UHF radio control for submarines (openLRS, LoRa, FSK etc. on 458Mhz and 433Mhz )

Hi All

At the Haydock Boat Convention over the weekend and spotted a 'Target' using a 433 transmitter (2 sticks and 12 switches) and receiver.
it was a bespoke special built by
http://timpdon.co.uk/timpdon/telec/index.php?x=54&y=10

Check out the site, some interesting information there (I've no connection)

John

Interesting. Presumably the use of 433Mhz is OK and legal in the UK if the control is not continuous - like car keys?

Thanks to all who showed an interest in openLRS at the Haydock Park show. It does help for people to get "hands-on". (Mind you I WAS impressed by Tim's video.)

It has been a good year in the UK for openLRS on UHF. (I haven't used 40Mhz since Spring.) The hobby is safe - thanks Tim!

(My plans for winter are to sort out this short range problem on one of my models and do some aerial experiments. I will write more in the months to come!)

David

you will also note by reading the descriptions of the different model transmitters:

* Up to 30 metre range with URX1 receiver in normal model or garden railway environment.


that equates to an output of about 10mw...

that might be some of the same regulations that Bart was coming up against in Belgium.

david f wrote:

...
It has been a good year in the UK for openLRS on UHF. (I haven't used 40Mhz since Spring.) The hobby is safe - thanks Tim!
...

David

as to this portion of David's post... I too am in the same "boat", having not used 75mhz since the spring... I moved all my currently operating subs over to 433mhz, with two more to go... one still in the construction phase, and the last one being my Alvin, which is currently in drydock awaiting a refurb.... including new radio, new ballast tank, and new switches for better lighting.

I also went the extra mile and upgraded my transmitter... you can read all about that endeavor on this supbirate's thread: http://www.subpirates.com/showthread.php?5258-FrSky-x9d-quot-Taranis-quot-Transmitter-Review

Hi all

The following was conducted after reading (and re-reading several times) all the information available on this thread, a similar thread on the Sub-Pirates Forum and an arduino thread on http://www.singlechannel.co.uk/ and I must express my thanks to Tim Senecal, David Forrest and Phil Green for all the background work that has been done. Thanks Guys...


Here are a few notes and photos regarding my initial attempts at 458 UHF control.

At first, I could not locate any transmitters that would take a UHF module on the second hand market so decided to try converting the hardware of a basic transmitter.

I picked up a Futaba T4EXA dirt cheap as the owner had somehow managed to let all the smoke out of the circuits.
Removed back and extracted RF board and encoder board which was then sawn in half to allow me to retain the original switch, charging plug and battery supply socket.



encoder board after "modifications" - Can you tell that I'm not an electrician?


By now I had obtained an Arduino  programmed as a 7 channel encoder from
Phil Green who produces them for aircraft use.
Further details are available on  http://www.singlechannel.co.uk/
Note header pins, designed for servo leads.


A small mounting plate was cut from plastic to carry the Arduino shield and glued into the main case.


Next modifications to the case were blanking unwanted holes and adding additional holes for switches and the module.
Trainer switch replaced with button to calibrate sticks and two switches fitted. One is channel 7 on/off andgives a slow acting servo (originally for aircraft retracts) and the other is a spring loaded centre off (on)-off-(on) which will be used to drive the reversible ballast tank pump.


On the other side of the case I fitted three switches wired to 4 resistors which will give seven different settings on one channel.This idea came from
http://www.reseau.org/arduinorc/index.php?n=Projects.AnaSwich
and will need a PIC on this RX channel to decode the switch-able options - ANOTHER winter project!




Final hole in case was in the back and sized to take the UHF transmitter module.



The beauty of the arduino back board used is that all the connectors are via servo leads which simplify the connections to the transmitter.

excuse the untidy wiring. (Have I mentioned that I'm not an electrician)
The two potentiometers shown at the bottom of the case are to set Rate and Expo if needed.


Two points of noteworthy interest for anyone looking at Phil's Arduino C code is that the throttle has NO REVERSE (not needed on an aircraft) and I used channel 5 to give forward/reverse throttle and that the encoder contains built-in range checking software.

