Resistance of a carbon fiber mounting plate

[drodrig]

Hi there,

This is my first post in this forum, I'll try to be as clear as possible.

We are designing a mounting plate to place some kind of heavy stuff with screws. Then it is supported from back with some rods.

The original idea is using aluminium, but since it will be placed at -25ºC one needs to take the shrinking into account, which makes things more difficult. Plus, the plate is full of holes making the available space for rods more complicate (they should allow displacement for the temperature shrinking, complex mechanism).

Please, have a look to the attached pictures

https://www.dropbox.com/s/go96zldf9lio97x/front view.png

https://www.dropbox.com/s/8palgtbacsrokce/detail front.png

https://www.dropbox.com/s/fe0nqaom5mw79f5/side view.png

In the "front view" picture one can see all the big holes (for cables) and small ones for screws (every 4 is like a box screwed). The outer diameter is like 750mm.
In the "detail front" picture one can see better.
In the "side view" one can see a sketch of the idea. The purple support are the rods. The green boxes the weight to be taken by the mounting plate (like 500 Kg). There are like 60 of these boxes.

One colleague came up with a new idea; making the plate in carbon fiber. With different layers of thick prepeg (0.8 mm cured ply thickness) we could reach 15-20 mm; since it won't shrink with the low temperature, the support rods will be simpler.

This we could do ourselves, pressing with two plates and making vacuum with some bags.
Then we could machine it. It seems hard to machine but possible
Also we could make four quarters to decrease the size

We are not very experienced with carbon fiber, so I'm a bit afraid this won't work properly. Carbon fiber is not orthotropic and with so many holes many fibers will be cut. There will be a lot of shearing forces in the screws, also with the torque the head of the screws will focus much force in a small area.

What do you think? Is it possible to make a carbon fiber plate for this case?

Thank you in advance
Regards,

 

[cloa]

I get errors for the 1st and 3rd images.

 

[drodrig]

Hi there,

I don't know why it is not working, in any case here they are uploaded to another page:

http://www.subirimagenes.com/imagen-frontview-8388010.html

http://www.subirimagenes.com/imagen-detailfront-8388013.html

http://www.subirimagenes.com/imagen-sideview-8388015.html

thank you
regards,

 

[RPstress]

My network nanny won't let be see any of those. If you could get them on Engineering.com it would help.

General advice, for the least expensive sort of carbon with a good mix of fibre directions, so-called quasi-isotropic, almost like metal but E = 45 to 50 GPa and CTE maybe 3 to 5 µε/K. So it's low but still only something like 1/3 or 1/5 of steel (Al 23 to 24 of course). You could get a lot lower CTE with ordinary carbon with careful design of laminate and restraints.

 

[drodrig]

Strange about the pictures; here they go attached.

This quasi isotropic idea looks very nice; but what about the holes in the plate? We have many and I'm afraid it looses strength.

http://files.engineering.com/getfil...-43fa-9a03-c3cf5eac61ac&file=detail_front.png

http://files.engineering.com/getfil...f8-40dd-a0b2-8fe99565f7ea&file=front_view.png

http://files.engineering.com/getfil...f47-413f-b38a-85e518ebe16c&file=side_view.png

thank you
regards,

 

[RPstress]

Thanks. I can see them now.

If all those holes have rads of about one inch or more then very roughly the strength will be reduced by a factor of three over unholey material (net strength, you must allow for the missing material). Smaller rads actually give some relief (it's still best to have a rad rather than no rad).

The basic in-plane material strength would be roughly 500 MPa to 750 MPa or 70 ksi to 100 ksi, without any consideration of damage tolerance. The through thickness shear (interlaminar shear) might be 50 MPa/7 ksi of you're lucky. You'd have to account for the knock-down due to the holes as well. For rough work the stress concentration is often ignored with just the loss of area accounted for. Coming up with a size for Al alloy is actually a sensible first step. I suspect that a simple-ish FE model (plate elements) would be a way. Fastener holes are not usually modelled.

I assume you have a good idea of accelerations/shock loads. An allowable crush stress underneath fastener heads/tails including the through-thickness CTE would probably be about 250 MPa/35 ksi. This is quite cautious. Carefully applied crush stresses of over 800 MPa/120 ksi are achievable, but the edge of a washer will cause some local crushing at much lower stresses.

I don't know about inserts to take screws. Maybe someone else can comment on effectiveness. Through-penetration is usual for me but a lot of commercial inserts seem available.

Note that the through-thickness CTE will be about 25 to 30 µε/K (yes, ~10x in-plane quasi-iso CTE). Through-thickness E is roughly 8000 MPa/1 Msi (yes almost 1/10th in-plane quasi-iso E).

The cost of material is going to be very roughly $25/ib. Compare commercial Al, maybe $3/lb? Also the manufacture will cost more than machining Al. If you were to make a thick laminate and then machine it, it will cost more than machining steel (and use a lot of carbide cutters). Perhaps someone else can comment on practical methods to make a thick laminate with a lot of holes.

A possibility is a huge load of bulk-moulding compound with short carbon fibres. CTE will be more and strength less. You could presuamblly find a way to remove the material in the holes and reuse.

It may be cheaper to use Invar. It's maybe $12–15/lb for a limited quantity, as hard to machine as soft stainless steel (quite difficult; it's gummy) and about as strong as good Al (460/260 MPa UTS/yield or 67/38 ksi). CTE ~1.8 µε/K -200 to +200°F (-130 to 90°C). Sorry for mix of units. E ~145 GPa/21 Msi.

 

[mrsdesign]

Looks like it would be a good candidate for water jet cutting. There may be some tolerance issues as there tends to be a divergence of the stream with increasing thickness, but it may be an option.
You could also think about kitting material and using tooling blocks to mould in the main pockets in the laminate, leaving just the holes to machine.
As RPStress mentions, there are lots of good inserts for taking fasteners.

 

[RPstress]

Clean forgot about water jet cutting even though we used it a couple of years ago. Very effective even for 3-axis machines with some variation in thickness (quick and relatively cheap—just don't get in the way!). Thanks.