Joint design assistance required - PLEASE!

General Composites Discussions Engineering Talk
#1

Hello Composite Central! This is my first post, so be gentle! :stuck_out_tongue:

I am an enthusiastic amateur when it comes to composites. I have done some repairs/modifications to a boat using vinylester/tapes/Q-cells, and used bondo/bog on car repairs. But my next project is a little larger - like a 12m bus shell!

Due to space/weight constraints (not enough room for a 12m mould, and the panels would be very difficult to move/store/install) we have decided to manufacture approximately 30 individual panels - 10 for the roof and 20 for the walls - which will be joined and flushed when the panels are installed on the buses frame. Each panel is approximately 2.4m by 1.2m.

With some help from the supplier of our raw materials we plan to construct the panels from polyester resin, a single 3mm Coremat layer, 3 x 450g CSM layers and a 450g biaxial fabric/225g CSM combination fabric layer. We think we might have over specified the panels construction - what do you think? :wondering:

We have produced some one square metre panels using the materials mentioned above and did some very non-scientific impact and bending tests on, and were very impressed. For interest, the largest span between steel members in the buses frame is approx 500mm.

I understand that a lighter panel of equivalent strength could be constructed with vinylester or epoxy, but cost is more of a driving force than weight - and the weight of the metal “skin” we are replacing was substantial!

To enable the individual panels to be joined - we have access to the external surface ONLY - we came up with a multi-step recessed edge. I think a single stepped edge is a known as a joggle - I guess you might call this a triple joggle? We are planning on using epoxy to join/glue the panels together and, at this stage, at least a single layer each of 50mm, 100mm and 150mm tape. I say at least a single layer, because more layers might be used to end up with a close-to-flush final tape layer, to minimise the filler needed to flush the panels external surfaces.

Is this triple joggle overkill? Would a double joggle be sufficient? Perhaps using 100mm and 200mm tape?

I look forward to your replies and suggestions! :slight_smile:

#2

Hi Mark,

It is difficult to say if things are over-specified, with the info you give.

One of my customers is making busses, and he uses a much heavier laminate. I must admit that these busses do NOT have a chassis. Just the rings he makes, a front and a rear, a couple of wheels, and an engine.

For just panelling it seems rather heavy, I would suggest a thicker core, and less glass. The rear could even be without mat, just a biax 450.

As for the joint: You could joggle things, but also just taper the ends. You need body filler anyhow.

#3

Gday Herman,

Thank you for taking the time to reply. What other info would be required to work out if the panels are over-specified?

The bus that these panels are for has a space-frame chassis that is very rigid. The bus could easily be driven without an external skin fitted - if it didnt defeat the purpose of a motorhome! :wink:

The bus has a front and rear composite shells that we will be recycling to use with the new side and roof panels.

You mentioned a thicker core and less glass - Coremat is also available in 4mm - would this be thick enough? Or would you suggest another product, in a thicker section?

We thought a joggle joint would be best because of the simplicity of incorporating it in our mould. Our method would also allow easy repositioning of the mould pieces that form the joggle, in the event we need narrower/shorter sheets. How could we taper the ends of the mould simply and allow repositioning?

Thanks again for your help.

#4

OK, you got me on the joggle. They ARE easier to build.

As for specs: A picture of the bus, and panel size of the largest panel would do, basicly. For just enveloping a space frame, the lightest possible solution would do. Factors that are of more concern are vibration and impact resistance. Is it weight that you are trying to optimise?

#5

What angle should I take the photo from? I guess you are interested in seeing the distance between the steel framing to work out the unsupported span of the panels. The largest panels, as they will come from the mould, is 2.4m x 1.2m.

The finished height of the walls is approx 3.2m - 2.1m above floor level and 1.1m below in the storage bay area. The doors to the storage bays obviously wont be bonded to the wall panels. The surrounds around the wheel wells(in front and behind the storage bays) wont be bonded to the upper wall panels either - when tyres go then can easily destroy bodywork - so we want to be able to easily remove the damaged panels. The joins between the fixed wall panels and the removable wheel surrounds will be sealed using a flexible sealant. So there will only be a few areas where panels extend from the roof to bottom of the vehicle.

