Poor results with polyester products

General Composites Discussions Composites Talk
#1

I’ve done virtually all my composites work over the years with epoxy based products; tooling epoxy mold surfaces, epoxy laminating resins and woven reinforcements. So I have a project on the bench that has no structural requirements, is fairly large surface area (compared to what I typically encounter) and would be well served being infused with a clear finish already applied in the mold surface. Since I didn’t need high accuracy in the mold or subsequent parts, I figured an inexpensive tooling gelcoat, polyester tooling laminating resin and woven+chopped strand glass would be sufficient. For the parts, I planned on clear gelcoat, polyester laminating resin and a variety of reinforcing fibers (infusion after the gelcoat set).

Mold construction went OK, but spraying the tooling gelcoat cost me an old/cheap HVLP gun as even catalyzed at the low end of the recommended range, it kicked before I could get the gun cleaned. So that pissed me off a bit, but I used a junk gun figuring this might happen.

Moving to part construction. The clear gelcoat from US composites is like a can of peanut butter. So I catalyzed it at the low end, thinned it with Styrene, and after about 4 minutes of very poor spray performance through one of my favorite HVLP guns, it started to set; quickly. I rushed to clean the gun and I’ll hopefully be able to salvage it.

So now, with a thin/incomplete layer of gelcoat in the mold, I ran to Harbor Freight for a couple cheap HVLP guns. Second batch of clear gelcoat was expected to be sacrificial over what was already in the mold. Sprayed a bit better with the HF gun, but after about 10 minutes; BAM, another gun headed for the trash barrel.

In all the years I’ve been working with composites, I’ve always fabricated my laminates with epoxy based products, then cleared them; generally with DuPont automotive clear. At this point I’m so pissed about the lack of control with some of these polyester based products, that I’m tempted to box up every bit I have and take them to the dump.

While I’m sure it will be pointed out that these challenges are more about my process and lack of expertise with these materials, it’s still frustrating that the chemistry has this many variables. I don’t like the idea that cure time is variable based on catalyst amount, and the volumes used are too small to be controlled with high enough accuracy to guarantee cure times. Give me epoxy any day over this stuff. Component quantities large enough enough to measure, the exact same ratio every time, small variation in gel time based only on ambient temps and surface to mass ratio in the container.

Sorry for wasting your time with my rant :slight_smile:

#2

You are simply using too much catalyst , I use the very same products and 2% will give about half hour work time. I build full size airplane molds and get excellent results and cheap.

#3

Yeah, I’ll need a much better way to measure/dispense these small volumes than “drops”. I think I was expecting a little more forgiving catalyst ratio. I was shooting for 1.25-1.5% due to high ambient temps, but can’t say with certainty that I was hitting that.

Might try to pick up an actual gelcoat cup gun before I have another lash at spraying that stuff.

#4

And buy yourself a catalyst measuring dispenser while your at it. Essential for what you are doing.

#5

You are absolutely correct. Your process is flawed and your expertise is lacking. With the right knowledge and tools you can build large molds with polyester. I’ve built hundreds of molds up to 40’ boat hull molds with polyester. If you are making relatively small molds, a cup gun is a good option. Using any gravity feed or siphon style spray gun is nothing but trouble. Fluid tips aren’t big enough and the gel is too thick to flow.

For larger molds, a pressure pot system a much better choice. I have a 2.5 gallon pot attached by a ten foot hose to a Divilbis spray gun. I never mix more than a gallon at a time. With the setup I have, I can spray the entire gallon in a little less than 6 minutes. I never go less than 1% hardener. If it’s really hot or the mold has lots of geometry, I’ll mix less gelcoat rather than reduce the cat level below 1%.

As a final note, off-the-shelf gelcoat may not have a long enough gel time for what you are doing. You need to know how long you’ve got to get the material out of the gun before gel occurs. If you can’t empty the gun quickly enough, mix less material.

#6

All good advice, thank you!

I knew I should have gone with a cup gun, but I wasn’t sure how much ongoing need I would have for it so I was hoping to get away with what I had for some prototyping.

Can anyone point me to measuring dispensers appropriate for what I’m doing?

Thanks again!

#7

Here’s a link to Bodi. Most any composite distributor will have these:
[URL=“https://www.bodico.com/collections/mixing-spraying/products/adjustable-volume-dispenser?variant=1032084683”]

#8

OK, I’m just about to throw in the towel on this project and cut my losses.

