Epoxy Casted Heated Bed

[tjschweizer]

Just checking to see if anyone has ever made an epoxy casted heated bed. I've searched around, and haven't found much.

What I'm thinking is use a high temperature casting epoxy (Smooth-on has the Epoxacast 670 HT - says it has a heat deflection temp of 177°C) and encasing a silicon heating pad in the epoxy itself. The surface of the epoxy should be pretty self-leveling, providing a decently flat surface for printing on. This would eliminate the need for clamps (and possibly PEI, depending on how well printing on hardened epoxy works).

Has anyone tried this? I might just buy some epoxy and give it a shot.

[Glacian22]

I think you'd be the first, so you'll have to let us know how it goes ^^

[tjschweizer]

Will do. I just ordered some solid copper wire for my heater instead of a silicone pad, and some of that epoxy. Hopefully I get everything by next weekend!

[Xenocrates]

I don't think copper wire is a good idea. Nichrome would be a better choice, due to the higher resistivity. Also, make sure you cast it oversize so it can be machined down somewhat, as I doubt it will self-level without major thinning. I also suggest adding the copper metallic powder to the epoxy, to increase thermal conductivity.

[tjschweizer]

I was thinking about Nichrome wire, but I was concerned about localized hotspots causing issues inside the epoxy. I couldn't find much info on using Nichrome for lower temperatures (ie. 60°C) - although I'll be honest I didn't look for too long. I may try this on the next "prototype" version.

Based on my quick math, the Onyx Rev 8 heated bed has a resistance of ~1.1Ω, and handles the 12V input. That's approximately 10.9A max (although this varies with the output voltage of the power supply, obviously). 18 gauge wire has a rated amperage of somewhere between 10-16A (based on Wikipedia).

18 gauge copper wire has a resistivity of 20.95mΩ/m, so roughly 52.5m is necessary (approximately 172ft) to achieve the 1.1Ω. Using This spiral length calculator, a tape thickness slightly larger than the wire diameter (1.024mm dia., thickness 1.2mm), an external diameter of 280mm and internal diameter of 20mm gets about 51m. So basically I'll just make a spiral of copper wire that should provide very even and consistent heating.

I am curious to see how well the self-leveling will work, the mixed epoxy has a viscosity of 6000 centipoise, which is somewhere between mayonnaise and molasses. I might try using a vibrating table to encourage leveling. Otherwise I may have to machine it (or use a large glass plate to lap it)

Using copper powder is a great idea, I'll add that to the list of things I want to try!

I was also considering making removable "plugs" in the center of the bed to house the thermistor, as well as a thermal fuse if possible.

[tjschweizer]

Thought I'd throw in a quick update. I purchase some of the epoxy, and will hopefully make a mold today. If all goes well, I will have my first prototype bed done this weekend.
Some features I hope to put in this or the next prototype:

• Embed several thermistors in the bed, spaced relatively evenly. (First prototype)
• Embed a simple plug connector in the bed, so the power and thermistor cables can be unplugged with ease. (First prototype)
• Place a thermal safety fuse in a replaceable "plug"- in line with the heater (Second prototype)
• Include mounting/locating points for FSR leveling (Second prototype)

In the meantime, however, I have been thinking about the whole FSR probe situation. Since I switched from the accelerometer probe to an FSR system, my probing has improved significantly. I still there there is more room for improvement though. I do quite a bit of work with strain gauges, and I'm going to run a few simulations to see what kind of strain we are dealing with in the arms during a probe.

Speaking of which, does anyone happen to know what material the injection molded arms are made of?

[Glacian22]

I think they're some grade of glass filled nylon.

[geneb]

Yep, the arms are glass-filled nylon.

g.