Bellcranks (Again)

[Curare]

Hi guys, as seems to be the norm for me at the moment, I've been doing some deep diving into our aircraft controls, and while there is a plethora of information not only on here and SSW, a few questions have popped up and are either extremely obvious to everyone except me, or are so esoteric as to be excrutiating minutae.

Ok, lets begin.

1. bellcrank positioning in relation to the leadouts. In the past I've always set up my bellcrank with the leadouts in line with the aircraft centreline, thinking that this would give me equal travel at the flap/elevator, but there's seems to be some conjecture as to whether the bellcrank should be "pointed" at the leadouts. Wouldn't this impart an offset to the controls? Would we be better making a bellcrank that has the pushrod hole on centre and leadouts offset to match the line rake?

2. Bellcrank lateral positioning. Looking at some of the plans i have nearby, it seems like the lateral position for the bellcrank is up-for-grabs, some on centreline, some off centreline. In my mind I feel that having the pushrod aligned to aircraft centreline at mid-travel would make the most sense, so the offset to the pushrod is split between up and down. Am I right? if not, why not?

3. Pushrod height. This is borne out of what I think was a throwaway comment by Brett Buck in previous bellcrank discussions https://stunthanger.com/smf/stunt-design/bellcrank-position-effects/msg339848/#msg339848, I think I understand the premise of reducing the angle of the pushrod to the horn. If that's desireable, why don't we have bosses on our bellcranks to raise the pushrod up and away from the bellcrank plate?


Sorry if this is all common knowledge to everyone, and I may well be doing the "Junior Engineer" thing here, and zeroing in on one or two issues without having a comprehensive understanding, still I am curious!

[Brett Buck]

Hi guys, as seems to be the norm for me at the moment, I've been doing some deep diving into our aircraft controls, and while there is a plethora of information not only on here and SSW, a few questions have popped up and are either extremely obvious to everyone except me, or are so esoteric as to be excrutiating minutae.

Ok, lets begin.

1. bellcrank positioning in relation to the leadouts. In the past I've always set up my bellcrank with the leadouts in line with the aircraft centreline, thinking that this would give me equal travel at the flap/elevator, but there's seems to be some conjecture as to whether the bellcrank should be "pointed" at the leadouts. Wouldn't this impart an offset to the controls? Would we be better making a bellcrank that has the pushrod hole on centre and leadouts offset to match the line rake?ng, still I am curious!

You want the bellcrank pointed at the leadouts, put the pushrod hole offset to get symmetrical response at whatever angle that needs to be.

    Brett

[John Skukalek]

What does it mean "the bellcrank is pointed at the leadouts?". This is a question from a limited experience modeler.

Thanks

[Tim Wescott]

What you really want is for the controls to move freely, and for the elevator and flaps to move the same amount up and down for equal up and down movement of the lines.

(and you don't want it all to rip out of the plane in a pull test -- that would be bad).

Doglegs in the flap arm, offsets in the bellcrank hole and all of that just serve that purpose.  It turns out there's no way to get the motion perfect, so it's all just fiddling to try to get things as good as possible.  Howard Rush has a spreadsheet that calculates it for you, once you've entered in the essentials about your control system geometry, I'm sure others have worked this out, too.

A way to (probably) come close to this is to have the bellcrank pointed at the leadout guides, then put the output hole of the bellcrank square to the centerline of the aircraft (so the three holes don't make an isosceles  triangle).  Then make a dogleg flap arm, etc., etc.  It all gets complicated and difficult to really know what you're doing unless you do a pile of math -- hence Howard's spreadsheet.

[Tim Wescott]

What does it mean "the bellcrank is pointed at the leadouts?". This is a question from a limited experience modeler.

Thanks

The flat face of the bellcrank is "looking at" the leadouts.  Put another way, at neutral, with the leadouts taut, they describe an isosceles triangle.

[Howard Rush]

It's a 3-dimensional problem.  Give me your email address, and I'll send you an Excel-VBA program. 

[Geoff Goodworth]

I agree with your second point, Greg.

That said, anybody with a plan that I've drawn will note that the bellcrank is positioned laterally in the wing to achieve equal movement either side of a line parallel to the fuselage centreline.

I assume 45° movement each way but both Brian Hampton and Brian Gardner did experiments years ago to establish the amount of actual movement required to fly the pattern.

They came at the question from different perspectives but, if memory serves me correctly, they found that a minimum of about 15° was required.

One thing I know for certain is that up movement in my ARF Oriental is limited to about 30° and it flies the complete pattern in the hands of a competent flyer. I replaced the M2 clevises and pushrod with 4-40 ball links so the ball link on the elevator pushrod fouls the flap pushrod and limits total movement.

I have seen some classic designs with quite high levels of angularity on the flap pushrod in plan view and I believe that this generates significant asymmetry in the controls.

