Questions regarding large (200mm dia) CFRP tube

General Composites Discussions Composites Talk
#1

Hello everyone,

I have read a lot regarding different techniques and “how-to”'s on the internet. But I do still sit om my chair with some questions that seems pretty big for me. I will first tell a bit about the project, and then ask the questions below.

In my bachelor thesis this spring I’d like to engineer, build and test a Under-Water Scooter(DPV - Diver’s Propulsion Vehicle). Built to whitstand the pressure of 100m depth(approximately 10bars) and a lot of rough handling. It will consist mainly of two plugs, and one tube as a pressure hull. Plus one large cowl for protecting the propeller/diver. The inside of the tube must be very good, because of O-ring sealings. So I thought to use an old Dive cylinder as a mould. I also thought it’s smart to put it in the lathe and make a fine cut to make the surface as good as possible. Please look at the attached photos of some really bad CAD drawings(just for me to think out the solutions to the real CAD drawing and model).

Questions:
#1 How would you prepeare a mould that must be circular(ID:195-210mm) and have a length of approximately 4-500mm.

#2.0 Would you use prepreg, vacuum infusion or shrink tape?

#2.1 In case of prepreg: Is it then possible to implement some kevlar or CF/Kevlar to improve impact resistance of something that can get some abuse.

#2.2 In case of vacuum infusion or shrink tape: How can I start to lay the CF and kevlar and still get it tight enough, so it doesn’t buckle under vacuum or shrink? This is what I’m most worried about.

#3 Is it anything “bad” to use an excessive amount of woven CF rather than the right amount of uni-directional CF? I need a good safety factor.

#4 Will it actually help to implement some layers of kevlar to the construction to improve toughness?

#5 How many layers would you approximately use to get a wall thickness of 5mm? And would this be very over-kill?

#6 (sorry guys, many questions) Will it have any purpose to use uni-directional tape and build some ribs on the outer layer? To improve both stiffness and durability.

Thanks in advance for the answers! I promise that I will keep you guys updated. I just got material samples from Easy-composites, and I will make an experimental hull for a dive torch or something to get used to the materials.

Best regards,

navigas

#2

Forgot the pic’s of course…

-navigas

#3

Hello!

1)Just take an hydraulic piston of suitable size: it has to be absolutely smooth.

2)I would use prepreg tape. Because of the pressure from the outside, the 90° fibers have to be applied as smooth as possible. Any wrinkle is a good chance of local buckling. Use an amount of fibers for the 90° that is double compared to the 0° one. Symmetric laminate 90°/0°/90°. Make at least one intermediate debulk every mm of thickness.

2.2) Vacuum bag for the intermediate debulks and shrink wrapping for the final cure in oven.

2.1)/3) Kevlar is useless, especially in compression, unless you need break through resistance (I don’t think this is the case).

  1. Use only tape because it’s stiffer and easier to drape whithout wrinkles is you put a good tension while wrapping. Ideally the best way to wrap it is by filament winding.

5)The number of layer is given by the thickness of the prepreg you are going to use. The thickness required has to take into account any possible flaw into the laminate wrapping/compaction. 5mm. could be ok. but it’s all up to the quality of the laminate.

  1. Any rib is useless because the pressure is constant on the whole surface. Any discontinuity of the pressurized wall leads to a stress concentration.

  2. The cure have to have a peculiar ramp (apply as quickly as possible the heat necessary to shrink the tape, than, once shrunk, lower the temp. to the minimum temperature allowed for a cure with the resin system you are using). Then, dwell until the metal mandrel has reached the oven temperature. Now you can raise the temp. very slowly (0.2°C/min) up to the final cure temperature. Increase then the curing time to allow the inner fibers to reach such temperature time too.

  3. After the cure, let it cool, and after, put it in a freezer overnight. So that it will be easier to remove the mandrel. DONT’ USE ANYTHING BUT SEMIPERMANENT RELEASE AGENT!!

#4

Thank you very much Roberto! That just made my life a lot more easier!

Just to make it clear for me as a newbie:

intermediate debulk, do you mean that for each mm +/-, I should stop wrapping on new material, but vacuum bag it in room temp, and then continue with more tape?

Where can I buy prepreg tape that are of a good quality? I considered buyind from Easycomposites but, it seems they don’t offer any prepreg tape.

What is a peculiar ramp? If it is the process you explained in the following (), I understood.

And one last noobie question, semi release agent, what is special about that?

I do really appreciate your answer!

Take care! And excuse my english

-navigas

#5

  1. Yes!
    2)I have no idea of a Norvegian supplier. I buy the tape at http://www.delta-tech.it/

  2. OK
    4)Frekote 700NC for example. It doesn’t leave any residue on the surface like a wax or PVA could do, so it will make the slipping of the mandrel out of the tube easier.

  3. I’m Italian, so my English is worse than your :wink:

#6

Long time, no see.

After a while with a long design process and lots of hard work, I have finally made my first CF-part! And it’s quite large I think. There are a total of 14-15 layers, 1xcarbon/kevlar(for a brighter inner surface), 10xCF@200g/sqm, 2xCF/Kevlar and finally one layer with “pro” finish CF@200g/sqm.

As you can see on the pictures, there are some wrinkles and it is still on the mould. Just got the peel-ply off. I didn’t get the peel ply in a smooth way in the front, that was probably a mistaske. It will though be a hole with 135mm dia in the final product.

Is it something I can do with the surface finish where the peel-ply have been? I’d like a shiny surface for the presentation. I’m not shore if it is anything to do with the wrinkles, but it is a prototype product.

It is wet laid as I got some difficulties to get some of the products I would need for a prepreg or infusion.

Best regards,

Mathias Stabell

PS! The pictures

#7

Oh cool, Good work!

Yah i bet you learned a lot!

Wet layup is a pain… if you used pre preg you would have a much easier time in the layup. Still for being new at layup, looks like you have a part!

#8

I have learned a lot! And yes, I have a part. BUT it will not come off, so I will probably need to machine it off and do it all over again. But I will try every trick to get it off first, without destroying it.

Regards,

Mathias

#9

Crap… I didn’t catch that it stuck…

Do you have some draft on the cylinder? It would make things easier… or have a multiple piece mold to remove it.

Well, still a good forward movement.

#10

Nice project.

If the top hole was sealed / plugged before you built the tube , it would have been possible to tap the bottom hole and use a air fitting for compressed air to pop it off.

#11

or make a very loud bang ! :wink:

instead of cutting it off, id laminate carbon/kevlar tow to make a loop on the front of the cylinder, tig weld some loops on the back of the tube and use a car to drive it off …

as mentioned, leave it in the deep freeze over night to shrink the tube, that makes a massive difference

#12

FINALLY!! IT IS OFF!!! :smiley: :cool: (sorry for Caps)

I’m in heaven right now. It got off with a large bang and a freezing mold(it’s only 0 degrees Celsius here in Northern Norway right now), so we used a spray can of CRC -50 “freezer” and a hydraulic press(it came off at about 3-4Metric tonnes) so I’m really satisfied. The surface is very good. See for yourselves. :slight_smile:

Now is it just to get the outer surface as shiny as the inside.

Any tips to how I shall treat two small pinholes?

Again, please excuse my poor english.

Best regards,

Mathias