Frsky problem.

Problems, experiences or just something to say about RC gear? Say it here.

Re: Frsky problem.

Postby RCModelReviews » Mon Jul 19, 2010 10:18 pm

I've been talking with FrSky about this and the problem (as I suspected) is not to do with resolution but to jitter caused by the need to run the RF frame rate asynchronous to the PPM data stream.

I'll do a full article to explain this but suffice to say it doesn't affect resolution. What's happening is that, at some rates of stick movement, the servo will move in slightly irregular steps (caused by occasional oversampling).

With a standard servo this is not really discernible because a standard servo stops driving the motor shortly after each pulse arrives however, digital servos continue to move to the commanded position because they effectively memorize the pulse.

This means a standard servo effectively integrates (smooths) the pulses but a digital one responds to each individual pulse and, because there's an occasional oversample, the movement has some jitter associated with it.

Does it matter in the air?

I would wager that it wouldn't but I'd be keen to hear from anyone experiencing this jitter as to whether they can discern the effect in the air.

Can FrSky fix it?

I'm talking to them at the moment and will be making some suggestions. One is to increase the level of oversampling so that the amount of jitter is markedly decreased.

I've been flying hardcore 3D with big surfaces, big throws and digital servos for several months now using the FrSky system and haven't noticed any problems. I'll be reviewing the telemetry system shortly and will look to make sure the jitter isn't greater in this product.
RCModelReviews.com, just the facts.
User avatar
RCModelReviews
 
Posts: 2120
Joined: Tue May 04, 2010 3:40 am

Re: Frsky problem.

Postby mvvs35 » Mon Jul 19, 2010 11:24 pm

Hi Bruce thanks for finding out what the problem. effectively normal servos not noticed the problem. Another problem I found is in my Trex450 with
GP750 Giro sometimes do not take it. and I have to plug and unplug the Lipo. Of Anyway I made several flights and the problem is not noticed in the air. Talk
through MSN Jonathan Zhou, FrSky seller and told me not to notice any problems. It that makes me mad enough.
I wonder if they will correct this problem. otherwise I will buy one Spektrum DSM2 module. although it is vulnerable to interference. But the movement
servo is clean.

Thanks!

Greetings!
mvvs35
 
Posts: 9
Joined: Thu Jul 15, 2010 3:09 am

Re: Frsky problem.

Postby mr.sneezy » Tue Jul 20, 2010 2:13 am

I did some checking today with a high speed digital servo (500 heli tail servo) on the FrSky V8 and the FrSky Two-Way verses an old school FM receiver. Yes, the FM is smoothest in movement, there is no doubt about that. There was a roughness to the movement of both of the FrSky 2.4Ghz systems when compared to FM, but neither one stood out as being more rough or steppy than the other. The roughness is very speed dependent, but can be seen at most stick speeds to some degree on my gear too.

I noticed though, if I loaded down the digital servo as it wold be in flight that the effect was damped. I have all digital servos in my 500 size helicopter, and until this thread started I'd never noticed any issues in flight and I still don't. I suspect there are none for most pilots.

The effect of the problem is a bit like having a dead pixel on a LCD TV, to anyone not familiar with the particular TV they would not even notice it, but the owner of the TV will always be looking for it, as it gets 'into his head' so to speak.

I think the issue will be fixable in firmware, but the cost may well be some latency. I gather that some other 2.4Ghz brands have high latency already, and that may well be caused by some of the extra time needed to smooth out the sync issues between the PPM frame and radio frame.

In the meantime I WONT be selling my FrSKy and buying Spektrum.
A. I can't afford Spektrum.
B. I've seen too many crashes on Spektrum DSM2, and repaired poorly designed circuit board soldering issues.
C. I fly 90% analogue servos in park sized models and there is no issue with them at all.

Martin
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Fly's airplanes & helis, glow & electric.
mr.sneezy
 
Posts: 52
Joined: Fri May 07, 2010 3:38 am
Location: Adelaide Australia

Re: Frsky problem.

Postby takilara » Tue Jul 20, 2010 7:14 am

Another Confirmation
Two way Rx (D8R) with the Futaba Module (DFT) in a Hitec Optic 6 using Align DS610 servos has very choppy movement (seems like low resolution, but I will trust your statement that it is not). There is also quite abit of jitter aswell.

I have not tested it in air yet as build is still not complete.
I also have a V8 Rx that I will test for similar behavior.
I am running 6V to the Rx and servos
The price of getting what you want;
is getting what you once wanted...
_____
T-REX 600 ESP
CopterX 450AE V2
Multiplex FunCub
Belt-CP
User avatar
takilara
 
Posts: 144
Joined: Mon May 17, 2010 7:53 am
Location: Stavanger - Norway

Re: Frsky problem.

