SeamusG

I am trying to improve the resolution of the aileron servos following guidelines suggested in the Pattern forum. My goal is to locate the servo arm clevis in a hole closest to the servo arm hub and locate the control horn clevis in the hole farthest from the control surface. In addition to this basic control rod location method I'm using the DX7 capability to increase the servo travel from 100% to 140%. With both the control rod and radio changes Iwant to keep the total control surface at the recommended 5/8".

The SIGapproach of mounting the aileron servos is with the servo mounted to the backside of the servo hatch with the gear axis parallel to the spars. In this configuration as the servo arm swings thru its arc the hatch interfers with the clevis as the arc approaches about 30 degrees from neutral position. To achieve a clean sweep of the arm (forward and backward) the clevis needs to be installed in the arm hole farthest from the servo arm hub and servo travel reduced to 120%. With this compromise the aileron travel is a whopping 1 1/8" (compared to the recommended 5/8").

Have any of you mounted your aileron servos so that the servo gear axis exits the hatch vertically? What kind of structural changes did you make to the servo "bay" to support the servo?

TIA,

Campgems

I've built and flown two of the 4*60 and I liked the aileron setup so much that I've converted other planes to that method. I use Futaba S3151 digitals and it is all you need. The only inprovement is to make sure the clevis is tight to the servo arrm, and the clevis is tight in the aileron horn, so theonly movement is gear lash.

The4*60 isn't a pattern plane and going nuts trying tomake it oneisn't productive. It is a great sundy flyer and will do about all you ask of it shy of3D stuff, but in the end, it still isn't going to compete with a well designed pattern plane.

Don

FlyingPilgrim

My Feedback: (26)

I second the opinions stated by Campgems, well stated, Don. Don't stress over the suggestions you read on the pattern set-up. Yes, to clear the hatch cover, you can move the linkage to the outer most hole, but you will lose torque by doing so. While it is best to have straight control linkages, you could also bend the linkage in order to allow the use of the inner control horn holes and make use of as much torque as possible. With a sport plane of this size, the short linkage, and use of standard servos, there should be no problem with a slight bend in the rod. I would also consider dropping the servo travel to 100%. Yes, you can get more travel by increasing it past 100%, but you also run into geometry problems, where the horn goes too far around the axis of the servo, causing interference (as you discovered already).

Always try to have the horn on the control surface (your ailerons here) as far out as possible, then move the servo horn linkage in and out until you achieve the desired throw. Fine tune it with the radio servo travel. What you DON'T want to do is to reduce servo travel too low so the servo hardly moves, as you are not making good use of your servo power.

SeamusG

I'm not stressing over anything.

I have a character flaw - I try to improve things when I can. I upgrade shocks, springs, tires & brakes on cars and motorcycles allowing me to stop and turn better than the manufacturer's specification. Hey, I even upgrade the hardware provided by kit suppliers like SIG .

If I can upgrade something as simple as the installation approach of the aileron servos to improve the control of a 4Star60 (my 2nd btw) - I'm gonna do it. I can't tell you how well these same types of changes improved the performance of my Somethin' Extra (2nd) and Big Stik. I plan on using the same approach on my Hog Bipe (needs to be covered).

I installed a new set of hatch covers that meet my goals. Scrap plywood, scrap hardwood blocks and some scrap covering. A little time & effort and the change has been completed.

planebuilder66

My Feedback: (8)

Man, your the first person I've seen to use those hitec adjustable horns, I personally don't trust them, they don't seem like they will be strong enough for much more than a 20 sized plane. But good job on the modification, I run into the same issue, sometimes I just open the slot up and put a 20-30 degree bend in the control rod so the clevis can go into the slot hole on the forward pull one the travel. But now that I've been thinking of new ways to set the alerions up, I think the servo in the wing and the linkage inside with a small slit for the rod to protrude to the horn is best. This way you can hide the servo and the servo linkage in one. Of course, you'll need a thick wing to do the internal linkage.

