My Devo 7E 4-in1 + 2x3 Build (and Syma X4 Issues)

All,

Firstly thank you to the community (specifically victzh, PhracturedBlue and hexfet)! – You guys do amazing work and I can’t believe how useful competent and feature rich a ‘cheap’ tx can be!

I’m pretty new to the RC hobby, and starting to become a little obsessed with flying, and specifically FPV flight, so I’m doing an awful lot of learning, and paying a fair noob-tax along the way. I wanted to document my progress with my 7e, first to remind myself what I’d done, and second to put all the information in one place in case it helps anyone else (I had to do a lot of searching to find the information I used to get a working 7e! – and found some awesome resources along the way).

First Build (skip this if not interested, it’s here for completeness and background to my current troubles – more on that later)

After a little reading around, I found that DeviationTx with a NRF24l01 would be able to fly my cheap XinXun X39 – This later turned out not to be the case as my X39 is a newer revision, and I’ve not got round to trying to figure out the protocol / it was easier to swap the flight controller for a £6 SymaX5 R6 board that is far more responsive and fun! – More on that later.

I picked up the Devo 7e from eBay for ~£38, and picked up a NRF24l01 ~ £6 from bang good from reading the comments on hacksmods.com
This is the exact one I got - banggood.com/DEVO-Transmitter-Tuner-Modified-for-FBL80-FBL100-WLtoys-V922-p-89812.html

I downloaded and installed DeviationTX following the instructions in the getting started section.

Installed the module as per the instructions and pinout on hacksmods.com (linked above) and set up the hardware.ini to use the TMS pin (A13).

All went well – But I couldn’t bind to My XinXun X39 or by now I’d picked up a cheap Syma X7 too. – I tried a whole lot of things, including separate 3.7v regulators, voltage smoothing capacitors and thicker gauge wires. – Nothing would let me bind!

I went through two more nrf24l01+ boards, one with poweramp and LNA, one without:
www.ebay.co.uk/itm/161862594141
and
www.ebay.co.uk/itm/161844826577

After killing the new one with PA & LNA wiring it backwards to my home built 3.3v regulated power supply, I started trying other protocols with the non-pa version. – This is when I realised that I could fly my X7 with the YD 717 X4 protocol variant. – But it was very laggy – you push up on the throttle, and in something between 100ms to 2 seconds, the quad will respond, but all input seems to have this variable lag.

On a hunch that maybe my original rf24l01+ PA + LNA board actually still worked, I swapped that back in. – This time it worked without issue. No lag, and the X7 flies well on it!

Long story short, the information on the Wiki (wiki / supported_models) Isn’t correct – for my models at least.

The Syma X7 – Binds and fly’s using the YD717 Syma X4 protocol, not the SymaX protocol listed on the wiki.
The XinXun X39 r3 board is new and I can’t fly it with any protocol I’ve found yet – It was on my to-do list to try and reverse engineer it – but it was cheaper and more fun to replace the flight controller with the Syma X5-1 v6 board and fly it straight off.

New Antenna and Failed Range Mod:
The next surgery my 7e had was to parr off the existing antenna, and replace it with a bulkhead RPSMA pigtail. Again this was pretty straight forward, and I didn’t follow any specific tutorial, but my method pretty closely matches HappyHarr’s antenna mod here:
deviationtx.com/forum/how-to/5955-7e-ultimate7e-adding-the-devo-version-4in1#49844

Then very gently flame polishing the raw cut plastic of the original stub to match the shiny finish of the 7E everywhere else – all worked well!

Next I picked up a RX601 as a friend offered me a fixed wing aircraft that just needed a rx and charger, so I decided to try for the extended range mod of the stock CRF6936 – I picked up some silver conductive ink from Maplin ~£8 and gave it a go. – It was terrible, and I got the ink everywhere. – So I tried soldering, and managed to destroy the diode and get something too hot I think. Needless to say it wouldn’t bind to the RX601 at all after that. >.<

Time to start looking for a replacement CRF6936!

