Thread Tools
Nov 05, 2008, 05:52 AM
Go fast, turn hard
Tailslide UK's Avatar
Thread OP
Mini-HowTo

Linkages


Ok guys, I am sat here today while the carpet cleaner guy is attacking the house. I am trying to catch up on some promises I made to people on the board for more details. A couple of weeks ago I made a comment in the 80/20 thread about helping out new(ish) pilots. The comment I made pertained to helping new builders sort out the linkages driving their control surfaces. In my experience, this has had the greatest impact on the flying quailties of the models owned by people I have helped. The comment generated more than a few PM's asking for more, so here goes.

If you have a coule thousand posts and have been doing this since rotary escapements were used for servos, close this thread now and move on. If, on the other hand, you have trouble getting control surfaces set up that always return to nuetral, don't load the servo and are slop free then read on.

Nothing I will present here is my own invention. A wise and experienced modeler took me under his wing when I had been bashing planes about for a couple of years. He taught me the simple basics of linkage set-up and I have been doing it the same way for 25 years. I will say that getting the control surface to work right made the biggest single impact to the flying quality of my models.

I will cover one part at a time and use photos to illustrate as I go. Rather than one long post, I will break it up into several. When the carpet guy gets to the office, I will have to go off-line until tommorrow.

Let's get started then. Also, please feel free to add to this thread with your experience or ask specific questions. No one here will flame you for trying to improve your models (right guys?).

The first post will be up in a couple minutes.
Sign up now
to remove ads between posts
Nov 05, 2008, 06:12 AM
TonyC
CaterhamTone's Avatar
Look forward to this
Nov 05, 2008, 06:18 AM
Go fast, turn hard
Tailslide UK's Avatar
Thread OP

Geometry


Getting the linkage right starts with putting the servo in. I will focus on wings, but all the concepts apply to fuse mounted servos as well.

Before you cut your servos in, look at where and how you will place them. There are a number of considerations.

1) It is best to have the servo at the mid-point of the surface (spanwise). This is not always practical, but I do try for at least 1/3 of the way out for an all moving TE aileron.

2) Position the servo far enough forward so you have room for your linkage. If you use a solder link on one end and a clevis on the other, make sure the distance from the hinge line to the servo are is at least 2 clevises long.

3) Now that you know where to put the servo spanwise and chordwise, get out your square. Align the servo so that it is at a right angle to the hinge line. On a highly swept TE you can fudge a bit for less drag, but the ideal is 90 degrees. Have a look at the pictures. My triangle is clear, so you will have to open them to see it.

4) I cut the servo holes with a dremel and router attachment. Trace the servo and cut undersize. I then open the hole in small increments until the servo will push in. Zero up everything , put on the arm and install the servo. I use silicone to mount the servo, but wrapping the servo in 1 layer of masking tape and fitting with epoxy also works well. If there is any slop in the servo pocket, shim it with basla. The servo must not move in the pocket when you are done.

Pictures:
Nov 05, 2008, 06:37 AM
Composites Kid
Alex.Schweig's Avatar
Looking forward to the next part! Ive always had trouble with linkages and model set-up.
Nov 05, 2008, 06:46 AM
Go fast, turn hard
Tailslide UK's Avatar
Thread OP

Part III - horns


Now that the servo is in, the horn on the control surface needs fitting. This is where I see the most problems and the most slop/binding getting into the system. The position of the horn is the single most important step in a good linkage! There is only one right place for the horn. The pin of the clevis needs to be ON the hingeline. Not forward of it, not behind it - on it.

This means that in most installations, the horn must protrude forward as it exits the control surface. The photo shows the horns I use, but any horn will work as long as it is postioned correctly. I like these horns for a couple reasons. They are big enough that most shapes can be cut from them and the hole for the clevis pin is the right diameter. I get them here:

http://www.hilaunch.com/Aileron%20Horns.htm

To the first point, almost every horn you fit will be bespoke for that particular control surface. You will need to do some cutting and shaping to get the geometry right.

As to the hole diameter, the horn should have a hole sized so that it is difficult to get the clevis pin into it the fist time. The clivis should be stiff to rotate when you first put it in. A few minutes spent rotating the clevis in the hole will loosen it up. Time will also creep the material a bit and you will find the clevis easier to reotate after about 24 hours.

If the clevis has any slop in the horn, bin it and get another.

Now, using your square again, draw a line on the control surface that is in line with the servoa arm and 90 degrees to the hingeline. Cut a slot or mount your horn so that the hole in the horn is directly over the hingeline.

