2.4 modules discontinued!

Like B&Q for homebrew radios
GarydNB
Posts: 282
Joined: 15 Feb 2018, 23:12

Re: 2.4 modules discontinued!

Post by GarydNB »

I've got a DHT I'll never use if someone wants to spend a tenner plus postage..
Gary
User avatar
Shaun
Posts: 1049
Joined: 15 Feb 2018, 21:49
Location: West Yorkshire

Re: 2.4 modules discontinued!

Post by Shaun »

I'll have it mate.
Tobe
Posts: 665
Joined: 16 Feb 2018, 06:19
Location: Varberg or Stockholm, Sweden

Re: 2.4 modules discontinued!

Post by Tobe »

Cheers,

Tobe
GarydNB
Posts: 282
Joined: 15 Feb 2018, 23:12

Re: 2.4 modules discontinued!

Post by GarydNB »

This is all very complicated.....couldn't one of our resident geniuses just clone a frsky module? Or is it harder than that?
Martin
Posts: 744
Joined: 16 Feb 2018, 14:11
Location: Warwickshire

Re: 2.4 modules discontinued!

Post by Martin »

GarydNB wrote: 28 Jan 2020, 15:03 This is all very complicated.....couldn't one of our resident geniuses just clone a frsky module? Or is it harder than that?
Pretty easy to make them but getting FCC/CE type approval is another matter.

Someone could buy approved multimodules, test each module individually to discover the optimum tuning using a Taranis or similar, and then reflash that module's firmware so that the default 'tuning' is spot-on. It would take maybe five to ten minutes per module once they had their 'production line' set up.

The source code is freely available so it's just a matter of editing the default tuning value, recompiling and reflashing - it's similar to programming an Arduino.

FrSky and other manufacturers of 'fixed' modules must do something similar - though of course they will have automated test and assembly equipment on their production lines that can do the job automatically in seconds. I'm surprised that the more reputable manufacturers of multimodules don't already do the same.
User avatar
Mike_K
Posts: 669
Joined: 16 Feb 2018, 06:35
Location: Hertfordshire

Re: 2.4 modules discontinued!

Post by Mike_K »

GarydNB wrote: 28 Jan 2020, 15:03 This is all very complicated.....couldn't one of our resident geniuses just clone a frsky module? Or is it harder than that?
It's already been done:

https://www.rcgroups.com/forums/showthr ... RX-Modules

You buy a CC2500 module with amp from Banggood or eBay and connect it to an Arduino, program it and you've got the equivalent of a DHT. And I'm certain it could easily be integrated together with the Phil_G encoder so you wouldn't need two Arduino, total cost around £15. But as Martin says, getting it CE approved (or FCC) is another matter. And from 2015, the FrSky DHT didn't meet the latest CE requirements for LBT and technically only equipment already in the supply chain was exempt (as far as I understand) and I think somebody must have turned a blind eye, for the supply to have continued until 2019. But it still meets US FCC regulations to this day.

Martin wrote: 28 Jan 2020, 15:23 I'm surprised that the more reputable manufacturers of multimodules don't already do the same.
I'm as surprised as you Martin. But how would they do it? I've had a quick look through the Multi-Module firmware and nowhere that I could see does it read a factory calibration value. Yes, you can set a frequency offset in the multi-protocol module firmware, but it's not in EEPROM and the next time you update, the value would be over-written. When you calibrate with a Taranis, I think the value is stored on the transmitter, not the module? So it would require a multi-protocol module and OpenTx firmware update to achieve? Probably a question for an OpenTx forum, not here!

Mike
Martin
Posts: 744
Joined: 16 Feb 2018, 14:11
Location: Warwickshire

Re: 2.4 modules discontinued!

Post by Martin »

Phil_G wrote: 28 Jan 2020, 18:01 No the offset can be stored in the module Mike, its incorporated at compile time,
otherwise it couldnt be done in PPM mode ;)
I think what Mike means is that the calibration value isn't stored independent of the code (in EEPROM) so it wouldn't survive a subsequent firmware update.

People who upgrade vintage radios to 2.4GHz are unlikely to ever want to reflash the module firmware so it wouldn't matter much to them, but ordinary users of multiprotocol modules are likely to reflash (for example when new protocols are released). That's probably why the module manufacturers don't bother with factory calibration.

Presumably the modules sold by FrSky that can be flashed with firmware updates but don't require "tuning" afterwards do store their calibration values separate from the code.