I abandoned the idea of gelcoat altogether to just focus on the other aspects of this laminate. So I prepare everything in the mold to infuse a test part. I’ll itemize the laminate stack in a minute. After about 3 minutes I can tell there’s a catastrophic problem. Do a little testing and find that the perforated release film I used is either defective or just plain unsuitable for infusion. The perforations are incomplete and the film is basically impermeable. Dammit! OK, well there’s a wasted attempt headed for the barrels.

So I get new film from Fibreglast; this stuff I know has large perforations and works fine.

Attempt number 2: infusion begins OK, but resin draw seems a bit slower than I’d hoped for, but still should be sufficient. 4 minutes in and resin basically stalls at less than 1 sq. ft. and I’m thinking “yep, this one’s a goner for sure too”. At 7 minutes I pull the plug and go to dump the resin cup. Sure enough, it’s gelled. Screw it, this isn’t worth the aggravation. I only needed 2 of these parts, they were cosmetic anyway and certainly not worth the thousand dollars wasted on top of the couple hundred I figured for 2 completed parts.

#9

So the materials that have been repeatedly failing me are as follows (I’ll post some pics shortly).

Mold surface = tooling gelcoat/fiberglass. about 3 sq. ft. total area. frekote release
4 layers 5.7oz/yd. twill weave carbon fabric (lightly held with airtac spray)
1 layer peelply extending beyond laminate edge and covering spiralwrap vac. tubing (which is touching edge of laminate stack)
1 layer perforated release film (holes 0.045" DIA on 1/4" centers)
green flow media to within 1" of part edge to allow resin brake
enkafusion channel under resin fitting (fitting uses 3/8" DIA tubing)
Full vacuum 29+" with negligible leakdown at 1hr. catchpot inline to pump
Resin = USComposites clear surfboard laminating resin
16oz. resin catalyzed at 1.0% MEKP (4.5mL dispensed with a 10mL graduated transfer pipet)
mixed 90 second in 32oz plastic cup

vac. channel forms horseshoe around mold flange with 2 - 1/4" DIA vac lines to pump

#10

oh, on the first go 'round I used the red flow media thinking I wanted to keep the resin front from getting too stretched as the resin penetrated the laminate. Trying to get the damn thing infused as fast as possible on attempt 2 I used the green flow media. I’ve never particularly liked the way either of these behave vs. the enkafusion blue stuff, but they are easier to position and hold in place on highly contoured surfaces

#11

I’m not a carbon or epoxy guy so others may chime in that are more expert on those materials. I’ve done a ton of infusion with polyester and vinylester and glass laminates. The three keys to infusion that I was taught are viscosity, permeability, and pressure.

You’ve got to have a low viscosity resin. Typical laminating polys are about 600-700 CPS. Infusion resin should be down in the 125-150 range.

Your material need to be permeable. Resin has to have a space to flow to. Most materials will allow some flow but when you are working against the clock, having material designed for infusion is a big plus.

I’m sure you already know about pressure, or more accurately pressure differential. If you didn’t, we wouldn’t be talking about infusion.

From the looks of things, I’d say your biggest problem is viscosity. Permeability may play a part as well but if you are not flowing across flow media then my guess is your resin is not suitable.

#12

Here are some photos of what I was working on.

I ran into an interesting situation on that last attempt and was able to actually salvage the effort with mostly luck.

The resin stalled about 4 inches from the edge of the enkafusion mat. I was a little puzzled by this (and still can’t confirm what happened). In desperation, I released the vacuum enough to poke a small hole in the bag at the edge of the stalled resin front. I snaked a length of spiral tubing attached to another feed line into the bag and around the edge of the resin saturation line. Pulled the vac. again and infused another batch of resin. This time it raced around the laminate and infused completely in a few minutes. Completely saturating the remaining laminate before the resin gelled.

I haven’t completely figured out why the difference in results, but theories include the original feed line somehow getting sealed off at the enkafusion mat. Whatever the issue was, the resin flowed and saturated the laminate so quickly and completely on the plan B infusion, I suspect viscosity and permeability may not have been the original problem (I had suspected that as well).

#13

So the part was a “success” in that I ended up getting it completely infused. I’ll make a few adjustments for the next try and see if I can get closer to an actual satisfactory result from end to end. Thanks for the tips!