1/2 CCPM on my Xcell

1/2 CCPM on my Xcell

Post by Ian Atkinso » Thu, 22 Apr 1999 04:00:00



Hi,

Just to be different, I've discarded my "tilting" aileron servo on my
gas Xcell and I am using 2
servos to directly drive the bell-cranks either side of the frames. I am
using the 2 free mixers on my
8UHF to produce the desired pitch/aileron mixing (hence my earlier
post).

However I have noticed undesirable coupling between the aileron and
pitch operation.  Its like this:

1) I set it up so that at 0 pitch, washouts arms are horizontal, bell
cranks are horizontal and servo arms are
90 degrees to push  rods.

2) At 0 pitch aileron causes servo arms to move in opposite direction
tiltiing the swash plate as required

3) As I change pitch, both servo arms move in the same direction (i.e.
opposite rotation) and the
swash plate moves up and down whilst remaining horizontal.

The problem is when I'm at non-zero pitch and I use aileron.  At zero
pitch, each servo arm moves an
equal angle causing the push rods to move an equal and opposite distance
causing the
swash plate to tilt without any net pitch movement.

But at no zero pitch, the servo arms are already off-centre.  Thus any
aileron input causes one arm to
move towards the centre of  rotation and the other to move away from the
centre of rotation.  Now
this no longer corresponds to equal and opposite linear movement of the
push rods  ( the one
moving towards the centre position travels further).  This causes the
aileron stick input to be
coupled slightly (very small) to pitch.  I can reduce this effect by
using longer servo arms and using
lower ATV's, but it doesn't seem possible to completely remove this
effect.

Question?  It would seem to me that this problem is inherent in
electronic CCPM systems, whether
120 or 90 degree.  Is this the case??  Do CCPM equipt radios use a
"arcsine()" function to
translate linear rotaional movement to linear pushrod movement (which
would fix this problem)

Or, is it just nto a big enough problem to worry about?

Ian

 
 
 

1/2 CCPM on my Xcell

Post by Peter Christ » Thu, 22 Apr 1999 04:00:00


<Just to be different, I've discarded my "tilting" aileron servo on my gas
Xcell and I am using 2 servos to directly drive the bell-cranks either side
of the frames. I am using the 2 free mixers on my 8UHF to produce the
desired pitch/aileron mixing (hence my earlier post).>

Neat idea!

<However I have noticed undesirable coupling between the aileron and pitch
operation.>

Er, yes!

<Question?  It would seem to me that this problem is inherent in electronic
CCPM systems, whether 120 or 90 degree.  Is this the case??  Do CCPM equipt
radios use a "arcsine()" function to translate linear rotaional movement to
linear pushrod movement (which would fix this problem) Or, is it just nto a
big enough problem to worry about?>

The problem IS inherent, and CCPM radios do not incorporate any
compensation. You can achieve it by using free mixers, but don't ask me to
do the maths!

The ideal solution is to use linear output servos direct to the swashplate.
This is how the big Yamaha helicopter (built by Hirobo) works.

The effects can be minimized by using the longest arms possible on the
servos/bell-cranks, though obviously this reduces the amount of thrust
available.

In most cases this solution will result in the effects being minimal in
flight.

Pete


 
 
 

1/2 CCPM on my Xcell

Post by Stephen Be » Fri, 23 Apr 1999 04:00:00


Ian,

        I tried the same idea and you need longer arms with reduced ATVs to
rminimize the effect.
        The stock  mechanical setup will have less cross-coupling unless
something electronic could be done to remove the differential you
speak of. At the extremes of the stock X-cell collective pitch one can
see the swashplate ball rise and fall slightly with cyclic due to
differential in the lateral/roll belcranks.
        CCPM guys accept the interaction and adjust to it like we do the
aerodynamic crosscoupling effects every helicopter has.
        If someone build a linear servo things would improve. There is a cable
rap device which planks use to make the servo linear but it would not
be good for helis. Someone clever and with ambition/time could make a
rack/pinion device to adapt for a helicopter.

