. Perhaps Joe, Phil, or someone else who has done the modification(s) can clarify all of this for us.

73,

Rob W1AEX

I found the first two images below on the Apache Labs Facebook page and I added the F1, F2, F3 labels with Photoshop. To the best of my knowledge these are the top and bottom of the ANAN Hermes board.

Image
Image

I found the two images of the TAPR version of Hermes using a Google images search and I added the F1, F2, F3 labels with Photoshop. To the best of my knowledge these images show the top and bottom of the TAPR Hermes board.

Image
Image
"One thing I am certain of is that there is too much certainty in the world."
VE3YTZ
Posts: 4
Joined: Mon Apr 24, 2017 12:28 pm

Re: Hermes Protocol 2 firmware v10.3 released

Postby VE3YTZ » Sun Apr 30, 2017 9:47 pm

Totally excellent post, Rob. Extremely informative. Thank you.

I know its a lot to ask, but to simplify things and to avoid errors by me and others, if anyone could provide the Mouser or Digikey part numbers for replacements of F1 through F3, that would be totally awesome.

Thanks in advance

Greg/VE3YTZ
User avatar
W1AEX
Posts: 425
Joined: Sun Apr 09, 2017 6:17 pm
Location: Connecticut, USA
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby W1AEX » Sun Apr 30, 2017 11:04 pm

Greg,

Keep in mind that my assumptions in the post above are based solely on conclusions I have drawn from all kinds of information and probably some misinformation so I would definitely wait for more information from someone who actually knows! I would hope that Apache Labs would have some recourse for the ANAN Hermes owners as the boards have been advertised as 1G capable for several years now. If it turns out they are not due to a component limitation seems to me like it should be addressed. At any rate, the link to the PDF posted by Scott in the first message of this thread has a complete rundown on possible replacements for the resettable fuses:

https://apache-labs.com/community/viewtopic.php?f=17&t=2354

At the bottom of the PDF are links to Mouser for each of the Bournes resettable fuse devices. I suppose it would be a matter of debate as far as which would be the best choice for replacements. Hopefully, someone will have some ideas on that.

73,

Rob W1AEX
"One thing I am certain of is that there is too much certainty in the world."
VE3YTZ
Posts: 4
Joined: Mon Apr 24, 2017 12:28 pm

Re: Hermes Protocol 2 firmware v10.3 released

Postby VE3YTZ » Mon May 01, 2017 2:19 am

Thanks again Rob and I understand where you are coming from. Also, thanks for the link to Scott's post.

I will continue to monitor the forum and once all the info seems in, I'll proceed with a with a fuse replacement if it seems warranted.

73

Greg/VE3YTZ
User avatar
DL1OF
Posts: 28
Joined: Tue Apr 11, 2017 12:38 pm

Re: Hermes Protocol 2 firmware v10.3 released

Postby DL1OF » Mon May 01, 2017 10:24 am

W1AEX wrote:If you have the Apache Labs ANAN version of Hermes Abhi has stated several times that the onboard SMPS supply can handle the supply demands of the 1G protocol and if you are running the ANAN-100 you already have a separate 5v supply mounted on the base plate. Again, Abhi has indicated that it can handle the current demands of the 1G protocol.

That is good to know. Never took a closer look.

W1AEX wrote:That leaves replacement of the 3 resettable fuses with F1 and F2 mounted on the top of the board and F3 mounted on the bottom of the board. I'm uncertain if all 3 would need to be replaced or if only F3 is involved but it looks like a fairly simple job to lift them off and replace them and it appears that Mouser (and probably Digikey) carry them.

Doesn't sound to complicate but still customer have to open their radios and do the mod. to get the Gbit capability! And those who aren't able to do that? No reprimand, just a thought!
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby Tony EI7BMB » Mon May 01, 2017 11:43 am

Scott thanks again for the info. I'll try running at 384
User avatar
w-u-2-o
Posts: 5525
Joined: Fri Mar 10, 2017 1:47 pm

Re: Hermes Protocol 2 firmware v10.3 released

Postby w-u-2-o » Mon May 01, 2017 11:44 am

DL1OF wrote:Doesn't sound to complicate but still customer have to open their radios and do the mod. to get the Gbit capability! And those who aren't able to do that? No reprimand, just a thought!
Not much you can do about that! The hardware was designed for the software/firmware that was available at the time, and is the oldest variant. No doubt no one had thought about GigE rates back then, nor their demands on the hardware. The dev's have seen fit to make available a 100Base-T variant of the code for that hardware, which is mighty nice of them because every extra build is a lot of work! Although it's a good trade-off in that by doing this ultimately only Thetis will need to be supported as a client.

73!