SeamusG

Good point PB66 - I have used the adjustable horn to prove the "geometry" on the control rod and surface movement. A lot easier to move the flexible arm back-n-forth on the servo than opening / closing the clevis. I'm visualizing how these things could flex resulting in them opening up the center of the arm and watching it separate itself from the servo[:@] at the MOST inappropriate time - I don't want to imagine the possible results.

I will replace these flexi-fliers with circles or arms shortly!

Cheers,

Follow on comment ...

I believe that form follows function - not the reverse. IF the cosmetic beauty of the original servo hatch approach allows for the most effective functioning of the servo and control surface - GREAT. However, this is not the case. Functional compromise was necessary with the hatch approach. If one finds the functional compromise within reason - fine. If not (like me)change the form to better allow for function.

Pic is the original servo hatches ...

Updated ...

Pic of "improved" servo arm installation ...

Campgems

One last idea for you for an improvement. This is a photo of my Ruperts Dad wing. I used a mount similiar to the stock 4*60, S3151 Futaba Digital servo witha a long heavy duty Dubro servo arm with one side cut off. Iused a Dubro ball link with a short piece of #2 push rod epoxied into a carbon fiber rod, On the other end another short piece of #2 push rod and here I used the Robart ball link horn with the steel clevis. The hinge line to the ball length is aproximatly the same length as the Servo horn. I get a full 30 degrees of deflection on the servo arm without binding. With the equal length, that equates to 30 degrees of aileron if I want. I never fly with that much travel dialed in, but I could. There only back lash in this setup is the gear wink in the servo and there is no side to side binding.

The elimination of back lash is important.

Don

SeamusG

Campgems - I like your setup:
. the CF tube with threaded extensions epoxied. What size is #2? Is that comparable to 2-56, 4-40, bigger?
. the Dubro ball link - the only issue that I see is that the nylock nut/spacer requires that the swing of the servo arm hole clears the body of the servo. I'm thinking that I'd put the clevis at the servo arm and the ball link at the horn.

I understand engineering a parallelogram resulting in a 1:1 ratio for servo arm mount hole travel and control horn hole travel. Servo arm swings 60 degrees at 100% throw -> 30 degree travel at the control surface in each direction. My goal is to achieve 10 degrees of control surface movement with about 40 degrees of servo arm movement.

Em, what do you mean by "back lash"? Is that the potential slop in the clevis-to-horn mount (nut-n-bolt), ball-link-to-arm mount (again, nut-n-bolt), and the gear-to-gear gap inside the servo?

Campgems

# 2 is the 2 in 2-56. I used the short threaded on one end rods and cut the unthreaded portion about 3/4" frrom the thread. I then used a 100grit grinding wheel and kissed the unthreaded rod just giving it a rough finish for the epoxy to bond to. Clean everything with denatured alcohol and then epoxy into place.

If you look at what the 3D guys go for, especially in the larger sizes, is that 1 to 1 ratio. With the precession of the digital servos, and the end point and expo in most transmitters today, there is no need for doing the 3 to 1 travel you are asking for. You are still not going to gain or lose torque by changing the ratio. 50 in/oz is 50 in/oz. It doesn't matter if it is a 50 oz load one inch from the fulcrum, or a 1 oz load 50 inchs away.The servo cant tell the difference. With the analog servos especially the older one, they had a wide dead band, somaking the servo move more in relation to thesurface reduced the dead band effect. on the surface. Withdigital servos, this is less of aproblem.