Since finishing this build, I've located a Futaba FP-T8UP which had a 35mHz module and which takes a 458 mHz UHF module as a direct replacement so I now have two working UHF sets.

Next build is a type IX uboat to use this new gear.

John

John,

curious what you mean by the following:

"Two points of noteworthy interest for anyone looking at Phil's Arduino C code is that the throttle has NO REVERSE (not needed on an aircraft) and I used channel 5 to give forward/reverse throttle and that the encoder contains built-in range checking software.
"

where is "Phil's Arduino C code" listing located?, i have taken a quick scan of http://www.singlechannel.co.uk/ and see no mention of this 7 channel arduino encoder that you possess, so that i might take a gander...

...as to why the need for "reverse" on the throttle. a single PWM servo pulse has no concept of "forward" or "reverse"... and a train of PWM servo pulses that make up a PPM frame therefore also have no concept of "forward" or "reverse".... your need for setting that up has me scratching my head.

thanks

tim

Tim.

Sorry for any confusion with my interpretation. Phil's coding is set to prevent the possibility of reverse direction of his aircraft's propeller as a safety feature. Modification of the software is of course a possibility but I'm still on a learning curve with Arduinos and C coding and found it easier to just swap two leads over.


Code and schematics from the following:-
http://www.singlechannel.co.uk/

ARCHIVE tab

P16 and P 17 for schematic and Arduino C code

The code is fully commented and will explain all.

cheers
John

Thanks for posting the detailed links, John.

(I too could only find stuff that was too retro for even my tastes!)

The revamped TX looks very good, John.

So we have got it all now from use of the Trainer plug to input to an Orange 433Mhz openLRS module to a completely gutted and re-built Transmitter.

No excuses now, chaps!

I'm about to try some submerged aerial tests at Barrow.

I've also received (this morning) a couple of Arduino Nanos
http://www.ebay.co.uk/itm/311057315455?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT

Arduinos look very promising and these worked "straight out of the box."

David

A couple of aerial trials to see if the receiver aerial is best to be vertical or horizontal  in the WTC.

This is the plastic tube "Radiobuoy" I'm using. Just sealed with a battery and RX and sunk at the bottom of the lake, about 3 feet deep.



This is a view of the Barrow lake (Thank you Google Maps!) It shows the numbered locations where I made measurements of signal strength using the openLRS system at 458MHz and using the Telemetry option with an FRSky unit.





Here is the data:

VTX means the Transmitter signal strength with the Radiobuoy vertical. So hRX (1) means a receiver signal strength, Radiobuoy horizontal , Trial 1 with the tube at right angles to the shore.

Data:

Aerial orientation trials. Barrow in Furness.  1.5 inch diameter plastic tube (Radiobuoy) resting on bottom at about 3 feet deep.

Trial at 18/10/2015. Radiobuoy at Location (1) Horizontal (1) is at right angles to the shore. Horizontal (2) is parallel with shore.
Loc  vTX  vRX  hTX (1) hRX (1) hTX (2) hRX (2)
1 84 82 91  90 91        90
2 54 0 54  4    85        85
3 54 0 49  0    83        82
4 58 0 58  58   53        30
5 63 63 84 83  84        83
6 58 0 54  0    84        84


Ave 62 24 60 39    80        76

Trial at 2/11/2015.  Radiobuoy at Location (5) Horizontal (1) is at right angles to the shore.

1 31 0 85 85
2 86 85 92 91
7 36 0 63 0
8 54 18 84 83


Ave 52 26 81 65

Conclusion:
Sorry the columns don't line up but the key thing to note is that hRX is always larger than vRX so you always get a better signal strength with a horizontal aerial.


Discussion:
I must admit that I didn't expect a horizontal aerial to be superior!

