Prevost XLII Skin Delamination Repair

[cmillsap]

The first photo shown in David’s link above showing the jigs attached to the sides of each panel was interesting to me. I have always wondered why I could see 2 parallel lines in the SS panels about 3 inches away from both sides of the vertical edge of the panels. These lines separate and distorts any reflections or mirror images reflected by the polished ss panels. It appears these lines were made by the edges of the jig as it compressed the ss panel to the vertical forms behind. All panels on my bus display this condition. Looking straight on at the sides of the bus, those lines can’t be seen. You must view the bus sides from an acute angle and in certain lighting conditions to see them. I don’t know if this can be seen on any other bus or not. I do know that I have looked at many XL2 and XL3 models but have not been able to see these lines.

Now, I am wondering if this has had a positive effect on the fact that my 15 year-old bus has its original panels which to my knowledge have never been re-glued. I have searched the repair records from Prevost and Marathon and can’t find any evidence that the panels were ever re-glued.

Could it be that “over” compression of the panel edges to the forms by the jigs created the condition but improved the adhesive’s ability to remain adhered, I’m only guessing? I also don’t know if these same type jigs were used during the original assembly. But if anyone else has seen this condition on an XL2 or XL3, I’d be interested in knowing about it.

A photo showing the condition on my bus isn’t possible as my camera can’t pick up the distortion lines. But here’s the interesting photo of the jigs I mentioned earlier.

[davidbrady]

Buckling FEA results: Remarkably, adding ribs increases the buckling strength of the flat stainless steel panel by a factor of 7. Double remarkably, the buckling critical load of an H3 fiberglass panel is very close to the that of a ribbed stainless steel panel. Granted, I'm estimating the dimensions of the ribs; the actual size could put the buckling loads on par with each other. Notice how slight the buckling displacement is, it's in the hundredths of inches.

Here's the model. Forces are applied longitudinally on one end. The other end is constrained to resist the forces.



Flat stainless steel panel. Buckles at 748 lbs applied longitudinally:


Ribbed stainless steel panel. Buckles at 5201 lb, applied longitudinally, parallel to the ribs:


Flat Fiberglass panel. Buckles at 6830 lbs:


What does this exercise tell me? It says that differences in panel characteristics between the Fiberglass H3 panel and the ribbed stainless XLII panel aren't critical in the glue failure, and it tells me the glue stresses are low. The buckling strength of the panels puts an upper limit on the forces induced in the adhesive and that's approximately 200 lb/in. So if the bead is one inch wide, the shear stress in the glue is at a maximum of 200 psi, which is well within the elastic range of the usual adhesives, and is well below the endurance stress levels; i.e., the glue should last a good long time before failure.

----EDIT 5/6/17---
The stiffness of the SS panel is two times the stiffness of the FRP panel. Greater stiffness leads to more load supported which leads to greater stresses on the adhesive.

[Hisham Amaral]

Great job David, I love this program fast results. The results are at room temperature and not 100Deg F, when factoring in the high temperature the adhesive expands and thus failure can be induced.
Please look up the adhesive properties at 100 and 150 deg. F.

[davidbrady]

(05-04-2017 15:45)cmillsap Wrote:  Could it be that “over” compression of the panel edges to the forms by the jigs created the condition but improved the adhesive’s ability to remain adhered, I’m only guessing? I also don’t know if these same type jigs were used during the original assembly. But if anyone else has seen this condition on an XL2 or XL3, I’d be interested in knowing about it.

Hi Chuck,

I've seen those depressions on XLII's from time to time. The jig does predate the adhesive failures and fixes. If you go back to early versions of the CatBase or early versions of parts documentation you'll find that jig. I've seen it in documentation that came with SpongeBob which tells me it's a factory jig which turned out to be useful for panel repair.

Again, I couldn't agree more about applying the right pressure to the adhesive to ensure a good spread and to allow curing. Not too thick and not too thin. To thick and you risk curing issues and less bonding surface area; too thin and you risk venturing outside the elastic limits of the glue as the panel tugs at the interface.

(05-04-2017 22:57)Hisham Amaral Wrote:  Great job David, I love this program fast results. The results are at room temperature and not 100Deg F, when factoring in the high temperature the adhesive expands and thus failure can be induced.
Please look up the adhesive properties at 100 and 150 deg. F.

