This page documents the carbon fiber successes and lessons learned while working at Artisan’s Asylum on 22A. The people involved in this process are Michael Chang, Jake Alpert, Sofia Varner, and Casey Sauer.

Test Layups Round 1

We did one test of the curvatures of the seat on a mini seat foam mold, and we did a test for bodywork with copper mesh strips and tiny bumps.

Day 1 (2/4/2024): Gel Coating

On the first day, we put the gel coat on the test piece foam molds. This was our first time going in the composites lab. Chang, Jake, and Sofia were present this day.

Open

3 sexy mfs ready for the apocalypse

What Went Well

• We were pleasantly surprised by the supplies and resources at Artisan’s. Everything was also very easy to use. They provided different sized popsicle sticks, gloves, safety glasses, aprons, scale, paper sheets, and plastic sheets. They also have their own vacuum pump.

  • Here is an email from Myra (the shop lead) with some info on what to use:

    Open

    

• There was a shelf there to leave stuff to cure (Myra needs to be emailed ahead of time to use this)

• We packed everything the evening before which meant we could leave campus faster.

• There was a nice little tutorial on how to use the spray booth.

• The spray booth had plenty of room to do the coating (for scale, the larger test piece is 15”x15”)

• The plastic in the roll they provide (next to the tool shelf) was nice to lay the gelled parts on since they did not stick to it like they did with the paper.

What didn’t Go Well

• THE FUMES. The gel coat fumes were way worse than the epoxy (from day 2).

  • Sofia and Jake’s respirators were fine, but the disposable ones we purchased were for particulates and not fumes. We need to order better ones for the future or make sure everyone has their own fumes one.

  • When we left, the room was still very smelly. We had left the parts in the spray booth for about 15 minutes, moved them to the curing shelves, and then ran the spray booth vacuum for a few extra minutes. We left the room quite smelly. Myra had one of the deskies go in to turn the vacuum back on, and when we got there the next day realized they had moved the parts from the shelf into the fume hood, left the vacuum running over night, and propped the door open with a shop fan in it to get fresh air in. We should be doing this ourselves for all gel coating in the future, but make sure to get Myra’s permission in advance. Also make sure to leave tags on the spray booth cover AND the vacuum switch saying to leave the vacuum on since a project with fumes is curing.

• We forgot to buy hardener for the gel coat which delayed our start date by a few days. The gel can be sold with the hardener or separately; we should be buying them together in the future. The hardener comes in teeny tiny bottles, but the shipping is still very expensive.

• Do not mix the gel/hardener in a plastic cup. The cup will dissolve. The plastic buckets were just fine, but these should be considered consumables since they are cheap and we do not want to have to wait for them to cure with the other parts. We were able to pour the hardener into the cup, but as soon as we started mixing them the cup started to dissolve.

• The dowels did not hold the foam pieces together. We also need something to go in between the layers of foam.

Other Info

• We did 3 coats of gel coat on each mold.

Day 2 (2/5/2024): Layup and Curing

We went for about 1.5 hrs to do our layups. The people present were Sofia, Chang, and Casey.

What Went Well

• Applying the mold release was chill. Would have been nice to have more foam rollers but we lived.

• We sealed the vacuum bag ends with the double sided sticky tape stuff. It worked okay, but there were some small leaks from resin getting onto the surface of the bag. It worked well enough while the vacuum was running.

• We just brought a strip of CF rather than bringing a whole roll. This was chill and easy to clean up after we laid it out on the table.

• Casey gave up on using the roller and just started using his gloves at some point for the seat test mold. This worked pretty well.

• The white stuff was nice for vacuum bagging. We did not bring enough.

• We put the vacuum pump in the spray booth with the vacuum bag and dilly dallied for a bit and had no issues/weren’t releasing the fumes from the vacuum into the space.

• It was nice to have 3 people so one person could work with the non contaminated stuff while the other two worked on the layups, but this could be done with just 2 people if needed (but it would still be good to have one person touching contaminated and stuff and one touching only non-contaminated stuff).