There may be an explanation in a very good paper by Butler available online (I've referred to it before and I'm grateful to John R for bringing it to my attention . I quote from Page 7:)

http://users.tpg.com.au/users/ldbutler/Underwater_Communication.pdf

"There is also the question of polarisation and directivity. According to Moore(ref.2), a submerged horizontal electric dipole
is equivalent in its field to a weaker vertical antenna at the surface. Most of the energy, radiated upwards from the
antenna, is refracted at the surface into a vertically polarised, almost horizontally travelling wave, above the surface. This
phenomenon helps to explain the technique used in Figure 8 to transmit signals horizontally above the water surface and
to receive them in the reverse process."

So refraction may be the reason?

(Other matters - range issues. The Charlie class sub is back in the water and the range is still very good. So it is just the installation which goes into the Resurgam and Nordenfelt subs which has low range issues. It is probably caused by screening in such a packed WTC but the work continues to sort it out this Winter. This is the only sub with range problems we have come across so far. I would think that any installation in a roomy WTC with the aerial at the top of the WTC and HORIZONTAL should be fine.)

David

My take (Thanks Phil Green, Tim S and John W) on an Arduino based modified transmitter for 458MHz over here:

https://www.theassociationofmodelsubmariners.com/t375p45-pic-and-arduino-microprocessors#9860

david f wrote:That's very good news!

Let us know how you get on.

David

Got the question what gifts I want for under the Christmas tree....So I ordered all the necessary 433 MHz parts…….will keep you informed about the progress.

• FUTABA T9CP (bought 2nd hand 4 months ago)
• OrangeRX Open LRS 433MHz Transmitter 100mW
• OrangeRx Open LRS 433MHz 9Ch Receiver
• SparkFun FTDI Basic Breakout - 3.3V

Grtz,
Bart

Nice to see that Santa is on his way, Bart!

When you are setting up, it may be helpful to know what the configuration files look like. (The Chrome configurator is quite friendly and menu driven but you can save text configuration files as below):

These TX and RX configuration files are with failsafe and also the beacon feature(signal loss transmits to a walkie talkie) which I have found useful. Choose a different (legal) rf_frequency but I don't imagine that we will get interference across the North sea!

{"type":"TX_single_profile_backup","firmware_version":901,"configurator_version":"0.54.1","obj":[{"tx_config":{"rfm_type":0,"max_frequency":463000000,"flags":1342177280,"chmap":[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]},"bind_data":{"version":56,"serial_baudrate":115200,"rf_frequency":458750000,"rf_magic":1090660067,"rf_power":7,"rf_channel_spacing":5,"hopchannel":[65,54,75,25,71,57,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0],"modem_params":2,"flags":2}}]}

{"type":"RX_configuration_backup","firmware_version":901,"configurator_version":"0.54.1","obj":{"rx_type":1,"pinMapping":[33,0,1,2,3,4,5,6,7,38,38,36,37],"flags":2,"RSSIpwm":255,"beacon_frequency":446043750,"beacon_deadtime":0,"beacon_interval":20,"minsync":3000,"failsafe_delay":100,"ppmStopDelay":0,"pwmStopDelay":0}}

Some aerial tests on 458Mhz - braving Cumbrian floodwaters but at least the pond is full!



Trial (1) on 12.12.2015. Using Futaba 9C TX with FrSky telemetry. RX in plastic tube 3 feet deep about 30 feet away with stock aerial.

                                    Vertical              Horizontal

(a) Stock aerial                  70%                 4%
(b) Commercial from Flytron 38%                  50%
(c) Commercial from Flytron 40%                  63%
(d) Home made wire 176mm 80%                  0%

Trial (2) on 21.12.2015. Using Arduino based converted 35Mhz Futaba TX with FrSky telemetry. RX in Charlie class sub steering around white buoy about 50 feet away - stock aerial.

                                              Vertical

(a) Stock aerial                          50-90%
(b) Commercial from Flytron          50-90%
(c) Commercial from Flytron          50-90%
(d) Home made wire 176mm long    50-90%

Some conclusions.

Not really a lot to choose between the aerials apart from the fact that the commercial aerials seem to have less polarization. This is presumably an intentional part of their design?

