Control Rod Geometry

[Garf]

I would like to try a different setup on my next model. I would like to try having the flaps move about 1/2 the travel of the elevator. Getting this with the conventional method of a rod from the bellcrank to the flap, and another rod from the flap to the elevator would cause a lot of stress on the rear rod. I would like to use 2 pushrods, one from the bellcrank to the flaps, and another from the bellcrank to the elevator. The trick would be getting clearance for the rods at the bellcrank. Would this require an X shaped bellcrank, or is there a way to set this up with a standard bellcrank?

[Zuriel Armstrong]

Designs like the Neptune and Atom (I'm sure their are others) ran the pushrod from the belcrank to the elevator, then back to the flaps.  An easier way would still be to use an adjustable flap horn and dial in the ratio that works best.

[Kim Doherty]

I would like to try a different setup on my next model. I would like to try having the flaps move about 1/2 the travel of the elevator. Getting this with the conventional method of a rod from the bellcrank to the flap, and another rod from the flap to the elevator would cause a lot of stress on the rear rod. I would like to use 2 pushrods, one from the bellcrank to the flaps, and another from the bellcrank to the elevator. The trick would be getting clearance for the rods at the bellcrank. Would this require an X shaped bellcrank, or is there a way to set this up with a standard bellcrank?

The easiest way would be Fly-by-Wire   

Kim.

[Allan Perret]

I would like to try a different setup on my next model. I would like to try having the flaps move about 1/2 the travel of the elevator. Getting this with the conventional method of a rod from the bellcrank to the flap, and another rod from the flap to the elevator would cause a lot of stress on the rear rod. I would like to use 2 pushrods, one from the bellcrank to the flaps, and another from the bellcrank to the elevator. The trick would be getting clearance for the rods at the bellcrank. Would this require an X shaped bellcrank, or is there a way to set this up with a standard bellcrank?

I think that the additional stress of what you are considering would still be within the strength limits of the carbon fiber rods we now have abailable.  You could use a flap horn with holes at 1" and 1.5", and an elevator horn with hole at .75" (or a slider) and put the elevator rod in the top hole of the flap horn, instead of the normal bottom hole.  That will get you a 2-1 elev/flap ratio.  You would probably want to use a shorter than normal throw for flap rod on the bellcrank. 

[Howard Rush]

Getting this with the conventional method of a rod from the bellcrank to the flap, and another rod from the flap to the elevator would cause a lot of stress on the rear rod.

How do you figure that?  The elevator would have to provide the pitching moment it does with 1:1 flaps, less that required to balance the extra flap pitching moment.

[Garf]

Control rod pressure is higher when you use the hole closest to the pivot vs the hole farthest away from it. Maybe the lessened force on the flap rod from decreased deflection angle would help things a little. 

[W.D. Roland]

I use a rod from bellcrank to elevator and another from bellcrank to the flaps on all my newer airplanes.

This gives me a larger range of adjustments, now learning what to do with said adjustments.

This also eliminates slop as parts wear and reduces loads on flap rod and the single hole used on bellcrank.

Alot of slop can develop  at the Bellcrank, Flap horn X2 places, Elevator horn, that's 4 places that add up to lots of elevator slop.

I have done this with std bell cranks placing flap rod in inner hole and elevator rod in outer hole, location of flap horn is critical.

I have done this with X shaped homemade bellcrank with one rod inboard of bellcrank pivot and one rod outboard so that both elevator and flap horn can be placed on bottom for easier access and adjustments.

David

[Serge_Krauss]

Would there be a problem just using ball-links attached at the top and bottom of the bellcrank arm by one screw through the same hole? The top rod  could go to a flap horn above the flap, while the lower one could go to the elevator horn below the elevator. Rate adjustments could be done at the horns. While I don't understand the original conjecture about added stress in a normal set up, this method would keep the two rods far apart and at least partially balance twisting forces at the bellcrank arm. Am I missing something?

SK

[Larry Fulwider]

I think that the additional stress of what you are considering would still be within the strength limits of the carbon fiber rods we now have abailable.  You could use a flap horn with holes at 1" and 1.5", and an elevator horn with hole at .75" (or a slider) and put the elevator rod in the top hole of the flap horn, instead of the normal bottom hole.  That will get you a 2-1 elev/flap ratio.  You would probably want to use a shorter than normal throw for flap rod on the bellcrank.  