• The fumes were very chill. We were able to turn off the spray booth before we left and bring the vacuum bag back in Casey’s car.

What Did Not Go Well

• The coarse copper mesh was not usable for the tests since they are very loosely woven. Cutting a 1” strip just made the mesh fall apart.

• The mylar was a bit too thick and stiff to mold over the seat contours and stay. This could be a problem since last time we used thinner mylar, it shredded when we tried to get it off of the cured carbon fiber.

• We should not have put more than one part in the same vacuum bag. The parts just got sucked into each other which doesn’t look like it could be good.

• We did not give the parts enough time to sit with the vacuum running. The corners of the CF over the seat mold were still mushy, and air entered the vacuum bag immediately after we disconnected the vacuum (even when we plugged the hole with more sticky stuff). We probably should have had someone stay there for 3-4 hours to make sure nothing went wrong with the vacuum and had Casey pick them up after class.

Awaiting Results

• We started by pouring epoxy onto the molds, rolling it out, then smushing the carbon fiber onto it. We then rolled some more epoxy over the first layer and laid on 1” strips of two of the copper mesh samples.

Test 1 Results

…not good. The surface was not even and for both the seat and panel, and the mold was difficult or impossible to remove. There was an uneven distribution of resin across both parts, with some areas being fully saturated and some areas being bare cloth.

[ADD PICTURES. I DON’T HAVE ANY PICTURES]

Potential Errors

• Super leaky vacuum bag

  • Caused by resin accidentally wiped on edge of bag, making the tape ineffectual

  • Change gloves before bagging! Or have someone else bag it.

• Too strong of a vacuum

  • Would pull out too much resin

• Not vacuum bagging long enough

  • Not fully allowing the part to cure under ideal conditions

  • In combination with a very strong vacuum, could cause uneven distribution of resin

• Vacuum port placed directly over part

  • Sucked out all the resin on its immediate surroundings

  • Place vacuum port away from parts, but make sure there is breather fabric traversing from the part to the port

Test Layups Round 2

Day 1 (2/10/2024): Gel Coating

[ADD PICTURES. I ALSO DON’T HAVE ANY PICTURES. WE SHOULD GO TAKE PICTURES MORE.]

A second round of test layups was required. The molds were flat ~4”x6” foam panels. Fumes were mitigated with proper respirators and area ventilation (gel coating was done in the fume hood and a fan was placed at the door, blowing fresh air inside).

Because of the low temperature of the composites shop (didn’t have a thermometer but it was sweater weather fs) we were not able to cure the gel coat at the recommended 77°F. However, the instructions did recommend upping the gel:hardener ratio by up to 50%. We used two layers that consisted of 3 and 5% (as opposed to 1.5-2%) hardener by weight, respectively, and called it a day.

Day 2 (2/13/2024)

Gel coat was still tacky when we arrived. Waited a bit and it was still tacky. Decided to just send it with a few samples.

Four of the molds were used to test release methods with a varying combination of release wax, PVA release film, and mold sanding. [Jake idk what the exact samples were, can u write them down here]. Sanding the molds was a major pain in the ass and we should definitely get a gel spray gun to waste less of it and produce a smoother coat.

Resin:hardener ratio was messed up a bit and a little too much hardener was added. Popsicle sticks were used to spread out the resin onto the molds and worked very well. Last time we used too much resin and it was a mess, so this time we only applied resin to wet the mold. Pressing fabric on top of it saturated the fabric and gave it a slightly wet appearance. This resembled what layup tutorials on youtube looked like, so it must’ve been a good thing. Each sample got four layers.

The fume hood was definitely crowded with four people working simultaneously tho; in the future we should keep it to a maximum of two (2) ppl actively working on a layup.

We then vacuum bagged it (no resin this time, so good seal) and ran the vacuum at the lowest setting possible with the new regulator we got. We then let it sit for an hour, however, when we went to check on it, it had not set at all (the resin in the mixing bucket had not even begun to harden yet). Because the seal was good and there was not noticeable leakage from the bag, we brought it back, vacuumed it one last time, and left it on the bay shelf to pot.