Thanks Hish. It was fun working thru this exercise. The attached paper, page 7, shows the results of 200 psi on a metal to metal 1" square joint adhered with MMA (Methyl Methacrylate like Loctite H8000) at 120 deg F, and 140 deg F. Their test was simply hanging a weight from the sample which actually turns out to be the worst case for adhesives. Fatigue loading is more pronounced as the frequency decreases - creep sets in. The results are good at low psi and higher temps as shown on page 7.

BTW, the results of the FEA agree remarkable with theory:
Fiberglass:
Theory: 6686 lb, FEA 6830 lb

Flat Stainless Steel:
Theory: 731 lb, FEA 748 lb

The ribbed stainless panel is too complex for theory; numerical methods are required.

[Hisham Amaral]

David
You are having too much fun with this, you have to remember that at .2%strain you are in the none linear ranger or start of the yield as you continue to strain the spacemen. Overall great results I was trying to add stress strain curve but was to taking the copy past. I am sure you can get the plots I am talking about. from the internet.

[davidbrady]

(05-05-2017 01:01)Hisham Amaral Wrote:  David
You are having too much fun with this, you have to remember that at .2%strain you are in the none linear ranger or start of the yield as you continue to strain the spacemen. Overall great results I was trying to add stress strain curve but was to taking the copy past. I am sure you can get the plots I am talking about. from the internet.

Good point Hish. Here's the strain percentage of MMA from the paper I posted. The paper also has plots:



Here are the displacements of the Ribbed Stainless and FRP panels at buckling:

Ribbed:


FRP:


The fiberglass panel deflects about twice as much as the stainless panel: .02" versus .01". I contend we don't get into buckling but I could be wrong. What are your thoughts? You know this stuff better than I do.


As an FYI:
Here's a close up picture of my fiberglass panel. You can see there's an extruded track that's bonded into the panel to hold the adhesive. This channel ensures a minimum depth of adhesive. IOWs, you can't squeeze out the adhesive by applying too much pressure.



Here's a pic of the side panel on Spongebob. This is the panel above the bay doors that corresponds to the stainless panels on the XLII:

[Hisham Amaral]

David
From the plots you show the strains are well below the .2% thus the parts are in a linear range.
Did you notice the pop rivet on the plastic laminate?
Again great results and thanks for sharing David. Like I said before lost of memory and gray hair are on the same path one losses color and the other I not sure what!!!!!

[davidbrady]

(05-05-2017 13:13)Hisham Amaral Wrote:  David
From the plots you show the strains are well below the .2% thus the parts are in a linear range.
Did you notice the pop rivet on the plastic laminate?
Again great results and thanks for sharing David. Like I said before lost of memory and gray hair are on the same path one losses color and the other I not sure what!!!!!

Thanks Hish,

One final takeaway from all of this is that the stainless steel (SS) panel is stiffer than the FRP used on the H3; the critical buckling loads are about the same but the SS panel deflects half of the FRP amount. This means there's more stress on the glue with the SS panel. If we assume the tubular spaceframe wall has the same stiffness as an FRP panel, then the FRP panel and the wall share the loads equally. Given that the SS panel is around twice as stiff as the FRP panel, in this same scenario the SS panel will support 2/3rds of the loads and the wall 1/3rd. Stresses in the glue get the same treatment. The stresses in the SS adhesive would be 33% greater than in the FRP adhesive.

[cattfuu]

Please understand that I know nothing, but we are about to take the plunge on our first 2002 XLII Prevost, and are shaking in our shoes.
Regarding this de-lamination, which manufacturing years have been mostly affected by it, and as most of the coaches during those years were built with slides, how has this failure compared with the few non-slide coaches produced during that time period? Could the additional pressure from retracting the slides, even though the door and windows were open, have put additional stress on the panels? Just trying to make sense of it all.

[davidbrady]

Hi Cattfuu,

Probably the best thing for you to do is to take any XLII that you're considering buying to a Prevost Service Center and have them examine the coach for panel separation. Only Prevost knows exactly what the issue is and why it occurred. In the years you're looking at panel delam is a reality. You asked, "which manufacturing years have been mostly affected"? I'd say, and only based on anecdotal evidence, the years from 2000 through 2010. It may go beyond that but reports are sketchy. No one but Prevost knows if slided coaches are more afflicted or if 45 footers are more afflicted than 40 footers, and so on. It's more than likely a process phenomenon related to panel preparation and adhesive application conditions and restrictions. There are several fixes that Prevost is offering, but I'm afraid there won't be any financial help on a coach over ten years old. If you can get Prevost's opinion on the bus you're considering purchasing and if you can get a quote from Prevost for the cost of repairs, that's a very good bargaining chip in negotiating purchase price. Good luck!