Bonding aluminum hub to steel power transmission shaft

[PCS74]

Gentlemen, I need some help with concept feasibility. I am looking for feedback, advise, recommendations, opinions, experiences and thoughts on the bonding aspect of this concept and not other ways to join the parts.

The details are as follows:

7in outside diameter steel power transmission shaft
356T6 cast aluminum hub
170in^2 bond area
Torque is applied to the shaft through the hub
Variable toque load from 0 to 1025lb-ft
Design factor of safety of 3.3
Weibull torque distribution (wind power)
750 million cycle / 20 year life requirement
Joint is critical and can not be allowed to fail within specified life requirements
Outside operating environment
-20F to 160F operating range
Minimum overlap shear strength of adhesive 2000psi
Joint prepared per adhesive manufacturers specifications


Statically:

1025lb-ft * 3.3 = 3382lb-ft

3382lb-ft * (12in / 1ft) = 40590lb-in

P = T/r = 40590lb-in / 3.5in = 11597lb

s = 11597lb / 170in^2 = 68psi

Margin (design load) = 2000psi / 68psi = 29


Looking at the loads statically this is no problem.

From a life perspective I do not know what to expect. I do not have any experience calculating the expected life of adhesives nor have I seen any literature from adhesive manufacturers that would provide me with the information I need to do this. I have spoken with representatives from several manufacturers, both sales and technical representatives, and all claim that the life requirement is attainable. However, I have not been able to find any case studies documenting such life or accelerated testing that would indicate such life in the public domain(I am not ready to spend a fortune on papers and books at this point).

As stated before, I would greatly appreciate your feedback, advise, recommendations, opinions, experiences and thoughts on the feasibility of this idea.

In addition, if you have any specific recommendations on resources (books, papers), specific adhesives or manufacturers I should look into or contact it would be very helpful.

Thank you,

Chris Skarzenski

 

[MikeHalloran]

I'd worry some about thermal cycles.



Mike Halloran
Pembroke Pines, FL, USA

 

[btrueblood]

Reasons not to use a taper-lock bushing or similar?

 

[btrueblood]

Along to the OP more specifically...

Most adhesive manufacturers will give you test data for standard test coupons and standard test conditions. Life testing is generally up to the end user. A large aircraft manufacturer I once visited had a huge room set up with salt spray, thermal cycling and wet/dry cycling tanks, all testing various parts made of various aluminum alloys and glued with various adhesives, all busily cycling along.

In addition to Mike's (as usual) excellent advice, I'd worry about wet/dry cycles and freeze/thaw cycles too (depending somewhat on the adhesive I chose).

I'm a bit at a loss as to how you think you will apply adhesive and verify no voids/bubbles in what sounds like a slip fit radial joint.

The closest product I can think of to what you describe is an automotive "harmonic balancer" or damper pulley. You might look at how those are built...generally by directly overmolding and bonding the rubber onto the steel. This is typically done with very high mold pressures to ensure no voids in the rubber...

 

[MiketheEngineer]

Heat??

 

[MintJulep]

Minimum overlap shear strength of adhesive 2000psi

This number applies to specific test coupon construction and a specific test loading. The test coupon materials are unlikely to be the same as your materials. The test loading is unlikely to be the same as your application.

Joint prepared per adhesive manufacturers specifications

This is much much more than putting at note "Joint prepared per adhesive manufacturers specifications" on the drawing. Adhesive failure is almost always due to improper surface prep. It is essentially impossible to do surface preparation correctly 100% of the time. You need to count on poor surface prep and design for the bond that will be achieved with poor prep.

Adhesives can fail at the bond plane with either surface, or within the adhesive.

I suggest that you read ASTM D6465 Standard Guide for Selecting Aerospace and General Purpose Adhesives and Sealants. You will discover a long list of "Oh! I never thought about that" items to consider.

 

[TVP]

Critical joint and adhesive bonding usually are mutually incompatible. I concur with btrueblood about taper-lock bushing, etc. Have you looked at polygon shaft-hub connections? Here is a link for more information:

http://stoffelpolygon.com/

 

[racookpe1978]

Why not a taper fit of shaft and hub, but with a key between the two to positively transmit the torque under all thermal and vibration conditions? Then lock the taper fit with simple threaded nut on the shaft clamping the two together.

Think of a Morse taper fit in any of ten million drills or milling machines, with the Morse taper parts held by friction but driven apart by a wedge when the drill bit or milling adapeter neds tobe changed. In this case, you prevent the movement by the nut on the shaft, rather than driving them apart with a wdge.