4-in-1 Devo Mod.
While looking into replacement modules for the one I fired, I came across the 3-in-1 and 4-in-1 modules thread.

Let it be said now that I can be obsessive, and yet impatient. I really liked the Idea of victzh’s daughter board, but as I said, I can be impatient and thought if I can get hold of the Devo 10 daughter board being talked about, I’d give that a go – All four modules, one antenna and only one wire to solder to the MCU. Sure let’s give it a go!

Enter a new world of dealing directly with China!!

I’m lucky that I work for a large household name computer games company, and we have a translations department. Talking with them, and then organising shipping via taobaofocus.com I managed to get hold of the 4-in1 Devo 10 daughter board. (It’s the new 13 pin version of the 4-in-1 and is the Y220.00 ‘blue’ option here: item.taobao.com/item.htm?id=27232304992)

Following the pinout here (thanks HappyHarry) - deviationtx.com/forum/how-to/5955-7e-ultimate7e-adding-the-devo-version-4in1?start=20#50816

All was installed and works well! – Bound to the RX601 fine, no module errors on the TX and have been happily flying an X5, X5-1 Rev 6, and Cheerson CX10-W!

And just reminding myself here that the chip enable pins are connected too:
TMS is Port A13
TCK is Port A14
Pin 50 (second from the bottom on the MCU) is Port A15

My only issue is that the YD 717 X4 Protocol is now again being really laggy! – There’s between 100ms and 2seconds of lag between human input on the transmitter, and the Syma X7 responding  - I’m really tempted to just replace the flight controller with another X5-1 Rev 6, or even a DSM2 Brushed Naze32 controller ~£24.59 banggood.com/Micro-Scisky-32bits-Brushed-Flight-Control-Board-With-1_25mm-Plugs-For-DIY-Micro-Quadcopter-Frame-p-1067976.html that would let me start to experiment with ACRO mode cheaply!

2x3 Switches mod.
Possibly the simplest mod I’ve done, but included for completeness; drilling out the 2 spare 6mm holes on the front and wiring up two three way switches as following Strafferr’s diagrams here:
rcgroups.com/forums/showthread.php?t=2468549

Current Issues (Help Appreciated):

• The Syma X7 is listed as SymaX protocol on the supported models page. – I can’t get this to bind using that, only the YD 717 (Syma X4 Protocol) – but seems buggy on different NRF24l01+ implementations. Works well on some, laggy on others.
• The YD 717 (Syma X4) Protocol seems to have issues playing nicely with the NRF24l01+ that’s included in the new 4-in-1.
  I’ve confirmed this isn’t an issue with the quad, as it flies fine on the original NRF24L01+ PA + LNA with this protocol.
  The 4-in-1 NRF24l01+ seems to be working correctly other than this protocol – it flies SymaX for a 5C and a 5C-1 Rev 6 perfectly. Along with a CX10W (Cherson CX10W protocol – blue variant)
  This seems to be the same issue that rilot encountered here:
  www.deviationtx.com/forum/6-general-disc...with-7e-and-nrf24l01
  But I don’t really want to have to take out the 4-in-1 that’s working well for everything else, just to be able to fly the Syma X4 variants.
• I haven’t found a protocol that works with the XinXun X39 r3 – The original is supposed to be the YD717 SkyWalkr protocol, but I can’t get this to bind. As mentioned, I’ve swapped out the flight controller to get it back in the air, but it would be nice for my Son to be able to fly his if anyone has updated info on this model?

Haven’t found a protocol that works with the XinXun X39 r3 – The original is supposed to be the YD717 SkyWalkr protocol, but I can’t get this to bind. As mentioned, I’ve swapped out the flight controller to get it back in the air, but it would be nice for my Son to be able to fly his if anyone has updated info on this model?