We are trying to optimise weight and keep costs down. Hard to to do two things at the same time! We dont want to produce bulletproof panels - just strong, reasonably flexible, light to medium weight panels, with good resistance to vibration. I imagine vibration resistance will be strongly linked to the method/product used to bond the panels to the frame. We plan on bonding with a Bostik Simson product, either 70-03 or 70-05, Silyl Modified Polymer adhesive/sealants designed for transport applications. Impact resistance is much more important on the front shell than the sides, and the existing front will be recycled, so thats not a huge issue in regards to the panels we are talking about.

#6

Just take one or 2 pictures of the bus, so that gives an idea of what we are talking about. Also one or 2 pictures of the steel frame that the panels will bond to is a good idea.

What is the maximum span between 2 steel frames, or in other words, what is the maximum unsupported panel size?

#7

I have a 4 photo composite image of the side of the bus, but I cant use the attachment feature here to post it. Nor can I create any albums - possibly because I am a new user?

Herman, could you PM me your email so I can send you the photo?

The largest unsupported panel size for the roof is 980x360. The largest unsupported panel size for the walls is 1300x430. Many/most of the spans are smaller than this. I could easily cut the max. wall spans in half by installing five extra 50x50mm tubes in the wall.

#8

hbrouwerjr at brandscomposiet dot nl

#9

here is a picture

#10

What is your exact goal? An ultra-smooth seamless side, or can you live with some seams?

You could just make a couple of sections, say 120cm long (most fabrics are that size anyhow) and make them seamed, or with your joggle make them seamless.

As for thickness: To me it seems that the thinnest option strengthwise should already be enough, now just add some material for impact resistance. The whole structure can be glued in place with MS Polymer caulk.

#11

Thanks to Herman for posting up the image I sent him. :slight_smile:

We are aiming for an ultra-smooth seamless side. I dont expect to achieve it - but we will work towards it. :wink:

Is Coremat a good choice for this application, or is there another core material (3-5mm thickness) suitable for use with polyester resin, of a similar finished laminate price? Perhaps Nidaplast/Nidacore/Polycore would be more suitable?

thinnest option strengthwise should already be enough, now just add some material for impact resistance

If a panel constructed of 2 layers of 600gsm CSM and a single layer of 3mm Coremat is sufficiently strong enough, what weight and specification material would you suggest to add impact resistance? A layer of 565gsm quadraxial stitched fabric?

Would we be better off using a layup of 852gsm quadraxial / 3mm Coremat / 852gsm quadraxial layer stitched to a 225gsm CSM (reduces print through I believe?) - more strength, less weight, less work?

I seem to be coming up with more questions - not less! :frowning:

#12

I personally would go for a 3mm core, and a 565 gr/m2 quadrax on both sides. On the exterior a (not stitched) CSM225 would take care of the majority of print. This exterior side would be your mould side.

Try and source sanding gelcoat, which will save you some hours of sanding. Do not gelcoat the joggles, but if you have the oppertunity apply peelply instead. Saves sanding again.

Fix your panels by bedding them in putty, and check fairness with a long straightedge. Push the panel further in the putty where needed.

Then after cure, remove the peelply from the edges, and fill them up with a strip of biax 450 gr/m2, and another (I would say 2 layers would be enough. Make the joggles such that after the 2 layers of biax450 (each some 0,5mm thick) you have another 0,5 left which you need to fill with sanding putty. Use longboards for sanding things smooth, in diagonal strokes. (and become familiar with the term “torture board”…

Now things are ready for primer and paint.

Keep in mind that for the fixing putty (MS Polymer) to be effective, it needs a certain thickness, say some 3-5mm.
Do some shopping on the MS Polymer. Bought palletwise it should not cost more than 3-4 euro (4-5 AUD) per cartridge. Sausages are cheaper, but get yourself a pneumatic caulk gun. (same goes for cartridges)