Structural Adhesive - Rear Sway Bar Frame Brackets?

As you may or may not know, my Barracuda recently came home from paint. I have a Hotchkis rear sway bar kit that I would like to install, however requires welding brackets onto the frame. Since the rails have been throughly coated inside and out and the car is finish painted, I would like to avoid any welding whatsoever on, or near, the car.

This Hot Rod article sparked my interest in using an adhesive for this application. I have never used structural adhesives but I know some people swear by them and I would like to hear some input.

These brackets are expected to be welded as shown to the frame rail. I am proposing using the linked adhesive and running a bead around the entire bracket to the frame. I can't help but think that it will be more than sufficient for this application...from what I hear. Also the failure mode, if it did not hold up, should be far from catastrophic. One bracket would separate from the frame and the bar would then see no load, other than moving with the suspension until repair/removal [welding].


Thoughts?


https://www.amazon.com/3M-Resistant-Stru...ctural+adhesive

Instructions for the stuff I found quickly online do mention...



So if you have to strip the paint for it to adhere properly, would it be just worthwhile to weld it?

Why not just drill/bolt them in, that's how the factory installed the upper bar brackets?

So you plan on stripping the paint and adding the adhesive at the bracket area ?? The adhesive will only be as strong as the paint adhesion if you leave it on there... Why don't you just :
A}run a self tapping screw into the "rosette weld" holes like the factory pieces had or
B}Bolt straight through the frame/brackets

Fish a spacer in there so the rail does not crush down under the bolt.

Yes grind and adhere. It would keep the inside of the frame coating that I sprayed intact. I also don't want to worry about melting much else, slag, etc.



That's certainly an option. I wasn't aware the factory mounts were bolted. Was the frame slugged? I can machine some spacers on a lathe quickly to keep the frame from crushing.

I'd be inclined to either use nutserts in the rails, or through-bolt them.
If you use Nylock nuts, you don't necessarily have to crush the rails. Or drill the holes just enough oversize to slide in some tubing, through which you run the bolts.

Bolt or weld- if you use structural adhesive it will fail. Guaranteed.

If you ever intend to actually put to use the sway bars, either weld or through bolt the brackets onto the frame. The reason they started welding them on was because the bolts could pull through the frame. The adhesive doesn't stand a chance, those brackets are under a tremendous stress if you using the bars to their full extent. Gene

He is correct.

Through bolts with spacers is my vote.

Even with self locking nuts, the rails will collapse eventually and start to bang around and wallow out the bolt holes.

when using spacers, you only need to open the one side of the rail to insert the spacer. the opposite rail only needs the hole big enough for the bolt to pass through. make the length just slightly shorter than being flush with the rail [approximately .010-.025 should do].

If it was me, I would through bolt them. I really don't even think you need to worry about the frame crushing. Those brackets are likely thicker steel than the frame rail - so when you tighten the bolts, the bracket will take most of the compression load.

That adhesive is only used in production cars for things like gluing the hood support to the skins and similar applications.

I would just bite the bullet and weld them on. It's not like you'll be driving the thing in salt or parking it outside 24/7 where there would be water trapped in there.

that crap doesn't belong on older cars.
its for people who cannot weld..lol

If welding is absolutely off the table, I think RivNuts would be your best option - personally I would add 2 more into the bottom of the frame rail.

Just a tid bit. My boys new Dodge truck had a tree fall on it, and the roof needs replaced, and they are going to glue it on instead of welding. Also I used to own a Grumman airplane that was completely glued together.

The factory just ran a course thread set of self cutting thread bolts to retain the brackets, I generally thru bolt them in with grade 8 hardware as I like to sling my cars around the roads/tracks and give them a good trashing, it'll take a lot to collapse a "solid" frame rail by thru bolting them, bonded on with adhesive is not wise IMHO, nor is welding, they'll eventually shear the welds in any form of spirited course driving...

its ok to use it that way,
for parts that will experience stress like suspention parts
no flippin way,you are just inviting trouble

I love it when people say something will "never" work.

Structural parts of a chassis are certainly under constant loading, like a suspension part? No. However a sway bar, especially a rear, is not the highest stressed part down there, not by a long shot. If the factory was able to get away with self tapping course thread screws into a ~1/8" frame rail, that is quite telling of how the component is loaded. There are three forms of dampening before loads hit the chassis [axle mount, end link to bar bushing and endlink to chassis bushing], which helps its modal response.