When calculating possible control hookups, it is important to calculate all of the ratios. Yes, that would give the desired 2 to 1 elevator / flap ratio, but an overly sensitive handle to elevator ratio.
       That is, using, say, the inner hole of a Brodak 4" bellcrank (.464 per my data sheet), the elevator movement is 93% of handle rotation. Most modern controls have the elevator rotation at only 60% to 65% of the handle rotation. That setup would have an even faster (more sensitive) elevator movement than the most sensitive of the old 3" bellcrank classics with small elevators.  

Nothing is wrong with the two pushrods. An alternative is to use the same flap horn hole for both the bellcrank pushrod and elevator pushrod. **Edit: Same as what Serge posted when I was writing off line** (Ron Burns does this sometimes). At 1.5" flap horn and .75" elevator horn, the inner bellcrank hole as above gives an overall 62% of elevator movement to handle movement (typical modern ratio). The stresses, total ratios, are in the "normal" range, as many folks use the the inner bellcrank hole (although usually because of shorter horns, not for the reasons here)

       Larry Fulwider

[Allan Perret]

At 1.5" flap horn and .75" elevator horn, the inner bellcrank hole as above gives an overall 62% of elevator movement to handle movement (typical modern ratio).

       Larry Fulwider

That is what I was suggesting when I said   "You would probably want to use a shorter than normal throw for flap rod on the bellcrank."

[Larry Cunningham]

One of the things I think is overlooked by many is the non-linearities introduced by flap and elevator horn arms of different drive radius (on conventional setups). It may not be terribly important to most people, so much of non-linearity is insignificant or is adjusted to by our human adaptabilities.

But when I see fliers absolutely convinced that other small features (say stretchy vs rigid flight handles) have major effects on their patterns, I wonder how they manage to ignore the non-linearities in our control systems, and readily fiddle with a slider on an elevator control horn.. 

If you believe in the necessity of an absolutely linear relationship between flap and elevator displacements (say F = -(n x E), where F == flap rotation angle and E == elevator rotation angle and n = a constant, - since our flap and elevators rotate in opposite directions) what are you doing about it? Are you using a tilted bellcrank, or a neutral offset flap horn, or magic geometry, to improve linearity? Naw, not really, huh?

I've considered a mechanical solution using a "shaft drive" (a la BMW motorcycle, say), but being an electronic engineer and all, I'm inclined to believe the very neatest solution would to be to use servos, with their operations mapped to whatever we might think we want. Just a little look-up table in ROM does the mapping, and we could easily fart with it, to find out objectively what works best on our particular models. IF we were so inclined, and wanted to make hypotheses, construct repeatable experiments, and other scientific method stuff. Heck, we might even come up with a Theory of Control Systems for Controlline Stunters, a body of interesting/useful knowledge!

Anyway, my apologies for not making any valuable contribution to this thread.. But it's easy enough to set up a spreadsheet and vary geometries and plot and watch their effects. If you're interested.

L.

"My grandfather invented Cliff's Notes. It all started back in 1912.. Well, to make a long story short.." -Stephen Wright

[bob branch]

Dennis Adamisin did a thread on electrifying the Oariental a while back, I think in the electric forum and he uses 2 pushrods off the bellcrank quite often. Think he did in this piece and there are pics.

bob branch

[Larry Fulwider]

That is what I was suggesting when I said   "You would probably want to use a shorter than normal throw for flap rod on the bellcrank."

Yeah, I know, I saw that. However, the "shorter than normal throw" in this case is only 5/16" (Bellcrank pivot to flap pushrod hole center) which is not a practical dimension. (I wouldn't want to try it)

The throw would have to be that short (5/16") to get an overall handle-to-elevator ratio down in the .63 range using your 1" hole in the flap horn. Even at the 1.5 inch flap horn (common) hole, it still requires the innermost hole bellcrank hole to get a decent ratio.

I don't mean to be overly picky about the issue, but control ratios are such that unless you do "all the math", you can make a system that is unusable or impractical for one reason or another. In this case, the "shorter than normal throw" needs to be calculated to see if it is, indeed, practical.  One wouldn't suspect (or I wouldn't) that the throw had to be that short without actually doing the calculations.

       Larry Fulwider

[Allan Perret]

Yeah, I know, I saw that. However, the "shorter than normal throw" in this case is only 5/16" (Bellcrank pivot to flap pushrod hole center) which is not a practical dimension. (I wouldn't want to try it)

The throw would have to be that short (5/16") to get an overall handle-to-elevator ratio down in the .63 range using your 1" hole in the flap horn. Even at the 1.5 inch flap horn (common) hole, it still requires the innermost hole bellcrank hole to get a decent ratio.