Pictures:
Nov 05, 2008, 07:12 AM
Go fast, turn hard
Tailslide UK's Avatar
Thread OP

Part IV the linkage


This part really comes down to personal preference. I use what I have found to work right every time. I will not go into other brands, but I will say that I use Sullivan rods with one end threaded and the rest bare wire. Whatever you chooses, keep this in mind:

1) Some rods are not hardened. They are soft metal and take a bend easily. I would avoid them

2) Some brands of rods have the threads rolled on instead of cut. If the clevis has a lot of back and forth (axial) movement when screwed on, this is most likely the reason.

I won't link to rods, they are avaialbe everywhere.

I tend to use the same set-up on about everything. I use a Du-Bro 2-56 solder link at one end and a Sullivan 2-56 gold-n-clevis on the other. DS stuff is the same set-up , but the size is upped to 4-40. I use a sleeve on the solder link (the one that comes with it) to keep it tightly closed. At the adjustable end, I always use a 2-56 jam nut. I can't stress this enough. The jam nut removes all of the slop in the system. Do not omit it.

A couple of other points. Don't drill out or open up the hole in the servo arm. It is meant to be tight the first time. The plastic will creep over a day or so and everything will loosen up. Keep your links straight if at all possible. Bends in the links will flex a bit under load and add slop. Sometimes you have to, but avoid it if possible.

The Solder link should be attached using acid core silver solder. Please do not use electrical solder! It just is not up to the job.

Pictures:
Nov 05, 2008, 07:18 AM
isoaritfirst
isoaritfirst's Avatar
Nice thread Paul,
just an adendum to;
"There is only one right place for the horn. The pin of the clevis needs to be ON the hingeline. Not forward of it, not behind it - on it. "
I'm sure you know but for the thread

--more correctly stated - At surface nuetral point the angle should be 90 degrees from pushrod through pivot point to hinge point.--

--the statement is correct for your installation where the pushrod will be running parallel with the bottom wing skin, but if the pushrod was angled, perhaps due to using different size servo/surface horns or more so with a inverted servo installation, then the horn needs to such that the pushrod to hinge creates 90deg.
Nov 05, 2008, 07:29 AM
isoaritfirst
isoaritfirst's Avatar
Easy as a second look so please take this in the spirit it is offered -

May be worth adding a bit to post 462 about the importance of making both wing sides identical in servo and horn position, and rod lengths.
And how you add the second surface horn (jig)
Nov 05, 2008, 07:39 AM
Go fast, turn hard
Tailslide UK's Avatar
Thread OP

Part 5 - gear ratio and servos


I use a wide variety of servos, both analog and digital. The key thing I look for in a servo is head slop. If given a choice, I will always choose a BB servo over a bushing servo. If the servo arm has slop, so does the control surface. In general, buy the best servo you can afford and will fit. I won't go into digital vs. analog, but I will make one point. If an analog servo is displaced a small amount from nuetral, it apllies a smal amount of power to rotate it back. If a digital servos is displaced a small amount from nuetral, it applies the full force of the servo to push it back. This is why digis center better.

Now that you have your servo, it is time to get the most out of it. This is another dark area we can shed some light into. Most servos will give you about 90 degrees of total travel out of the box before you start progrmming. That is, everything in the TX still set to default 100%. (travel adjust, dual rates, endpoints, etc). However, we usually only use about 25% of this. We are wasting power, precision and repeatability by not using everything we pay for. I am talking about the gear ratio of our linkage set-ups.

If the sevo arm is long and the horn is short, a little movement of the servo gives huge deflections of the surface. On the other hand, if the opposite is true, the servo has to move a lot to generate the same deflection. This is a very good thing. Short servo arms and long horns mean that the servo uses its whole available movement to deflect the surface. This gives more transmitted torque, more prescesion and more repeatability. It also adds to the abilty of the servo to get the surface back to the same nuetral point every time. This one simple thing can greatly enhance the "flyability" of your plane. If you have a plane that is "jumpy" or too sensitive, have a look at your geometry. If you are flying at 30% travel on your DR, you are throwing away a lot of torque and precision that you paid for in that servo.

I won't go into the math, I will use the photos to illustrate the point. Basically, use the shortest arm on the servo that will give you the desired amount of travel. Use the TX to increase the travel before making the arm longer. Use longer horns and balance control response against increased drag. Only you can decide where this balance lies. The concept can be applied when making your decision.

The best way to illustrate this is with my Zero Gravity Python. The horn is long to get clearance for 3D deflections. The transmitter is set at max throw. As you can see, the servo moves a lot to generate the deflections. The clevis started out near the botom of the arm (short) and was moved out in succesive stages as the plane was tuned in. It is easy to see from these pictures how the gearing discussion applies. Note that the pivot point of the horn is on the hinge line and everything is square.

