Composite soak time over exposed at 250°F 4

[Cisnerjo1973]

Good morning Gentleman,

I have a problem that I would like your support with if possible in order to place a good rationale justification to accept some parts that were oven cure for a longer period of time than what the specification request.

Heres the situation we cured some parts on the oven, and the customer specification request to be cured at 260°F +/-20°F for 60-90 minutes unfortunately the recipe was set up to start the soak time once the lagging Thermocouple reach 250°F instead of the 240°F. so after the lagging TC pass the 240°F the load lasted 105 minutes instead of the maximum 90 minutes per Specification, we tested the test samples and they all came up on good condition they exceeded the customer minimum results required, but unfortunately this doesnt care to them. For my short expericed I know this condition wont hurt the properties of the parts, but they want me to write a good and rationale justification so they can accept those parts. Does any of you could please help me to write a formal and good rationale justification so I can provide to my customer?

Thanks a lot for your support.

 

[Cisnerjo1973]

SWComposites's
Compositepro

Is it possible that you could please help on this topic? I really appreciate

 

[Compositepro]

You did violate a process specification, so I guess it is not unreasonable to request some justification. Why was a max of 90 minutes in the specification? Obviously it was an arbitrary limit put in place because tolerances are expected on all process parameters. A more appropriate process tolerance would have been "minimum of 60 minutes" rather than "60-90" minutes. Remember this when writing process specs in the future. Always ask yourself "why is this limit needed?" for any written limit.

Thermoset resins cross-link though chemical reaction. Reactions take time and asymptotically approach 100% completion. They cannot go past 100%. There could possibly be other degradation reaction that slowly occur, but this would be very unlikely at a materials intended curing temperature, particularly since it is only 250F, in this case. Some materials could be designed to give certain properties at a certain degree of cure and may become brittle if over-cured. This would be pretty rare because then the material properties would not be very stable. In any case, the test of whether a material has be properly cured is whether its mechanical properties are with specification. You have done the testing to show that this is the case.

 

[Cisnerjo1973]

Thanks a lot for your response Compositepro:

The Specification request to Soak from 60 to 90 minutes at 260°F +/- 20°F once the lagging thermocouple has reach the minimum cure temperature which on this case should be 240°F but instead it was placed 250°F on the recipe, the recipe was set up to soak for 60 minutes once the lagging TC reach 250°F, these parts are cured along with three process test coupons which are tested for flamability , drum peel and hardness, which all three exceed minimum specification requirements.

As inmediate action the recipe was modified according to the Specification but as mentioned before would like a proper way to mention this over time exposure at soak temperature did not represent any adverse effect on the parts based on test coupons results, but also would like to know if theres something else that could guide me or help me to validate this?

Really appreciate your valuable help as always.

 

[SWComposites]

1) don’t expect an answer here in a few hours. We all have real jobs.
2) what is the part configuration? Laminate? Sandwich panel? Adhesive? What is the resin system?
3) flammability test wont tell you anything useful for this
4) peel test might be useful if its a sandwich panel
5) hardness test is probably not sensitive enough to be useful
6) suggest running Tg tests. Compression tests might are be useful depending on the part type and usage.

 

[Cisnerjo1973]

SWComposites (Aerospace) Below are a few answers : Really appreciate your support !

1) don’t expect an answer here in a few hours. We all have real jobs.
2) what is the part configuration? Laminate? Sandwich panel? Adhesive? What is the resin system?
This is a sandwich sidewall panel, PREPREG, GRAPHITE EPOXY, FILM ADHESIVE NB106HC
3) flammability test wont tell you anything useful for this...
Yes thanks !
4) peel test might be useful if its a sandwich panel
Yes it is a sandwich
5) hardness test is probably not sensitive enough to be useful
The panels have some laminate areas were the hardness readings were obtained besides the test coupon results the two panels were tested for hardness showing positive results beyond minimum requirements from applicable Specification
6) suggest running Tg tests. Compression tests might are be useful depending on the part type and usage.
I believed we don't have enough left over material from the test samples or the parts to get the test samples for compression test, do you know what would be a good sample dimension? and what results should be obtained?

 

[moon161]

Coming from medical device validation, I would ask, how was the original process specified and validated? The answers hopefully are 'based on component mfg recommendations and recorded tests'

The resin and fiber mfgs are probably a good place to start about process specs / recommendations.

If it was determined previously [how?] that the process [60-90 min soak @ 260 ± 20 ] produces a quality product, I think that validating a revised process [>60 min soak @ temp][the same way] with a range of times [60 min to as long as you care to].
You probably want to replicate whatever tests validated the original process because then you have direct comparison to the original validation but with a longer soak. If that's the case and customers/regulators accept it, then you can revise your process spec as supported by your testing.

 

[SWComposites]

Ok, sidewall panel should not be too critical structurally.

Most epoxies are insensitive to extended cure times, within reason.

Is this the first parts for the customer, or have you been making these for some time and the oven got mis-set this one time? If the latter, the can you make SPC plot of past process panel data and show this discrepant cure data is within the control limits, not just above the spec? Might give a better rationale. If you haven’t kept and plotted the data, well, no sympathy here. If the former case, were these FPQ or FAI parts? If so, just scrap them, learn a lesson, and repeat fab correctly.

Have you discussed with the customer’s M&P dept regarding what data they want to see?

 

[SWComposites]

Ok, sidewall panel should not be too critical structurally.

Most epoxies are insensitive to extended cure times, within reason.

Is this the first parts for the customer, or have you been making these for some time and the oven got mis-set this one time? If the latter, the can you make SPC plot of past process panel data and show this discrepant cure data is within the control limits, not just above the spec? Might give a better rationale. If you haven’t kept and plotted the data, well, no sympathy here. If the former case, were these FPQ or FAI parts? If so, just scrap them, learn a lesson, and repeat fab correctly.

Have you discussed with the customer’s M&P dept regarding what data they want to see?

 

[hpaircraft]

I am quite certain the parts are just fine.

But you did violate the customer specified procedure, which I assume was part of the original negotiation of the price and delivery, and they therefore have cause to reject them in the absence of a valid justification.

The most cost-effective thing might be to just scrap the parts. If they're large or expensive, you might go to the support engineers for the firm that supplied the prepreg and adhesive films and get them to write up a justification for variance. They might want some money for sticking their necks out, or they might do it just to support a loyal customer. You won't know unless you ask.

If it was me, I'd simply cancel the contract and tell the customer to go pound sand.