The part being fully epoxied around it's perimeter with nice large radii [minimal stress risers] is ideal. If we were filling out a DFMEA on this, it would have quite a low RPN, easy to detect, minimal severity, and the occurrence is debatable. Meaning, it's not a significant risk to use [or try].

Open wheel racecars commonly run epoxied ends in pullrod/pushrod suspension applications [carbon pull rods] and they survive quite well.



Epoxied carbon axle shafts certainly exist as well. [Nissan GTR]



These parts are not drilled and tapped, they are held together strictly via epoxy and see significant more loading than a rear sway bar would ever see.

What about a drive shaft?

[quote=GoodysGotaCuda]I love it when people say something will "never" work.

Structural parts of a chassis are certainly under constant loading, like a suspension part? No. However a sway bar, especially a rear, is not the highest stressed part down there, not by a long shot. If the factory was able to get away with self tapping course thread screws into a ~1/8" frame rail, that is quite telling of how the component is loaded. There are three forms of dampening before loads hit the chassis [axle mount, end link to bar bushing and endlink to chassis bushing], which helps its modal response.

The part being fully epoxied around it's perimeter with nice large radii [minimal stress risers] is ideal. If we were filling out a DFMEA on this, it would have quite a low RPN, easy to detect, minimal severity, and the occurrence is debatable. Meaning, it's not a significant risk to use [or try].

Open wheel racecars commonly run epoxied ends in pullrod/pushrod suspension applications [carbon pull rods] and they survive quite well.



Epoxied carbon axle shafts certainly exist as well. [Nissan GTR]



These parts are not drilled and tapped, they are held together strictly via epoxy and see significant more loading than a rear sway bar would ever see.


Then epoxy it and let us know how it works.....

How do you think they put together a CF driveshaft?

Glue

Fancy glue, but glue nonetheless.

My MW driveshaft is glued, I better get some bolts in it.

I would if possible plate the brackets, as any bare metal is future source for corrosion and joint failure initiation. Additionally, the highest performance epoxies normally require a heat cycle cure, a luxury you don't have in this application. These guys have some tough (expensive) stuff that I have used.

http://www.aremco.com/

PS Pay attention to viscosity selection in your application, unless you can still rotate the chassis.

I will. Challenge accepted.

Well, so be it then but when do you think this bracket is going to let loose ??? Not going down the highway at 35 MPH...You'll probably be deep in a corner when it lets lose and hopefully ready for a ride...That experimental 50 dollar tube of epoxy isn't worth your build..

My money's on Goody

I have been in collision repair for 25 years and there is no way I would rely on any structural adhesive for this situation, whether that adhesive comes from Fusor, 3m, or Plio-Grip.

Do yourself a favor and weld/bolt it in, and if you are worried about inside the frame rail, buy a 360 degree wand and re spray the inside of the rail with epoxy primer.

PS, I have experience with 3M impact resistant structural adhesive and it is not everything it is cracked up to be. Do a test for yourself with some scrap metal.

Epoxied parts in a factory or manufacturing process is totally different than diy in the garage. The bonding of aircraft wings or drive shafts is done under controlled temperature, pressure, humidity etc. Read the disclaimer on the panel/structural adhesive products. They do not suggest using it in highly stressed structural applications.

Bullseye...

I've had ny-locked bolts holding OEM rear sway-bar brackets in place for over 20 years and 20,000 miles...

Zero issues...

While true, Formula SAE students do it all of the time in a standard shop without fancy pressure/temperature/NASA chambers, and they hold up quite well for axle shafts and pull/pushrods.

I keep coming back to the component in discussion, it's a 3/4" tube of steel with 18" or so levers on it. The loads are not that of an upper control arm, strut rod, tie rod, etc.



Valid point. The concern for a catastrophic failure could be mitigated by using a few rivnuts/nutserts with a fastener size smaller than the bracket would normally require. Putting those bolts in with no load on them would be a sufficient failsafe and it would clunk pretty harshly if the epoxy failed.

How about epoxy and self tappers?

Question to the OP, does your frame rails not have holes already drilled from the factory. My 71 B body did, even though a sway bar was not aviable on a 71 B body.

Are both rails drilled on your car???


Pass side only on mine...

My rales were drilled, but not the right size holes. I took them out the rest of the way. I went to a dealer, and bought the brackets, and hardware needed.

that is wussing out..

LMAO...


(I see why it says what it does under 'flypaper'...