I don't mean to be overly picky about the issue, but control ratios are such that unless you do "all the math", you can make a system that is unusable or impractical for one reason or another. In this case, the "shorter than normal throw" needs to be calculated to see if it is, indeed, practical.  One wouldn't suspect (or I wouldn't) that the throw had to be that short without actually doing the calculations.

       Larry Fulwider

OK, I see your point now.  I would not have thought that it would need a 5/16" bellcrank arm to give a normal overall sensitivity.  Question, when you did the calculations what handle line spacing are you using, 4" ? 

[Larry Fulwider]

. . . Question, when you did the calculations what handle line spacing are you using, 4" ? 

Yes. I have a spreadsheet with the following variables:

Input Data

   Line Spacing (at the handle)
   Bellcrank Size
   Bellcrank Hole (flap horn rod to pivot)
   Flap Horn (hinge line to flap pushrod hole)
   Flap Horn (hinge line to elevator puhrod hole)
   Elevator Horn (hinge line to elevator pushrod hole)

Results

   Flap movement as a percentage of elevator movement (degrees of rotation)
   Elevator movement as a percentage of handle movement (degrees of rotation)
   

       Larry Fulwider
   
 

[Brian Hampton]

I've posted it before on here but my system uses the normal(ish) internal bellcrank to rotate it's support shaft to drive an external bellcrank where both pushrods connect on either side. That gives independent adjustment of flap and elevator travels plus the balljoints connect with turnbuckles to independently set pushrod lengths. One hatch in the bottom of the fuselge gives access to all adjustments.

[Allan Perret]

I've posted it before on here but my system uses the normal(ish) internal bellcrank to rotate it's support shaft to drive an external bellcrank where both pushrods connect on either side. That gives independent adjustment of flap and elevator travels plus the balljoints connect with turnbuckles to independently set pushrod lengths. One hatch in the bottom of the fuselge gives access to all adjustments.

What are you using for the shaft, material and size ?
What is your set up for the shaft bearings ?
Are you using a commercial bellcrank or your own custom ?
How are you locking the bellcrank and drive arm to the shaft, just setscrews ?

[Garf]

Interesting, but I've decided that a systems reliability is inversely proportional to its complexity. In other words, I believe in simplicity.

[Larry Cunningham]

Yes. I have a spreadsheet with the following variables:

Input Data

   Line Spacing (at the handle)
   Bellcrank Size
   Bellcrank Hole (flap horn rod to pivot)
   Flap Horn (hinge line to flap pushrod hole)
   Flap Horn (hinge line to elevator puhrod hole)
   Elevator Horn (hinge line to elevator pushrod hole)

Results

   Flap movement as a percentage of elevator movement (degrees of rotation)
   Elevator movement as a percentage of handle movement (degrees of rotation)
  

       Larry Fulwider
  
  

However, shouldn't your spreadsheet inputs include (3D) coordinates of the axes of the bellcrank and both control horns? And any neutral offset of the control horn arms (from perpendicular to horn wires)? The whole point is that the 3D geometry of our conventional setups perturbs linearity.

Some time ago a gentleman named Salvatore Cabitza sent me a fairly comprehensive analysis (although I'm not sure he dealt with tilted bellcranks, or exponential bellcranks), 17 pages worth with many illustrations and equations, which seem applicable for a serious spreadsheet. It even talks about human hand and control handle geometries. I believe the article was submitted to PAMPA but I'm not sure if it was published. I'm attaching it here as a PDF file.

The flap to elevator link is a 2D problem (normally) and fairly simple to analyze, and observe effects of control horn arm drive radii, control rod lengths, and "magic" geometries. In itself, that analysis will demonstrate effects of differing drive radii of flap and elevator control horns. (I'm not sure there is a practical way around the geometry effects..)

As a general rule shorter control horn drive radii and longer flap to elevator control horns tend to be more linear (reason: less "tilt" associated with the control pushrod).  Similar comments for the 3D geometry associated with the bellcrank to flap drive. Although it is (correctly) argued that a bellcrank can be moved fore or aft inside the wing, distortions associated with the control system geometry will be affected as control rod lengths change.

In the end, you may discover why few people really are all THAT interested in the finer points of the geometry problems..  