Postby takilara » Tue Jul 20, 2010 8:51 pm

Tested the V8 (V8R7) with the exact same setup, and experienced the same behaviour, choppy and jittering movement.
I tried to move the cyclic stick full range left->right, and i counted 92 distinct movements, and this does not seem like jitter. What happens is that i move the stick a fraction of a millimetre, and nothing happens, another fraction and the servos move a little bit, move the stick another fraction and nothing happens, another fraction and the servos move, this feels ALOT like poor resolution, and not oversamples.

If i use my hand to "hold the head of the heli", the jumps seem "smaller", but they are still quite distinct, i suspect i need to get out my oscilloscope to take a look at the pulsetrain
The price of getting what you want;
is getting what you once wanted...
_____
T-REX 600 ESP
CopterX 450AE V2
Multiplex FunCub
Belt-CP
User avatar
takilara
 
Posts: 144
Joined: Mon May 17, 2010 7:53 am
Location: Stavanger - Norway

Re: Frsky problem.

Postby RCModelReviews » Tue Jul 20, 2010 8:58 pm

I still haven't had a chance to examine the production telemetry version for resolution -- hopefully in the next day or so.

I'll keep everyone informed once I've tested it.
RCModelReviews.com, just the facts.
User avatar
RCModelReviews
 
Posts: 2120
Joined: Tue May 04, 2010 3:40 am

Re: Frsky problem.

Postby tommy » Fri Jul 23, 2010 1:47 pm

I have the Frysky DIY module and have also noticed the choppy and jittering servo movement. I am using HS65 servos (analogy)
and can still tell the difference between the FM signal and the Frsky. It happens if I am doing slow or fast servo movements.
Hence it does seem like poor resolution to me as well but it’s hard for me to confirm this as I do not have the testing gear that some of you guys have.
I will be very interested to hear what your conclusions are. I am concerned that this issue will make the frsky a very poor system for when I upgrade to
all digital servos. I hope I haven't chosen the wrong system
tommy
 
Posts: 39
Joined: Sat Jun 19, 2010 9:46 am

Re: Frsky problem.

Postby Whiteprop » Fri Jul 23, 2010 5:16 pm

tommy wrote:I have the Frysky DIY module and have also noticed the choppy and jittering servo movement. I am using HS65 servos (analogy)
and can still tell the difference between the FM signal and the Frsky. It happens if I am doing slow or fast servo movements.
Hence it does seem like poor resolution to me as well but it’s hard for me to confirm this as I do not have the testing gear that some of you guys have.
I will be very interested to hear what your conclusions are. I am concerned that this issue will make the frsky a very poor system for when I upgrade to
all digital servos. I hope I haven't chosen the wrong system


Do the servos jitter in the power reduction mode..(Range test mode)?
I had a problem similar to this.
Whiteprop
 
Posts: 23
Joined: Thu Jul 15, 2010 12:59 pm

Re: Frsky problem.

Postby takilara » Fri Jul 23, 2010 5:56 pm

Whiteprop wrote:Do the servos jitter in the power reduction mode..(Range test mode)?
I had a problem similar to this.


I'll check that this evening.

I'll do a few other test this evening as well.
(test analog servos, test the DS610's on FM, connect the servo leads to a oscilloscope and see if i can see anything..) I guess the best way to determine a resolution problem, is to measure the pulselengths, and see how many "distinct" lengths one get, or how much they increase/decrease. I doubt i will have time for that this evening though.
The price of getting what you want;
is getting what you once wanted...
_____
T-REX 600 ESP
CopterX 450AE V2
Multiplex FunCub
Belt-CP
User avatar
takilara
 
Posts: 144
Joined: Mon May 17, 2010 7:53 am
Location: Stavanger - Norway

Re: Frsky problem.

Postby takilara » Fri Jul 23, 2010 6:11 pm

I got so curious that i had to test in power reduction mode right away.
No real difference that i could spot.

As others have said though, if i add weight to the servo, it moves a little smoother, but i can still feel the individual steps though.
Also the servo's are incredibly noisy in "neutral". I guess the Rx is switching back and forth between two "steps". Very seldom do i manage to put any of the sticks so that the servo does not make any noise.
The price of getting what you want;
is getting what you once wanted...
_____
T-REX 600 ESP
CopterX 450AE V2
Multiplex FunCub
Belt-CP
User avatar
takilara
 
Posts: 144
Joined: Mon May 17, 2010 7:53 am
Location: Stavanger - Norway

PreviousNext

Return to Radios and Servos

Who is online

Users browsing this forum: No registered users and 2 guests