                                Stephen

Quote:

>Hi,
>Just to be different, I've discarded my "tilting" aileron servo on my
>gas Xcell and I am using 2
>servos to directly drive the bell-cranks either side of the frames. I am
>using the 2 free mixers on my
>8UHF to produce the desired pitch/aileron mixing (hence my earlier
>post).
>However I have noticed undesirable coupling between the aileron and
>pitch operation.  Its like this:
>1) I set it up so that at 0 pitch, washouts arms are horizontal, bell
>cranks are horizontal and servo arms are
>90 degrees to push  rods.
>2) At 0 pitch aileron causes servo arms to move in opposite direction
>tiltiing the swash plate as required
>3) As I change pitch, both servo arms move in the same direction (i.e.
>opposite rotation) and the
>swash plate moves up and down whilst remaining horizontal.
>The problem is when I'm at non-zero pitch and I use aileron.  At zero
>pitch, each servo arm moves an
>equal angle causing the push rods to move an equal and opposite distance
>causing the
>swash plate to tilt without any net pitch movement.
>But at no zero pitch, the servo arms are already off-centre.  Thus any
>aileron input causes one arm to
>move towards the centre of  rotation and the other to move away from the
>centre of rotation.  Now
>this no longer corresponds to equal and opposite linear movement of the
>push rods  ( the one
>moving towards the centre position travels further).  This causes the
>aileron stick input to be
>coupled slightly (very small) to pitch.  I can reduce this effect by
>using longer servo arms and using
>lower ATV's, but it doesn't seem possible to completely remove this
>effect.
>Question?  It would seem to me that this problem is inherent in
>electronic CCPM systems, whether
>120 or 90 degree.  Is this the case??  Do CCPM equipt radios use a
>"arcsine()" function to
>translate linear rotaional movement to linear pushrod movement (which
>would fix this problem)
>Or, is it just nto a big enough problem to worry about?
>Ian

 
 
 

1/2 CCPM on my Xcell

Post by Steve Simpso » Fri, 23 Apr 1999 04:00:00


Ian,

You beat me to the punch! Sort of.

I plan to convert my Gas X-cell to CCPM  down the road! I will use three
servos instead of only two as you have done. This will eliminate the
elevator carriage as well as the rocking servo.

To test the concept and get a little experience with CCPM, I converted a
Century Falcon which took a LOT of modifications. The X-cell will be very
easy to convert.

You can obtain very nearly linear push rod movement via the Exponential
functions. Using longer arms (I used Super Strong Arms from DuBro.)  is not
a problem because even though you are cutting back on power, you have two
servos pushing collective. I ended up using a 75 degree sweep to obtain 20
degrees collective with 6 degrees cyclic.

I set the linkages to be 'square' at 4 degrees positive so that hovering was
precise and most of the cross input came in at extreme negative where I
spent little time lingering anyway.

The real problem with CCPM is that today's crop of servos are really
marginal for the task (my opinion). With a mechanical setup, a deflection in
the collective servo will merely make the heli drop or rise slightly, but in
a CCPM setup, deflection of one of the two collective/cyclic servos can
result in cyclic pitch changes. Pulling a lot of collective to stop a
descent, for example, can cause a servo to deflect resulting in an
unexpected aileron reaction.

There are some other problems too, but the new digital servos should have
the holding power and centering to make CCPM really work well. That's what I
am waiting for.

Check to see if your radio already has provision for CCPM. I am using a JR
8103 which has 2,3,and 4 servo CCPM as a standard (but undocumented)
feature.

 
 
 

1/2 CCPM on my Xcell

Post by Ian Atkinso » Sat, 24 Apr 1999 04:00:00


Hi Steve,

See below:

Quote:

> You can obtain very nearly linear push rod movement via the Exponential
> functions.

But don't you need the exp after the CCPM mixing is applied??

Most radios do:   STICK--->EXP--->MIXING--->SERVOS

for max non-linearity conpensation you would need:
                  STICK--->MIXING--->EXP--->SERVOS

Quote:

> I set the linkages to be 'square' at 4 degrees positive so that hovering was
> precise and most of the cross input came in at extreme negative where I
> spent little time lingering anyway.

> The real problem with CCPM is that today's crop of servos are really
> marginal for the task (my opinion). With a mechanical setup, a deflection in
> the collective servo will merely make the heli drop or rise slightly, but in
> a CCPM setup, deflection of one of the two collective/cyclic servos can
> result in cyclic pitch changes. Pulling a lot of collective to stop a
> descent, for example, can cause a servo to deflect resulting in an
> unexpected aileron reaction.

> There are some other problems too, but the new digital servos should have
> the holding power and centering to make CCPM really work well. That's what I
> am waiting for.

> Check to see if your radio already has provision for CCPM. I am using a JR
> 8103 which has 2,3,and 4 servo CCPM as a standard (but undocumented)
> feature.

My Radio is Futaba 8UHP, non-super version which does not have the specific CCPM
features the
new 8UHPS has ;-(

Anyway, I'm going to test fly it tomorrow, so wish me luck.

Ian