Thread Tools
Aug 01, 2020, 10:04 AM
Registered User
Something sounds wrong with the mod.
If you can point me to a full diagram, I'll take a look.
It sounds a little bit like E and C swapped on a transistor but without details I don't even know if there is one to get swapped.
Sign up now
to remove ads between posts
Aug 01, 2020, 10:52 AM
Registered User
Markran's Avatar
Quote:
Originally Posted by Mike Blandford
I've been looking at possibilities to allow the use of a MPM, operating in receive mode, to output trainer data to FrSky radios. By adding a transistor inverter, driven by the normally unused Tx pin (USART1), I have some things actually working. Due to differences between FrSky transmitters, not all possibilities work with all transmitters, although using pin 2 of the module connector should be OK for most, if not all.

The Aloft Linker outputs either SBUS or CPPM to pin 2 of the module bay, I believe CPPM is more generally used as there was a problem in openTx at one time when using SBUS.

So, here is a list of the possible way ahead.
1. Do nothing!
2. Use Time Division Multiplexing on the SPort signal - would need support in radios, and ACCESS radios appear block the SPort signal.
3. MULTI_TELEMETRY to pin 2 via inverter, pin 5 off - would need support in radios.
4. SBUS to pin 2 via inverter, pin 5 off - may not be supported on some radios (I have this working on a X10S Express and a X9D+).
5. CCPM to pin 2, possibly via an inverter, pin 5 off - widely supported.

The Tx pin also has a timer 1 output available, so this could be used to generate the CPPM signal.

One reason for using the transistor inverter is that some older radios (non-FrSky) output either a supply voltage or a RS232 level signal on pin 2 and the transistor inverter I have handles such inputs without damage. Also, ACCESS radios appear to have an inverted USART input on that pin.

Option 5 should only need changes to the MPM code, and a single wire added if not using the inverter, and then work with many FrSky radios.

What does anyone else think?

Mike

Mike,


First, thanks for doing all the discovery on this. Since I'm not a hardware guy, the most interesting option to me would be #5 since I feel reasonably comfortable doing one-wire patches. If it gets into adding electronic parts I'd probably pass on doing the mod but that's just due to my knowledge level.
Latest blog entry: Champ S+ Fly Away
Aug 01, 2020, 04:56 PM
Registered User
Quote:
Originally Posted by BeanieBots
Something sounds wrong with the mod.
If you can point me to a full diagram, I'll take a look.
It sounds a little bit like E and C swapped on a transistor but without details I don't even know if there is one to get swapped.
It isn't discrete components, it is just a single SN74LVC1G04 (or equivalent) inverter. It has 3.3V/GND, with the TELEM pin from the Multiprotocol module connected to "A" (input) and the STM32 in the Taranis connected to "Y" (output ). I don't think there's any way to mess that up, but don't know much about electrical engineering so I'd love for you to take a look. Photos of it in place are in installation instructions:
https://blog.seidel-philipp.de/fixed...and-frsky-qx7/
Aug 01, 2020, 08:13 PM
Registered User
How do you set fail safe in this module please? i don't appear to have the option to do this and without it, the multi protocol module will be of little or no use . Thanks Taggarc
Aug 01, 2020, 09:18 PM
65 years of RC flying
Daedalus66's Avatar
What protocol are you talking about?
Aug 02, 2020, 01:35 AM
OpenTX core team
3djc's Avatar
Quote:
Originally Posted by taggarc1
How do you set fail safe in this module please? i don't appear to have the option to do this and without it, the multi protocol module will be of little or no use . Thanks Taggarc
If you use OpenTX, the failsafe option should be available as usual for protocols that do support it
Aug 02, 2020, 03:32 AM
Registered User
Quote:
Originally Posted by CapnBry
It isn't discrete components, it is just a single SN74LVC1G04 (or equivalent) inverter. It has 3.3V/GND, with the TELEM pin from the Multiprotocol module connected to "A" (input) and the STM32 in the Taranis connected to "Y" (output ). I don't think there's any way to mess that up, but don't know much about electrical engineering so I'd love for you to take a look. Photos of it in place are in installation instructions:
https://blog.seidel-philipp.de/fixed...and-frsky-qx7/
As you say, little chance of messing it up but there are a few possibilities.
Two things to check for:-
1. When the transistor was removed, double check with a good magnifier that no tracks were lifted or cracked.
2. From your description of voltage and symptoms, it sounds like insufficient drive which made initially think CE swap. Similar characteristics are possible when a chip does not have its power supply connected. It is possible for a chip to be powered by one of its inputs (called parasitic power). When this happens, function can become erratic and sometimes even go unnoticed. Typical clue is reduced output voltage.
If you have access to a DVM, check (right on the chip pin) that 3.3v is actually getting there.
Aug 02, 2020, 04:44 AM
Pascal
hpnuts's Avatar
Quote:
Originally Posted by mog_joe
Where can I download the BIN.File? X9D STM OpenTX
Here: https://github.com/pascallanger/DIY-.../tag/v1.3.1.49
Aug 02, 2020, 04:50 AM
Pascal
hpnuts's Avatar

Multi V1.3.1.49 Released


Multi V1.3.1.49 Released on GitHub.

