Whizbang

I created an Excell spread sheet to calculate the offset needed in the servo arm tocompensate fora crossed wire setup. It shows you graphicly what is happening to the wire tension (lengths). I just keep poking in values of offset ( the run, servo arm length, and rudder horn length are known ). It assumes the rudder horn pull points are on the hinge line. Here are couple of plots. The first has no offset. At 45 degrees the loose cable is almost 300 thousandths too long. The second shows what happens with proper offset. The loose cable is only 10 thousandths too long at 30 degrees and actually tightens back up again at 45 degrees. Notice that the cables become too short after 45 degrees. If you want more deflection than that you would use a different offset. I hope this helps.

Whizbang

jaka

Hi!
Cannot find it!

Gray Beard

My Feedback: (-1)

A couple years back a club member was showing me his pull/pull set up, I had never seen anyone use anything like this before. On his servo was a brass round servo arm that had a groove channeled around the center. Looked like a servo hub adapter. The wire went around the center groove twice, maybe thrice?? Then went through the fuse as normal
and out to the rudder. Very cool set up and not a lick of slack in the wires during rudder deflection. I had found one of these brass items at my LHS on the wall one day but for the 12 of 15 bucks I didn't buy it. I haven't seen the item again but it was slick. Maybe someone here knows something more about it. There was also a brace that went around the servo output shaft {I think that's how it went?} and connected to the servo case to keep the tension off the bearings. I was told that was a Heli item? Never have seen that item again either. Anyone know anything about this system??

scale only 4 me

My Feedback: (158)

Hanger9 makes them, or used too, I have only in my Kingfisher,, they are some what small,, good for 60 size planes


Rudder horns are not accessible in this plane so I used nyron and the adaptors to adjust tension,, I haven't had to touch it in 15 years

kenh3497

That looks like just the ticket for a pull/pull system. I've seen similar 'wheels' used on large stuff like farm equipment and such. Easy on the cable and less connections to make as in cable crimps.

Found them.....[link]http://www.horizonhobby.com/products/pull-pull-wheel-jr-air-large-HAN3512[/link]

scale only 4 me

My Feedback: (158)

Servo City sells a plastic Version Pulley that mounts to a servo,, with a little engineering it could serve the same purpose

http://servocity.com/html/hitec_pulley_56314.html

They have a futaba version too,, no luv for you JR guys

Gray Beard

My Feedback: (-1)

That be the critter. They work slick. I'm kicking myself for not buying the one I spotted.

Bozarth

My Feedback: (15)

Are you guys implying that it will eliminate the geometry issues previously discussed?

Kurt

kenh3497

Wasyou talkin' about.........

Lnewqban

It will reduce a 50% of the trigonometric problems,............or 100% if applied to each end.

Any pulley is infinite servo arms in one.

Bozarth

My Feedback: (15)

I don't believe the trig introduced with a pulley corrects the errors of a pull-pull setup with incorrect geometry. Since the pull-pull cable is fixed to the wheel, you are technically not using it as a pulley. It has no benefits, especially if you don't match it on the control horn end. In fact, you might make the problem even worse. (i.e. even tighter or looser tension on the non-pulling side)

Using a wheel with the cable wrapped around it at the servo converts the output from exponential to linear. This linear movement of the cable results in an exponential movement of the control surface since it is moving around a hinge. If the control horns are not aligned with the hinge line, we will still have cable slack/tension on the non-pulling side, just as we would without the wheel on the servo. Our normal systems produce a linear movement of the control surfaces.

Kurt

Lnewqban

Kurt,

As I understand it, is just like the difference between walking (servo horns) and rolling (servo pulley).

You are correct about the different relation of movement between a lever and a wheel.

The first one is sinusoidal (y=sin x), being the displacement max around 90 degrees; while the second one is continuous or directly proportional to the angle of deflection.

Pulley to pulley completely eliminate any slack of the line.

Check these links:

http://www.mindspring.com/~rellis2/rcpattrn/ppull.htm

http://www.espritmodel.com/pull-pull...planation.aspx

http://www.rcuniverse.com/forum/fb.asp?m=9227808

(Edited to correct formula y=sin x and to insert diagram)

Bozarth

My Feedback: (15)

Good links - all are good illustrations of the need for symmetry. But I didn't see any mention of your idea of a "pulley" or the benefit of using the wheel shown earlier in the thread. IMO it's because geometrically there is none.

