DIY 433mhz long range system

Hi guys, I like it here more than RCGroups, nothing wrong with RCG and it's a great resource but were a tight little group where everyone knows everyone plus this is a radio forum so I'm posting my project here...

This is a replacement board for the [no longer available] OrangeRx LRS modules. Running the HopeRF RFM23BP module at 6V this board is a full 1W transceiver capable of accepting PPM input or having a PPM output (via same pin depending on which FW you load: RX vs TX). Also a fully transparent serial port is also available so the same system can pass RC commands from ground to aircraft at the same time as passing 2-way data comms' for telemetry / GCS software (such as MAVlink control). The board is designed with Ultimate LRS in mind however as it's an exact clone of the Orange module it will also run OpenLRS NG (which BTW I'm looking more into now as it's directly compatable with Taranis unlike ULRS)

The board accepts up to 16V in (4S) and can actually be ran on up to 40V with one slight change (a trace cut and an air-wire). There is also a FDTI compatible header at the top (same thing as the serial port on the side but w/o the resistors in-line to facilitate easier FW flashing with an FDTI).

All components are labeled along with their values except for C4 and C5 (which one, are not actually required and two, will need some testing to decide on exact specs) please see the LM317 datasheet for official TI recommendations. R13 is also unmarked, its a protection resistor for the switch, probably 100-500ohms is good


V1.01 available for order on OSHpark TESTED WORKING


FlipFlap's site ITluxembourg

Arduino pro mini clone $2

Pair of Hope RFM23BP modules $16

LM317-N SOT223 package 20pcs $6


For antenna I've build a Moxon square on a piece of coroplast following this tutorial and will be using monopoles on the receivers. If you're not down to DIY your own antenna VAS offers a very nice Moxon for only about $35 or you can get decent performance from an ebay monopole for say a BaoFeng radio, just make sure you get the correct port for the connector you use to build it.

Dont forget all your passives, I didnt list any cause I didnt need to buy any cause I had it all on hand. The pads are technically 0805 size but I use passives from the SparkFun eagle library which is known to be very large, almost oversized (I regularly use 0603 size pads for 0805 parts cause the sparkfun 0805 pads are so large but I wanted to make it a tad easier for others to solder)

Looks very interesting ! I'm happy to read about this
Just wondering, any possibilities of SBUS output ?

Absolutely, infact it's already supported in a more advanced FW build for the official orange module (which again, this is a direct clone of meaning it will run on this too), I'll be working with the more basic FW at first but for sure, I require sbus output lol.

I think my biggest question is who here can do the protocol work to use the HopeRF module direct in devo? It is SPI controlled after all!

I also like to add 433 to my devo. Possibly could 433 be better to fly inside buildings obstruction, or for very long range FPV etc The modules can work also 900Mhz band depending on configuration. Also telemetry is there.
However I do not anymore want to use ppm in my tx. Added Latency and jitters and why should we deviation direct digital is the future!
If the ULRS project is documented and open source, I hope that some master coders as guys like Goebish, could have a look at and deviate it, so we can hook up a RFM/hoperf module over SPI in our DEVO's.

The RX side we could use the SBUS, or servoboard over I2C, basically the Ultimate LRS design, or your board.
I am happy to be part of testing, as I have two of the 433 orange modules here, but not yet used them yet.

this is great news! I'm off on holiday for 2 weeks next Friday but I'll make an order while I'm away that should arrive once I get back home. thanks for sharing your work bud!

HH if you ordered already please EMAIL OSHPARK RIGHT AWAY! I found a major issue (dead short from 5V to GND). It has been corrected now and is g2g. If you did order email now and you'll be able to catch it, if you dont see this and you dont catch it in time you can fix the short with one slight razor blade cut.

thanks for the heads up bud but when we're away there will be no one there to collect the mail so i'll not place my order till i'm actually away, that way it won't arrive till after i'm home and there will be no issues with it sitting about in the mail box unwatched

Spl input to TX and SBUS output is absolutely necessary.