Back lash is a term used to describe the slop in a drive. For example you have a milling machine and you want to cut a 1" square. If you crank it forware on the Y axis 1", and then crank the X axis forward 1", you have cut a 90 degree angle with two 1" long sides. The next step would be to crank the Y axis backward 1". This time, the 1" of travel isn't really 1", and on a well worn machine, it may be only 0.0970". Now you crrank the X axis back 1" and again loose the 0.030" of "backlash" The same applys to the the servo linkage. with the servo locked at a setting, how much movement you have in the control surface is the back lash in the linkage. It comes from loose fitting attachments to the servo arm, the control horn, and the hinges. These are the ones you can cure. There is also some backlash in the gear train of the servo. This is necessary toprevent theservo from struggling with just turning the gears let alonemoving the surface. Some servoshavemore and some less. But what ever you have is whatyou have to live with. So,given you can't change the servo, you need toeliminate and back lashin the linkage. The bestway iswith a setof linkage that fits tight. A Z bend is the worst of the lot. You can't get a zero back lash linkage due to the fact that the bendhas togo through the hole in the arm. The bigger the wire size is, IE#4 vs #2, the more slop you will have. A standard metal clevis has a 1/16" pin, and most of the servo armsand horns have 1/16" holes. So far so good. If you are using a setup where the servo arm and the horn are moving on the same plane, IEthe stock 4*60 setup forailerons, then you can setup a zeroback lashlinkageby using two metal clevis on the linkage,one soldered to the wire andone screwed onallowing for some adjustment in length. Where things start to go a bit bad is if you havethe servo and thecontrol surface moving on planes that are90 degrees apart. You now have to have some slop in the linkages to allow for the side to side movement as each arm or horn rotates on its axis. The ball linkage is great for this setup. It gives a near zero back lash and little resistance to the movement.

OK,I've got to go slice onions and tomatos for tomorrow's fun fly.

Don

SeamusG

Don - Thank you for your excellent response. I get it. The thing I love about this hobby is "as you gain more knowledge and understanding you realize that there is so much more you that you DON'T know." I sleep well when I learn something new.

Have fun at your fun fly!!!!

Cheers,

Campgems

Seamus, I'm sure my post will draw some different views. Don't ignore them. Sort out what makes sense for you and use it.

Onions and Tomatos sliced. Man, tomatos are more expensive than RC stuff. Six big hamberger size tomatos cost me $16.27 today. Istill havent recovered from the sticker shock. I remember paying $4.50 a bushel for them.

Don

ATVFMX

heres how mine are in my 120

SeamusG

Campgems - something that is not readily apparent about your setup is that you use a clevis at the horn and a ball joint at the servo arm. I was at the field today (before the tornado-grade storm rained on our parade [:@]) and had the opportunity to talk with a couple of pattern fliers. They too used a setup similar to yours. I asked whether they use a ball joint on the horn side as well? The answer was "not unless a double horn is used with the ball mounted between the two horns".  "Why?" I asked. "If only one horn is used with the ball joint mounted on one side of the horn, the force applied by the servo will distort the horn towards the ball joint distorting it to one side and causing unwanted flutter".

Yet another learning experience!

Cheers,

Campgems

SeamusG, my setup in the photo has a ball on each end. The Robart Horn has a steel ball built in and the clevis spans both sides of the ball. I wouldn't try a stand off ball like Ihave on the servo though. Their explanation on why it isn't good sounds good. I just never gave it much though as to whyI didn't think it would be good but new I didn't like it. The straighter the linkage is and the closer to the horn or servo arm, the less likely that it will twist or deform. Hummmm, Now if Robart would just come out with ball end servo arms.

I think the fultter issue would be compounded with the horziontal / vertical mix of servo and surface axis.

One think I picked up here on RCuniverse, is to use a flat washer on the head of the socket head cap screw in cale the outer part of the ball linkage popps off the ball. With the washer, it will not come completely off but without the washer, it is possible for the plastic to seperate from the ball and loose all control. Just something to think about.

Don

SeamusG

Me thinks that you're onto a good idea with the Robart "ball link" horns. Using these will require doing a bit of homework to come up with the appropriate horn length before installation. And I understand your dee-zire for "ball link" servo horns.

My friend at the field uses your approach using CF tube with #2 or #4 threaded inserts. Depending on the overall length he will double or even triple the CF tubing thickness for the majority of the control rod's length allowing the ends to "taper" back down to the size that holds the #2 or #4 inserts.

Do you have a brand-specific preference for servo arms? The original equipment horns that comes with the servos are just plain basic (dare I say) junk. I never thought of it before because of my "bore any old hole in the sky Sunday" flying techniques.

Appreciate your input ...