Kurt

scale only 4 me

My Feedback: (158)

It's not limited by it's radius and the servo travel??

kenh3497

By using the pulley, the radius from the center of the servo output to the point where the cable "leaves" the pulley never changes. Assuming you were using a "control horn"..... If one were to draw a line from the servo output to the hinge line at the control surface, as the servo rotates the distance between your control cable and the imagery line would decrease. Using a pulley at each end the distance would stay constant as the effective radius never changes. Using a pulley on one end would decrease this error by 1/2 as Lnewqban stated.

If you want to experiment, this would be easy to set up using two of anything round, They don't even have to be the same diameter as different diameters would just change the ratio. Then do the same thing with a couple of balsa sticks to see the difference.

For those of you that ride motorcycles. Pretend the swing arm doesn't move for this exercise With the sprockets the chain moves smoothly around . Now imagine the sprockets had two or even four points. The chain would flop like crazy and the tension would change continuously. We have the same basic set up with a pull/pull system as a chain and sprocket setup.

This all makes perfect sense to me. Hope it works for you also. Tell me if I'm wrong please!!

Ken

Lnewqban

There is at least one, Kurt: freedom of movement without the restriction of the lever.

http://www.slopeaerobatics.com/tag/madflight/

Bozarth

My Feedback: (15)

Not true. Converting the servo output from sinusoidal to linear, linear to sinusoidal, linear to exp, exp1, decreasing radius or not, etc. (even through tx programming) does not remove or induce error. The error (slack/tension differnce) is a result of how well the control horn geometry mirrors the servo geometry. "Fixing" the servo side by installing a "pulley" isn't fixing anything unless the control horn geometry matches it.

I don't think describing your concept as a "pulley" is correct. Anyone can google pulley systems and see the difference.

Kurt

Bozarth

My Feedback: (15)

The system on the right is not a "pulley system."

Rodney

Here are some pictures of set ups I have used successfully.

D.J. Erlenbach

Most of you are all full of crap !! If the distance from center to end where the cables are hooked are the same on the servo arm and at the rudder or elevator horns, whether they are straight or crossed, the tension WILL NOT CHANGE !!! GEEEEEZ, forget your computer generated junk and just look at plain mechanical physics!!! Cross them, if you have to, it won't make a bit of difference, IF the center to end is the same at both ends. If they are not the same length, YUP the tension will change !!!!! I'm just a pliers & baling wire mechanic but I can sure figure that out quick!!

Bozarth

My Feedback: (15)

GEEZZ - you are wrong. What you are describing will get you a one-to-one ratio of servo movement to control surface movement - it doesn't address the slack/tension issue we are talking about. Here is the issue you and most are missing. See the pictures.

The trig supports it. If you do the math with the following set up: 1" control horns on the control surface, control horns are 3/4" behind the hinge line, plastic clevises extend 1/4" forward. This puts the holes 1/2" BEHIND the hinge line (see the slack photo). A 45 degrees deflection of the surface requires the pull-cable to move 0.8535 inches. However, the non-pull side only moves 0.78 inches, so you have SLACK - 0.0735" to be exact.

Kurt

kenh3497

It looks to me as if this discussion is getting way to technical. We are after all, just playing with toys..... very expensive toys

Bozarth

My Feedback: (15)

kenh3497,

Help me understand what a pullley, or sheave, at each end looks like. I am obviously not following you. Thanks.

Kurt

Bozarth

My Feedback: (15)

[/quote]


Not too technical for some of us who enjoy getting to the nitty-gritty technical details that accurately explain the issues - toys or not. Call it a sickness[8D]

Rodney

D.J.E., I'm sorry to say you are wrong in your comments in your post above. If you refer to my post above (post #43) you will see pictures of the pulleys I use at the servo end and also the way the cables are attached at the surfaces. When you use a pulley at the servo end, you will always pay out as much line at you retract. What determines whether or not you get slack is the way the cables are attached at the surfaces. If the attachment is aft of the hinge point on the moveable surface you will get a slackening on the line not being pulled as you move away from neutral; if the attachment point is forward of the hinge point, the lines will tighten (a very bad thing) as the surface is displaced from neutral. If the attachment is directly over the hinge point, no change in tension will occur. You can change the amount of deflection by moving the attach point of the horns at the surface being deflected in and out within reason.