If someone is going to calibrate the modules then it would be a good idea to write the "tuning value" on the module as well as editing it into the firmware - that way, if a user ever did need to do a later firmware update they'd know the correct value - but it would still mean recompiling the source rather than just downloading and flashing the binary file. The latest modules make it very easy to do the latter as they have a built-in USB port, and updater programs that can be used by non-experts.
jackdaw
Posts: 165
Joined: 16 Feb 2018, 20:30
Location: Wet and Windy North Wales

Re: 2.4 modules discontinued!

Post by jackdaw »

Being an owner of a Jumper T16 Pro Hall Tx(on OpenTx 2.3.5) with a built in MPM and a plug in Jumper TP4in1 MPM(latest version without dial, just two lights, a micro USB and installed bootloader) that I bought for my Taranis perhaps I can provide a little more info. I've done this 'tuning' on a couple of genuine FrSky Rx and they do vary from Rx to Rx. The value of 'tuning' obtained is stored in the model memory on the OpenTx plus if you give the Rx a number it is related to that. If useful I could run a few additional tests or answer questions within my ability. Can't report on the plug in module that I bought for my Taranis as I have yet to update it from 2.0.13 to 2.3.5, a bit concerned by the big jump.
Note: both the internal and plug-in MPM modules are on the latest firmware, 1.3.0.53 and non EU Open Tx.
User avatar
Mike_K
Posts: 669
Joined: 16 Feb 2018, 06:35
Location: Hertfordshire

Re: 2.4 modules discontinued!

Post by Mike_K »

jackdaw wrote: 28 Jan 2020, 19:56 I can provide a little more info. I've done this 'tuning' on a couple of genuine FrSky Rx and they do vary from Rx to Rx. The value of 'tuning' obtained is stored in the model memory on the OpenTx plus if you give the Rx a number it is related to that.
That has confused me (though it doesn't take much to confuse me nowadays). If FrSky Rx's vary from each other, I wonder how Taranis (or our original DHT for that matter) work with any Rx you throw at them, without having to do any tuning? They all "just work". Does FrSky allow a bit of leeway from centre? It's a long time since I looked at GFSK, but if I remember, the curve is "bell-shaped", so you could be a bit off-centre and it should still work fine?

Tuning each TP4in1 MPM before it is used for conversions is do-able, but tuning one to every FrSky Rx you'll ever use with a conversion wouldn't be. But the two Rx I tested when tuning were near-identical, did I get lucky or maybe there are some rogue Rx out there?

But until we all sort the FrSky "tuning" issues out, we can still use the MPM with DSMX and FlySky as it works fine with them, so not all is lost.
User avatar
ozrs
Posts: 114
Joined: 16 Feb 2018, 15:21
Location: West Australia

Re: 2.4 modules discontinued!

Post by ozrs »

Some thoughts:

It seems that genuine Frsky TXs, RXs & modules are factory matched sufficiently to not need any tuning.

The Grayson Hobby video linked earlier suggests that Jumper branded products (4in1 module and RXs) may also be factory matched.

The issue may arise when mixing various manufacturers. In our case, this being the 4in1 modules with rotary switch (to use in PPM mode), mainly the Irange IRX4 and Jumper (if the older Jumper module with rotary switch is still available?) and genuine FrSky RXs. There is obviously some allowable tolerance, and it would need to be determined how close you need to be before binding and range are affected. (The video indicates that maybe a range of +-40 units may do to cover differences).

I think most of us could readily determine the offset for a new 4in1 module fairly easily if we have an OpenTX radio or by accessing a friends X9D+ or T16.

The difficulty for most is likely to be compiling the code to implement that offset.

I envisage 2 solutions:

1) Someone with the required skills compiles a variety of firmwares each with a different offset (from -XX through to +XX at set steps) so that users can select the appropriate (closest) one for their particular module and flash that.

2) Since most of us are probably only going to use the module for FrSky RXs (& perhaps SpeKtrum DSM2, DSMX or Flysky AHDS2, AHDS2A), the firmware could be compiled that instead of giving the module 31 different protocols, it only gives say DSM2, DSMX, AHDS2, AHDS2A plus 13 D8 protocols (at various offsets) and 13 D16 protocol (at various offsets). Thus once you know you offset, you could select the closest via the rotary switch.
Richard
Post Reply