Chang- I really think it’s because it was too cold. We can’t put the pieces into an environment far below recommended and expect it to pot as if in ideal conditions. Def should allot an entire day to let a sample piece sit to get an idea of how long it would take under Artisans conditions.

Things to change

• bring more gloves

  • Bring hand lotion too, the gloves dry ur hands out

• allot more time for vacuum bagging

• use spray gun to apply gel coats

• Figure out optimal vacuum setting

  • Youtube dude recommended 20% vacuum which equates to 80 kPa; not sure if that’s absolute or relative

• Figure out how to clean buckets

  • Not very economically friendly to throw out buckets at the end of every layup. Acetone or paint thinner should do the trick.

[bunch of testing occurred that nobody documented]

4/2/2024 - Seat and Nosecone Side Panel Layups

Instead of gel coat, the last of the mylar film was spray-tacked to the molds. For gentle contours, the mylar was cut to allow conformity to the curves. Parting wax was not applied on one of the side panels except for the masking tape applied to cover the seam in mylar. The vacuum bag was prepped to accommodate all three pieces and left in a box and the carbon fiber layers were cut. Because there was no more mylar, a bagging setup consisting of the mold, nylon peel ply, release film, and then breather.

Layups did not happen on 4/1/24 as intended due to the resin looking like slush. It was very opaque, granular, and hard to stir. Application of a heat gun seemed to help, although the integrity of the resin was still at question. Therefore, a very simple test layup was conducted without bagging and left to sit overnight.

The next day the test layup was still tacky on the exposed surface, but clear and fully cured on the surface in contact with the mold (presumably because we did not bag it).

We decided to go ahead with the layups- bodywork did not need structural strength and seat was very easy to remake if done wrong. Layups were too big to take place in the fume hood or table, so we flattened a cardboard box and did it on the floor. The roller and wide paintbrush worked very well at spreading the resin; the roller and stippling of the brush worked well to drive the resin into layers. However, the layups took longer than anticipated simply due to the size of the parts, side panels especially. We were rushed toward the end because we were worried the resin would cure as it had been longer than an hour since it had mixed, even considering the cold temperatures in the shop.

The scissors available were kitchen scissors and weren’t effective at cutting the nylon peel ply. Fabric shears would be nice. In order to get the massive vacuum bag open enough to slide the pieces in, a small fan was turned onto a low setting and used to inflate the bag, then turned off and the pieces slid in before the bag deflated. The scramble to get everything into the bag resulted in lack of time to ensure the fabric was not rolled up at the edges- seat and one of the side panels were fine, but the second side panel had the edges peeled up. There was also a big leak in the seal that could not be fixed; this was likely due to the tall seat layup being placed at the opening of the bag, distorting it and making sealing difficult. In the future, arrangement of the pieces should be considered carefully, and two bags used if necessary.

Layup + vacuum was ubered back to campus; the bag was vacuumed one last time and then left on the main bay table (unfortunately it was too big to fit anywhere else and there were concerns about setting it on the outside makerspace).

The next day the parts were unbagged. The vacuum port was unfortunately thrown out with the rest of the bag. The peel ply came away fairly easily, and had left an opaque finish with a nylon fabric texture finish- this finish is ready for further bonding, such as paint, but was sanded off in preparation for a repair gelcoat layer anyways. The edges of the bodywork were not complete, likely due to the rush, and the copper was broken in several areas. The seat had a big seam in it from an unknown source, and a similar finish from the peel ply.

4/4/2024 Gel Coat Repairs

Gel coat repairs were conducted as advised on the product sheet. To prevent tackiness and also ensure a smooth finish, the gel coat was brushed on with a paintbrush and roller, taking care to fill in the big seam in the center, and then a film of mylar placed over to prevent a reaction with the air that would cause tackiness, as well as flattening out the gel coat.

In the future, usage of a hi-gloss additive as per the product sheet will make spraying possible. It also allows for air curing.