Best,

L.

"People don't go there anymore. It's too crowded." - Yogi Berra

[Brett Buck]

Control rod pressure is higher when you use the hole closest to the pivot vs the hole farthest away from it. Maybe the lessened force on the flap rod from decreased deflection angle would help things a little. 

   Presumably you make the all the same except the flap, which is longer. Or make them ALL longer.   That also helps the slop.

    Brett

[Garf]

I should know better than to ask a simple question here. It quickly turns into an analysis worthy of Albert Einstein, but beyond the comprehension of 98% of the people out there. Not everyone is capable of grasping this stuff. For some reason, you all don't seem to want to understand that.

[Garf]

Would there be a problem just using ball-links attached at the top and bottom of the bellcrank arm by one screw through the same hole? The top rod  could go to a flap horn above the flap, while the lower one could go to the elevator horn below the elevator. Rate adjustments could be done at the horns. While I don't understand the original conjecture about added stress in a normal set up, this method would keep the two rods far apart and at least partially balance twisting forces at the bellcrank arm. Am I missing something?

SK

The only problem with this is clearance because I use platform mounted bellcranks, not post mounted cranks.

[leoflyboy]

Both Larry and Serge have shown a willingness to go the extra mile in past posts, and Brians setup is elegant in its simplicity. That being said, Igor Burger has done some work on exponential ratio control with good effect,but the problem persists. To get us into a true power strg mode where nuetral band is forgiving but accurate, and boost is applied with more input ie,lke our car PS is gonna take a btter effort!! A while back Don Ogren showed an elevated bellcrank that he has used,it took some of the BC to flap error out. I built a proto of BC with elevated pivot, approx .75 with tiny ball brgs, put it in a Skylark ,couldnt tell difference. Top line guys could surely feel it, side benefit was really nice easy no effort BC movement, now Im gonna put Ball brgs (2) in BC main pivot,see what that gets me. Combining all good ideas is fine in theory but you gotta start somewhere,the little brgs come from RC car guys, and quite reasonable, wonder if a whole system approach would show good rsult?? Next airplane will be Ares testbed,what btter benchmark? If this has all ben done and Im just polishing eggs, will somebody clue me in? Hats off to the the above for their many great caring posts on other subjects,you guys are great..I have to get Bob Harness to do my test flying as he has the exp and interest, also a flight circle in his front yard,good feedback,good guy too,wish all luck at Nats!!  Rob F

[Larry Fulwider]

However, shouldn't your spreadsheet inputs include (3D) coordinates of the axes of the bellcrank and both control horns?  . . .
 . . .

No    As you have pointed out, as well as others (Serge did a thread some time back on SSW), non-linearity is a "specialty" that some might want to pursue in its more complex forms, looking for improvements. That was not my purpose here, at all.
       The goal here was to look at Classic controls, how they have changed over time, modern norms, the differences between flapped and unflapped systems, the suitability of kit / ARC / ARF hardware components, the advantages of different size bellcranks, and . . . things like that.
       Sometimes, drilling a hole in the right place at the right time can avoid odd handle spacing -- more important after reading Brett Buck's data on line stretch (in SSW). Installing a control system that does not require narrow line spacing at the handle is one key, I think. The wider the handle spacing, the less the line tension differential at any given handle rotation.

For all of these purposes, the calculable ratios around neutral are as deeply as I chose to look. A "Plain Jane" tool. My numbers do not give a picture of linearity at all, leaving that for others to refine.

      Larry Fulwider
       

[Garf]

       The goal here was to look at Classic controls, how they have changed over time, modern norms, the differences between flapped and unflapped systems, the suitability of kit / ARC / ARF hardware components, the advantages of different size bellcranks, and . . . things like that.
      Larry Fulwider  

The question was to see if anyone here could see a problem with using twin pushrods and to see if anyone had a better way to set the system up for a 1 to 2 flap/elevator ratio.

[john e. holliday]

Garf,  they do get deep here don't they.  If you have seen or been where Tom Morris has his wares, you will see a mock up of his control system.  It is very easy to make your own mock up with bellcrank and horns.  Especially if you  have the slider type adjustng horns.  You can sit an play with the mock up until you get what you want.  Then duplicate in the plane.  In some kits for some reason I have had to swap the pushrods at the flap horn to get the elevator to move more than the flaps.  Looks right on paper but, doesn't work in actual use.  That is one of the reasons for external controls when you want to expierment.  

