Tube construction: Alternatives to conventional welding

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,778
Location
Saline Michigan
I see much design with body-bound press fit fasteners or tapered pin joints ....mainly in wing attach fittings and wing carry through fittings or spar caps....What might be the reasons for that considering your statements......So should I not be concerned if I have a slip-fit wing attach bolt connections vs press-fit wing attach bolts?

and weather this should be a separate thread or not....I do not know,...found these posts here so I kept it here.
My first comment is going to be predictable - Is there history in any of the airplanes that have the slip fit bolts with failures of these joints. If there are a bunch of airplanes flying and there are no problems, I suspect that the practice, as executed and with the FOS used in the successful designs, is pretty darned good. If there are failures, there is some evidence.

Let's get a little further into it. First thing to know is that bolts have places that will commence fatigue cracking if the bolts see much load cycling in use. They are protected from fatigue by good design and achieving enough preload that cyclic loading of the joint is really not seen in the bolt. If you have a broken bolt in a well designed joint, the preload was low. If the joint is poorly designed, all bets are off. Now if you want proof that a new joint is or is not moving, put it together, load it in service type loading, then remove the bolts and look for fretting or any other signs that the joint has shifted or moved in use. Use strong light (sunlight over your shoulders is great) and a hand lens to examine the joint faces. Shifting is usually accompanied by fretting, smearing, or multiple sets of marks. No shifting evidence and it is tight enough.

For a detail discussion of how bolts and rivets work in joints, go through the chapter in your copy of Shigley on fasteners.

While Shigley does a detailed description of bolted joints, rivets have many of the same issues in design of joints, with an additional issue - rivets are intended to be upset and completely fill the hole they are in to further protect against shifting. Still, if they do not have appropriate preload, and have much load cycling, the joints will shift and fret, further loosen, and the heads will come off through fatigue. In between you will have rivets that "smoke" or spin in use.

The biggest reason for using close fit bolts or taper bolts or pins is to precisely set the position of a joint and make it go back there again later. If there is just one big bolt forward and another aft and they have clearance, that wing's angle of incidence can change quite a bit if the bolt is ever loosened and retightened... Rigging with the bolts loose while you get everythign set can be a huge headache if there is any positional slop in the joints.

And yes, sometimes the engineers do not design a joint for friction to hold everything, but instead count on the bearing strength of the holes and shear strength of the fastener shaft to carry the loads.

Going to the prop examples, yes friction is keeping your prop from shifting or you will see fretting on the interfaces. Those of us running wooden props who find darkened wood against the hub or plate are looking at charred wood from inadequate clamp loads (low bolt torques) - the prop is cycling back and forth on the hub with considerable torque on each cycle making heat that burns the wood and further loosens the joint... Remember that the prop does not just see average torque from the engine. The crank and the prop with it see firing cycle accels that put several times the mean torque on the joint between crank flange and prop.

The prop bolt bushings have a couple purposes that may not seem obvious:
They help with repeating centering and repeating alignment (for balance, etc) of the prop when removed and replaced, which helps with balance, tracking, etc;
They keep the prop on the airplane a LOT longer when the bolts are not tightened enough or moisture related contraction of wood loosens the prop. Omit the bushings with the joint slipping on firing cycles, and the prop and bolt damage will proceed much faster and result in you flying a glider. Yeah, it is a bit of belt and suspenders, but sometimes redundancy has been found to be a good thing...

Bill
 

BBerson

Light Plane Philosopher
HBA Supporter
Joined
Dec 16, 2007
Messages
13,226
Location
Port Townsend WA
Well, if an alternative to welding is sought and the alternative is drilling holes for rivets and bolts, then the issues of drilling holes needs to considered.
So on topic for this thread.
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,558
Location
NJ
Now if you want proof that a new joint is or is not moving, put it together, load it in service type loading, then remove the bolts and look for fretting or any other signs that the joint has shifted
like it....I imagine the aluminum will turn black....like smoking rivets?
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
4,948
Location
US
Well, if an alternative to welding is sought and the alternative is drilling holes for rivets and bolts, then the issues of drilling holes needs to considered.
So on topic for this thread.
That makes sense to me.

