Hi Guys

I am using a servo for each elevator half setup with a JR PCM10sxII, however I have noted that the travel on each half is different - i.e. one elevator half deflects more than the other during transition even though the linkage setup is identical. Full deflection is however identica.

your comments are appreciated- thanks

you are seeing a problem which is a result of the tx design -- you can not eliminate it .
I did "fix" it on mine by assigning /mixing both elevator halves as AUX (2&3) and nothing plugged into elevator.
The new DX7 does not do this .
the last version of the 10x still did it if you mixed EL with aux 4
most if not all PCM radios do it to some degree.
the least is the new Fytaba 14.
The technical reason is that the info for EL is fed at a different time than the later AUX channels and at a different rate in some cases.

By tweaking the mix rates and the ATVs you will be able to get it very near perfect. Worst case senario you can do an elev-elev multi point mix to correct the difference mid travel.

Shawn

point to point will get close but relative speed is the problem - early on 10 series radio was a 512 setup and then the 1024 was better Fubas 2048 pretty good !
I have had all of the series - the problem is built in -just like on many Futaba also -- and other radios - the time lag is part of the basic system.
The DX7 is done differently the signals run as a pair.

On my 6102 I can dial in the point to point on high rates and then when I switch to low they differ. If I dial in the low rates then the high rates differ. Pain in the ***.

or you can put the elevators on a matchbox also

for those who REALLY hate this problem-- the DX7 is spot on - with no adjustments high /low expo makes no difference
until someone starts jiggering around with sub trims to correct lousy linkage and goofy servos .

I will try to eliminate the ele channel as suggested by dick and do a seperate mix for the elevator.

DX7 seems to be promising in most areas - I wonder if the spectrum module will have the same effect or will have to wait until a more versatile 10 channel sys becomes available.

What? Do you work for Spectrum or something? I have the 9C and have not noticed this. perhaps it's a matter of matching up servo speed. I know they may be rated the same, but some are slightly faster or slower than the rating. It's really only a problem if you fly precision.

Bob

the latency thing will not be cured thru adding the module to the 10--it is in the basic guts of how the TX assembles info -lines it up and feeds it thu .
ditto for the 9C
maybe some never see this "problem" but if you are doing stuff that calls for fast accurate matching - it's there - How much it will affect your flying -- that's an unknown.

I agree the matcing will always be a tad off, how much this affects the way the airplane flys is anyones guess. I remember 1/2A pylon racers with a single aileron and half an elevator flying pretty well. Not exactly apples to apples. On my current setup I put 18" sticks on the elevator halves that come together at the end. Through the entire travel I get no more than 3/32" differential 18" out. I agree that we are on the verge of some serious improvements. I just would never be able to give up my 10X for a hunk of plastic DX7.


Shawn

I'd look real hard at the mechanical set-up, even though you thinks it spot on several varaibles affect the final outcome. Did you utilize a throw meter or similar to set up the model?

Servo speed or "latency" is not a contributing factor to this type of problem, IMO. If you stop the elevator halfway through the travel and they are off, it is not because one servo is slower than the other. I'm with Mglavin on this - it is in your geometry or just that the two servos are not exactly the same, internally.
It only takes being off a fraction of a millimeter anywhere in your geometry to have assymetry. Even if your servo arms are both exactly at 90 degrees relative to the servo case, your pushrods are exactly the same length, and the lenghts of your control horns are exactly the same, doesn't mean your geometry is perfect. The variances in the thickness of each elevator half, the location of the hinges on the hingeline, the location/position of your servo in the servo tray, the location of your pivot points in relation to the hingeline, or amount of friction in each elevator halve's hinges/ball links, etc. will affect their travel...

Latency is not -as you know -servo speed -it is signal mismatch.
My own elev setups are done off the model -on a bench with the two servos lying flat, facing each other and the arms first positioned on the splines to make for identical "0" centers .
With my DX7 this is so far -spot on every time with any of my servos - sometimes I have to select arms to use but I always start at 0 both servos
then I watch track thru full throws - - - I don't use any servos except JR and JRsport so obviously -I can't speak to any other brands .
This "pair" is put on the model ,then all other adjustments are done with linkage or throw settings - no re-centers - no mismatching of relative end points -no sub trims
This takes more time than fiddling with electronic correction devices but so far -it is still the most accurate method I have found.
the electronic sub trimming methods never worked well for me -
how you can get matched travel and throw speed with mismatched electronic setups is beyond my puny skills.