Swash plate motion problem

[ibutcherjp]

Hi,

I recently bought a Belt from a friend. After solving a cheap blade problem, I face now another one with the swashplate motion. With the heli freshly set up (servos and control rods at 90deg) and the Tx reset (no subtrim, 100% on all channels, 50% on the 3 swash mix channels), the swashplate moves perfectly horizontally with the pitch command, on the whole range. Motion is smooth, the axle is lubricated. But when I move the front/back cyclic, the plate center moves up and down. No problem with the lateral cyclic. I just connected the swashplate, the rest of the head is not connected yet. Here is the illustration of the problem :

http://img166.imageshack.us/img166/2345/beltqt2.jpg (http://imageshack.us)
http://img166.imageshack.us/img166/2345/beltqt2.c50315b9b1.jpg (http://g.imageshack.us/g.php?h=166&i=beltqt2.jpg)

I have been spending few days trying to solve this, reading different topics, but I cannot figure out how to set up the plate system correctly .... :(

Do you have any idea ?

[Jonnyheli]

Hi, I don't think I know how to solve your problem but I know that others here will be more than likely to help out.

Just letting you know that I cannot see the pictures. Maybe it's my computer but normally it's not.

Good luck with the swash plate. Im sure a few people here will be more than happy to help.

-Jonny

[ibutcherjp]

Thx for the answer ! ;)

ImageShack seems to have some problems tonight ....

Here is the direct link :

http://img399.imageshack.us/img399/4136/beltnl5.jpg

[another_finn]

I'm not sure that is a problem to begin with. Your pictures don't just illustrate the effect, they also do a pretty good job of showing what's causing it. When you input cyclic, the servos all move about the same distance. With cyclic roll, your aileron and pitch servos are spaced equally in relation to the main shaft and the motions cancel each other out at the center of the swashplate. With cyclic pitch, the axis of rotation falls naturally aft of the main shaft (look at the apparent distance of the ball links as viewed from the side). This causes the swashplate to rise and fall.

You could probably tune it out with very advanced mixing - my DX7 can't do it, but I suppose some of the $2000+ radios could. However, I don't think you really need to. The change is very small, and you will be working the collective whenever you change cyclic pitch anyway. I don't think you would ever notice the effect of that slight imperfection in flight.

[ibutcherjp]

Ok ! I came to the same conclusion but thought that they would be a way to correct this. I am not yet really familiar with CP machines :)

Indeed, I noticed the problem when flying : when the Belt flies in stationary position, if I move forward, it raises and backawards it looses altitude pretty hardly. Maybe there is another problem ....

[Real_Sparky]

Maybe i'm not understanding the problem bit it looks like your swash mix is the wrong way round.
Your evelvator when pushed forward should dip the front of the swash, in your pictures if I understand right, it raises. If this is the case reverse the swash mix. If it is + make it - and vice versa.
If not I dont know.

Darin

[Ian 777]

Hi
It sounds like that is your problem then. If you are using the stock TX the ball links may be in the wrong holes on the servo arms, If you have a computer TX then you can adjust it using the servo travel adjust function.

I don't understand why you said your DX7 couldnt fix it Mika???

[ibutcherjp]

The front of the heli in on the right of the photos.
Left : cyclic command back, plate back, pitch slightly down, heli goes to the ground.
Center : cyclic command centre, plate correct, pitch neutral, heli gently hovers.
Right : cyclic command front, plate front, pitch slightly up, heli goes up.

This is what I noticed during my last flights. ;)

Maybe my (poor) settings of the heli amplified this natural phenomenon. I will put all back together following the videos I found here, this time, and then try again.

[ibutcherjp]

Hi
It sounds like that is your problem then. If you are using the stock TX the ball links may be in the wrong holes on the servo arms, If you have a computer TX then you can adjust it using the servo travel adjust function.

I don't understand why you said your DX7 couldnt fix it Mika???

I have a cheap computer Tx. Since the pitch command works correctly through the whole range with standard settings, I am sure that the servo arms are set correctly (i use the default arms with no settings possible anyway).

I tried to play with the travel adjust : I managed to correct the position of the rotation point of the plate. But then, the pitch command was altered : through the pitch range, the plate was changing angle, since the travel distances of the 3 servos were not the same anymore ....

It is slowly driving me crazy ! :lol:

[Ian 777]

Oh I see.... The tx is set to120 degree ccpm I take it?

