Launch Post-Mortem

[John Laidler]

I've been working with Ross G6GVI to set up a Heltec Wireless Tracker board to run Horus Binary V3 using the code from Roel https://github.com/RoelKroes/TBTracker-ESP32/releases/tag/V0.5.3

The code needed a little bit of tweaking including a few lines needed to turn the GPS on and of course it uses different pin numbers for the radio and GPS. I'll post on here the modified code in a day or so. The boards were bought from AliExpress and cost under £25 but beware, I have had 2 boards, with defective GPS. You can buy them in the UK for a higher price but at least you could return them more easily than trying to send them back to China. 

The photos show the payload, which is made out of 3 pieces of 25mm thick expanded polystyrene. The design gives 25mm of insulation primarily to keep the batteries warm. The batteries, 2 AAA Lithium Energizer are in a small unswitched holder. After installing the batteries and checking it was working I wrapped Kaptan tape around it to hold everything together. 

The antenna is a vertical dipole soldered directly to the antenna connection on the board with the two elements set 180° apart, emerging from the top and bottom of the housing. 

Tests had shown the integral GPS antenna on the board worked well when orientated vertically, in other words it doesn't have to face upwards which makes the design of the payload easier. 

The parachute, an 18" Spherachute, was too long to fit between the balloon and the payload if I was to keep within the UK sub-2m rule so I hung it upside down and attached it to the balloon as shown in two of the photos. This was risky with opportunities for tangling but it must have worked as the descent was gentle. 

The payload and parachute weighed 60g and with a Hwoyee 50g balloon and party gas helium a neck lift of 250g it should have reached around 9,000m. 

The filling and launch went more easily than expected although I was slightly surprised by the sixe of the balloon after filling. I did the filling indoors and it only just fitted through the door. One of the photos shows the filled balloon with a bottle of water in the parachute to keep it on the ground. 

The balloon flew west for a short time then east, all as expected by the CUSF Predictor. The expected landing was in the south east corner of Dartmoor so I headed towards Haytor Rocks and watched as it just kept climbing! 

When it looked like it was going to land in the sea I set off home. 

In the end the balloon reached nearly 18,000m, twice what was expected. It did return to land but it is a four hour return journey from my home and I suspect the batteries will have run flat by the time I arrived so I have left it there. The payload has my name and telephone number written on it so perhaps someone might find it. 

Receiving equipment was:

Base station - Yaesu 897D with Diamond X-30 antenna on a 6m mast connected to the PC running Horus Binary GUI. 

Chase car - Icom IC-705 connected by USB to a Windows Surface Go 2 tablet, Horus GUI linked to the Internet by an EE 5g router. Antenna was a little 434 MHz mag-mount on the car roof. I also had a 3 element Yagi but didn't need it the stubby mag-mount worked well. 

Many thanks to all those who tracked the flight. I was very impressed by the distances seen. And especial thanks to Ross G6GVI for his help including today when his websdr setup was perfectly positioned for the light. 

I will try another flight some time and may try for a higher altitude with a smaller parachute and a bit of weight loss on the board - the LCD screen and GPS antenna can be removed bringing the board down to 15g. The integral GPS antenna being substituted for a small dipole. I suspect 20Km+ would be achievable with the Hwoyee 50g with a light payload and low ascent rate. 

John

M0WIV 

j

[Ross G6GVI]

That was brilliant John, thanks for an entertaining morning and a very comprehensive write-up.

I wonder if this was the first flight-test of one of those UC6580 GNSS units? Now we know that they are good up to 18km at least.

What a good job we found that bug with the number of satellites >31 (overflowing their 5-bit value) last week, otherwise we'd have lost all the transmissions after the first hour:

All the Sondehub data can be viewed on this Grafana page: https://grafana.v2.sondehub.org/goto/stEg9STDR?orgId=1

I do hope that you get the payload back - at least if didn't end up in "Davy Jones' locker"!

[Morseman]

How do ground stations monitor the downlink, please?

I’d like to set up a ground station to track any that come within range. 

Dave (G0DJA)

--
You received this message because you are subscribed to the Google Groups "UKHAS" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ukhas+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/ukhas/5bc0fd78-3594-4a97-8b3d-37b39f007a0dn%40googlegroups.com.

[Ross G6GVI]

Hello Dave, fancy meeting you here!

For "Horus" FSK transmissions, you just need an SSB receiver (I was using my IC910 today), a sound-card and a PC:

https://github.com/projecthorus/horusdemodlib/wiki/1.1-Horus-GUI-Reception-Guide-(Windows-Linux-OSX)

and of course a 70cm antenna (most balloons use Vertical polarisation).

There are also a number of round-the-world (solar-powered) flights using WSPR on HF and/or LoRa APRS on 70cm, but a standard terrestrial system for these modes will pick up the balloons automatically.