"I keep coming back to the component in discussion, it's a 3/4" tube of steel with 18" or so levers on it. The loads are not that of an upper control arm, strut rod, tie rod, etc. "

Never paid much attention to an e body OEM rear sway bar, what are the load paths, ie are the brackets always/mostly under compression (the ideal for epoxy?) on the glue joint? And there are really no direct impact loads, all loads transmit thru the somewhat flexible 18"? lever arm and resisted by the length and diameter of the sway bar, against the opposite flexible leaf spring. These are not IMO, excessive loads, but enough to eventually cause failure, I think not, if done correctly. Push the envelope. And in case of a sudden failure, if at maximum load, ie 10/10ths cornering, with a stock dia bar, the result will be a surprise, and a change towards understeer handling, and a great story to share on moparts.

PS technically it is a bar of steel not a tube?

Prime the inside and outside of the bracket and plug weld the holes (A couple holes on all three sides should do it so drill more holes if you have to). Then clean the welds up and carefully back mask the bracket and paint it yellow with a Preval, or airbrush. No visible tape lines if done right.

Moral of this story is

When your only tool is a hammer (welder) then all your problems are nails (things that need welding).

Or

Old dogs cannot learn new tricks.

Or

"Cause that's the way we always dunnit".

But ultimately what it boils down to is this

"I dunno how a swaybar works, so I don't know how the forces are applied and therefore cannot comment cogently".

Sway bars in general typically have a few bushings in them. The bar mounts are a bushing of sort which will distort before transferring any load between sides. The end links, typically, on both ends have bushing/rubber as well. Loading will be in tension on one side and compression on the other during roll. In heave the bar is just along for the ride and rotates about the bar bushings.

This car really will not see 10/10 cornering at a high rate of speed, if I am pushing it, it'll be at an Autocross. Cones don't hurt too bad.


And yes, these bars are tubing, typical OEM is bar.





This does sum it up quite well. Change is scary and sway bars are wizardry.

I thought it was interesting that whomever did this Hot Rod article used epoxy for their front chassis stiffening components as well as the subframe connectors and most of the torque boxes.

http://www.hotrod.com/articles/dont-drop-hemi-4-chassis-mods/

During the production run of 3rd or fourth generation Camaros, the tooling was getting old and some parts turned out to be pretty flexible with just spotwelds. Spotwelds plus structural adhesive on the length opf the joints stiffened things up considerably.

Mention anything new and you ALWAYS get the doomsday sayers, "put it in the worst possible condition and it will fail." Green bearings are a great example. Yet, how many failures have actually been documented by Board members?

Those pieces see NO impact loads. First of all they are loaded in either tension or compression as they attach to the "frame". Second as Goody has mentioned there is a lot of rubber in the system. Third, if it DID fail "at 10/10ths in a corner' the suspension won't suddenly collapse, instead it'd allow more roll and probably put the car in an understeer condition, meaaning it would go straighter than it should. Adding more rear bar makes the rear end rotate out, oversteer, but removing rear bar makes the car front end not turn in enough. Not at all like the Jeep I saw on Interstate 70 yesterday lying on its side, one rear wheel sheared completely off.

In the same way I'd glue and screw a wood joint, I'd suggest gluing and screwing the bracket. The screws hold it together and aid in the clamping. The glue prevents the movement that "wallows out the holes".

R.

Unless you assembled the whole car with ever last accessory, you will likely run into this issue again of needing to modify an area that was pre-painted. Especially with the amount of fabrication in your build.

Whether or not the adhesive will work, its the bottom of the car. Install and weld as Hotchkis suggests. Use the opportunity to learn to blend new paint/clear into the old. You're a pretty creative person, it shouldn't be much for you to figure a paint blend. Ans its the bottom of a car, no one is going to see this.

its not the location as some would love to think


you gotta love all these experts
about a product they have never touched in their lives!
the one single person who responded in this thread with FIRST hand experience
said not to use it for the application.
that holds more weight to me then some stupid magazine article
and if you are screwing and gluing
that tells me you don't trust it yourself!

it costs a lot of money to build our rides
and i'm not rich enough to gamble it on the word of some stupid magazine article
that does no follow up on failures or has any clue of the longevity of this product or has done any real world testing.
its not a matter of embracing new technology
its a matter of not gambling on it
and being the guinea pig until its proven for the use.
at least this makes for some good winter reading
if nothing else..

fixed it for ya...:)

he was going to do what he wants to do but posted here in hope to find someone to tell him what he wanted to hear.