As Howard says, it is a 3 dimensional problem but I just draw the flap and elevator horns so that there is a 90° angle between the pushrod and the virtual line between the flap or elevator hinge line and the pushrod link pivot in the elevation.

[Curare]

It's a 3-dimensional problem.  Give me your email address, and I'll send you an Excel-VBA program.

Thanks Howard, I'll PM you.

[Curare]

OK, next stupid question. I've always built models with the elevator horn being driven via the flap horn, but In my little discoveries I noticed the Stiletto 660 plan has the flaps being driven from a 'dropper' pushrod.

I've seen this is a few plans over the years, but I don't see what benefit it would have, if any. Why would you chose this way over the other?

[Steve Helmick]

OK, next stupid question. I've always built models with the elevator horn being driven via the flap horn, but In my little discoveries I noticed the Stiletto 660 plan has the flaps being driven from a 'dropper' pushrod.

I've seen this is a few plans over the years, but I don't see what benefit it would have, if any. Why would you chose this way over the other?

I don't think many, if any, would build their pushrod like that these days. I'd say it's 'old fashioned' at least, since it'd be about impossible with carbon tube pushrods, which are very typical now. And then, there's that 3" bellcrank. Rule of thumb is to put the largest bellcrank you can fit in the latest & greatest. If you want bigger than 4", you'll probably have to roll your own, but some will and get the benefits.

What you absolutely don't want is the too fast controls from those old plans. You can easily make a handle with wider spacing to speed up the response, but being forced to use narrower handle spacing if the controls in the model is absolutely not what you want.  Steve

[Curare]

Steve, I don't plan on using it, have no fear!

[Dave Hull]

The "spliced-direct" pushrod system probably only has a couple of differences structurally. It doesn't react the elevator aero moments (that cause a pushrod force) into the flap horn, instead routing it directly to the bellcrank. And, you have one less bearing joint. Neither seem like a significant factor with normal geometries and setups. I haven't heard of the bellcrank arms twisting or failing except at the braze joint if not done well. And I assume that the bearing joints are holding up for people. So no real downside to the current layout.

One advantage of the old system before all the ball links and adjusters were available:  you get the controls in, then shim both flaps and elevator to get neutral (or whatever droop you felt was going to be needed) and solder. No need to bend the elevator pushrod to a super accurate length. No need to put in a split coupler. Less joints=more reliability, all other things being equal. One thing to watch out for though would be if thru the range of travel you are bending any portion of the pushrod you might end up with some non-centering spring forces. I doubt if it would be very large unless the angles for a particular plane were large.

If you look at it in the context of the day, the designs often made very good sense....

Dave

[Dave Hull]

On your question #3 about molding a spacer onto the bellcrank for pushrod height, think about the design this way:

--A perfect bellcrank no pun intended) would be exactly flat and infinitely thin. There would be no "out of plane" bending loads
--The more offset (ie. a big spacer) that the flap pushrod attachment has, the more twisting moment is applied to the arm on the bellcrank
--More twisting loads cause more rocking of the bellcrank and less positive controls, and
--To avoid the rocking, a better fitting bearing is needed and the pushrod attachment may get more critical.

All that said, the bellcranks we are generally using are pretty robust. But it isn't just about them not breaking. With excessive twisting loads you can get faster wear on the pivot, and also on the pushrod attachment unless it is a ball joint. I like to set up my controls with a bent rod for the flaps and avoid the whole plastic reliability issue. But there are pros and cons for everything in design....

Dave

PS--On some CL racing planes (especially the old B-Team Racers) the plans would show bending the aluminum bellcrank arms down (out of plane) so that you could fit it and the leadouts into a very thin solid balsa wing. With as hard as some of those can pull, you were putting a lot of stress on the bellcrank bushing and there were people that strongly recommended against doing it--with good reason.

[John Park]

Since we commonly use fore-and-aft adjustable leadout guides these days, I can't regard the precise orientation of the bellcrank vis-a-vis the leadouts as terribly critical.  If it were, we'd have to incorporate some arcane adjustable features to compensate for different positions of the leadout guide - and even then, I doubt that the effect of such adjustments would be noticeable (certainly not by anybody of my level of competence!).

[Ken Culbertson]

... I doubt that the effect of such adjustments would be noticeable (certainly not by anybody of my level of competence!).

John you make a good point.  We all have an intended location for the CG and bench trim it to that spot before flying and place the leadouts accordingly.  I make my bellcrank square to that layout.  Then I fly it.  In the building of 32 PA planes, I cannot recall one that didn't require an adjustment to the CG or leadouts (since the universal adoption of adjustable leadouts).  One of the things I incorporated into Endgame III was an adjustable flap horn that actually lets me realign the bellcrank to the leadouts.  What I learned from that exercise was that "close counts".  After the first trim flights my CG moved nearly 3/4" forward and with it the leadouts.  When I hit the "sweet spot on the CG" (Matt, it is probably still tail heavy for your taste) I re-alligned the bellcrank and adjust the handle accordingly.  Maybe it is age, maybe it is a lack of skill, maybe it was the logarithmic horn, maybe it is that we rarely use much more than half of our control movements, but I felt *absolutely* no difference, and yet I can feel a 1/16" difference in line length in inverted level flight. 