73 de Ross

P.S.  more info here: https://www.qsl.net/g6gvi/habtracking.html

[John Laidler]

Dave, 

I only know the manual approach - tune the radio to the correct frequency and connect it to a PC running the appropriate software. 

It is possible to use a Raspberry Pi to run a program which will decode LoRa signals using an SDR dongle such as the RTL one but you still have to tell it what frequency to listen to. I used a dongle attached to my PC during testing of a payload. The PC had SDR# running and the Horus Binary GUI. I wasn't too impressed by the sensitivity of the dongle but you can get inexpensive pre-amps. 

I'm not aware of any "skimmer" software which could monitor an entire band as you have for CW and the RBN system but it would be nice if there was. 

Several of those tracking me today used websdr systems and this was very successful and may be a way forward for you. I'm certainly going to try it soon. 

John

M0WIV 

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJfTFSa%3DuFXd-awmdqLUWuoa-mpJ1fFXn8yizT-kWCXMmDh7Kw%40mail.gmail.com.

[Morseman]

Hi Ross,

I have all those things (computer that I run programs like WSJT-X/Fldigi, IC9700, three band vertical etc.) I need to know what software is needed…

Cheers - Dave (G0DJA)

--
You received this message because you are subscribed to the Google Groups "UKHAS" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ukhas+un...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/ukhas/2cb05d2c-d6b6-4ff5-814d-38069619ae8dn%40googlegroups.com.

[John Laidler]

Dave,

The Horus GUI program Ross linked to works very well. The only problems I had was the usual fiddle amongst the Windows sound settings for it to hear the radio and of all things turning it on!  There's a "Start" button which needs pressing before it does anything. I didn't see this for a while... 

John

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJfTFSaEFce1iVnMqxH657%2BfFUWa78NRGAZWKKmDoY1GAH5KuA%40mail.gmail.com.

[Steven Boonstoppel]

Re the balloon - as I (the "Ichthus College") plan on doing a launch May 9th with the HY-50, I'd like to double-check your helium volume. I assume it's 0.35m³ given the 250g neck lift and 18km altitude, is that right?

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJn3SCoFLNnzDyCcRHWyixCx8Y_AGHx7rncrV2ypHQGeb4cM1Q%40mail.gmail.com.

[John Laidler]

Stephen,

Using the calculator on the Random Engineering website it shows a volume of gas of 0.3m3 for a necklift of 250g and an ascent rate of 4.8m/s. As the inflated balloon only just fitted through a door, so was about 0.75m in diameter, this works out at a volume of 0.2m3 so given it was under a bit of pressure it probably was about 0.3m3 at standard pressure.

Not very scientific but the decimal point is probably in the right place. 

John

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CA%2Bf_UVUnjEXozH9cCktSm1A-rpgmD5u-MEX31Bzv4g8Efitbog%40mail.gmail.com.

[Nick McCloud]

On Sunday, 26 April 2026 at 15:32:54 UTC+1 John Laidler wrote:

In the end the balloon reached nearly 18,000m, twice what was expected. It did return to land but it is a four hour return journey from my home and I suspect the batteries will have run flat by the time I arrived so I have left it there. The payload has my name and telephone number written on it so perhaps someone might find it. 

Litter bug 😉

It will be where it landed, with a bit of luck dangling in a lower branch of a tree on the edge. Or in the field. 

On Sunday, 26 April 2026 at 19:30:16 UTC+1 John Laidler wrote:

 

Not very scientific 

We can make it scientific! What was the capacity of the party helium tank (usually displayed as number of 9' balloons) and did you empty all of it in to the balloon?

On Sunday, 26 April 2026 at 15:46:56 UTC+1 Ross G6GVI wrote:

I wonder if this was the first flight-test of one of those UC6580 GNSS units? Now we know that they are good up to 18km at least.

Sorry, not the first, the RSGB bowling ball with spike payload used the same board.

The docs for the GNSS does make reference to portable vs static mode, defaulting to portable, so I suspect the RSGB got away with it - given no code or other details were released.. To contrast, U-Blox has static, pedestrian, automotive & three flight modes.

 

[John Laidler]

Nick, 

I can't estimate the volume of gas used. I bought a new large bottle for 50 * 9" balloons but I already had two part used 30 * 9" bottles. These latter I exhausted then had to use the big bottle for a final top up. 

I don't think a single 30 * 9" bottle would be enough. They say they are 0.23 cu m so you would need two or a single large one which are claimed to be 0.42 cu m. I'm always a bit distrustful of their claimed volume so like to have a spare bottle on hand. Hydrogen is the answer but I can't justify the cost based on my irregular launches 

John

[Ross G6GVI]

Yes Nick, I'd forgotten that LoRa repeater flight (or maybe blotted it out?).

But that RSGB flight only got up to 45,000 feet, whereas yesterday John reached 59,000 with the UC6580.