Does the OP have access to a welder? What's the problem with welding them? It's the bottom of the car. Weld them on, squirt some left over paint on them and call it good. I'm not sure what the big deal is here.

exactly.
and when he didn't get the answer he wanted
we get mocked as being old school
and not embracing technology..

Mocked?

Nope.

Just the facts.

Even the factory didn't weld them in. Why should he?

a few bolts would be just too easy..
then you have to think also
what if he wanted to take them off and move them in the future??

Well, I don't think I'd glue them on either, mostly because I have no hands on with it. He has experience with it and knows what to use to do it. So why not?

If no one ever did anything new we'd still be slime in a pond somewhere.

not taking sides here but i would think his car would stick to the road a lot better when he is finished with it then ever imagined when new.

Op is an engineer, lets crunch some numbers. Assuming just static loads for arguments sake, as in gigo. We know the lever arm length, we know the diameter (still puzzled over a solid bar vs a hallow bar ie wall thickness , we know the effective length, we know? maximum travel, we know in use temp of bond, we know/measure the glued surface area, both loaded in shear and tensile, some HP epoxies can achieve 14K psi tensile, and we can assume a perfect bond for starts, curious what the back of the napkin loading would be.

Before you sail off into the wild blue yonder, what's the loading in a stock setup?

No side loading to speak of.

No fore an aft loading to speak of.

No vertical loading to speak of.

No horizontal loading to speak of.

All the loading in a properly working sway bar setup is torsional and at the ends where they connect to the rear axle.

If there was enough loading at the frame mounts to stress a bond then the rubber bushings would be hammered out long before it stressed the bond.

The loading vertically will be equal to the load and direction seen at the end of each link I believe.

Think of our TB's up front, ie why a K member needs to be so beefy.

A couple things:

1) Goody, I think I'd strip any zinc plating off the inside of the bracket and get it down to a good prepped bare steel. I'm sure the bracket plating is good but I'd feel better with a properly prepped bare steel for the epoxy to grab.


2) Shut the naysayers up. Get me some of that adhesive. I'll load test it (instrumented to 10,000 lbs, max capacity 40,000 lbs) in the shop and post a video of the results. An incremental load test is quite different than an impact test but that is what I have available and will show a comparison of strength.

Aren't joints glued together stronger than the metal itself? Failure of the glued joint was away from the joint itself, somewhere in the base material.

And as said, it's sway bar mounts. If it fails, you won't die. You will learn something and have the opportunity to make it better. If you came here wanting to glue on your wheels and not use lug nuts, I'd call you a special kind of stupid.

The glue probably is or at least could be stronger than the base material...and I would consider that to be a requirement for this application.

I'm an mechanical engineer myself, and I used to work at a company that made the "body shops" where they did all the welding for OEM car sheetmetal. This was only about 6 years ago, we weren't seeing any of this other than for skins, hence my reply. So yeah, I don't deny that it could work. Is it cost effective? Probably not for a production car.

I'd also ask the same question here. Considering I already welded on the same type of brackets to my car with a really basic MIG welder in about 45 minutes and I bet I used less than $2 worth of wire and gas....and I know I can trust it...and if my weld cracked I could very easily repair it. Not sure what you'd do with adhesive with a partial failure.

The only issue I could really see is if you didn't get it clean enough for a good bond. So go ahead and try it, worst thing that could happen is it fails and causes understeer. Sub-100 RPN anyway. lol

He asked.
We responded, maybe not everything he wanted to hear.
He responded with his reasoning.
We debated, and still are. (Does that really surprise anyone that has been around Moparts?)
He will do what he was probably going to do in the first place.

So now, the question is, how much time must pass before this gets the stamp of success or failure? 6 months? A year? 5 years? His (or our) lifetime?

When do we get to say "I told you so!" or "Gee, I guess that really did work." Gene

4 years seems about right to me.

100,000 miles

Gene I would think sooner than latter...He gets his out...drives it...and even plans on abusing it a little...
That beats the hell out of some of the show poodles around here...

A lot of good stuff here.

My question to add to the mix is how different is today's steel for which that glue was designed for in comarison to the old school metal.

Most of the new steel in cars I have seen is thinner and less ductile than the old stuff. Less flex in the new stuff may lead to better long term adhesion.

The factory just had four 3/8 fine thread self tappers per side. IIRC So I would try it, and beat the snot out of it in a flat (skidpad like) arena several times before I got it in a curb that could be devastating if it let go.

