splice for a 4 ft 4130 aileron torque tube control shaft. 3

[booshambo]

Hello,

I have to cut a 4 ft .035 wall 4130 control shaft to install some support bearings near the ends.

I can then splice it back together and install it. (Experimental small aircraft).

I can weld it which is sound, but I'd like to consider some kind of fastening that is removable for bearing maint reasons.

The shaft is a roll control and also reacts pitch inputs in compression/tension as it mounts the control stick pitch fulcrum. It is driven by one hand of the operator via an 8 to 10 inch control stick. It never rotates more than about 50 degrees and everything is unboosted.

Question is, what is a good way to couple the two shaft segments (subsequent to the cut)? I just need the joint to be secure against everything a primate might do with one hand while excited. The interior of the shaft has to remain clear for the pitch control rod to run within.

Is a simple clamp/clamshell like coupler a bad idea on a hollow (vs solid) shaft?
Is there need to have some kind of pin or other feature running radially as a torque path? I'm considering a sleeve that is perhaps clamped and also has a few
rivets radially drilled to connect the sleeve and shaft segments. I also don't know how long a sleeve should be proposed in such a situation.

I'll proto and test whatever solution I consider but I'm hoping someone can say what best practice on such a system might be and whether there are examples of such connections I might reference.

Also any references that mention problems like this would be of great interest.

Thanks,
b

 

[rb1957]

a couple ideas, the basic idea is a splice with cross bolts.
1) slide a tube inside the cut tube, align the tube pieces, then fish the splice tube into position. 2 cross bolts in both 1/2s should be plenty, #10 or 1/4", more if you prefer.

2) splice plates on the outside of your tube. Cut the splice tube length-wise. Align the control tube pieces, cross bolts .. lick o'paint.

My main concern would be crushing the control tube. I'd suggest a relieving cut length-wise into the control tube pieces, maybe 1/2" long so that the free edges of the cut tube can deflect.

another day in paradise, or is paradise one day closer ?

 

[booshambo]

Hey,
Re: cross bolts

I like the idea but I have to keep the interior of the shaft empty because the pitch control rod runs down the middle of it.

Roger that on crush being an issue. I was also considering an interior bushing at the clamp point on each shaft segment to support the clamp force. It still has
to be open enough for the pitch control rod to pass through.

b

 

[MintJulep]

Have you considered a split plastic bearing?

Rulon, Torlon, PERK, PTFE, UHMW.

Seems a shame to cut a perfectly good part.

 

[booshambo]

Hey,
Thanks very much for the list of slick plastics - Trying those is certainly on my list of stuff but if I can get a reasonable solution using a rolling element
that is my current preference/prejudice. My application is known for high roll control forces at high speed,
so rolling element comes first to mind. Still, I will check out the "plain" bearings you mention with a test model.

I do have to object to one thing however, and that is the characterization of the current thing as "perfectly good" ... If it were perfect it would have
a ball bearing at each end! Seriously though, it is extremely nice otherwise.

b

 

[WKTaylor]

Sooooo….

My curiosity is Killing me: what OD is the steel-tube You are splicing?

Also any 'thru bolts' for splicing will be impossible, since You have that peculiar rod in the center of the tube.

Why can't the 4-ft X-tube be made larger with internal plastic guide-blocks for the pass/cross-thru rod??

other ideas...

Install [weld] male-splined ends onto both-tubes [installed before cutting mid-point to assure absolute teeth alignment]... and a female splined-shaft coupling [sliding tightly-over the male splines, then somehow secured in position], to join the segments.

A variant of this would include splines with meshing end-teeth and the coupling. Naww

Or perhaps external mating flanges [one each end at the cut joint]… that are mate-pinned/bolted-together for the joint... if room permits.

Regards, Wil Taylor
o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

 

[booshambo]

Hey,

The splice is for a 1.125" OD .035 wall shaft.

I'm unclear on the plastic guide block advantage but in any event the space the thing runs through is already very tight, so larger is a problem.

The spline ideas (with variants) are worth trying, thanks.

Regards,
b

 

[3DDave]

The plastic blocks help if splitting them to install them avoids cutting the tube.

 

[mfgenggear]

do you have drawing

 

[booshambo]

Hey,
Thanks for all the suggestions.

I may be able to fix this all with an interior steel plug with a slot milled in it to acommodate the interior pitch rod.
The plug is the splice reinforcement attached by radially oriented screws at each end.

This immediately suggests to me using several small screws threaded into said plug and arranged radially around the 1.125" control rod.

AN only goes to an3 and I'm thinking that's too big, but don't know.

Also don't know how many to suggest - load path, tear out etc. nightmare.

Question: what references might mention this kind of thing? I'm familiar with a lot of stuff that gives best practice for rivet spacing and edge distance in Al
but this is perhaps a bit unusual. Is there something that mentions this kind of thing? AC43-13 and Standard Aircraft Handbook are familiar to me,
as well as Machinery Handbook (not the whole thing...) and some Machine Element texts, but I don't think this question is addressed directly in those places.

Regards,
b

ps - still working on drawing (rather a machine readable drawing).

 

[rb1957]

I would hollow out the "slug" (rather than a slot) unless the slot helps you to align it.

NAS #8 screw (#6?) should be plenty ... look for NAS602-08, -06, maybe -04

another day in paradise, or is paradise one day closer ?

 

[booshambo]

Thanks for the NAS numbers - that helps a lot.
The slotness is only so that there's more meat to thread into. That part will be considered and annotated "hog out as necessary". The
alignment is completely done by the attaching ends.


Thanks again.
b

 

[Tmoose]

"My application is known for high roll control forces at high speed.."

Is that proven to be from control rod bearing friction, or the result of aerodynamic forces?
If from aerodynamic forces, I wonder if applying vast amounts of mechanical slipperiness can reduce that.

Regardless, I'd be most reluctant to splice the control rod resisting the "high control forces" with pins or set screws or even a clamped joint. Any loosening would seem to invite deteriorated control feel, or worse.

 

[booshambo]

I expect that the high forces are aerodynamic and short of some kind of boost that will remain.

You make some excellent points but the remark regarding high roll control forces is that they exacerbate any static friction in a journal bearing arrangement vice a rolling element approach. The goal is not to reduce force but rather to avoid the increase in static friction effects under higher load so it feels heavy but smooth.

Sadly the current approach (of aluminum to steel journal bearing) exhibits very bad feel properties even at zero aerodynamic loading (which may be my fault for some
other implementation reason - hence interest in plastic bearing alternatives remains).

The other problems you mention have to be qualified by destructive test (I guess - nothing else comes to mind) but that's a problem too as knowing the max expected
or required torque is a bit of a mystery. I think the usual approach is to bang on it until something more expensive breaks and call it good. I wish I had a handbook
of "pilot strength when they are really afraid" for example. In such cases in the past I've
made a sane estimate, rigged up a pull test and given up after the joint held up after breaking my fixture at 5 x saneMaxEstimate. Then again that was an "on
the ground only" control lock.

Good point on "or worse"...

Regards,
b

 

[RoarkS]

AC43.13-1B Figure 4-37 or 4-38?

 

[WKTaylor]

RS...

As I understood the problem, this this control-tube has-to be re-assembled in place in the aircraft; and that a secondary component is operating thru the middle, end-to-end. Welding wasn't an option.

Regards, Wil Taylor
o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

 

[berkshire]

Booshambo,
Instead if trying to splice this sidestick controller rod, would it be possible to install an access port in the side of the fuselage, to enable you to install the rod then install the bushings with the rod in place?
B.E.

You are judged not by what you know, but by what you can do.