Loctite Sprocket Mount

[Underthefloor]

Has anyone tried to loctite their sprocket mount to the diff rather than using bolts or splines?

Do you think that this could work if you had over 6 sqare inches of contact area between the collar and shaft?

Loctite specifies that their 680 compound can withstand up to 4 ksi in shear.

[Underthefloor]

Has anyone tried to loctite their sprocket mount to the diff rather than using bolts or splines?

Do you think that this could work if you had over 6 sqare inches of contact area between the collar and shaft?

Loctite specifies that their 680 compound can withstand up to 4 ksi in shear.

[Jersey Tom]

I definately wouldn't.

Even if an adhesive claims it can withstand 4ksi in shear, that's BEST case scenario. If you don't prep the surfaces PERFECTLY it will go down to well under 1ksi easily.

Plus, Loctite is designed to resist shearing forces.. not necessarily to be transmitting engine torque. And you've got the fact that it isn't gradual easy loading. Lot of vibration, shock loading (shifting?), possibly a fair amount of heat radiating off from the engine.

I wouldn't do it.

[James Waltman]

I wouldn't use that particular Loctite product for that job but I wouldn't hesitate to use a proper structural adhesive. Something more like Loctite E-120HP. It should also be good for about 4000psi shear strength. I wouldn't count on its elevated temperature performance because it is a room-temperature cure product. I've also had good experiences with Loctite's 9394 and 9430.

At WWU our testing of 3M's Scotch Weld 2214 showed an average shear strength of about 5300psi. Our lowest test sample was over 4500psi and higher than 6500psi shear strength was common. These aren't ˜best case' numbers either. They are actual samples that were bond prepped and bonded by us. I'll admit that our (about) 30 test samples weren't an exhaustive study of 2214 but it was enough to give us confidence in the product.
The 2214 is a one part heat cure epoxy adhesive that should have great performance up to about 250?F. 3M also makes a high temp version of the same product now that has consistent strength up to 300?F (although lower strength than the regular version).

Six square inches of bond area should give pretty good strength. What kind of loads do you expect to see and what kind of safety factor do you want?

You'll absolutely need to do some tests if you decide to bond important structures. It can be very effective if you know what you're doing.
I happen to know of an aircraft manufacturer that bonds all the major structures of its planes.

[Jersey Tom]

Bonded frames are one thing. Those are proven, even at the FSAE level. But using an epoxy right near your engine to be transmitting all the power your car puts out seems sketchy to me.

The aerospace industry is also VERY knoweldgable to the materials it uses, the prep, the fabrication, QC, service life..

College undergrads late night painting some adhesive on a sprocket and slappin it on a differential.. just aint the same level of quality.

I personally just wouldn't do it. What you gain is minimal if not negligable. And you'll have to put in a LOT of hours testing.. time spent better elsewhere IMO.

[Travis Garrison]

First off, James posts on a bonding question, and I follow up? Shocking http://fsae.com/groupee_common/emoticons/icon_smile.gif

Tom, bonding is not that tricky you're making it out to be worse than it is. Ask yourself this, would you trust someone to weld a portion of your drivetrain? Welding takes serious skill, a good bond joint just takes a kindergartener, some sand paper and some acetone...maybe some supervision so they don't drink the acetone.

I personally would hesitate to do it because it sounds like a bear to work on, are you going to be able to swap sprockets easily? or are you just bonding your sprocket adapter?

If I were to do it here's how I would go about it.

-Make sure you have a taper (say 2-5deg) on you your bonded surfaces and that you have ~5 thou gap between walls

-Definitely do a physical test of a dummy part, at temp with matching materials (so aluminum on steel? at say 250F)

-consider sealing the bond (with a compatible paint maybe) in order to protect it from environmental unknowns (oil, occasionally brake fluid spill, etc..)

-do a thermal analysis of you parts to make sure different cte's aren't going to overstress your joint, if you've got an aluminum diff this probably won't be much of an issue, but if you're modifying the university special you could be in trouble here

-invest in an acid etching pen for the bond surface prep..not really necessary but hey why not do it right?

Getting a good bond with the proper static strength is not at all the issue, making sure it can handle the thermal cycles might be the real trick...

[Jersey Tom]

I actually no longer trust student-welded drivetrain components that much anymore. The welded aluminum differential housing we had on a previous car was done by one of our better welders last year, and it recently cracked. Wasn't even that stressed. Professionally done and subsequently heat treated, sure no problem.

