Foam sandwich composite construction: Practical Guide

[UK-Corlett]

Hi All

I have posted "UK-Corlett New design New build" in March.
This is not my build diary but some details of what I have been playing with.

I want the build to be as light a possible and to curve on all sides. So just cutting the sides from 12mm ply is not on. (yes it will all be in metric, I guess on Tips and Trick you can cope)

I have chosen to skin in 4mm ply with a fine studding frame and outer skin in 4mm ply. I intend to shoot the gap with expanding foam. As I have no idea how to do this I have run some trials and found some things out which I wish to share.

First I got two strips of 4mm (actually more like 3.5mm) ply and filled the gap with polyurethane foam. This made a beam 45mm x 45mm x 1.5metre long. These photos show up close and tested.




Its simply supported over 1metre with 1kg in the middle and the defection was only 2mm. I feels really stiff and is less than 300g in weight.

This is Soudal Foam Gun Canisters 750mL tins. My rule of thumb is now 100g measured weight loss form the canister gun assembly makes 4000ml or foam. 1 litre required cavity requires a weight loss of 25g. Which costs £1 for 40litres of foam.

I then played with making a trial panel with the intended max curve. Its about 500mm sq and 50mm thick. It has an internal volume of 15 litres, which becomes important later


This is the finished panel shot with foam. I miss read the data and shot enough foam for a 30:1 expansion ratio but its actually 40:1. So I got quite a high internal pressure build up which bloated the panel. The timing is interesting. It took 5-10min to fill the panel but kept on growing for a approx 24 hrs. So it expanded 30:1 initially and then slowly to its full 40:1. NOTE to self, don't panic initially and put more in. Its immensely strong and solid but also amazingly light.


I then ripped it apart to see inside. This system will only work if the foam sticks to the inside of the skin. You can see that the bond strength is greater than the strength of the foam. The plywood failed and de-laminated trying to get the skin off. Also because the exit holes were not large enough some larger bubbles developed. NOTE to self, bigger exit holes.

I have tried two part polyurethane but it reacts too quickly is more expensive and expands less, but is much stronger. I will stick to what I have.

So on the whole encouraging with some more skill this could work. I hope you find some of this interesting.


Final test just for fun.


PS
Something I missed. When the foam is full hard (24hrs) you can push your finger into it, but only up to you nail root (say 10-13mm) and not further, it really is quite strong.



Clive

[noseoil]

This looks like a very good system for a structural panel which is light, strong & durable. Hard points could be any blocks inserted where fasteners are needed. It looks like the weakest point is just the foam itself, if there are too many voids & large air pockets trapped during expansion. Would vibrating the panel during curing affect the air bubble size with this system?

[lrrowe]

Yes this does look very interesting. I worked with similar foam in a shipping department and and familiar with the expansion qualities. I am going to build two groups of sandwiched walls for my CT. One section will enclose the shower area and the other to separate the galley section of the back of the trailer from the living area. I want to make these light weight walls instead of using 3/4 in plywood for the obvious weight saving reasons.

But I think I will stay with 3/4 in EPS foam and 1/8" paneling because it is simplier and I can add blocking where needed.

But please keep us posted on this great experiment.

[UK-Corlett]

Hi Noseoil

Thanks for that vibration! good idea.
However even with the air pockets and say an 80% filled panel its still super strong.

When I shot my first trial panel and got it completely and only put half enough foam and foam of a DIY quality and yet is was still very strong.

Hi Irrowe

You are correct this is not a new system, boat and small aircraft builders have been using it for years. Flat panels of paper foam paper /aluminium and wood can be got.
Its just may be new to the TD builders. Its the fact that you can add the strength after the shape that attracts me.

Clive

[UK-Corlett]

Hi

I have stared a light weigh constructed TD, which will curve on all sides
I have opted for 1/8”plywood- 36mm polyurethane foam - 1/8”plywood sandwich construction.

I have now started construction and have something to share. The basic construction is like a dry wall where the insulation is added after the wall is made. I get to shape the wall as I want and by injecting the foam later I insulation and enormous stiffness. However I knew I could be in trouble as my early experiments showed that you can get internal pressures which could distort the panel.
Its as though the foam continues to expand and its viscosity tends toward stiff. Thus it stops flowing while continuing to expand.
First go. These shots show the dry wall and foam injection. I know I know I put too much in.





It all went well however the wall in the flatter section did bulge. You cant see it but you can feel it. I think when the aluminium cladding goes on you would also see the bulge. So I eat humble pie and cut it out.




I was encouraged to see the quality of the foam. I was expecting a few big bubbles but the constancy we very homogeneous. BIG idea was to put less in and say in 3 or 4 goes and have more bigger exit holes. After re-work the bulge was significantly reduced.



Side two was done using the "only put a bit in at a time" technique and again although there is a little deformation you have to know its there and go look for it.



The roof was going to be a bigger problem the panel size is much bigger. I have experimented with bonding little towers to connect the outer skin to the inner. Although there looks like a lot of towers and a lot of holes it works out to be less than 4% of the total area, which also accounts for the low incident of towers clashing with holes (the sum of the added towers adds ~1mm to the thickness of the panel). This photo shows foam injected at the edge only, rest of the foam to follow. Good news is that there is no bulging on the inside or out.





I think the number of towers I used is ridiculous, As I become more experienced I think there will be an idea number density. Multiple foam shots at 1 hour intervals and bigger holes.

Thanks for listening, Clive.

[CanuckShooter]

Very interesting, how has it been progressing?

[RonS]

It all went well however the wall in the flatter section did bulge. You cant see it but you can feel it. I think when the aluminium cladding goes on you would also see the bulge. So I eat humble pie and cut it out.

I was watching a home improvement show one time and this is exactly whey they said NOT to attempt spray foam insulation in this manner. Either you don't get enough in to fill all the void, in which case you may as well have not done it at all, or you get too much and end up blowing out the walls....

[UK-Corlett]

Hi RonS

Sorry for the late reply I have been on holiday (Southern France) in my Teardrop. Its why we build them.


Agreed, there are significant technical issues. But because its difficult could be a challenge. Also the technique allows complex curves which can't be achieved unless you go 100% GRP. ( I am crap at GRP)


I did a lot of trials before attempting this technique and I would recommend anyone doing the same. You have to develop your personal skill by trial and error. I have long term test samples in the garden, shed etc. to monitor changes in density, bond, shrinkage. Some of these are 18 months old and showing no signs of deterioration.
I have concluded that you can have lots of holes with no strength lost in the panel. Also large holes 1-1/4" minimum. 1-2% holes is a lot of holes. Inject at many sites, and multiple visits. Wait till the first shot stops expanding before adding more foam. If you attempt to inject in one go you will be in serious trouble.
Also if you get a few voids it does not significantly reduce the strength of the panel (so far I have few voids) but 10% void is still not much in the whole panel.
Design the panel to be small, this way its much easier to monitor the fill.
Curved parts have inherent stiffness and are not prone to 'blowing'

Although I will be cladding in Aluminium I have elected to plug my fill holes this epoxy car body filler.

Thanks for the contribution.

PS
The build has significantly moved on so I should post to my build journal. Problem is I so much time building it and working to pay for it I don't get round to documenting it. These are the last photos I have, but it is more complete than this as one of the sides is in.