On Sunday, 26 April 2026 at 21:27:27 UTC+1 Nick McCloud wrote:

[David Ackrill]

I have found the Horus GUI program, installed it and connected to the sound card inside my IC-9700.  It's showing wavy lines and occasional SNR value on 70cm. I'm also being plotted on the SondHub webpage.

I did a bit of digging on the internet and found that 434.417 and 437.600MHz appear to be the popular frequencies for 4AFSK in Europe, but it looks like there's some payloads using other modulation systems on different frequencies. Including LoRa on 433.775MHz.

I have a couple of 70cm ESP32 boards, one on 439.9125MHz running LoRa APRS in the car (G0DJA-7) the other was being used as a repeater but there's another LoRa repeater local to me and we were both just duplicating the coverage area of each other, so I switched mine off.  That means that I have that board available to reflash, if there's a way to program it as a receiver for LoRa type balloon transmissions?

Dave (G0DJA)

[Nick McCloud]

On Sunday, 26 April 2026 at 21:52:26 UTC+1 John Laidler wrote:

I don't think a single 30 * 9" bottle would be enough. 

It wouldn't, but a 50 x 9 is as that's what was used on the Ichthus launch with a slighter higher burst altitude.

Once we get past a couple of more controlled launches, one coming up at Ichthus with a slightly lighter payload, we'll have more data and I can join the fray now the weather in the NW isn't predicting here, there & everywhere. 

[Nick McCloud]

On Monday, 27 April 2026 at 08:28:33 UTC+1 David Ackrill wrote:

That means that I have that board available to reflash, if there's a way to program it as a receiver for LoRa type balloon transmissions?

https://github.com/daveake/HABBase/releases/tag/V1_7_10 - uses a LoRa board plugged in to a Windows machine - HABBase is like HorusGUI for LoRa.

You'll find many different variations on how to send telemetry - particularly for Pico (round the world) balloons.

UKHAS, Dave Akerman, et al came up with the LoRa format that can include images - but there aren't many of us left flying Pi-in-the-Sky that does it all, so you'll see a mix in the UK but mostly HorusBinary in the EU.

[Nick McCloud]

On Monday, 27 April 2026 at 10:51:33 UTC+1 Nick McCloud wrote:

 I can join the fray now the weather in the NW isn't predicting here, there & everywhere. 

Um, still a bit random ...

 

[Rick Hewett]

On Mon 27 Apr Ross G6GVI wrote:
> But that RSGB flight only got up to 45,000 feet, whereas yesterday John

> reached 59,000 with the *UC6580*.

Having seen video of the RSGB launch (on a horribly wet day) I expect
anyone who's done a successul(-ish) launch might have a dozen or more
ideas why it didn't get that high...

--
73, Rick, M0LEP (aka lazyleopard on #highaltitude IRC)

[Morseman]

I’ve had it running the last couple of days. It seems to stop working after a while and needs to be stopped and restarted. 

Dave (G0DJA)

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJn3SCoFLNnzDyCcRHWyixCx8Y_AGHx7rncrV2ypHQGeb4cM1Q%40mail.gmail.com.

[Ross G6GVI]

My favourite UKHAS LoRa receiver is the old "Pi LoRa Gateway", which can run standalone or linked via TCP to the HABBase. 

I'm still using the one which I built on a Pi2B back in 2019 to follow Dave Akerman's "Apollo 50" flight: https://www.qsl.net/g6gvi/pigateway.jpg

Dave's original version used WiringPi, but thanks to Tom Wardill there is an update with lgpio:

https://github.com/teamtoadprojects/lora-gateway/tree/lgpio-port

and I managed to build one of these with a PiZero 2W and an Ra02 recently.

[David Brooke]

On Sun, Apr 26, 2026 at 07:32:54AM -0700, John Laidler wrote:
> were bought from AliExpress and cost under £25 but beware, I have had 2
> boards, with defective GPS. You can buy them in the UK for a higher price

I don't know in what way your GPS was defective but I've had a couple of
boards where the GPS patch antenna pin had not been soldered, and thus
was unable to receive, so it might be worth checking for that as it's an
easy fix.

David G6GZH

[John Laidler]

David,

Thank you. I've checked and the patch antenna seems to be soldered
correctly. I also used an external antenna which works very well with
the good boards so it should have been able to receive signals from
satellites. I've left them running for hours but without success.

On booting up, both boards perform a GPS comms check and this seems to
go correctly and looks the same as what the good boards show although
it is very fleeting.

John

> --
> You received this message because you are subscribed to the Google Groups "UKHAS" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to ukhas+un...@googlegroups.com.

> To view this discussion visit https://groups.google.com/d/msgid/ukhas/ae-Jg5IBsahopqxW%40cubitt.dbrooke.me.uk.

[John Laidler]

My conscience got the better of me and this morning I drove to the area where the payload was last heard from. I spent a little over two hours scouring fields, trees and knocking on a few doors to ask to look in back gardens. 

I drew a blank. 

