RPstress
Aerospace
- Jun 4, 2003
- 846
We have an automotive part with high volume fraction (about 68%) of UD IM carbon and a service temperature of 220°C (430°F). It is currently all 0° fibres in a ring about 300 mm across.
Being automotive it's quite sensitive to cost and needs a fairly low cycle time. So, a lot of the usual resins we would use in aerospace are ruled out (BMI, cyanate ester, PI). At the moment a two-part epoxy is used, but the long, stepped cure makes it too expensive and its Tg is too low.
In addition it sees significant interlaminar shear at the max temperature, and needs an ILSS of about 100 MPa (14.5 ksi). We could change geometry to reduce the shear a bit but 75 MPa would be awkward and 50 MPa would be pretty much infeasible; it might be a goer but the size increase would make 50 MPa a very poor choice. So, a low porosity is desirable but not essential as long as it doesn’t compromise the strength too much.
On the plus side it doesn't need to be tough and damage tolerant and doesn't need resistance to the usual chemicals, 'just' fairly standard automotive lubricating oils. Also it doesn’t matter if it’s prepreg, infused, wet filament wound, dry wound, microwave cured, etc., (braiding might work but we don't think a non-0° UD layup would be useful), and one part, two part resin doesn’t matter. In fact a thermoplastic could be used and it doesn’t need to have good bonding characteristics except to the fibres. The fibres will probably need to be continuous and circumferencial, but you never know, short fibre filled resin mght not add too much weight. However, we do need that temperature resistance and a shear strength of 50 MPa would be very problematic. Also, it’s concealed and so effectively can’t be seen, so the ugliest brown phenolic would work if only it was strong enough (and an orange benzoxazine is ok). Also it doesn’t need to have a low coefficient of thermal expansion so something like a polyester might be ok in spite of (rumours of) misbehaviour on cure.
A reasonably fast cure, temperature resistant, epoxy would work well. The right benzoxazine might be very appropriate. Anyone know of any possible candidates that we could look into?
PS: I said that BMI, cyanate ester and PI were ruled out; that's only true if they are indeed too much money. My cost data is often approximate and at least two years out of date, but at the moment does indicate that those resins are a bit too dear. Each ring will weigh probably about 250 g/10 oz (depends on properties) and about 2000 or 3000 are forecast to be needed in 2014 and maybe 70000 by 2017.
I think that sums it up passably well. I haven't been able to share details of the usage. Apologies.
All suggestions welcome (most especially stuff we haven't thought of).
Being automotive it's quite sensitive to cost and needs a fairly low cycle time. So, a lot of the usual resins we would use in aerospace are ruled out (BMI, cyanate ester, PI). At the moment a two-part epoxy is used, but the long, stepped cure makes it too expensive and its Tg is too low.
In addition it sees significant interlaminar shear at the max temperature, and needs an ILSS of about 100 MPa (14.5 ksi). We could change geometry to reduce the shear a bit but 75 MPa would be awkward and 50 MPa would be pretty much infeasible; it might be a goer but the size increase would make 50 MPa a very poor choice. So, a low porosity is desirable but not essential as long as it doesn’t compromise the strength too much.
On the plus side it doesn't need to be tough and damage tolerant and doesn't need resistance to the usual chemicals, 'just' fairly standard automotive lubricating oils. Also it doesn’t matter if it’s prepreg, infused, wet filament wound, dry wound, microwave cured, etc., (braiding might work but we don't think a non-0° UD layup would be useful), and one part, two part resin doesn’t matter. In fact a thermoplastic could be used and it doesn’t need to have good bonding characteristics except to the fibres. The fibres will probably need to be continuous and circumferencial, but you never know, short fibre filled resin mght not add too much weight. However, we do need that temperature resistance and a shear strength of 50 MPa would be very problematic. Also, it’s concealed and so effectively can’t be seen, so the ugliest brown phenolic would work if only it was strong enough (and an orange benzoxazine is ok). Also it doesn’t need to have a low coefficient of thermal expansion so something like a polyester might be ok in spite of (rumours of) misbehaviour on cure.
A reasonably fast cure, temperature resistant, epoxy would work well. The right benzoxazine might be very appropriate. Anyone know of any possible candidates that we could look into?
PS: I said that BMI, cyanate ester and PI were ruled out; that's only true if they are indeed too much money. My cost data is often approximate and at least two years out of date, but at the moment does indicate that those resins are a bit too dear. Each ring will weigh probably about 250 g/10 oz (depends on properties) and about 2000 or 3000 are forecast to be needed in 2014 and maybe 70000 by 2017.
I think that sums it up passably well. I haven't been able to share details of the usage. Apologies.
All suggestions welcome (most especially stuff we haven't thought of).