The newly Arduino converted old Futaba transmitter works very well "on the pond." What am I going to do with all my redundant 40Mhz equipment now!? (Three Futaba 9Cs at last count, plus innumerable receivers.)

I also realized that the telemetry works faultlessly in this model with no trembling servos. So I have put the FrSky sensor hub in to give me battery voltage and motor temperature. There is space for it in this model but I look forward to news of your simpler, smaller data hub, Tim.

That's about it for now folks. Time to get stuffing that turkey and downing a bit of festive spirit! Happy Christmas to one and all and see you in the New Year.

David

david f wrote:
I also realized that the telemetry works faultlessly in this model with no trembling servos. So I have put the FrSky sensor hub in to give me battery voltage and motor temperature. There is space for it in this model but I look forward to news of your simpler, smaller data hub, Tim.

David

the build thread on subpirates is getting closer.... a few more posts and there should be something capable of doing what you need.

Well worth pointing out on here that Tim S is running a master class over on SubPirates on providing Telemetry in a small and very useful Arduino package based on openLRS technology (Tim's original openLRS posts are on Subcommittee):

http://www.subpirates.com/showthread.php?5271-Custom-Frsky-Telemetry-Hub&p=53624#post53624ials

What would be really good is if the sunken sub could transmit a GPS location back via Telemetry. As far as I know, the GPS signal does not get through any depth of water? Any thoughts, Tim?

So I've been doing some more aerial trials over at Barrow pond:



It occurred to me that a simple vertical aerial which transmits a horizontal "donut" shaped zone is OK for aircraft but less good for boats and submarines. The donut is transmitting as much backwards as forwards. Again OK for aircraft but not for us.

We need something rather more directional in a forward facing direction hence the small "Yagi" aerial above.

This is based on the approach shown here using elements made from an old tape measure to make it easier to stow away:

http://www.instructables.com/id/433-MHz-tape-measure-antenna-suits-UHF-transmitte/

There is a lot on the net about this type of portable aerial which is used for radio direction finding "Foxhunts" (No Foxes are injured!) mostly in the US, it seems.





It folds up quite neatly. (A good use for an old Pringles tube!)

This aerial worked very well with the openLRS system on 458Mhz. Good range. A little bit clumsy on the back of your transmitter and it attracted a lot of questions from passers by at the Pond!

However it was directional and would be useful for finding a lost submarine. It picked up the normal telemetry signal and the emergency beacon signal from openLRS. (As long as the battery lasted anyway.) So some competition for the audio "Pinger."

If you want to use it in this way, I found it most useful just using the transmitter with Telemetry. You can see the bar graph signal strength and the audio beeper is useful as you scan the Yagi aerial over the water. The Pringles tube and the folded aerial now have found there way into my "Subsunk" rescue bag along with the hydrophones, waders and grapple! (The bag is always in the car when I go sailing subs.)

But I digress. A decent compromise aerial for everyday use is shown next.  This is a a half wave dipole. I made the length of each limb as 158mm - a bit longer than theory would suggest. This gave as much range as you would want i.e so that you can see which way the sub is pointing.

It is very easy to make by soldering some stiff copper wire to a right angle SMA connector.

Sorry I haven't been posting anything on this topic recently. Partly because I have been working on the Pinger 2 (See separate post.)

I have had Tim's Telemetry working in my sub and that included depth and compass sensors.

But immediately after that I hit the limited range problem again with openLRS.

So I have taken all the software back to a "Vanilla version" until I can find out what is going wrong.

David J is till zipping around on the pond with several subs on openLRS which is heartening. He was also very helpful last Sunday by allowing me to swap transmitter modules etc. This pointed to the fact that the original "Commercial" aerials seem to give better range.

I wonder if my aerial experiments have overloaded and damaged the RF modules?

Is there any easy/cheap way of measuring RF power at this frequency? (Current draw by the module?)

I am trying to reproduce the problems on the bench.