The problem is the area is a deep and heavily wooded valley running north to south and the tracker was last heard from as it crested the eastern side heading west at 10 m/s at an altitude of about 180m above the ground. A critical factor was the parachute at 18" was too big for 60g and it was descending at only 4 m/s. My basic geometry suggested it should have landed on the west side of the valley but this was difficult to access and heavily wooded. I searched where I could but to no avail. 

I need to work on setting up a receiving station which is more portable than what I have now which uses an Icom IC-705. I think I'll try an RTL dongle plugged into the Ms Surface tablet running SDR# and Horus Binary GUI. I can leave the keyboard at home and this should be light enough to use on foot when searching for downed payloads. This is of course where I went wrong yesterday in not being on the ground with a suitable receiver while the batteries were still alive. I found the RTL dongle less sensitive than the IC-705 and Yaesu 897D but at short range I'm sure it will be fine. 

I'm not sure when the next flight will but I want to try again with the Hwoyee 50 and perhaps see how high it can go.  

John

M0WIV 

[Radiocose]

Hi John, when you mention the Hyowee 50, do you mean their 50 g/120 cm balloon? If yes, could it be suitable for HAB experiments?

Thanks.

73, Daniele IZ5WWB

To view this discussion visit https://groups.google.com/d/msgid/ukhas/fe58de87-ed13-4503-aa4c-e0e5486d63b0n%40googlegroups.com.

[Nick McCloud]

On Monday, 27 April 2026 at 19:25:19 UTC+1 Radiocose wrote:

the Hyowee 50, do you mean their 50 g/120 cm balloon?

 

Yes, that the balloon, available from Amazon. Look at the recent topics, this was the very same balloon type used for the Ichthus launch. 

If yes, could it be suitable for HAB experiments?

It is being used for HAB experiments, so yes, it is suitable. But it depends on what you want to do.

At present we haven't got reliable figures for burst calculations but it can be filled using a 50 x 9' balloons party Helium cylinder which is very convenient. It can lift ~100g + a plastic bag used as a parachute.

[John Laidler]

Yes, I bought mine from Amazon UK. You may be able to contact the seller and see if they can ship to Italy. I'm sure they can.

https://www.amazon.co.uk/Professional-Meteorological-Entertainment-50g-120cm/dp/B0CTCMKNPR

John

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAES4b1bj_Kg7WPQ%2BhP3Q4vnqt-b5XhvpBuyOQjtMtn%3Dj120yWQ%40mail.gmail.com.

[Radiocose]

Hi John, it is available through Amazon Italy as well pretty for cheap, compared to alternatives.

73, Daniele IZ5WWB

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJn3SCopYBPCUuxdzhj0ks%2BmumS4gYQL%3D7CXeiYKuAFyGWbeLg%40mail.gmail.com.

[Nick McCloud]

On Monday, 27 April 2026 at 22:16:36 UTC+1 John Laidler wrote:

Yes, I bought mine from Amazon UK. You may be able to contact the seller and see if they can ship to Italy. I'm sure they can.

https://www.amazon.co.uk/Professional-Meteorological-Entertainment-50g-120cm/dp/B0CTCMKNPR

They are shipped from China so I doubt they care which Western country they end up in. Ichthus College purchased via Amazon.NL on account of them being in the Netherlands. You should be OK getting them from your national Amz site.

[Ross G6GVI]

John, have a look at the mobile RasPi-RTL receiver which Mark VK5QI has developed for use in his chase car: 

https://github.com/projecthorus/horusdemodlib/wiki/1.2--Raspberry-Pi-'Headless'-RX-Guide

it links to his ChaseMapper: https://github.com/projecthorus/chasemapper

which is viewed on a browser running on a phone or tablet.

I've not tried it myself, but it looks like the guys Down Under have been using it for ages.

[John Laidler]

Daniele,

That is good to know it is available in Italy. It isn't expensive or needs a lot of gas so this should keep costs down. I'm sure 20,000m+ is possible with one of these balloons.

John

M0WIV 

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAES4b1bSx%3Dp2FfPTAXMe-4DzDiueaPuqQmUuHvh-UgbGSFWtvg%40mail.gmail.com.

[John Laidler]

Ross,

That looks very good. My first launch I tracked with an RPi and Dave Akerman's LoRa gateway software which as you mentioned earlier was excellent. 

John

--

You received this message because you are subscribed to the Google Groups "UKHAS" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ukhas+un...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/ukhas/6569153c-9e54-49b4-b0e2-87870b847a2en%40googlegroups.com.

[Mark Jessop]

For a 'handheld' receiver, another option is to use WebHorus: https://horus.sondehub.org/

This will talk directly to a RTLSDR on Android phones (we use WebUSB, which iPhone does not support). On non-android phones you can also feed audio in, but the RTLSDR approach is definitely more convenient.

