Stressted Panels

[MikeWaggoner at UW]

Anyone done a torsion test w/o panels, w/panels, and w/panels after a few months of driving? I'm asking because some of the uw panels seemed less well attached on the old car(s).

[MikeWaggoner at UW]

Anyone done a torsion test w/o panels, w/panels, and w/panels after a few months of driving? I'm asking because some of the uw panels seemed less well attached on the old car(s).

[flybywire]

As soon as enduro is over the warranty expires. http://fsae.com/groupee_common/emoticons/icon_wink.gif

[Frank]

lol @ flybywire

im wondering what are the prefered methods of attaching panels?

is it in a book somewhere?

would anyone care to give a "how to guide"?

[alfordda]

Try searching www.f2000.com (http://www.f2000.com). There is a thread on Hysol Adhesive, that discusses how to. I've never tried it, but if it works for the Formula Ford guys it should work for us too.

You could try this link too.

http://www.formularacingservices.com./Reynard_Techic.html

[Kevin Hall]

The Hysol E-20HP is awesome stuff. 24MPa on Aluminum. Thats plenty for most joints, and typically holds more than the panel itself can deliver. We have been working on some tests, and found that when we joined a section of honeycomb panel to a small space frame, the welds in the space frame gave way before anything even came close to moving in the composites.

[James Waltman]

Kevin,
Do you mean E-120HP? Its pretty remarkable stuff. We did extensive adhesives testing last year to choose what we would bond our whole chassis together with. We ended up choosing the E-120HP. They recommend Hysol 9430 in some of the links that alfordda posted. We tested that as well and it was good but averaged about 73% the shear strength of the E-120HP. I can confirm Kevin's claim of 24MPa (3500psi) - we were in that area with aluminum/aluminum and carbon fiber/aluminum (actually a little higher). The guys at Loctite have been very good to us. I think it worked because they got several of us hooked on their products. The static mixer and hand pump make it so easy.

We are also using a one part elevated temperature cure epoxy from 3M. It was also very impressive stuff. We bonded aluminum inserts into our A-arms and we tried to find out when the bond would fail. The results were similar to what Kevin reports. The bond didn't fail – we destroyed the tube first.
http://dot.etec.wwu.edu/fsae/images/A-arm%20test%203_jpg.jpg

Some of the guides that alfordda linked to also suggested that the rivets don't really contribute anything to the strength of the panel. I think this is wrong. It's true that the shear strength is nearly all from the epoxy. The big problem is that adhesives are horrible in peel. The rivets are pretty good at preventing peel.

Mike,
Are the panels at UW laminated on? I can't remember.

Are there other teams that are laminating composite panels onto the steel frame? I seem to remember that Toronto was doing this. Is that right Vinh? Does anyone want to share how it was done?

[MikeWaggoner at UW]

I'm pretty sure the UW glued 'em.

The problem with rivets is that they maintain a continual tension at the point of contact, so they induce creep, which are a problem with composites... I think it's still a good idea to hit the corners with them though, for fatigue/peel reasons.

[jerktaculus]

Aerospace epoxy - $$ - i know but it works and continues to work. I am not appreciating hysols E20HP these days cause they are not incredibly tough.

3M 3524 structural 2 part expoxy - it is mainly glass microbaloons in a really tough structural epoxy so it is amazing stuff. We usually use < 300 gm for the whole car. Try to find out of spec from composites specialists.

Jeremy Koudelka
University of Toronto Formula SAE

[Kevin Hall]

James

I mean E-20HP.....the number is the amount of workable time in minutes. They make a 20, a 60 and a 120. The guy at Loctite suggested the 20. I found that to be more than enough, unless you really want to be cheap with your mixing tips. The glue itself cost $100 for a case (500ml/16oz) and the mixing tips are $1 each, so not a big deal.

I should find my pics from our impact testing. We notched a 1" honeycomb panel, and folded it to about 40?, sandblasted the area for bonding, and bonded a strip of 20 guage 6061 along the joint (about 1" on either side of the cut). Our impact test was done with a 35lb driver and 1" ball nose. When dropped from 4ft (200J of energy) onto the crown of the joint, there was a nothing left of the honeycomb, but the joint remained intact. In comparison, a sample of honeycomb was dessimated by the same impact test, with the hammer blowing right through. The glue never showed any signs of cracking.

On a funnier note, there is a piece on concrete missing about 3ft from our shop door where a janitor peeled a quarter off that we had bonded in place after etching the surface with acetone and laying down a good bead.

Overall, I have never been so impressed with bonded joints. Make sure you clean your joints well. Even when a surface looks clean, you have to make it better. Sandblast, and have a mix of epoxy and acetone nearby. The mix cleans off the grit from sandblasting that you don't get when you blow it off. The eppoxy in the mix seals it so that the layer of oxidation doesn't form on the surface. A new surface holds 24MPa. An old surface (sandblasted and acetoned and sat for a week) on holds 10MPa, and then peels off the oxidation.

Good luck, and happy bonding.

PS - it can and will destroy small parts when pranking other groups around the shop. I recall a top for a fuel tank being very difficult to pick up one day and then being cut from a new sheet.

[James Waltman]

I am pretty familiar with the product line but I forgot about the fast curing stuff. We've used a lot of the E-120HP and a little of the E-60HP. In all, we tested nine different adhesives before we decided what we would use on the car. Once we made our decision we tested processing variables: cure temps, surface prep, mix ratio, etc. We wanted long work times because we were bonding our entire chassis together in one shot. We don't have any welds on our chassis – all bonding (except the damn steel roll bars). If we ran out of work time it would have been disastrous. I'm pretty sure that if we ran through the assembly process a few more times we could do the entire chassis with the H4500 Speedbonder (8-10min work time) and have plenty of time to spare.

We conducted one experiment with the H4500 to determine cure time variation due to the addition of external heat sources. This experiment involved filling the keyhole of a padlock with H4500 while the padlock was on a team member's locker. Rapid cure time was critical because the experiment had to be completed before the locker's owner returned (for obvious reasons). The results of this experiment showed that H4500 will cure in about 2 minutes if you use a heat gun to help it along. The lock was disabled.