Ok thank you

This is turning out to be a "Seemed like good idea at the time" situations

The stuff out of the can is not the same, it is a much lower density. If you're doing it for structural purposes, you need to be using higher density foam which behaves a bit differently. It does give you a slightly longer working time though, I know my GP6 has about 20-30 seconds longer working time than my GP330. However don't rely on being able to push the foam around, you can do this while the foam is still in it's cream phase (roughly 60 seconds for my GP6, depending on mix time) but once it begins to gel then a feathers touch causes the cells to collapse and the matrix falls in on itself. I would just pour the foam into each cell and then leave it to do its thing. If you get a couple of cells where the foam doesn't completely fill it, then top it up later once it has cured with another batch. 

That's what I thought lots of little batches. I could keep the A & B parts in the fridge to slow the reaction and give me a bit more time. I've used cheap PU foam in a can and found that it takes a little while for it to cure and in that time it can be pushed back in to shape. It becomes a lot denser if its squashed as its curing. As I'm sure you know cures with moisture so takes a little longer.

Neatness isn't too important as it will all be hidden behind the roof lining. How about laying a sheet of foil on a flat surface then mixing enough foam and spreading it quickly over the foil and just put the alu core on top and the foam will expand into all the cells?

Yeh PU foam isn't the most user friendly. But I think it could be done, in lots of batches. Just mix up 200ml or so, then pour it into the cells until you run out. Repeat until all the cells are filled. Wouldn't be tooooo difficult, would take a bit of time and a lot of cups trimming it neatly could be trickier than you think too. 

I hope you know that with expanding 2K foams you have like 15 sec to mix and pour and then it will expand like X10 so it is almost impossible I think to do that for all the cells

Matthieu Libeert
Founder MAT2 Composites X Sports
website: www.mat2composites.com

I thought I would use some sticky backed plastic on one side to seal the cells temporarily then use a 2 part pouring foam and just pour a little into each cell. After it had expanded the excess could be trimmed off.

Then the panel could be shaped to the curve of the roof sections and I would apply a couple of layers of glass mat and resin to the underside to sandwich the alu core and foam.

Then it would be bonded to the roof with either epoxy or polyester resin.

May well be alot of work but I have the time and it would be cheaper than a big pack of CLD and so much lighter

Do you have any idea how you will fill every little hole with expanding polyurethane? :p 

Matthieu Libeert
Founder MAT2 Composites X Sports
website: www.mat2composites.com

I'm well aware of Dynamat and Silent Coat and all the various methods of Constrained Layer Damping. All of which are expensive for what they are and also very heavy. The roof isn't going to have speakers in it and doesn't need to be treated the same as a door or floor pan where road noise or resonance from a high power speaker is the issue.

I believe my idea is cheaper and also much lighter. It would also provide insulation from hot/cold conditions.

What I need to know is if this idea will provide the benefits I require.

I think most people damp the roof rather than stiffen it.  From what I've seen a couple of layers of self adhesive dynamat is the method of choice although you can do a similar job with several coats of stonechip.  It will be significantly easier than adding composite structures.

I'm installing a lot of stereo equipment into my car. As well as making some custom enclosures for the speakers out of GRP, I also need to stiffen the roof.

My idea is to get some 10mm thick 3/4" Aluminium honeycomb and some polyurethane 2 part expanding foam to make some panels to bond to the roof.

Would this be a suitable way to add significant strength/stiffness to a metal panel. here's a pic of my car roof

Each of those section is 900mm x 300mm. Any advice or other ideas welcome.

Cheers