Photos:
Nov 05, 2008, 07:59 AM
If it can-it will!
cliffhanger's Avatar
Also just one other point regarding setting of the horn in regard to the hinge line.

For pull/pull systems popular on rudder control, the horn line is slightly behind the hinge line to introduce the "Ackerman effect"

More info here: http://members.cox.net/bdfelice/Ackerman/ackerman.htm

( just my 2p worth )
Nov 05, 2008, 08:05 AM
Go fast, turn hard
Tailslide UK's Avatar
Thread OP

Wrapping it up


I will end this little project with some examples. Some are solutions to getting the geometry right in akward places and some are just plain wrong.

My K&A P-38 was a right pain to get the elevator set up. I departed from the instructions and went with a scale cut for the elevator. I thought I would use a torqure rod to drive it, but later found this to be not workable. After a bit of head scratching, I came up with the solution shown in the picture. The connection is over the hingeline and geometry is right, but it looks a bit funny. I removed the ball link for the picuture.

Just becasue a model costs way on the north side of $1000, does not mean it will be right out of the box. I am currently setting up some Wizards that we did a group buy on this summer. Have a look at the elevator horn! I assumed it was a sick joke. The pivot is more that 1/2 inch behnd the hinge line. This is supposed to be a DS capable BPV. The other picture shows the modification apllied to fix it. Same horn as shown earlier and slotted in around the drag spar.

The shot of the two fins is also from the wizard. As you can see, the geometry of the "horn" to the hingleline is wrong here as well. Sometimes you just ahve to live with it though.

The last picture shows two things. One is that you should not bend a linkage unless you have to. The other thing it shows is the "cusp" that has been cut into the clevis. If you are using a really short arm on the servo, sometimes the clevis will hit the barrel of the arm where it goes over the output spline. As long as you leave some metal behind, you can cut a clevis like this to avoid this problem.

I hope that this thread has been helpful. It now belongs to everyone. Please add to it and correct me as needed. Post your questions and we can all pitch in to help out. The carpet man is now in the hallway, so I am need to go until tommorrow.

Cheers,
Tailslide
Nov 05, 2008, 08:07 AM
Go fast, turn hard
Tailslide UK's Avatar
Thread OP
Quote:
Originally Posted by isoaritfirst
Nice thread Paul,
just an adendum to;
"There is only one right place for the horn. The pin of the clevis needs to be ON the hingeline. Not forward of it, not behind it - on it. "
I'm sure you know but for the thread

--more correctly stated - At surface nuetral point the angle should be 90 degrees from pushrod through pivot point to hinge point.--

--the statement is correct for your installation where the pushrod will be running parallel with the bottom wing skin, but if the pushrod was angled, perhaps due to using different size servo/surface horns or more so with a inverted servo installation, then the horn needs to such that the pushrod to hinge creates 90deg.
Thank you for the picture! Well said description as well. Please, keep adding to this thread. I just banged it out, now everyone can add and clarify to make it complete.

Cheers,
Paul
Nov 05, 2008, 08:09 AM
Go fast, turn hard
Tailslide UK's Avatar
Thread OP
Quote:
Originally Posted by isoaritfirst
Easy as a second look so please take this in the spirit it is offered -

May be worth adding a bit to post 462 about the importance of making both wing sides identical in servo and horn position, and rod lengths.
And how you add the second surface horn (jig)
VERY inportant. I should have said that. Thanks Mike. Please keep up your input, we all benefit from it.

Tailslide

Okay, the carpet guy is now standing outside the office with his arms crossed. See you all tommorrow!
Nov 05, 2008, 09:24 AM
Registered User
Chief High Horse's Avatar
Paul,

Note on the solder. I know you said silver solder, but the silver content is also improtant. Some only have 2 or 3% silver content, I use 55% silver content, the higher the better.

This silver solder is for structural more than electrical use.

I would like to hear from an experienced welder/brazer tell us their tips for working metal.

one question is how to keep the strength of the rod and the spring properties of the clevis after heating them so much.

Dave
Nov 05, 2008, 09:33 AM
auto-tune remix
slopemeno's Avatar
Well, I was a gunsmith for four years, and we used silver solder for lots of different jobs. Brownells is a great resource for different types of solder, but I'm sure you local Airgas welding supply carries the same stuff under a more generic brand name.

Tips? Roll your silver solder out into thin ribbons, as the amount you really need is very small. Use a LOT of heat- cold joints are probably the #1 cause of failure.

Yes, you lose some of the temper when you use that much heat.


Quick Reply
Message:

Thread Tools