How do you plan to handle SBUS output? Bitbanging? Because it's 100000baud inverted 8E2 if memory serves well. I implemented SBUS with telemetry reader (reads SBUS, supplies fake telemetry, just for an exercise) on a PSOC4 which has small programmable logic. As far as I know, Futaba uses small discrete inverters to implement it. It's close to impossible to read SBUS without hardware support, but probably you can write it using bitbanging.

victzh wrote: How do you plan to handle SBUS output...

I don't. Flipflap keeps his code closed source and he alone is responsible for it. He has said recently he envisions future development moving over to teensy (and future modules beige based on it) however promises sbus output to be the final "big thing" for the 328p based modules (diy and OrangeRx/ other supported OEMs)

You don't have to invert it, there are a few diy receivers (2.4 I believe) that have non-inverted sbus outputs, that's why the CLI option [in cleanflight] to turn the inverter off exists.

All my parts are ordered for 6 modules, my only question now; when will devo be modified to take the rfm23bp module so there's no need for the ground side ULRS transceiver and we have native telemetey w/ RSSI?

Implementing the openlrsng protocol with an internal hoperf module would interest me as well. Nice thing about openlrsng is that it's open source so this could be a bit more feasible. I'm not sure what kind of telemetry openlrs does because only use it for fpv and have the dates displayed on an osd.

Anyways, just expressing my interest in this project

Cereal_Killer I've ordered everything bar the pcb's and the passives, have you confirmed the latest pcb design as functional yet bud? also could you write a shopping list for the passives?

HappyHarry wrote: also could you write a shopping list for the passives?

All 0805 (note I had 0603 470 resistors and plan to use those, they'll be small on the pads but do fit on the package pads I used)
Resistors
220
470
1k
1k8
2k2
3k3
10k

CAPS
150nf (that's 0.15uf)
1uf

Regulator
LM317D (SOT223 package, I got a 20ct pack from ebay for ~$6).

If you plan to hook anything up to i2c (direct at arduino pins) you'll either need to ensure your device has pull-up resistors on board or you'll need to add your own (there are no pads but plenty of space to do so). I2c devices include the Adafruit 16 servo breakout board (draws LOTS of power and will need it's own UBEC for 5v) [air side] or a small OLED display to show RSSI [ground side]. A 2-line LCD [ground side] is in the works too but currently only shows debug data (OLED doesn't need extra power, can run off arduino on-board 5v vreg but ALL it displays is RSSI)

Also, tho it's confirmed not needed (partly cause of how close I put the vreg to the HRF input) I'm likely going to add a decoupling cap for the output of the LM317, that's a small (don't know the package name) tantalum cap ~1-4uf. Didn't order them cause those are one of the parts I always make sure to harvest off old deal electronics and have a million.

I'm still waiting on my boards, if you want to wait till I receive and build them that'll probably be another week and a half or so...

many thanks bud, I'm off on holiday today so by the time I'm home you should have received and checked the boards so that works out fine, but I'll order the passives now so I've got everything ready for the pcb's. one of those little 1"x1" oleds for rssi sounds good so I'll probably get one of those too

i have use devo7e flashed with deviationtx, ppmout to oplink(use RFM22 module), also i have DIY a 1W(use RFM23BP module).
the big problem is, if i use high power level such as 100mW, when the antenna close to LCD, the menu will go to be mass, seems the RF EMI cause the firmware 's buttonscan function detect key press, and when use low power or be away, all OK.
Is it possible to improve the firmware to get rid of the RF EMI

First two built
Image link:

nice one bud!

A pity that FlipFlap, doesn't want to share the source. Hooking up the RFM module straight over the SPI lines into our devo's radios. So even a diy project needs reverse engineering.........
PPM days are over, full digital link is what we need and it is the power of deviation.

But Cereal on your work, nice job clean and compact long range RX !

I've actually spoke with flipflap [via PM] about that (deviating his protocol). He told me he would be willing to work with a Devo dev on his protocol (even knowing devo is 100% open source so his code would be shared publicly) but the issue raises from the advances he makes. Just recently he went from single channel with no need to bind to a hopping scheme with binding. All of this would mean the devo protocol needs constant upkeep and it's just not something I'm able to do.

If someone here was up for the job of protocol upkeep I really think I could get Ben (flipflap) to atleast provide some basic documentation, atleast to us, so the protocol wouldn't have to be reversed...

Anyone down for a new job?