[Larry Fulwider]

The question was to see if anyone here could see a problem with using twin pushrods and to see if anyone had a better way to set the system up for a 1 to 2 flap/elevator ratio.

Yes, and my suggestion to you (common hole in the flap horn) is in Reply #8. See the pic of the Ron Burns mock-up below.

Your quote above was a separate response to a question from Larry Cunningham in his Reply #18.

      Larry Fulwider

[Paul Smith]

My Chief, which has been flying a lot since 2005 has two pushrods going to a standard off-the-shelf nylon bellcrank.  The flap horn is offset a little inboard and the elevator horn is a little outboard.  The flap rod is in the middle hole and the flipper rod is in the outboard.

Like the original poster, I think running all the stress through the flap horn is double trouble.  People do it all the time, but at the price of heavy, expenisve flap horns, which need to be rigidly attached to the TE (more mass).  My flap horn is buried under the "do it right the first time" ethic.  The ratio and throw of the flippers is fully adjustible.

[Larry Cunningham]

No    As you have pointed out, as well as others (Serge did a thread some time back on SSW), non-linearity is a "specialty" that some might want to pursue in its more complex forms, looking for improvements. That was not my purpose here, at all.
       The goal here was to look at Classic controls, how they have changed over time, modern norms, the differences between flapped and unflapped systems, the suitability of kit / ARC / ARF hardware components, the advantages of different size bellcranks, and . . . things like that.
       Sometimes, drilling a hole in the right place at the right time can avoid odd handle spacing -- more important after reading Brett Buck's data on line stretch (in SSW). Installing a control system that does not require narrow line spacing at the handle is one key, I think. The wider the handle spacing, the less the line tension differential at any given handle rotation.

For all of these purposes, the calculable ratios around neutral are as deeply as I chose to look. A "Plain Jane" tool. My numbers do not give a picture of linearity at all, leaving that for others to refine.

      Larry Fulwider
       

Sorry Larry, I didn't mean to muddy waters for anyone, only to point out the potential complexity of the problem, as well as issues associated with differing flap and elevator horn drive radii.

Recall the last part of my post:

"In the end, you may discover why few people really are all THAT interested in the finer points of the geometry problems.. "

Best regards,

L.

"My mother never breast fed me, she told me she only liked me as a friend." -Rodney Dangerfield

[Warren Wagner]

Hi gang,

     "We don't need no stinking Excel..."  or for those that break out in a rash when formulas and spreadsheets are thrown at them, and for those "visual" people that think better in 3D, I offer this photograph for your consideration.

What I was considering, was the amount of handle movement vs. the amount of flap deflection.   On the left is a rule that represents the linear line change at the handle ( as inputed to the bellcrank), and at the right is the amount of flap deflection in degrees.

I was considering only a bellcrank moment arm of 3/4", but you can drill additional holes and see what the difference is in out put.  Same on the flap horn....drill hole location for the commonly used moment arms.

Hey, it ain't 'rocket science', but it can give you all the information you need to know about *basic* control system geometry. 

Cheers,

Warren Wagner

[Howard Rush]

Hook those up to a couple optical encoders and a data logger and you're in business. 

[Garf]

Yes, and my suggestion to you (common hole in the flap horn) is in Reply #8. See the pic of the Ron Burns mock-up below.

Your quote above was a separate response to a question from Larry Cunningham in his Reply #18.

       Larry Fulwider

I have my Tutor 2 set up just that way. I thought that possibly 2 pushrods might simplify adjustability.

[W.D. Roland]

Garf
Twin push rod on std nylon bellcrank in the Chipmunk.
Smooth Flier, goes where pointed.
Inner hole is flap.
outer is elevator.

[W.D. Roland]

Bellcrank on the big yellow thing(Sharkie?)

Perfect bellcrenk was used for pattern.
Thickness is roughly double of perfect.
Flap rod spaced from bellcrank with wheel collar to clear lead outs. Lead outs go over rod( over in pic) even though it does not look that way in pic( Lousy camera)
The threaded clevis adapter at flap horn will be replaced with heavier.

3 flights on airplane indicate fantastic flying airplane.


Hope this helps
David

[Larry Fulwider]

I have my Tutor 2 set up just that way. I thought that possibly 2 pushrods might simplify adjustability.