So, in AL tube and gusset construction, does adhesive + rivets help or hurt vs rivets alone?
Pro goo case: If the bonding increases resistance to slipping, then the adhesive reduces cyclic loading of the rivets in shear and on rivets/holes. Also, the bonding over a large area could improve joint performance under unusual loading directions (crash, failure of other components, etc)

The case against goo:. If the adhesive provides a soft layer that reduces the grip strength of the rivets or allows vibration/cyclic loading of the rivets in shear (when new, or as the adhesive ages), it could reduce the long term life of the rivets/joint. Also, adhesive slows construction and makes repair/disassembly difficult.

Maybe, in addition to.prep.of the AL and selection of the adhesive, getting the right thickness of goo is important to keeping the rivets happy.
 
Last edited:

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,558
Location
NJ
Well one issue also to consider....a flat plate on a round tube does not have lots of square inches of bearing for friction.....might want to just consider the shear of the rivet and bearing of the rivet against the sheet or tube. I think in this case the glue might help but given the geometry of the joint (round against flat) and low shear of epoxy .....testing might be appropriate, and it is hard to glue aluminum.
 

rv7charlie

Well-Known Member
Joined
Nov 17, 2014
Messages
668
Location
Jackson
Prop bolts need to be retorqued periodically for a wood prop, otherwise the prop will move and the friction will char the face and then it gets looser and fails. My flying buddy had the flywheel bolts shear on his Jabiru. While he was installing new flywheel bolts I picked up his wood prop and showed him his charred prop face. He did not know about the need to retorque wood props (50 hrs or something) The loose prop had broke his flywheel bolts, fortunately while taxiing.
You are correct that prop bolt torque must be checked periodically. (Not necessarily tightened; I've run wood props on Lycs for almost 3 decades, I live in the deep (wet) south, and after the initial retorque on a new prop, I've rarely gotten any nut movement when rechecking torque). But at least on a Lyc, maintaining torrque has nothing to do with maintaining friction on the prop/flange interface. There are six big press-fit lugs in the flange, pressed at least 1/2" into matching holes in the back of the prop. There will be no friction affecting the back of the prop until loose bolts have allowed the prop to wobble on the flange enough to wallow out the lug mounting holes.

I suspect that your friend's bolts were broken because they were loaded in bending, as the prop wobbled.

Charlie
 

BBerson

Light Plane Philosopher
HBA Supporter
Joined
Dec 16, 2007
Messages
13,226
Location
Port Townsend WA
There are six big press-fit lugs in the flange, pressed at least 1/2" into matching holes in the back of the prop.
The 1/2" lugs are for tight fit in aluminum, I think. I doubt 1/2" of a 4" wood hub is enough. But who knows?
Older engines made for wood props had a splined hub front plate that kept the bolts from bending.
 
Last edited:

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,778
Location
Saline Michigan
Well one issue also to consider....a flat plate on a round tube does not have lots of square inches of bearing for friction.....might want to just consider the shear of the rivet and bearing of the rivet against the sheet or tube. I think in this case the glue might help but given the geometry of the joint (round against flat) and low shear of epoxy .....testing might be appropriate, and it is hard to glue aluminum.
Metal on metal friction is usually well described by a coefficient of friction times the clamping force. This applies until you get to yielding of one or both of the connected members - small faying surfaces (tube on flat qualifies) can reach yield fairly easily, which also results in loss of preload, so they have to be designed carefully. Remember also, that the region in compression around the hole usually extends only one diameter of the fastener beyond the fastener itself. Only in larger surfaces designed as gasketed joints do we spread the loads much further from the fastener itself.

Yes, getting adhesives to remain bonded and unoxidized on aluminum is tough to achieve, particularly in home shops.

Billski
 
Top