[sutty]

Interesting Mika, I had never noticed or even considered this. I just looked at mine to try to educated myself somewhat and was amazed to find that the front servo arm moves much further than the rear two. Now I have thought about it and observed it in action I can see why it would have to. The effective length of the lever on which it is operating is almost twice as long as the virtual level on which the rear two servos operate. Consequently the front servo appears to move about twice as far, which would seem to make some sense. If it didn't the rise and fall of the swash would be much more apparent. This must be automatically built in to 120 degree swash mixing. However since they don't know how big my swash is, how could they have known how much to build in? It doesn't have a setting for 450 size helicopters, etc, or are all swashes so similar in size, "one size fits all", so to speak?

Incidentally my swash does not go up and down at all, when I use forward or reverse cyclic, but even if it did I think you are completely right, you are managing collective in flight all the time anyway, so I guess a small amount wouldn't be a problem.

It still surprised me though, as I had never even thought about it. I think I'll go and have a troll around the internet to see what I can find on the subject.

EDIT: Mmhh, it probably doesn't matter how big your swash is, the ratio of those distances is probably always the same.

[another_finn]

It's getting kinda late, so I might be missing something - but I also think even if the swashplate size changes, the geometry and proportion of required movement remains the same. I suppose there is some universal built-in ratio for cyclic pitch mixing between the front and back servos that gets it close enough regardless of the size of your mechanics.

I don't remember really noticing much of an error on my swashplates when putting in cyclic pitch, but I'm pretty sure most of them have some unwanted movement. Nothing I would notice in flight, since there's no way to tell it apart from the usual effects of changing the lift vector, gaining translational lift and so on. Just too much going on with the collective at all times for it to make any practical difference.

[sutty]

I had just edited my post saying that about the ratio remaining the same, so I guess size doesn't matter.:lol:

[sutty]

You might have known this but this is what I just read, paraphrased a little. Because the front servo has to travel further in 120 degree ccpm mixing, extreme 3D pilots can notice the lag and the effect that it has on their helicopter. For slight or gradual stick input the effect will not be noticed as the servos can keep up with each other but for example if you suddenly apply full forward cyclic there will be a significant lag whilst your front servo catches up, causing disproportionate movement. There were several suggestions to improve this situation. One is much faster digital servos, so the lag will be less apparent. 140 degree ccpm mixing, which I hadn't seen before, which has all the leg lengths equal, thereby removing the issue, or modifying your elevator to pitch mix by roughly 35% Not sure about how you do this last one but it was one of the suggestions.

The 140 degree ccpm solution is the perfect one but I think a lot of radios may not support it. Here is a picture of a 140 degree swash.

54831

So it would seem that it is an issue for top pilots, but that most certainly doesn't include me. I would just think it was a little windier than usual.

[ibutcherjp]

The tx is set to120 degree ccpm
Yes, it is.
DIT: Mmhh, it probably doesn't matter how big your swash is, the ratio of those distances is probably always the same.

Yes. If you have the angle (120deg) and if you know that the 3 plate legs are on the same radius, the motion correction to apply to have a constant height swash plate is always the same, whatever its radius.

So, from what I read from you guys, I think I have a problem with my Tx .... :(
Should probably look for a better one and check the resulting motion. The plate should be steady during cyclic motions, from my point of view too.

EDIT : found this

http://img244.imageshack.us/img244/5470/f24ao2.gif (http://imageshack.us)
http://img244.imageshack.us/img244/5470/f24ao2.69a4f6b722.jpg (http://g.imageshack.us/g.php?h=244&i=f24ao2.gif)

http://img145.imageshack.us/img145/7673/f25fp2.gif (http://imageshack.us)
http://img145.imageshack.us/img145/7673/f25fp2.a4b3af6319.jpg (http://g.imageshack.us/g.php?h=145&i=f25fp2.gif)

http://www.iroquois.free-online.co.uk/ccpm.htm

I definitevely have a Tx mixing problem ....

[Ian 777]

Sounds like it because the compensation required by the Tx should always be the same. It seems strange for the TX to mix it wrongly rather than fail altogether,
It couldn't be receiver incompatibility could it?

[DierWolf]

sounds to me like your front servo isnt traveling the same distance as the others.

Cant see how this is possible if you say the pitch range is good and flat at all points.

[ibutcherjp]

I dont really understand neither .... I checked all the possible paramters, impossible to have both the collective and cyclic correct motion. It seems the CCPM mixing is defective on this Tx.

Anyway, cheap Tx will soon be replaced. I am checking now for its replacement.

[ibutcherjp]

I solved my problem : a brand new DX6i to replace my Tx. Swashplate moves now perfectly well :thumbup:


The mixing was bugged on the other Tx, nothing to do ...