Common electronic CCPM swashplate layouts include:
- 120 degree (forward or backwards pointing)
- 120 degree (left or right pointing)
- Uncommon. Futaba calls this HN3.
- 140 degree
- A variation on 120 degree (forward pointing) championed by JR, designed to eliminate CCPM elevator interaction. Futaba calls this H-3.
- 90 degree 3 servo
- Servos attach to the sides and either the front or back of the swashplate. Futaba calls this HE3
- 90 degree 4 servo
- Servos attached at 90 degree intervals all around the swashplate. Futaba calls this H-4.
"Mechanical CCPM" will always use a 90 degree layout (which Futaba calls H-1).
In the following discussion, A, E, and C are the commanded aileron, elevator and collective inputs, normalized from -1 to 1 inclusive. a, e, and c are the aileron, elevator and cyclic CCPM parameters. S1---S4 are the commanded servo positions, normalized from -1 to 1. Any servo reversing would take place at a later stage.
In the 120 degree layout, the swashplate control points lie on a circle centered on the swashplate. For all transmitters tested, 120 degree CCPM is implemented by:
S1 = cC + aA + 0.5eE S2 = cC - eE S3 = cC - aA + 0.5eE
The factor of 0.5 (=sin(30 degrees)) comes from resolving the position of the two diagonal controls (S1, S3) towards the fore/aft centerline. Logically, the aileron values should both be scaled by cos(30 degrees)=0.87... to give equal control deflections; however I have never observed this, probably for ease of implementation as it can easily be specified by the user.
The "sideways pointing" version simply swaps all elevator terms for aileron terms, and vice versa.
In the 140 degree layout, the control balls lie on two adjacent corners and the opposite edge of a square (or rectangle) centered on the swashplate.
S1 = cC + aA + eE S2 = cC - eE S3 = cC - aA + eE
As the diagonal controls (S1, S3) are now the same distance from the fore/ aft centreline as S2, no factor is needed in the mapping, and the swashplate should tilt more linearly fore and aft. Depending on the distance of S1 & S3 from the left/ right centerline, the a & e coefficients may need to be the same or different for equal control surface deflection. For a strict 140 degree layout, a = 0.84e; for a strict 135 degree (square) layout, a=e.
Despite the 140 degree design objective, some have argued that the 140 degree system may introduce CCPM interaction on elevator input due to differences in the servo loadings. That is, compared to the 120 degree layout, where each lateral servo will see approximately the same load as the longitudinal servo, but have to move half as far; in the 135 degree scheme each lateral servo will see half the loading of the longitudinal servo, but have to move the same distance.
As 140 degree mixing does not include the scaling factors on the diagonal control links, it runs a greater risk of running out of commandable channel travel than 120 degree on extreme combinations of inputs.
If necessary, a 140 degree mix can be synthesized by a 120 degree mix and a programmable mix, with a scaling to the elevator coefficient:
e' = e/0.75 C' = C + .25e'E/c
That is, the elevator coefficient is increased by a third, and a programmable mix adds to the collective input a proportion of the elevator input equal to a quarter of the new elevator coefficient, divided by the collective coefficient.
S1 = cC + aA S2 = cC + eE S3 = cC - aA S4 = cc - eE
In a 3-servo layout, one of these (usually either S2 or S4) is dropped.
Share your opinion