Controlling exotherm in thick solid laminate with infusion

I'd like to tell a story.

Some 20 years ago we were located in a manufacturing zone. We had one fiberglass manufacturer located in the eco zone. Their factory burned to the ground. With the money collected from the insurance, they built a brand new "modern" factory besides us. It burned to the ground 2 years later. With many pending orders (they are into mass production), they rented the building at the back of us. It burned also. 3 in a row. We used all our firefighting equipment to keep the fire spreading to our place.

The zone administrator became skeptical and asked us to do a forensic study because we said it was due to spontaneous combustion. But why during mornings only, about 10AM ish?

To demonstrate, we mix several pot of resin in a plastic container with varying degree of catalyst. The mixture with the most catalyst just heated up to the point where the mixture smoked and cracked. No fire but smoking hot.

We repeated the experiment using Metal Cans for the container. The last over catalyzed mixture heated, smoked, and suddenly, spontaneous combustion. We did it several times and had the same result. The metal can was adding to the exotherm of the resin causing thermal runaway.

So what was the cause of the fire? The worker were bumping up the catalyst to speed production and would take breaks while there is a hot batch of resin on the floor. The mixture will ignite, burn whatever is nearby and raze everything to the ground. There was nobody around. The shop floor is empty.

We had an added advantage. Our shop, though old, has fume extraction system with grated deep tunnels on the floor to catch heavy volatiles, and is temperature controlled in the main laminating area. We also do a lot of test on our laminations like Barcoll hardness and resin control. Overcatalyzing is a no no.

Steve, if you are worried about thermal runaway, you can do a similar batch experiment of about a square foot/meter replicating the same catalyst ratio, same thickness and glass content, same mold material you intend to use, and same ambient temperature. Using a thermal gun, see what peak exotherm you are getting. Keep fire extinguisher nearby. Be safe.

 

So, I have a dumb guy question.

I understand the concerns and that resin systems are a factor.

Could you use a water bag to temper the reaction or would that be worse than a fan blowing across the area?

It does seem like mitigation is possible. Certainly the heart of the reaction is where the greatest heat resides, but the mould would not allow heat to escape, so the heat leaves the mould on the top..

Just curious about mitigation..

 

Great question but how do you intend to use it? How do you reach equilibrium? As an example, a measured amount of water is used to absorb the heat in the engine block then fed to the radiator which is cooled by cold incoming air whereas air cooled engines have large radiating fins that is cooled by by an infinite supply of fresh cool air.

There are two zones that need to be controlled. The mold side and the laminate side. Metal molds absorb heat quickly and robs the resin mix of proper heat while the laminate side is radiating heat quickly due to its large surface area. This creates an thermal imbalance and creates warping in large flat laminates.

The most important thing is to follow the manufacturers recommended mix ratio to achieve the optimum mechanical properties of the resin. Most resin manufacturers will supply a data sheet that details the amount of chemicals to be mixed to reach a certain exotherm at different ambient temperature. There is a limited window to reach the optimum mechanical properties of the laminate. This can only be verified by using a Barcoll hardness tester.

But there is a lot of variables. The guys who were bumping up the resin were making relatively large thin laminates with some section in core. The large surface areas prevented overheating but disaster struck when they left a hot batch of resin in an enclosed metal container. The metal absorbed the heat but retained it much longer causing thermal runaway.

 

In most of the reading I did, the primary way to environmentally mitigate started with cool shop temps.

It is unlikely this will assist the OP, but I wondered after your story if those fires occurred in hot shops in summer. And perhaps OP knows he can start off at 55F vs 80F, or he can argue with mgmt about cooling the shop. 20-25 degrees may help when combined with the right resins

But I wondered if putting a 60 degree water bag over the reaction would mess up the reaction or not. Obviously, the water bag melting would be horrid, but we see water bags used in the absence of vacuum in tricky spots, so not like never been done. Or simply some fans..

 

Steve was worried about thermal runaway laminating thick in one go with an MDF mold. He has a reason to worry.

We had this experience. Our guy laminated a piece of about 20" diameter, 3/4" thick in one go. We had to make it in one shot because it has to be pressurized by adding a flat plate and weights on top after the last layer to meet the required resin content for the designed thickness. The laminate became so hot it cracked the mold. It is a manhole cover designed to be run over by a 24 ton truck. Cure was to laminate slowly.

Problem is Steve is infusing a thick laminate in one shot.

 

Its been done before. Pressure laminating with water and a rubber bladder. Proponent claims it is easier to pressurize water. Off the tap, he claims 30 psi.

 

Lots to think of. I have had a laminate get too hot and we cooled it down with snow. not something i want to repeat. One of the problems with this situation is that the mdf mold is a lot thicker than a splash mold would be so i will only be able to cool the laminate side where with a splash mold i could also blow a fan against the back side of the mold which would be helpful. With the mdf it is actually and insulator which is the opposite of what i would like. This is a keel section for a 42 ft powerboat and it is only about 2" wide at the bottom being much deeper at the aft end than the fwd end, its about 17ft long.. It is open at both ends so i will be able to set up a blower at one end and blow the length of it to take away some heat. I don't think exotherm will be much of a problem except in the bottom where it will be double the thickness where it laps so 20 plies x 36 oz/yd2. RX, i have done a lot of press molding with air bladders (snow boards) but not directly on the product. If we were directly on the product like some manufacturers did heated water could have been used. We have the mold together and coated with Duratec primer. I will be doing a vacuum integrity test next week and if it fails we will have to hand laminate so we could do it over a few days and excess exotherm would not be an issue.

