Jumpy channels DEVO12s

Good evening everyone, I'm having some issues with my Devo 12s running DEVO12-v4.0.1-583c468. I installed the multi-module a few nights ago and it is working without any errors. I can connect using all the modules. My problem is when using the FrSKY protocol, all the channels are very jumpy. I thought it may be an issue with the controller and the display, but the FC the rx is connected to is showing the jumpy behavior that is on the DEVO. Any help would be greatly appreciated.

Thanks.


EDIT: When I select the SKyartec protocol, which uses the same cc2500 module, i do not get the jumpy channels.

I spent some more time playing around with the transmitter last night and have come to the conclusion that when Telemetry is enabled in the FrSky protocol, the channels are very jumpy. Once telemetry is disabled, the channels do not jump and operate as normal. Does anyone know if this has been fixed in a newer nightly?

Time to read alot about setting up a build environment to build my own nightlies.

Thanks for the help in advance.

What kind of jumps are you seeing? I've been using the FrSky with telemetry with nightly builds for a while now, and haven't had any problems that would seem to be related to such a behavior. Of course, I'm just using it, not looking at the values from an FC of some kind.

The jumps are small, 2 - 3 points on the analog channels. This correlates into about 5 point jump that the flight controller sees. Which will start the motors on a quad after I arm it.

Now, I do remember PBR mentioned that the 12s was very noisy internally as it's running off a switching 3.3volt power supply. Being the channels that are jumping are all analog, This makes me wonder if there are some issues within the 12s itself.

Could be - I'm using a 10.

Hmmm, just got done looking through the forums to find where PB talks about noise. He fixed it by sampling 100 times and averaging out the result.

I have an extra cc2500 that I will wire in and remove the multi module. See what that does, and see how my wiring may be affecting the setup. And work on getting my computer setup to build the night lies.

Well, my experiments did not go as planned. I was hoping either upgrading to the newest nightly, or using the cc2500 by itself would solve my problem. Neither had any impact on the jumpy analog channels. I re-routed the signal wires as far away from the inputs of the channels, and nothing changed.

Any ideas would be greatly appreciated. The protocol I have wanted the most is the FrSky, and it's the one giving me troubles! Go figure, such is life.

Thanks

I will give that a try!

Was able to carve out some time last night to play around with the cc2500 and multimod with a separate power supply. When using the separate supply, Deviation had trouble detecting any of the modules. I went looking for the source file for the multi mod board to see if I could figure out how it is initializing the modules. I was not able to find it. Can someone help me out on where to find that?

I put my spare cc2500 into my Devo 7e and the there are not any issues with that module. It all seems to be coming down to issues internal to the 12s pcb. If I can't fix the issue by adding a bypass cap or two where I am getting my 3.3 volts from, then I may be reverse engineering the power circuit and build my own to feed the transmitter.

Thanks for the help!

If it's not doing it with other CC2500 based protocols then I'd start looking at the receiver end or the protocol it's self, not the HW... IDK, just seems to me if the problem doesn't exist on SKyartec then the HW must be capable of working right sometime I'd think.

I assume you're looking at the channel monitor of the devo and you see this jitter, not only on the FC side? If that's the case have you switched to the input monitor to see if you see it coming in from the sticks?

Where in the process of taking inputs and turning them into output's does the protocol first come into play? Not till all mixing is done and it's ready to go out or earlier?


Maybe to ease your mind, I fly my 200 Acro Tricopter with a old OrangeRx DSM2 receiver that has ~10 points of jitter when centered (ranges from 1496 to 1507) and it doesn't cause me any issues. Also note, it's a 3D copter (works like a CP heli where stick center is zero, above is + below is -) and even still, it doesn't have any problems flying. What specifically are you encountering?

I do not have a skyartec received, so I just tried setting that protocol up to see what happens. This one did not effect any analog inputs, though it could be because it was able to bind to a rx.

My initial results with Frsky looked liked it was only happening with telemetry on. Unfortunately it does it with telemetry off also when it is connected to the receiver. The 7e has been wired up and connects to the receiver with no jumpy channels. Both running the nightly version from the original post. This module was put in the 12s as a standalone unit and had the same behavior as the multi module. I have. Feeling the cc2500 draws more current fthan the power supply can handle, that will be confirmed later tonight if the external supply removes the analog noise.

The input monitor shows all analog channels jumping. I don't know where in the protocol it stats to send the data out, but if the inputs themselves are not behaving, I can't blame the protocol. I did try a completely basic setup with no mixing, just what the input read is what goes out.

I spun up the quad on my bench ( props removed for saftey) and turned off any self adjusting, pure manual mode and yu could hear the motors spiking up and down with the channels. It's a consistent 3 - 5 points fluctuation. Maybe i just see how it does in the air.

FDR-

I was not thinking and did not have a common ground. I have 120vac to 3.3 volt supply and jus had that connected straight to the module. I'll try with a common ground and hopefully get better results. If that works, I have some mini power supplies from pololu coming this week that I plan on using to use a larger lipo to power my fpv display, the transmitter and now maybe the multi module.

Try lowering the power setting in the model setup. If it's a power draw issue, that should help. PB says that 150 mw only gets slightly more range than 100mw, but uses a lot more power. Of course, the values displayed were chosen for the CYRF chip, and are "close" values on the other modules.

MWM-

I tried lowering the power to the lowest setting, 100uw and it still had the jumpy channel issues.

I just did the external power supply, except I made sure I had a common ground. The unit works flawlessly with the external power supply. I measured the current flowing to the module and with the cc2500 enabled and 150mw set, the unit draws 170mA nominal. When binding, it draws 220ma. The other modules draw 65ma when maxed out at 150mw and don't change consumption while binding.

I checked with the current of the cc2500 in the 7e, I get the same results as I do within the 12s. The only difference is the 7e has a more robust power circuit than the 12s.

I believe this was causing the issue as the 3.3 volt supply can't supply enough current and was causing fluctuations that were throwing the analog calculations off. Thanks for everyone's help, it is greatly appreciated!