I ran into a similar dilemma, not a sway bar though.
Underside of my car was fully painted, then as an afterthought I decided to install MP frame connectors. I didn't want to just bolt them in, but then if I welded them I'd have to do a lot of spot painting around a freshly detailed undercarriage. Similar to OP's situation.
I loosely bolted them in, traced around the mating surfaces & taped them off, & then removed the bars. Using a small strip disc I ground the mating surfaces to clean metal. I reinstalled them using 3M bonding adhesive and bolts. I figured, even "if" it's not as strong as welding it's gotta be stronger than just bolting them in alone. 6 years later they haven't budged.
Running a bead around the perimeter of the brackets isn't going to work, the adhesive isn't designed that way. You need clean metal against metal, and make sure all bare metal mating surfaces are covered with the adhesive or else you'll have a potential corrosion issue.
Yes, the tensile strength of the glue is supposed to be stronger than weld, that's why they do not recommend it to be used on any structural parts. Many cars these days are held together with glue & pop rivets.

I'd be a little leery of just gluing the sway bar brackets in, but combined with bolts it should be plenty strong.

30+ years auto body guy here. i am not a expert on the subject but i have taken several I-Car classes on adhesive bonding. i have also have attended 3M training classes on this stuff when it was introduced in the 1-1 mix and again when it went to the 3-1 mix. i have attended Norton classes on there panel adhesive also. the shop where i work uses this stuff almost daily.


it isn't recommended to be used in this way.
if it were my car i would weld it like the part manufacturer recommends.

now for a long winded rambling explanation on why.

the reason why carbon fiber drive shafts and other parts can be bonded is because of really good and consistent tolerances. this allows to keep the adhesive thin and even thickness thru the entire bond area. this is one of the most important things to get maximum strength. also as others posted they will have a much more controlled environment to use it in.

if you bond a sway bar bracket keeping THIN and consistent even adhesive thickness will be difficult. to thick of adhesive weakens the bond.

as others stated bonded areas will have greater SHEAR, direct opposite forces, than welding. but it will have poor PEEL, like peeling a banana, strength. you can hit a bonded seam with air chisel and the panel will separate much easier than a welded one.
when bonding on thicker metal every time it flexes it will introduce peel forces. also when it flexes the panels will start to separate. when they start separating this will allow rust to form. rust will introduce peel forces also.
the reason why bonding works on sheet metal is because when sheet metal flexes both sides of the bond area will flex at the same rate.
this adhesive is affected be HEAT also. i am not exactly sure of the temp, if my memory is correct it's 400*-500*. when removing a bonded panel you need to warm it up with a torch the panel almost falls off.

with the above stated in my opinion you could bond this bracket on. if it's done in the proper way.
you would have to bond and use screws or bolts together. i would use self tapping screws like the factory did.
when you bond you should also bond the area where the bracket and the bottom of the frame rail meet. this will help cut down on flexing between the bracket and the frame rail. i would also use a couple of self tapping screws on the bottom also.
grind the entire area where the panels meet with 36, or coarser, grit disc to give the adhesive something to bite into. make sure you grind the inside of the bracket also. on the outer edges i would use at least 4 self tappers.

hope this info help.

After my years of experience in the highest volume body shop in the nation I'll disagree with the above comment.

We held certifications for aluminum and composite construction from McLaren, Bentley, Mercedes, Rolls Royce, Nissan (GTR specialist), Lotus, Tesla, and several others.

We bonded cars together on a fairly regular basis. I was the one who sourced the materials required for the job.

I can tell you that the adhesive used for body panels on a Tesla roadster is vastly different than the adhesive used on a Nissan GTR or McLaren. Bentley required different aluminum structural adhesive than Mercedes.

There are hundreds of automotive adhesives out there. They are each designed for a different use.

I would not use a typical 3M 8115 panel bond in this application but the adhesive Goody chose would be more than sufficient for this application.

We already know the easiest way is to just bolt them on. It's stronger than what the factory ever did and doesn't require any grinding or repainting. Four bolts with nylon lock nuts and washers. You'll be done in twenty minutes and won't ever have to give it a second thought. Any other option is just creating more work just to prove a point.

I have seen on this thread a number of comments here that the "glue" is/needs to be stronger then the material (steel in this case). Not sure I agree with that, or even if its achievable. Sounds like a theory held over from wood working. The OP's 3M data on his selected adhesive is far from stronger then any common steel, not that it really matters, because his bond surface area looks to be adequate for the loading expected.

The factory bolted them in

So I am confused.

Well, the factory used self tapping screws on my M that I replaced with bolts.

Point still stands.