The reason I don't trust bonding that much is we had done some testing on pull-out strength of bonded carbon/aluminum A-arms.. used a couple different adhesives. For the most part there was a lot of sample variation and the pull-out strength was on average much lower than anticipated.

[Travis Garrison]

What did the inserts look like after they came out? Did they get a full bond? Tapering your bond joint will go a long ways towards improving the quality. What kind of prep work did you do on your inserts and tubes?

When we did test specimens up at WWU everyting was pretty much on the money...and I'm sure now that you've brought up a-arm inserts James is fighting back the urge to post pics steel tubes ripped apart via bonded aluminum inserts...

[Jersey Tom]

Inserts and tubes looked fine. Just pulled out and left a little ring of adhesive in there.

We used a slip fit for the two, so not much volume of adhesive in there. Surface prep.. half the samples were just machined and cleaned with isopropyl alcohol, the other half were machined, sanded with very fine grit paper, and then cleaned with isopropyl alcohol. Didn't have the time to get an etching agent.

None of the insides of of the tubes were 'prepped' by means of sanding or whatever.

[Travis Garrison]

Hmm..well that's unfortunate. What kind of psi did you get from you samples?

James is letting me down here so here's how ours turned out:

http://dot.etec.wwu.edu/fsae/images/A-arm%20test%203_jpg.jpg


You can make em as strong as you want. We were expecting a little more wiping of the epoxy...but instead we showed that we had over shot on the bond area. For inserts like this you really have to watch out for wiping of the epoxy. I've never tried alcohol for prep, I suppose it should work (it's aproved by the loctite specs), but I'm a big fan of acetone http://fsae.com/groupee_common/emoticons/icon_smile.gif

Also, fine grit sandpaper may be counter productive, depending on the grit. Medium grit emery cloth is the official recomendation..

Our samples were definitely not done in a profesional setting. It was one of us on a greasy countertop (we'd put down paper) hand sanding and then wiping em down with an acetone soaked rag. I don't think gloves were even used every time. We wanted to be sure we dind't inflate numbers with "perfect" prep work if we weren't going to be able to do it every time.

The official loctite bonding prep recomendations can be found here: http://www.aerospace.henkel.com/index.cfm?ID=117

Anyways good luck with the sprocket...

-Travis

[Jersey Tom]

I think at best we got up to 1ksi. We're gonna do some more testing though.

[James Waltman]

Sorry to let you down Travis. You beat me to posting that picture.

Sure the aerospace industry is knowledgeable. That information is with the engineers though. The people actually applying the bonds are just regular laborers. They know that they are supposed to prep and apply in a certain way. Travis mentioned that a kindergartener could do it and I would agree with that.
I happen to work for the aircraft manufacturer that I made reference to. At the risk of giving too much away, I'll say that bond prep is done by sanding, wiping down with acetone, and then a final wipe with isopropyl alcohol. Certainly no magic there.

If you didn't prep the inside of the tubes then you didn't really prep the joint at all. It's good to see that you are going to do some more testing though. Drop me a line if you would like some advice. Once you get your method sorted out I'm sure you'll join me and Travis on the Dark Side and soon you'll be talking up the advantages of bonding.

[Garlic]

Guys I think that bond might be shit. Not sure if you noticed but look what it did to that tube!!

[Underthefloor]

We were going to have the diff and hub splined but now our sponsor is backing out so I was just thinking of possible alternatives. I guess it would be risky to depend on loctite(or other adheasives) because of impact loading.

Are there other forms of shaft/hub couplings that have been successfully tried? We were also thinking of using dowel pins around the shaft which would act like a bunch of key pins.

I also looked at some friction couplers like those on fptgroup.com. It might be possible to use loctite along with a couple of FLK 300s in series to boost the torque capacity. http://www.fptgroup.com/downloads/fenlock_products.pdf

What do you guys think about either of these ideas?

[Wright D]

We have had a lot of success using multiple key ways. We used six 3/16 key for our sprocket-diff connection last year. They are easy enough to make, any descent machine shop should have the capabilities.

[deersound]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">None of the insides of of the tubes were 'prepped' by means of sanding or whatever. </div></BLOCKQUOTE>

I'm pretty sure that's your problem. Mill scale won't come off with isopropyl.

We did fairly extensive adhesive testing using one of the ASTM codes as a guide for surface prep, and our tests came out right on the money. We did the fine-grit sandblast / acetone combo and the tests came out damn near perfect.

[Jersey Tom]

Wasn't steel. Test was on aluminum plugs and carbon tubes.