SUGARCOAT - DIY ablative parts in the home shop

Safety disclaimer: This is for educational purposes only. Phenolic resin is nasty. Please use PPE and common sense.

Why not graphite? #

I initially got this idea because I wanted to make my own nozzles without a machine shop. I knew about the Aerotech and CTI molded glass-phenolic nozzles, and I’ve been dreaming of having access to such material. Very important to mention that getting these shipped to me would be a nightmare, and customs would cost a fortune.

Graphite is a problematic material to have machined, because many businesses will refuse to make graphite parts for you in fear of damage to their machines. I was considering building a machine designed specifically to cut graphite nozzles, but the idea of molding parts seemed more versatile for my use case. I also just wanted to mess around with phenolic resin.

3D printing allows for creating complex mold geometries. A low-cost, single-use mold designed to be softened with hot air and peeled off, allows for very complex parts, that can withstand thousands of degrees.

The resin #

The resin is Aerodux-185, which I discovered through a video by Derek Honkawa, where he demonstrated its ablative properties in a composite layup. I was interested in trying to mold a part with it directly. It is a 2 part adhesive designed to glue wood sheets together. It is able to be cured in room temperature and ambient pressure, meaning an autoclave isn’t needed to cure it.

First attempts #

The first attempts at molding nozzles didn’t go well. The resin cured and yielded solid parts that indeed charred and remained hard under a blowtorch. However, a few days after curing, the parts would shrink and deform significantly. By sheer luck, one nozzle was in a usable condition and I fired it in an endburner configuration. The motor ran at around 1MPa for 10 seconds, and there was no erosion in the throat. This made me keep digging.

Developing the formula and process #

Countering the shinkage #

My first guess was to add a filler material, so that there’s less matrix that is causing the contraction problems. Cork has been used in some ablative heat shields, and I figured that its heat resistance and elasticity would solve my problems. Indeed, it did. The addition of 5% cork solved all contraction and deformation issues. However now the mix couldn’t be poured into a mold, it had to be packed and lightly pressed to fill all the gaps and ensure that the cork is incorporated into the part.

This method worked reliably, and delivered repeatable and usable parts. As a bonus, the cork drastically reduced the density of the material.

I also added 25% of graphite dust into the mix. Graphite can handle extremely high temperatures, and becomes part of the char layer formed on the surface, increasing its hardness. It also glows white hot, improving radiative cooling. I was also told that it can act as a “ballast”, meaning it can store heat and take it out of the system as the particles break off and fly away in the plume. And of course it acts as an opacifier.

Mold design #

I use single-use, thin-walled 3D printed molds. My motors are single-use composite-cased, so I don’t care about o-ring groves, the geometries are simple (but don’t have to be). Once cured, the molds are heated with a heat gun and peeled off. PLA is in my opinion the best material, because it has a low melting temperature making it easy to remove, and it seems to have releasing properties from this resin. PETG for example will chemically bond to the phenolic and leave a nasty finish (don’t trust this bond though, I found it to be very unreliable. If you want to bond these parts, use epoxy or another adhesive).

A 1mm wall thickness seems to work well. Fill the mold completely, then close it with light compression from something like a vise. Single-use, 2 piece printed mold.

Same thing, but for the aerospike nozzle.

Oven post-curing #

One of my materials professors suggested that I try heating the material to counter the deformations, which might be caused by non-uniform curing within the part. Heat can help create a more uniform and repeatable material. After curing for 12-24h at room temperature, the parts can be loaded into an oven and post-cured for 3-4 hours at 120C, while still in their molds. This will make the parts significantly harder, and remove nasty phenolic smells. While in the oven, the parts will shrink by 4 - 5%

Another thing I noticed is improved surface finish if the ovening happens while the part is in the mold. Aerodux-185 reacts with air, and perhaps keeping it isolated from the atmosphere during heat curing improves its properties. This is just a guess though, I haven’t experimented with this that much. What I do know, is that its much easier to demold from a PLA mold after an oven cure.