For ChaseMapper, I have a full guide on setting up a Raspberry Pi with Chasemapper and Horus Binary here: https://github.com/projecthorus/chasemapper/wiki/Setting-up-a-Horus-Binary-Chase%E2%80%90Car-Raspberry-Pi

This is essentially what I am running on all of my balloon launches. I use a 7" Tablet on the dash to view the web interface.

73

Mark VK5QI

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJn3SCo0DMKkx%2B3xY_V%3DWe51xMfkzcFWYnoyqxnBzVKtbTSkBQ%40mail.gmail.com.

[John Laidler]

Mark,

WebHorus and ChaseMapper look excellent solutions. I'll have a look at both of those. Many thanks. 

John

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CABD2Lt0b02yndnDnRqK9_scPpPM-Z63JjnORsOYgcwHtGb5XCw%40mail.gmail.com.

[John Laidler]

Mark,

I've installed WebHorus on my Pixel 9a which is running Android 16 and Chrome. With an RTL SDR dongle connected it is receiving signal but doesn't seem to be decoding them. It uploaded the station position to Sondehub but nothing else. 

From the enclosed screenshots can you see any obvious errors? The same dongle works perfectly when connected to a PC running Horus GUI. The settings such as baud rate are the same. When the transmission is received the 4 hours position themselves correctly over the 4 peaks. I've tried minimum gain and moving away from the transmitter, in case the dongle is being over-driven but without success. 

Any help will be gratefully received as this is a perfect solution for chasing downed balloons. 

John

M0WIV 

On Mon, 27 Apr 2026, 23:48 Mark Jessop, <lennieth...@gmail.com> wrote:

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CABD2Lt0b02yndnDnRqK9_scPpPM-Z63JjnORsOYgcwHtGb5XCw%40mail.gmail.com.

[PE2BZ]

Hi Dave, the trackers flying with LoRaAPRS and run the software from PD7R, TBTrackerTX , do follow the LoRaAPRS frequency ranges defined by GPS position. So, when in the neighbourhood of the UK it switches to 439.9125 and rest of the EU 433.775  and poland and other areas to the frequency used there. 

I have flown 2 trackers with LoRaAPRS enabled and both appeared on UK receivers and showed 439.9125 as receiver frequency. 

Good luck with the HAB reception.

Ben - PE2BZ

[Mark Jessop]

Is the tone spacing any different to the usual 270 Hz? If so you need to adjust this in the Config settings, under Horus Binary configuration. 

Other than that, I'm not sure what else could be going on. 

73

Mark

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJn3SCqnTa%2By1_2j6teM-r7cU_wNA27HCsWZ0-hc%3D3Yja3t3Rw%40mail.gmail.com.

[Ross G6GVI]

Dave, if you want to leave something running unattended, try the "headless" Horus Demod on a RaspberryPi & SDR dongle:

https://github.com/projecthorus/horusdemodlib/wiki/1.2--Raspberry-Pi-'Headless'-RX-Guide

That will automatically acquire and track a signal within a 10kHz bandwidth.

[Morseman]

I saw a Dutch balloon on the website, and the blue circle came over me, but not the green one, so I didn’t decode anything. 

I saw yours earlier on the map but I don’t think it was launched?

I have an SDR dongle, somewhere, I’ll try to dig it out. 

Cheers - Dave (G0DJA)

--
You received this message because you are subscribed to the Google Groups "UKHAS" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ukhas+un...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/ukhas/2b048475-d667-4515-b8df-f1cf12e5528an%40googlegroups.com.

[Morseman]

There appears to be something on the map that looks like it came from the west of Ireland but didn’t get up very high?

Dave (G0DJA)

On Wed, 29 Apr 2026 at 11:09, Ross G6GVI <ross....@gmail.com> wrote:

--
You received this message because you are subscribed to the Google Groups "UKHAS" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ukhas+un...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/ukhas/2b048475-d667-4515-b8df-f1cf12e5528an%40googlegroups.com.

[Mark Jessop]

Where are you seeing this? Maybe post a screenshot, or a link to the tracker showing this?

I'm not seeing anything on SondeHub-Amateur.

73

Mark VK5QI

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJfTFSa8KcniVUE5M%2B5y%2BjT2x-CRz6OsvYSPab23UruY3N_e%3DQ%40mail.gmail.com.

[Morseman]

I tried to upload a photo but it failed. 

I’ll try again. 

Dave (G0DJA)

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CABD2Lt2qAnrSEhYm9Z_wH400E9dUKomP75ZzYoyUx9K%3DdskExg%40mail.gmail.com.

[Morseman]

David Ackrill

On Wed, 29 Apr 2026 at 11:43, Mark Jessop <lennieth...@gmail.com> wrote:

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CABD2Lt2qAnrSEhYm9Z_wH400E9dUKomP75ZzYoyUx9K%3DdskExg%40mail.gmail.com.

[Morseman]

Worked this time?

