noisy and fluctuating analog Inputs and TX Voltage

i got my X8R receiver today
Modded my Devo12s with cc2500 module from banggood.(Fine=0 coars=0)
the receiver bound on first try with hexfets test build.
Great work.
But something is going on with the analog inputs.
Whis a switch as src all is fine.
There is noise... +/- 2-3 digits on all analog prozessor inputs
Sticks, Pots, Slider and @voltage measurement too.
The Tx Voltage display fluctuates +/- 30mVSticks und Voltage fluctuate periodically with ~1sec up and down.
if the cc2500 comes active it messes up the 3,3V i think.
Could a capacitor help??

edit...some tests
Before binding there is no voltage or analog input fluctuating.
with 100uW sending power ther is almost no fluctuating.
the more power the more noise.
Could this be a shielding problem?!?

Unfortunately im no electronic expert

No noise with A7105 ore or cyrf6936 Module active

Fernandez and others have similar problems e.g. Link

any suggestions?

I also have a 12s that has the noise problems, it only happens when using any of the frsky protocols. The combined draw on the voltage regulator pushes it to the limits and creates noise on the power lines. (I have not verified this with an oscilloscope as I don't have one available to me at this time) I am using two small power supplies, LM2596s, to power the multimod seperately and the main board from a 3s battery. The noise on the analog channels is significantly decreased, there are no visible fluctuations on the input test screen. That being said, there are minor deviations on the signals that are showing up in cleanflight. The changes are 1 point changes. IE 1099 - 1100 or 1501 - 1502. That is very small compared to the 4 - 6 point change when I started this process.

Hope this helps. If needed I can send pictures later.

Pictures are yery welcome

I`ve a PC Oscillosco but not much skills.
How is the best way to test.
Only look for ripple on VCC?

As the CC2500 from Bangood is specified with input Voltage from 1,8V-3,6V should i trie to add a diode?
I dont think its only a power problem, because Fernandez has changed the Regulator to a bulletproof Step up/down regulator and the problem is not gone.

I am wondering if it has something to do with the code and handling of telemetry.....

I can understand that a high current draw etc might cause a voltage drop, but this fluctuation is relatively slowly and look like randomly changes to the displayed Tx voltage.....

I have tested with my old oscilloscop.
Look at the pictures(without any modification)
then i`ve tried with Elko up to 1000uF.
This would ca half the ripple.
Then i tried this in combination with Diode N4001 to VCC A7105 and CC2500.
Capacitor on TX(Prozessor Side) not that much better...
Hmmm....
Any electronic expert out there, who can tell me what to try?
Flysky with A7105




FrskyX with CC2500


DsmX with Cyrf6936




and last... no module active (Throttle not at min)

Hmm thats pretty significant so the peak to peak fluctuation is about 0.3v, is it correct. As it is spikes
indeed could happen, that form time to time it is sampled on top or in between etc, could explain why this is random.


Might try a coil in line with 3v3 vcc and a super cap direct at the module?

Can you please send me two links for coil and supercap?
What is the difference between Elektrolyt-Capacitor and Supercap?
And which coil should i try
Sorry ...i`m a carpenter

Can you do the same check on FRsky with different power levels(lower)?

FrskyX D16 150mW





FrskyX D16 30mW





FrskyX D16 1mW





FrskyX D16 100uW





FrskyX D8 150mW





FrskyX D8 100uW





Frsky V8 150mW





Frsky V8 100uW

Same for Flysky

Flysky 150mW





Flysky 100uW

@Fernandez
First test was at 100mW
the peak to peak fluctuation transmitting FrskyX at 150mW is about 0.4v

@wene001 what type of oscilloscope are you using? Would like to get one and the software and readings look good!

A super cap is used to store larger amounts of energy than a normal electrolyte cap. That's it in a nutshell. They take longer to charge but hold the charge for longer periods depending on the drain on them.

Scope is an 15 years old parallel port PCS500 from velleman.
amazon has them listet

can i try more than 1000uF without risk to the voltageregulator a startup?.

I bet that noise is at the hopping frequency rate, try a schottky barrier diode and a LC filter between the TX and the module power lines. It could also be leaking RF from the module causing interference. Where is the CC2500 located inside the TX?

I just put a 1000uF cap at the vcc of cc2500, but can't get rid of the fluctuation issue, although fluctuation is bit less. Also allready use polulo regulator instead of stock ldo. Also have a 3v3 line from polulo straight to the modules not via pcb traces.

Just to understand, what is used as the reference voltage for the ADC pin, maybe there is the cause?
A big capacitor at the ADC input pin, I tried that does not solve either, fluctuation stay, so it is not there.

@djdazzy
the cc2500 is mounted isolated laying on the cyrf module
can you pleas help me to find the right eletronic components for the lc filter

Here are the pictures of how I have my 12s setup. I will be doing a test flight this week when I get some time. Found a few minutes tonight to take the pictures and put them up.

Thanks for the pics
i Would like to stay with the 1S Battery.
Do you think there is no chance to filter the noise from the intern 3,3V VCC?

Please can anyone helb me with calculating a LC Filter to eliminate this noise on the power line.
Current peaks are in about 9mS

i tried to power the modules direkt from the 1S Battery in my Devo 12s ( Diode in series to reduce the voltage)
In this case all is perfekt. so i think its no rf leakage. But this is no Option, because in this case the modul is not powered off with the Transmitter.

i ve tested with my extern linear Power Supply too, and have seen pulsating power consumtion from cc2500. (about 80 mA)

Now the big question for me is, how to get the intern 3.3V stable??


Midelic sent two links about LC Filters in the FrskyX Thread

i tried to build such LC Filter with Elektrolyt (1200uF) + Ceramic Capacitor plus a diode and coil in series with not much success
i have no knowledge about coils so is testet a handful i had laying around.

Any help is welcome.

First calculate the pulse period and frequency of that pulse...and there is LC calculator on the net.
Imo but and exiting LC filter from the net ,buy also coild with inductanve closesest value to the one calculated and change the capacitor and coils
circuitcalculator.com/lcfilter.htm