David

David,

define "easy/cheap way of measuring RF power at this frequency"

i have one of these:

http://www.ebay.co.uk/itm/ImmersionRC-RF-Power-Meter-/121953926571?hash=item1c65053dab:g:h6oAAOSwZ8ZXDQrN

it is designed to do exactly that, and by using it i am fully aware that one of the RF modules i have for my taranis is in fact at about 65% of full power because it spent about 5 minutes transmitting with no antenna attached.


you may not consider it cheap (i didn't, until i saw how much "professional" units cost), but it is very easy to use, and works with the frequencies we use (35mhz to 5.8ghz)

David,

another reason you might not be getting the range you expect is because of the telemetry. remember, the receiver is "transmitting" at 100mw, and if you are using a 1 watt transmitter, there is a huge disparity between the range each has... the telemetry data will drop off before the failsafe kicks in.

I will be testing many things the next time i go back to "blue hole" new mexico, it is a perfect place to test these things, it has very clear water, so i can see exactly what is going on, but it has a very high mineral content, so the radio reception is on the bad side, last time i went i only got about 3 feet depth before the packets started dropping, i was using 100mw 433 gear only. the next trip i will be taking my 100mw transmitter, my 1 watt transmitter, and a bunch of identical 75mhz equipment with failsafe/glitch correction/pcm versions to see which works "best" in a harsh environment.

Thanks Tim.

Actually the RF power meter you describe is not that expensive when you consider what it does.

However I think I will have a go with the FrSky LCD display and the Spectrum Analyser for a bit. I'm really looking for differences between modules (i.e checking for damage.) (I have another 2 on order for comparison.)

It is the 100mw TX modules that I'm using at the moment. I was using the 1W module (turned down to  power level 4 to keep it legal) before my problems started and this module seems to give longer range now. (As well as I can measure it by stepping down the road. The neighbours are getting suspicious!)

But Dave J is having no range problem with his 100mw module.

By the way David - I checked the Orange TX module and the aerial does not seem to be connected. (No lands connected on the printed circuit board.)

David

Just to write up what I've been up to.

Well I'm pleased to say that I have got my range back on openLRS and this has been confirmed with 3 outings at Barrow and Bournville with both my Type XXI and the Holland.

I wish I could give a single "magic" solution but (as is always the way!) I have tried a number of things:

(a) Eliminating screening problems.

I have eliminated threaded steel rods inside my WTCs. Convenient but they can cause screening. I have converted to using plastic rods (Delron in my case, I had some handy.)

(b) Simplify everything - Go back to "Vanilla"

I have now accumulated 6 Orange receivers and 3 TX modules. They all had different software versions. I also had several DIY antennae fitted.

- So I refitted the simple commercial antenna.

- I standardised on one software version (3.8.2) set up for Telemetry (Telemetry is too useful not to use. Even the packet loss "beeps" are handy to have.) I have no evidence to say that more modern versions (Currently 3.8.8) were any worse. I have archived Version 3.8.2 + some RX and TX configuration files on the AMS Website. It occurs to me that openLRS is being developed for drones rather than subs, and our interests may diverge.

- I carefully checked over the equipment. One fault reported on the web is that the internal aerial lead on TX modules does not have the earth (ground) connected. I found this to be the case on the 1 Watt module I had. Annoyingly this was also the one on which I have done all my aerial trials! So I think my previous results need to be taken with a pinch of salt.

- I also found that 2 of the RXs had stopped working and just gave a red warning light. I found that this was because the bootloading software had become corrupted. The cure is reported on the Hobbyking website in the RX manuals section. You use a USB asp programmer(About £5 on eBay)  on the 3x2 pins on the board. Connect up and select load boot loader in the Arduino IDE on your computer. (I've just revived one of Dave J's RXs in the same way.)

(c) Range test "On the bench".

I'm saving up for an RF power meter (Thanks for the ideas, Tim) Until then I put the model (with a large piece of white tape on the rudder)in the front window of our cottage and walked over the green, across the main road, and walked over by the Church. If it still worked - the range is fine! After a day or so of this the neighbours were convinced I was mad!

This showed that all my equipment was undamaged and it gave range results close to those "on the pond."

