Disbond Creating During Heat Blanket Repair
Disbond Creating During Heat Blanket Repair
(OP)
I am trying to understand the mechanisms for disbond creation during metallic sandwich panel bonded repairs, specifically new disbonds in the unrepaired area of the part in contact with a heat blanket. For this conversation lets talk 250F cure film adhesives with aluminum facesheets and core. I know an obvious cause could be excess unremoved moisture turning into steam during cure, but I am more interested in the mechanisms unique to heat blanket use.
Is the issue caused primarily by unequal thermal expansion, where the area under the heat blanket is fighting with the area directly adjacent? I know there is less concern using an autoclave versus heat blanket, and I assume in the autoclave the repaired part would have more equal thermal expansion.
Is the issue caused by actual degradation of adhesive due to heat during repair cure? Probably not the case if an approved SRM repair is being performed?
Am I not even close in understanding the issue?
Thanks
Is the issue caused primarily by unequal thermal expansion, where the area under the heat blanket is fighting with the area directly adjacent? I know there is less concern using an autoclave versus heat blanket, and I assume in the autoclave the repaired part would have more equal thermal expansion.
Is the issue caused by actual degradation of adhesive due to heat during repair cure? Probably not the case if an approved SRM repair is being performed?
Am I not even close in understanding the issue?
Thanks
Talk To Other Members
RE: Disbond Creating During Heat Blanket Repair
How are You determining where damage is located, IE: by X-ray, tap, sonic-bond-test, US testing, etc???
Also... Is the Assy-under-repair more the say 20-YO... and was there a problem with atmospheric moisture intrusion [X-ray]?I discovered that old-moisture usually has a particular musty smell. IF so, open area under repair and apply low-heat/vacuum for at least 8-hours... preferably 12-to-24-hrs. Core must be 100% dry prior to repair-process. See comments in the next item...
Are You cleaning the core prior to repair... IF SO are You using aqueous or solvent cleaners? There must be 100% confidence that cleaning agents have been removed. usually low heat 140F-to-160F + vacuum [-4-to-8-PSIG] for several hours. Solvents and moisture residues under heat and vacuum can spread damage effects... moisture-steam or solvent vapors drawn towards a vacuum-leak source in adjacent core-cells... or in some cases low-grade fire-explosion events... blowing-out adjacent core areas.
IF you submit x-ray images of before repair bonding... and after-bonding... showing core condition before/after... that would be helpful.
Regards, Wil Taylor
o Trust - But Verify!
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation, Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", HBA forum]
o Only fools and charlatans know everything and understand everything." -Anton Chekhov
RE: Disbond Creating During Heat Blanket Repair
Probably not the case if an approved SRM repair is being performed? > not necessarily. some SRM repairs were only validated on essentially new structure. adhesive can degrade over time, particularly if thermally exposed and/or thermal cycled. what is the SRM cure temperature? if 250F then risky for an original 250F cure adhesive.
How well is the heat blanket calibrated? have seen lots of cases of highly variable heat blanket temps.
How many control TCs on the repair? insufficient or incorrectly located TCs can lead to heat blanket excessive temperatures.
Was the panel dried after removing the damaged facesheet? are there drain holes/slots in the core so that moisture can be dried out in areas outside of the damaged area?
Is the issue caused primarily by unequal thermal expansion, where the area under the heat blanket is fighting with the area directly adjacent? possible. depends on the specific part and repair geometry and cure temp.
RE: Disbond Creating During Heat Blanket Repair
Old adhesive that has absorbed even small amounts of moisture may degrade the glass transition temperature... which is the upper limit that can be withstood with strength, during repairs with the same/similar epoxy adhesives. So the same adhesive at the same temperature... fresh/new is 'no problem'. But just a few years later [15-to-25] with microscopic amounts of absorbed moisture, the epoxy resistance to thermal + vacuum can be dramatically lower. in this case I have seen core blowouts around perfectly good repairs.
Regards, Wil Taylor
o Trust - But Verify!
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation, Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", HBA forum]
o Only fools and charlatans know everything and understand everything." -Anton Chekhov
RE: Disbond Creating During Heat Blanket Repair
RE: Disbond Creating During Heat Blanket Repair
Is the structure uniform, or are there changes in thickness (spars, ribs etc). within the structure? If so, where were the thermocouples located? Many SRM use generic layouts which totally ignore substructure. The configuration of heat sources and theermocouples can greatlt influence the risk of overheat damage.
Next, did you examine the failure surface? Dis the adhesive separate from the face sheet or did it separate from the core? Did the adhesive fracture in the process of separation? Each of these failure modes may have a different cause. For example, separation of the adhesive from the face sheet is typical of interfacial degradation on the metal surface which results in extremely weak bond strength, so failure would be inevitable, no matter how you heated the structure. Next, fracture of the adhesive layer in the fillet regions is indicative of internal pressure caused by moisture within the cells boiling off. Separation of the cells from the adhesive layer without fracture of the adhesive is caused by interfacial degradation on the surface of the core material, and again no matter how you heated the structure, this would happen because the bond strength has been compromised. Weak fillet bond failure is very difficult (impossible?) to find using NDT.
If you could provide a close-up photo of the failure surface I could comment further.
I hope these links work.
Here are some of my papers on bond failures:
Link
Link
Regards
Max
RE: Disbond Creating During Heat Blanket Repair
Just a further comment. Repairs using autoclave heating is not the panacea that may be commonly assumed because all of the structural panel is heated, so ANY moisture trapped in the panel will result in damage due to the steam pressure created. I have seen structural damage well away from the repair in focus at the time when a panel was globally heated in an oven. If I could work out how to copy images from s PPT file onto this site, I would show you the failure along the ribbon direction which was caused by steam pressure from moisture present in an adjacent repair, away from the actual repair being performed.
So the question must be asked: Did you perform a moisture evacuation prior to heating the repair? Remove the disbonded skin, then as stated, check the location of the locus of failure. If it is a skin-to-adhesive interfacial failure, seriously consider a new part because interfacial failure is not a local issue. The part is totally degrading. If the failure is by interfacial degradation on the core, you MUST continue removing core until the core no longer separates at the interface. Then remove the degraded core and perform a core insert repair AFTER you perform a moisture evacuation. Insert dummy core, cover with glass mat, apply a heater system and vacuum and heat to 180F for six hours. Then perform a core insert and then a patch repair.
Please don't use scuff-sand and solvent wip surface preparation. That is as effective as P$554sing on the repair.
Happy to talk further. Let me state that I managed to reduce the repeat failure rate on F-111 aircraft from 40% in 1992 to 0.06% up until when I retired in 2007 and in each case, the failures were due to technician malfunction, and the quality management systems I had created could identify where the technician took a short-cut.
I always advocated a singular testicle-ectomy. He won't do it again, will he?
Regards
Max