Scott
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby Tony EI7BMB » Mon May 01, 2017 11:53 am

Indeed yes, I'm disappointed that Apache had not addressed this before I bought my radio as the Gigabit protocol was known about at the time.

DL1OF wrote:Doesn't sound to complicate but still customer have to open their radios and do the mod. to get the Gbit capability! And those who aren't able to do that? No reprimand, just a thought!
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby Tony EI7BMB » Mon May 01, 2017 12:01 pm

On the plus side too we will be able to use Simons SDR console as well as Thetis.
User avatar
W1AEX
Posts: 425
Joined: Sun Apr 09, 2017 6:17 pm
Location: Connecticut, USA
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby W1AEX » Mon May 01, 2017 4:27 pm

I would have to say that historically Apache Labs has been very honest and honorable with their customer base regarding support for the ANAN product line. If it turns out that there is a component deficiency that limits the Hermes based rigs to less than gigabit capability they will make it right. Note that the ANAN-10 and ANAN-100 have always been advertised as being gigabit capable as can be seen in the "Specifications and Highlights" section of the PDF brochure at the link below:

https://apache-labs.com/download_file.php?downloads_id=1001

It looks like about a 5 minute de-soldering and installation job to replace F-3 with whatever rating fuse would be appropriate to prevent the voltage drop to the FPGA. Of course it's about 30 minutes to get into the rig and 30 minutes to put it all back together, but that's part of the fun. After studying the specs it looks like this one (46 cents at Mouser) might be appropriate as the hold current is probably high enough although the trip current is rather high:

http://www.mouser.com/ProductDetail/Bourns/MF-MSMF260-2/?qs=%2fha2pyFaduiziRYNByQQcFVDHuBm7rGWll5yqyjpEdsveN3xXFGmXA%3d%3d

Then again, as John-AC0ZG mentioned in the F3 thread, when the trip current of the replacement re-settable fuse is +4A (or higher) is this any better than simply jumpering the existing F3 with a wire? I'd be very curious to know what was used in Angelia for the Cyclone IV FPGA.

73,

Rob W1AEX
"One thing I am certain of is that there is too much certainty in the world."
User avatar
W1AEX
Posts: 425
Joined: Sun Apr 09, 2017 6:17 pm
Location: Connecticut, USA
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby W1AEX » Mon May 01, 2017 7:06 pm

I downloaded the Hermes and Angelia parts lists ("Hermes production BOM" and the "Angelia production BOM") at the links below:

https://apache-labs.com/downloadsfiles/1014_HERMES_PROD_BOM_08112012.xls
Image

https://apache-labs.com/downloadsfiles/1010_ANGELIA_BOM.xls
Image

When I pasted the F3 part numbers from the Hermes and Angelia boards into the DigiKey search field it brought up the specs for the fuses used:

https://www.digikey.com/product-detail/en/bourns-inc/MF-MSMF110-16-2/MF-MSMF110-16-2CT-ND/662846 Hermes F3
https://www.digikey.com/products/en?keywords=507-1513-1-nd

As previously mentioned by the person that Scott posted for in the "Hermes F3 wrong part" thread, the Hermes F3 resettable fuse has a current hold=1.1A and current trip=2.2A.

What I found with the Angelia board (Cyclone IV FPGA), the F3 resettable fuse has a current hold=3A and current trip=5.0A

I would expect that the current demands of the Cyclone IV of the Angelia would be greater than the Cyclone III used in the Hermes but at 1Gbps I'll bet the Cyclone III would probably need an F3 hold current pretty close to 3A to avoid operating in its transition zone where heat and resultant increased resistance becomes a factor that results in voltage drop and FPGA instability. Just speculation on my part but it seems reasonable. At any rate, it will be interesting to see what the final recommendations are for this issue.

Note that these parts are listed at Digikey in a quantity of 1 at a cost of around 50 cents +shipping. Also note that the Angelia F3 part is listed as being obsolete but browsing the Bel Fuse and Bourns resettable fuses there are others with the same form factor that have similar voltage, hold current, and trip current ratings.

73,

Rob W1AEX
"One thing I am certain of is that there is too much certainty in the world."
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby Tony EI7BMB » Mon May 01, 2017 8:10 pm

Great info Rob, thanks
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: Hermes Protocol 2 firmware v10.3 released

Postby Tony EI7BMB » Fri Jul 07, 2017 1:18 pm

Just to update the thread,I did try running at 384 and two things I noticed are :

1. 384 khz does not give the same frequency coverage as 3 stitched receivers

2. Stitched receivers handle strong local signals much better IE they confine the noise floor jumping to the single receiver range so for example if there is a strong SSB station on 50.148 it has no effect on 50.276 whereas using 384 allows the noise floor jump to carry all the way up.

Return to “Protocol 2 Firmware (all radios)”