So all is fine again. I have standardised on the use of Tim S's Telemetry. It is so useful - for signal strength, battery volts and amps. A shame that I see that the FrSky fld-02 LCD screen is no longer available though.

My recent trip to Bournville and contact with my fellow modellers showed me a few things:

- I was still the only person using openLRS technology. (Everyone seems to be convinced that 40Mhz equipment is still available second hand or from Germany. I did notice some minor squabbling about frequencies, though - no pegboard was available.)

- The videos of Dave J and me using the equipment with their attendant "telemetry beeps" had rather panicked people into thinking that the range was very limited. Don't believe everything some folks tell you on the Internet, beeping does not mean that you have lost control.

Tim S has suggested muting the beep in the software, if it gets annoying. Until then I have adopted a "third world" solution - sticking some tape over the beeper in the TX module!

David

David,

glad to hear you have got your range where you want it. sad to hear about the FLD-02 display being no longer made, but it was inevitable. FrSky has moved to the Taranis and its associated "S-Port" devices, guess you will have to save a little more money for a taranis :) or build a FLD-02 replacement if yours ever dies. that shouldn't be too difficult. the library we use in building the telemetry hub tells us everything we need to know to build one using a graphic LCD panel, an arduino and a couple push buttons.... just a thought.

Some photos of the Telemetry hubs in 2 of my models. They are based on the nice and small Arduino Pro Micro.

The first photo is the original setup in my Holland. This was a modified wifi switch which gave a low battery voltage signal. It was PIC based and a little bulky.



This is the Hobbyking RX with the Pro Micro stuck to its back.



The back of the RX:



This is the more capable Telemetry hub for my larger subs. This sends battery voltage, current draw, depth and compass heading.



The back of the Pro Micro showing the compass sensor board.



The Arduino Pro Micro



These are both done using Tim S's software and construction details shown over on the SubPirates Forum:

http://www.subpirates.com/showthread.php?5271-Custom-Frsky-Telemetry-Hub/page9

A pm from Fred E (Deep Diver) which I am re-posting because I don't know the answer. Anyone?

Fred wants to move to openLRS but apparently the Hobbyking RXs are out of stock, at the moment, and he wants to know if the Flytron receiver will work with other TX modules, specifically the Turnigy 9xr pro. (I wonder if he could also  think about other openLRS RXs?)

The detail of his  message below:

"in the message I asked for your opinion on the Flytron receiver as the one from Hobbyking is still out of stock, I have read all the spec and to me they look the same.

I went with the Turning 9xr pro as used one of the first one's at work when on of our eng's used it for his home made drone, also I can get a 2,4 module if I decide to go with a 2.4 system for my surface boats."

David

David,

basically, anything on this list should work together:

https://github.com/openLRSng/openLRSngWiki/wiki/supported-hardware---feature-table

as long as the TX module and RX are flashed with compatible versions of the firmware, they should be able to talk to each other without issue.

I personally own the following brands:

TX modules:
OrangeRX futaba 100mw
OrangeRX futaba 1w
OrangeRX JR 1w
DTF-UHF JR 1 w

RX:
OrangeRX 9ch receiver
Brotronics 4ch remix
DTF-UHF/Hawkeye 1 watt long range receiver
DTF-UHF/Hawkeye 6 channel receiver

the only limitations i have encountered are the limitations imposed by the design of the device, for example, if you have s-bus output active on the Brotronics 4ch remix, you cannot have telemetry active (they use the same set of pins)

I have purchased the brotronics 4ch remix from this site, i will recommend them with one caveat, you need to watch them like any chinese reseller, sometimes they "forget" to send you things like tracking information, but so far they have delivered what i have ordered:

http://www.hobbiesfly.com/

they are listed as a preferred site on the openlrsng.org site, so that's why i went to them.

i noticed they have this receiver, which is "slightly better" than the OrangeRX 433 lrs receiver, but almost the same price, and is currently in stock:

http://www.hobbiesfly.com/transmitter-receiver/Receivers/hawkeye-9-channel-receiver.html

one note: this receiver does not come stock with an antenna, so you will have to make one yourself, or buy one, the same site does sell those as well.