 

If you are hand laminating; you could do it all in one day green on green with the right resin, or you could water bladder in thinner goes.

I'd love to get a few pictures posted here for edification of others for such a massive project.

 

I don't think you will have much of an exotherm problem with 8 mm thick. More of infusion problems.

VE as most say is the perfect infusion viscosity. There are other low viscosity resins but I think it is an overkill.

We had much success with slow cure formulation, low vac settings (20-24 in, 4 mm thick laminate), feed lines every meter apart. Full bulking materials on top.

The factory I have worked with, they would even cover the top of the very large flat laminating table with tarpaulin. Maybe to preserve heat.

 

Don't over think this.

There are many ways to accomplish the lower exotherm, the easiest is the catalyst route, it's a no Brainer for most applications.

You can build aluminum (any metal) tooling heated or unheated. The cost can be high, but you can customize it to your exact needs.

Fiberglass (any similar composite) can sort of do the same thing as metal, but typically not as efficiently. It can be heated or cooled In various ways to gain some of those benefits, but not to the same degree. Using aluminum pellets mixed into a slurry, with, or without water tubing to control the temperature is common.

Unheated metal is a heat sink, it can slow down production because your resin won't reach as high an exotherm, this could be good or bad. Fiberglass tooling is a much better insulator, so the exotherm of the part tends to be higher, again, this may be good or bad.

Many decades ago you couldn't buy blended catalysts, and cumene wasn't really talked about. We had production parts that were very thick and keeping temperature down was critical. At that time I had to make the blends myself, and once I started doing that I couldn't believe how good it worked, and that these blends hadn't taken over the market.

In this case the OP is going to use MCP75, it's a 25% cumene, 75% MEKP blend. You can get it as a 50:50 blend, or even straight cumene. These blends tend to take slightly longer to achieve a similar barcol as straight MEKP, but after about 24 hours the cure and barcol are normally better.

The blends work in gel coat too, but the cumene portion tends to yellow, so it's not really recommended for finished parts, but for tooling they're great. It's interesting that no testing has been done to determine if the yellowing is a real factor in gel coated parts, or just something that is easily see in the jug of material on its own.

 

I think the bigger problem is pulling a vacuum on MDF with just the Duratec as your only means of achieving that vacuum.

As long as the Duratec was applied at a sufficient thickness, it will hold a vacuum. The problem is the inherent weakness of MDF, without the proper bracing it will flex and move under vacuum, the Duratec can't move in the same way, so it may crack.

It works better on simple, or smaller MDF shapes, but on larger stuff keeping it from moving may be difficult.

The Duratec sealer helps in these situations because it soaks into the MDF and helps with the stability of the vacuum.

I'd say the exotherm is less of a concern than actually achieving a good vacuum at all. Even if the high peak temperatures are experienced, the resin is normally at point where it has gelled sufficiently to hold its shape even if the mold fails a bit and the vacuum is compromised.

 

Without seeing more about the manner in which the mould is constructed,I can't help concluding that hand laminating will be the simplest solution.A bit awkward to get into maybe,but the outcome is fairly assured.I think the OP has the experience to determine the best rate at which to proceed.

 

So, we did roll in 3 coats of Duratec sealer first until it stopped absorbing it, then did some fairing with polyester filler, then sprayed in the Duratec primer. We will not be sanding and buffing the primer because i don't want to risk going thru. It's the keel so it does not need to be like a topside finish, fair but not necessarily shiny and there will be fairing done after we glass the scarfs anyway. I will be doing a vacuum integrity test and if i'm not satisfied we will just hand laminate.

 

Seems like a very practical approach.If the two sides were joined,I'd be doubtful about the amount of coverage achieved near the bottom of the section.Before a vacuum test it might help to resin coat the outer surfaces of the MDF.I do have a couple of questions about the process of making the new section-will you be double gelling and will you apply a first laminate by hand prior to infusing?

 

WF,
I will not be using any in mold coating for several reasons. If i had done a regular splash mold like i wanted to i would have been guaranteed a vacuum tight mold in which case i would not have done a hand laid skin coat as infusion always ends up with a pinhole free surface which requires just a sand, and with VE or Epoxy there is no need for a barrier coat, so, straight to bottom paint. In this case though i will do a hand laid skin coat directly on the mold as an extra layer of resin rich vacuum barrier but it will be part of the part, not the mold. I'm sure there will be pin holes in the hand laid part no matter how well we roll it out, there always are on any boat i have ever worked on but they are behind the gelcoat and are small. In this case there will be no gelcoat so we will need to fill them but we will have filling and fairing to do anyway where we grind out and scarf the perimeter so we will be using a barrier coat before the bottom paint. By not using gelcoat we get to visually inspect the outside after we pull it from the mold and can fix any issues if need be. The mold is one piece and the glass will lap over one another in the bottom so we can work it into the corners to hopefully prevent bridging, the first feed line will be in the very bottom and i am planning on putting another one halfway up on each side even though my test piece showed it would draw the whole distance 25" from just the center line, i can make the decision on the fly whether to use them or not. Probably will.