Again, thanks to everyone, and I hope I can start to give back to the community that’s given me so much stress, interest and joy over the last couple of months!
Nathan.

i fell into a similar trap with the nrf24l01 module. the cheap ones with built in antenna are junk and no good beyond about 10m.

i replaced it with a slightly more epensive pa lna version with shielding over the rf chips, this works much better than my stock t which i suffered random signal loss even at close range (not very frequent but disconcerting hen heli didnt respond for a second)

it may turn out that something in the protoccols may need to be tweaked for some models hppefully in an updated build of deviation

I had lag problems
www.deviationtx.com/forum/6-general-disc...others-with-nrf24l01

I solved it by replacing the nrf module. I suspect the problem was due to a cloned chip not being a 100% clone. I did read that there was a mistake in the original data sheet and the clones accurately copied that.

Hi doctortim!

I found the info I think you're talking about:
forum.mysensors.org/topic/1153/we-are-mo...fakes-are-emerging/7

And that's kinda troubling if that's the case as its not a "cheap" (cost not nescaserily quality) chip I'm using, it's one of the "new" 4-in-1 modules, so of they're plagues with bad fakes, that's going to make quite a few people unhappy I think!

Cheers for the info, and ill have a pole at the code and there's some register responses that can be used apparently to identify counterfeit chips!

Cheers again,
Nathan

Hi Nathan,
It was actually this hackaday discussion hackaday.com/2015/02/23/nordic-nrf24l01-real-vs-fake/
Take a look at the replies from Jay Tyzzer, especially this bit....

Let me start by saying that I do have a dog in this fight as I work for Nordic Semiconductor.
The re-marked nRF24L01P (+) clones are not 100% register compatible. The issue with the counterfeit devices is that when they enabled “Dynamic Payload Length” (EN_DPL) in the “FEATURE” register, one bit get’s activated in the on-air payload (the NO_ACK bit) This bit should be active high (according to the Nordic datasheet), but it’s actually implemented the other way around. When EN_DPL is activated, the NO_ACK bit get reversed in the real nRF-devices. They did such a good job of cloning they cloned the datasheet error into the device!!!

cheers,
DrT

Hi doctortim / all,

So after a holiday and some time away, I managed to have a poke around and fix my lag issue on my Syma X7.

It does look like either the nrf24l01 that's included with the 4-in-1 module is suspect, or the LNA / Antenna switcher isn't working correctly.

If I disable Auto-Ack in the YD717 protocol, I can get my Syma X7 in the air again!

I branched the 5.0 source, and added a user option to the YD717 protocol to allow quick disabling of Auto-Ack. With AA enabled, there's significant lag, all the way up to several seconds, as soon as it's disabled, the Syma X7 flies as well as any of my other quads.

Not sure what this means for using the 4-in-1 "nrf24l01" for telemetry, but at least it gets me flying with my Syma again!

The changes I made are minimal but are here if anyone wants to see how little changed.
github.com/NathanBarguss/deviation/blob/...col/yd717_nrf24l01.c

And this is the difference dissabling AA made:


I'll attach my YD717.mod in case it's usefull for anyone, and thank you to everone who's helped out.. it's been a bit of a learning curve, but I've caught the bug for this!

Cheers,
Nathan.

Does the 4in1 devo from banggood has this issue?

Yes the BG modules are experiencing this issue. A workaround is to disable AA as demonstrated by Nathan. However if you don't fly quads assoicated with YD717 protocol then it won't affect you.

@ NathanBargus
on the same protocol i can use deviation 5 on my devo 7e but with nrf module standalone intalled same experience, every time want to fly must set AA bind and then unset AA.
this is my toy and must use y717 protocol
www.himotoracing.it/sito/prodotto/drone-mini-spider-himoto/
im waiting new cyrf module from walkera, then will try your file

i suspect is not only banggood module fault since i have a different one and as stated all clones have this bug.