A flying buddy (who prefers less flap movement in general) uses both the double pushrod and common hole solution with no problems. Depends on what else is around, or in the way, I suppose, as to which is superior in any given application.

      Larry Fulwider

[Larry Cunningham]

Hi gang,

     "We don't need no stinking Excel..."  or for those that break out in a rash when formulas and spreadsheets are thrown at them, and for those "visual" people that think better in 3D, I offer this photograph for your consideration.

What I was considering, was the amount of handle movement vs. the amount of flap deflection.   On the left is a rule that represents the linear line change at the handle ( as inputed to the bellcrank), and at the right is the amount of flap deflection in degrees.

I was considering only a bellcrank moment arm of 3/4", but you can drill additional holes and see what the difference is in out put.  Same on the flap horn....drill hole location for the commonly used moment arms.

Hey, it ain't 'rocket science', but it can give you all the information you need to know about *basic* control system geometry. 

Cheers,

Warren Wagner

That'd be too EASY, Warren! ;->

L.

"Bite the wax tadpole." -Coca-Cola name as originally translated into Chinese (later changed to "May the mouth rejoice").

[Garf]

Garf
Twin push rod on std nylon bellcrank in the Chipmunk.
Smooth Flier, goes where pointed.
Inner hole is flap.
outer is elevator.

I was afraid of the 2 pushrods interfearing with each other. I tend to use a large bellcrank movement.

[Garf]

Bellcrank on the big yellow thing(Sharkie?)

Perfect bellcrenk was used for pattern.
Thickness is roughly double of perfect.
Flap rod spaced from bellcrank with wheel collar to clear lead outs. Lead outs go over rod( over in pic) even though it does not look that way in pic( Lousy camera)
The threaded clevis adapter at flap horn will be replaced with heavier.

3 flights on airplane indicate fantastic flying airplane.


Hope this helps
David

This X shaped bellcrank is interesting. I think that you could get a similar effect with 2 standard bellcranks turned back to back.

[Garf]

Yes, and my suggestion to you (common hole in the flap horn) is in Reply #8. See the pic of the Ron Burns mock-up below.

Your quote above was a separate response to a question from Larry Cunningham in his Reply #18.

      Larry Fulwider

My version.

[W.D. Roland]

That sure is a lot of twisting load put on the horn.
probably will out last the airplane though.

David

[Serge_Krauss]

I suppose there's still some reason for "X"-shaped bellcranks and higher twisting torques, but I thought my suggestion addressed your question and was the simplest that satisfied what you said you wanted to try. The 2:1 ratio then is simply accomplished by placing the elevator horn hole half the distance from its hinge as the flap horn hole is from its hinge. Its only real disadvantage would come in smaller models with insufficient spar thickness for the necessary cut outs. That's assuming your side-to-side bias or horn angles are reasonable. Larry C. was right on that; these do make a difference, and I don't think that people always compensate well for these asymmetries. These can occur in ANY arrangement, and the most symmetrical control surface deflections often result from asymmetrical arrangements.

Edit: 'wrote this after a big-event evening and didn't think it completely through. On reflection, I think that the increased non-linearity of the shorter flap connection vs. the longer elevator connection might make the horn rod connection issue for a 1:2 flap-elevator ratio more one for experimentation, although the  ball-link stand-off heights from the bellcrank might help.

SK

[Garf]

That sure is a lot of twisting load put on the horn.
probably will out last the airplane though.

David

The horn is cut from an extruded aircraft angle. Either 2024T3, or 7075T6.

[W.D. Roland]

I was afraid of the 2 pushrods interfearing with each other. I tend to use a large bellcrank movement.

Yes Garf
Very tight.
The Chip was converted to this after flying with the designed installation.
It almost does not work due to clearance for the rods. It does have less slop(almost '0') to the elevator and that made the go where pointed thing very precise.
I like fairly large movement at the crank also, usually have to cut on the ribs to get enough movement to make me happy. May never need all the movement but its there if it is needed.

The X bellcrank was for several reasons.
More clearance and room for rods
Flap and elevator control horn are on bottom  for easy access and adj.
Making silly parts lets me play with the lath and mill and other fun and dangerous toys.(Flat Bastard File for example)

  There is no best way, yet at least, what ever I can make work and have fun making  when possible.

David

[Garf]

At the end of my aviation career, I made up a batch of bellcranks and surface mount control horns from whatever was laying around. Huck bolts, high shear rivets, extrusions, stainless steel. I even made some landing gear from titanium.