David Ackrill

On Wed, 29 Apr 2026 at 11:43, Mark Jessop <lennieth...@gmail.com> wrote:

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CABD2Lt2qAnrSEhYm9Z_wH400E9dUKomP75ZzYoyUx9K%3DdskExg%40mail.gmail.com.

[Morseman]

My iPhone is playing me up. 

I hope this is the photo of the screenshot this time…

Dave (G0DJA)

On Wed, 29 Apr 2026 at 11:43, Mark Jessop <lennieth...@gmail.com> wrote:

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CABD2Lt2qAnrSEhYm9Z_wH400E9dUKomP75ZzYoyUx9K%3DdskExg%40mail.gmail.com.

[Mark Jessop]

What you are seeing is a flight path 'prediction' of a payload under test.

Sometimes noise in the altitude data can make the tracker think it's ascending (albeit very slowly) and run a flight path prediction, which is usually going to be garbage.

73

Mark

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJfTFSYUs2oEg6PGb-u_AXEVnDUcUa4TJyVVADf%2BDqMRon%3D_5w%40mail.gmail.com.

[Morseman]

I’ve got WebHorus working after trying Crome instead of FireFox but the gain adjustment seems a bit touchy. I turned it right down and it said the SDR-Dongle had been disconnected. 

Dave (G0DJA)

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJn3SCp-xJKddtFZJpit0L%2BqmVJTC4sghF8CyktjBsxTZx75yA%40mail.gmail.com.

[John Laidler]

I've got WebHorus working and it just decoded a single packet and
uploaded it to SondeHub. Unfortunately, that one packet is all it
managed and I can't get it to do it again despite tweaking and turning
things on and off. But at least it showed that my settings are
largely correct as it decode the (one) packet correctly.

John
M0WIV

> To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJfTFSZfZRC9fbdy3neH6k4YvMdSrBK6mN2qKm6LqZjr0XS9Nw%40mail.gmail.com.

[Morseman]

Which signal was that, please?

Not seen anything on the map this afternoon. 

Dave (G0DJA)

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJn3SCrkNpfMoobUs9aqii%2B87ziZJO5xkohiyoZcV1dctyGzFQ%40mail.gmail.com.

[John Laidler]

It's just me playing indoors, the signal won't be reaching more than a
few hundred metres. It is showing on SondeHub as M0WIV-2 and is being
updated occasionally when it manages to decode a packet. This
currently seems to happen when I first turn on the dongle. It decodes
the first packet then nothing follows. I'm trying different gains but
without success. Very odd.

John

> To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJfTFSbGCTpkOmG5j2ihuyJjrPwwd%2BB92yfbXX3N_imX%2BjDZvg%40mail.gmail.com.

[John Laidler]

Mark,

Thank you. I've got WebHorus working now but not very reliably. It's better running on a PC than my phone and my guess is my Internet speed may be too slow for reliable decoding. Being web based it must need a reliable and fast connection I guess?

I'll give it another go with a better Internet connection and see if that improves things.

It's a very interesting project though! 

John

M0WIV 

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CABD2Lt1DKBMA6PtUpWcEmv5cu7jq%3DJwhOPLQ714hOyt5EQU-Dg%40mail.gmail.com.

[Mark Jessop]

WebHorus runs entirely in the browser on your phone. Once it's loaded, internet connectivity is only required to upload data to Sondehub.

So unfortunately I'm still not quite sure what issues you are having. I guess if it's an older phone you could be getting sample drops, resulting in mangled packets. It is fairly CPU intensive. 

73

Mark VK5QI

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJn3SCps%3Dw8We3XmwVP5oHnVf%2BR6JnvZuHGib3b3uhpn3VePxw%40mail.gmail.com.

[John Laidler]

Mark,

Thank you, that must explain why it worked better on the PC. 

John

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CABD2Lt0pT1iiJ0v65f7V%2BsSQsn0d551VY97OgSs4TGhMG4w5Sg%40mail.gmail.com.

[Morseman]

Ah, OK. Yes, I can see your plots on the map at times. 

How are you creating the 4AFSK signal on 2M, please?

I tried to get a 433/434MHz board to work but Arduino IDE would not compile the program, let alone let me upload it to the board.

I would like to test WebHorus with a signal so that I can get a feel for how it works. 

The UKHAB wiki appears to be down as well. I get an”out of bandwidth” error when I try to use it. 

Dave (G0DJA)

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJn3SCoqFq8N-CEeTR4xT3djv2ocnus%3DLFtGeYaJkrsBShJTZQ%40mail.gmail.com.

[John Laidler]

Dave, 

I'm using a Heltec Wireless Tracker running the program I linked to in an earlier post. I must find the time today to note which libraries it is using as this does compile correctly using the Arduino IDE. It's using 70cm not 2m.

John

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJfTFSawoVvifd5b4UQKx5%3DObsqM2OfVC7thJ0zCiY946WuycA%40mail.gmail.com.