One other point.  A portion of trim is perception.  If I feel that my plane is making equal turns with equal inputs, it may very well be my brain that is making them equal.  That is why flying with someone that is your skill level, or hopefully better, is important to progressing.  They will spot these things and having them fly your plane will give you a whole perspective you never get otherwise.

Ken

[Brett Buck]

Since we commonly use fore-and-aft adjustable leadout guides these days, I can't regard the precise orientation of the bellcrank vis-a-vis the leadouts as terribly critical.  If it were, we'd have to incorporate some arcane adjustable features to compensate for different positions of the leadout guide - and even then, I doubt that the effect of such adjustments would be noticeable (certainly not by anybody of my level of competence!).

   It's not super-critical, but, you darn well better know within a about 1/4" where your leadouts need to go before you start building the airplane. As Howard notes, the entire topic of setting up the control system is a 3D problem but that is not the case with the bellcrank angle, it is entirely planar. How you set up the bellcrank to flap pushrod definitely is.

       Brett

[Curare]

I've knocked up a quick schematic of the bellcrank, with widly exaggerated leadout rake, and aspect ratios.

The red setup is the basic set up that appears on virtually every plan. Bellcrank square to the aircraft centreline, and the leadouts sweep back at whatever angle they do.

The blue setup is what I understand we're looking for. Bellcrank arms "pointed" at the leadout guides, the holes perpendicular to the bellcrank/leadout centreline, and the flap pushrod hole square to the aircraft centreline.

Is my understanding correct?

[doug coursey]

On your question #3 about molding a spacer onto the bellcrank for pushrod height, think about the design this way:

--A perfect bellcrank no pun intended) would be exactly flat and infinitely thin. There would be no "out of plane" bending loads
--The more offset (ie. a big spacer) that the flap pushrod attachment has, the more twisting moment is applied to the arm on the bellcrank
--More twisting loads cause more rocking of the bellcrank and less positive controls, and
--To avoid the rocking, a better fitting bearing is needed and the pushrod attachment may get more critical.

All that said, the bellcranks we are generally using are pretty robust. But it isn't just about them not breaking. With excessive twisting loads you can get faster wear on the pivot, and also on the pushrod attachment unless it is a ball joint. I like to set up my controls with a bent rod for the flaps and avoid the whole plastic reliability issue. But there are pros and cons for everything in design....

Dave

PS--On some CL racing planes (especially the old B-Team Racers) the plans would show bending the aluminum bellcrank arms down (out of plane) so that you could fit it and the leadouts into a very thin solid balsa wing. With as hard as some of those can pull, you were putting a lot of stress on the bellcrank bushing and there were people that strongly recommended against doing it--with good reason.

SEEMS TO ME THE MORE THE ANGLE ON THE PUSHROD FROM THE BELLCRANK TO THE FLAP HORN THE MORE TWISTING OF THE BELLCRANK....THE TBIRD II CALLED FOR 45  DEGREES ELEVATOR TRAVEL FOR BOTH UP AND DOWN AS SHOWN ON THE PLANS

[doug coursey]

SEEMS TO ME THE MORE THE ANGLE ON THE PUSHROD FROM THE BELLCRANK TO THE FLAP HORN THE MORE TWISTING OF THE BELLCRANK...THE TBIRD II CALLED FOR 45  DEGREES ELEVATOR TRAVEL FOR BOTH UP AND DOWN AS SHOWN ON THE PLANS

[Brett Buck]

I've knocked up a quick schematic of the bellcrank, with widly exaggerated leadout rake, and aspect ratios.

The red setup is the basic set up that appears on virtually every plan. Bellcrank square to the aircraft centreline, and the leadouts sweep back at whatever angle they do.

The blue setup is what I understand we're looking for. Bellcrank arms "pointed" at the leadout guides, the holes perpendicular to the bellcrank/leadout centreline, and the flap pushrod hole square to the aircraft centreline.

Is my understanding correct?

   Yes

   Brett

[Miotch]

you darn well better know within a about 1/4" where your leadouts need to go before you start building the airplane.

       Brett

Well, I'm screwed !! LOL

[Ken Culbertson]

Well, I'm screwed !! LOL

you darn well better know within a about 1/4" where your leadouts need to go before you start building the airplane.

       Brett

And for those of us that don't, there is no kill like an overkill!

Ken

[Ken Culbertson]

Duplicate

[Curare]

And for those of us that don't, there is no kill like an overkill!

Ken

The graduations are a nice touch.