Change Log since v1.3.1.36
  • FrSky R9
    • FCC sub protocol added
  • V2X2
    • Fixed issue from last update...
  • Kyosho
    • New protocol
    • Surface protocol called FHSS introduced in 2017. Transmitters: KT-531P, KT-431PT, Flysky Noble NB4 (fw>2.0.67)...
  • Flysky AFHDS2A
    • New sub-protocol supporting up to 16 channels with LQI on channel 17
  • FrSkyX/X2
    • New "Cloned_8CH" subprotocols for 8 channels transmission (instead of only 16 previously)
  • RadioLink
    • New protocol
    • Surface sub protocol, TXs: RC4GS,RC6GS. Compatible RXs:R7FG(Std),R6FG,R6F,R8EF,R8FM,R8F,R4FGM,R4F and more
    • Air sub protocol, TXs: TXs: T8FB,T8S. Compatible RXs:R8EF,R8FM,R8SM,R4FG,R4F and more
    • Telemetry enabled
    • Only 64 IDs, cycle through then using RX number
  • Height
    • Previously called Flyzone but renamed using the real manufacturer name
    • New sub protocol 5ch: the one available so far
    • New sub protocol 8ch: upgraded version of the protocol to 8 channels
  • OpenTX Multi Lua scripts
    • MultiChannelsUpdater.lua and MultiChan.txt updated with all the latest protocols and sub protocols
As usual test and report.

Stay safe, Pascal
Aug 02, 2020, 05:24 AM
Registered User
Hello
what means the -- submenu in FrSkyR9 settings?
Aug 02, 2020, 05:36 AM
Registered User
Hi all. I would like to ask something. thinking to buy an whoop with frky spi onboard receiver, to bind to my radiomaster t16s internal multi. As i saw , binding process with this whoop (betafpv 65) needs about 1 minute and something. My problem is that radiomaster t16s , when binding with internal module on frsky protocol , go auto out of binding after about 5 secs. You think it will work the all thing ? thanx
Aug 02, 2020, 06:42 AM
Pascal
hpnuts's Avatar
Quote:
Originally Posted by weixelgeist
Hello
what means the -- submenu in FrSkyR9 settings?
Work in progress...
Pascal
Aug 02, 2020, 08:47 AM
Pascal
hpnuts's Avatar
Quote:
Originally Posted by mstranger
Hi all. I would like to ask something. thinking to buy an whoop with frky spi onboard receiver, to bind to my radiomaster t16s internal multi. As i saw , binding process with this whoop (betafpv 65) needs about 1 minute and something. My problem is that radiomaster t16s , when binding with internal module on frsky protocol , go auto out of binding after about 5 secs. You think it will work the all thing ? thanx
That shouldn't be a problem at all. The effective bind is extremely quick.
Pascal
Aug 02, 2020, 09:28 AM
Registered User
Quote:
Originally Posted by BeanieBots
As you say, little chance of messing it up but there are a few possibilities.
Two things to check for:-
1. When the transistor was removed, double check with a good magnifier that no tracks were lifted or cracked.
2. From your description of voltage and symptoms, it sounds like insufficient drive which made initially think CE swap. Similar characteristics are possible when a chip does not have its power supply connected. It is possible for a chip to be powered by one of its inputs (called parasitic power). When this happens, function can become erratic and sometimes even go unnoticed. Typical clue is reduced output voltage.
If you have access to a DVM, check (right on the chip pin) that 3.3v is actually getting there.
I thought that too, that it might have been pulling power from one of the I/O pins instead of the power line. Good idea to check it at the chip. I just did that and it has continuity from the chip's VCC pin all the way to another 3.3V pad on the Taranis so I think that's good. Where the transistor was removed, that pin on the inverter shows continuity from the inverter all the way to the STM32 pin so that's good too.

The inverter circuit on the iRange4 module is slightly different than the MPM schematic in git. Instead of the dual channel inverter chip, it just has this single channel chip. I'm not entirely sure what the second channel is doing in the MPM schematic, but perhaps that diode and 4.7k pullup resistor it has makes the difference. I think Jumper's MPM uses that dual channel chip as well.

I flew for half an hour with my kludge resistor in place and I never heard OpenTX complain about the telemetry so it may be ugly but it does work around the problem if anyone with the exact same hardware bits needs a quick fix.
Aug 02, 2020, 10:56 AM
Registered User

"Lite" version?


Full disclosure, I have an iRangex on order, so this is just a place holder.

TLDR, this thread is awesome, but HUGE!!

I am wondering if anyone has come up with a STM32 based "Lite" version of the 4in1?
I would love to build one just because I can.


Quick Reply
Message:

Thread Tools