[Nick McCloud]

On Wednesday, 29 April 2026 at 12:13:18 UTC+1 Mark Jessop wrote:

What you are seeing is a flight path 'prediction' of a payload under test.

Sometimes noise in the altitude data can make the tracker think it's ascending (albeit very slowly) and run a flight path prediction, which is usually going to be garbage.

Unfortunately this is my colleagues playing with tracking but not using a test callsign or turning off the uploader.

I'll remind them again to stop doing this.

[Nick McCloud]

On Wednesday, 29 April 2026 at 17:20:01 UTC+1 John Laidler wrote:

It's just me playing indoors, the signal won't be reaching more than a
few hundred metres. It is showing on SondeHub as M0WIV-2 and is being
updated occasionally when it manages to decode a packet. 

There is absolutely no need to upload it to Sondehub for local testing. If you see the payload decoded on your screen, it's all good.

Uploading means random "not" flights appear and it adds to the pay per byte/minute/transaction/request workload on the Amazon web servers that partly funded out of personal pockets of people here, without whom, no Sondehub. 

[John Laidler]

Good point! I'll avoid doing this. 

--

You received this message because you are subscribed to the Google Groups "UKHAS" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ukhas+un...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/ukhas/d31c61b5-2b96-4d55-a6e3-ffcc9ff8b8d6n%40googlegroups.com.

[Steven Boonstoppel]

Trying to decode Horus v3 for my upcoming launch... but no luck yet. Neither GUI 6.1.1 nor WebHorus show any sign of decoding.

Attached is also a .wav file of a transmission. Does it decode for others and is it just an error in my SDR/Horus decoding? 
I verified with the marimo decoder that the contents of one such message was valid.

Op do 30 apr 2026 om 10:53 schreef John Laidler <rjohnl...@gmail.com>:

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJn3SCosKzKWwwdA%3Djphx63B5ya3oo-GFDK8g%2BcWvuAa4%3DP_fA%40mail.gmail.com.

[Mark Jessop]

I'm trying a few things here, but am not getting a valid decode out of it. 

I do note that the tone spacing is ~240 Hz (typical of many of the lora chipsets), so as a start if you are using horus-gui you will need to disable the mask estimator.

I'm still not really sure what's going on however. There's possibly something odd going on with the symbol rate, it *might* be drifting down in symbol rate during the transmission, but i'm not 100% on this. For a signal that looks this good in a spectrogram, the modem is reporting a very poor SNR, which suggest something wrong with the transmission.

What was transmitting this?  

73

Mark VK5QI

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CA%2Bf_UVX6qeRpsjDGsgXciFMuJ48UviY9XzmJjmAcwXqJDsdk%2Bw%40mail.gmail.com.

[Steven Boonstoppel]

I should note that on my previous (and first) launch a few months ago I also had extremely poor HB reception while others received it perfectly fine. 

The radio is an Ra-02 ie repackaged SX1278. Hence the 244 tone spacing as I configured in the GUI. Disabling the mask estimator just now didn't change anything really, unfortunately.

My SDR is an official RTL-SDR v4 with the included antenna. I will find some SMA male/female adapters and try the antenna that I use to receive the LoRa telemetry from the same radio which works all good.

Op za 2 mei 2026 om 12:17 schreef Mark Jessop <lennieth...@gmail.com>:

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CABD2Lt3rdo6fZGyWHn4QiQxRE%3DXBQdXGjxb7rXSNPAspdku-2A%40mail.gmail.com.

[Mark Jessop]

Apologies, I didn't notice you had already set the tone spacing.

While radio might be a Ra-02, what's generating the modulation? This is what's going to set the symbol rate.

What else are you using on receive? What SDR application, etc?

Can you try using WebHorus with the RTLSDR - https://horus.sondehub.org/  (noting you will need to go into the config and adjust the tone spacing there too).

73

Mark

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CA%2Bf_UVVORmO9kaLaQCnqs6MLWgrzx2R3cDOd0tUPabutYGussQ%40mail.gmail.com.

[Steven Boonstoppel]

The microcontroller is the XIAO SAMD21.

I have to correct myself as I see that pretty much nobody received HorusBinary last flight (Grafana).

I am using SDR++ v1.2.1 piping into UDP.

Found some adapters and I now tried another antenna with no notable difference.

Webhorus: at 0dB gain (also tried AGC and 15dB):

Op za 2 mei 2026 om 12:32 schreef Mark Jessop <lennieth...@gmail.com>:

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CABD2Lt1aVEuEjKdbepqiazuKhXZL0cze-%2BvDsoYO387kSMog0Q%40mail.gmail.com.

[Mark Jessop]

On WebHorus you need to go to the receive tab to see any decoded packets.

That Grafana page appears to be of your LoRa payload - unless you used the same callsign for both the LoRa and Horus payloads? (I would not recommend this, it makes it very difficult to disentangle things in post-analysis). 

At this stage all I can think of is that there's something not quite right with your symbol timing. The modem is fairly intolerant of symbol rate offsets, and my guess is that it won't handle the symbol rate varying throughout the packet well either.

73

Mark VK5QI

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CA%2Bf_UVVGQTfMMmwz%3DNHN4UpgN0o6N_TC1y9%2Bc-CaV2z_TOAD5w%40mail.gmail.com.

[Steven Boonstoppel]

Yes, I just figured out that it was a different Grafana page. Sorry, I'm not that well versed here yet.

Here is the Horus Grafana page. Lots of people receiving it so I'd say that there is little issue with the transmission.

I did try putting the transmitter (which is at +2dBm) on the other side of the house one concrete floor up to no avail. 

Nothing on the Receive tab of WebHorus unfortunately.

Op za 2 mei 2026 om 13:01 schreef Mark Jessop <lennieth...@gmail.com>:

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CABD2Lt2ihTm5HdoD2xKCKhQSsaVLHdj%2BsF%2BKRb3A3mX7Cse72g%40mail.gmail.com.

[Mark Jessop]

OK... if it's the same transmitter, exactly the same same code, etc, then I'm really not sure what is going on here or why you're not receiving it.

Do you have anything else you can try and receive it with? I still don't really understand why your RTLSDR wouldn't work either through SDR++ or directly with WebHorus...

73

Mark

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CA%2Bf_UVWvmYyx_7QRG5J9CVamERZVN4q%2BtA_p2MT%2BM%3DWi6%2BPHcA%40mail.gmail.com.

[Ross G6GVI]

I think that the problem is more likely on Steven's receive side. 

Those audio clips which he sent certainly exhibit a lot of fast AGC action, which may be compressing the signal?

After lunch I'll send him a recording of one of my sources for comparison, to try with his decoders.

[John Laidler]

Steven, 

When I was testing my payload prior to the launch I tried slowing the CPU down to 20MHz as a means of power saving. It transmitted and the call looked valid in the GUI with the 4 peaks but it would not decode anything. Putting the CPU up to 40MHz cured it and every signal was decoded.

This may not be what you are experiencing but CPU speed might be something to look at if you can alter it.

John

M0WIV 

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CA%2Bf_UVX6qeRpsjDGsgXciFMuJ48UviY9XzmJjmAcwXqJDsdk%2Bw%40mail.gmail.com.

[Steven Boonstoppel]

Some further debugging...

Reverted back to the exact code from last launch (HorusV2). Works flawlessly, now with my SDR tuned much more scientifically than last launch thanks to Ross.

Patched in the HorusV3 code alongside V2. Both show up on the waterfall, both show up in HorusGUI, but it only decodes v2. WebHorus decodes neither.

I was able to confirm with Ross that HorusGUI is able to decode v3 with a recording he sent me. So that's not an issue either.

I did have to modify the named argument list for V3 ( { .sequenceNumber = horusCounterV3++ ..... } ) because the SAMD21 compiler doesn't like it, instead it wants asnMessage.sequenceNumber = horusCounterV3++ etc. But as I was able to parse bytes 2..numbytes (effectively removing CRC and filler bytes) as they appear in the device's Serial with the online decoder, this shouldn't be the issue. Besides that it's all unmodified V3 code from TBTracker 0.5.3...

If there are any ideas I'm happy to try them. Otherwise I'll dig around for a 433MHz ESP32 to try 0.5.3 without needing any modifications to make it compile. I'd almost say I would eat my shoe if that is the solution.. but anything is possible in the Dark Arts of Radio.

Op za 2 mei 2026 om 13:57 schreef John Laidler <rjohnl...@gmail.com>:

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAJn3SCqLeoVcG5dY4TMSx2np8yLjtFudwJ%3DnfzSvT725DKXVOA%40mail.gmail.com.

[Steven Boonstoppel]

Done and dusted... the Golay interleaver was modified with HorusV3 which only returns a different result for longer packages. It didn't surface for the 32-byte V2 packets so there was nothing that indicated an issue there. So for anyone updating from a HorusV2 setup (before V3 was added) to the new V3 version: also copy over the horus_l2.cpp file!

Thanks a lot to Ross and Mark for thinking along and helping out! My SDR reception has improved by a lot while trying to debug any reception issues as I learned how to do that, and it now looks like I can do V3 this Friday if weather/wind is all good.

Op za 2 mei 2026 om 15:50 schreef Steven Boonstoppel <steven.bo...@gmail.com>:

[Nick McCloud]

On Sunday, 3 May 2026 at 14:36:42 UTC+1 Steven Boonstoppel wrote:

My SDR reception has improved by a lot while trying to debug any reception issues as I learned how to do that

Do share the pertinent settings with the group - there's a fair number who are licensed here, but equally a fair number that would benefit from the details. 

 

and it now looks like I can do V3 this Friday if weather/wind is all good.

A new topic to make this clear with the protocols & frequencies in use will help the Northern Europe contingent catch up / get ready / be prepared.