The PEEK Performance Project Part 1: High Temp on a College Student's Budget

This is the story of how I raised our household's electricity bill by about $30 per month.

Many people have heard of the material called PEEK. If you've been around 3D printing for a long time, you may remember that old hotends used to have a thermal isolator made of PEEK. People with industry experience outside of 3D printing may also know that PEEK is widely used in aerospace, automotive, industrial, and medical industries, being very desirable for its strength, chemical resistance, and thermal stability.

Somewhat fewer people know that PEEK itself can be used as an FDM/FFF 3D printing material. In fact, some consider it the "Holy Grail" of FDM/FFF 3D printing. They also know there are a few good reasons it's not so commonly printed:

• It takes hotend temperatures in excess of 400C and ambient temperatures in excess of 100C to print.
• It costs $500/kg on a good day.

Not too many people print parts that actually demand PEEK's properties - namely, its 250C continuous use temperature or its chemical resistance. You can find filaments like carbon fiber nylon or polycarbonate which are similarly strong, and for general purpose printing (including the very common "printing upgrade parts for the printer") even PLA is often more than enough. I myself had no use for its temperature resistance, and being a college student with a part time job already trying to sustain a very expensive cycling hobby, I really couldn't handle the material or printer cost of attempting to print PEEK.

And that's exactly why I attempted to print PEEK.

The goal: breaking it down

PEEK is extremely difficult to print. It may be the hardest filament to print that currently exists. Yet, similar to ABS or PC, its main challenges can be broken down into a surprisingly simple and concrete set of printer requirements:

1. 400C+ hotend temperature. We'll say the hotend should be capable of 450C max, to give a bit of headroom.
2. 160C+ bed. We'll say 200C, again for headroom. (As we'll see later on, turns out this wasn't actually enough, but I'll get into that.)
3. 100C+ ambients. Really, you just want to push ambient temps as high as you can. (Above 180C or so you can start to forgo the requirement of a heated bed, but for now we'll assume 180C is not reachable.)
4. Some bed adhesion solution that will make PEEK stick.

That's it! It doesn't take some weird blackmagic printer tuning or a hotend capable of going back in time. It's just a simple, concrete set of temperature requirements.

Of course, that's easier said than done. Let's take a look at what each requirement entails:

1. 450C hotend temperature.
   
   This one is fairly easy - lots of commercial solutions already exist to accomplish this. Slice's Mosquito does, as well as Dyze's DyzeEnd X, and both designed to handle these temperatures natively. Arguably the high temp variants of E3D hotends (the plated copper blocks and nozzles, as well as the Nozzle X) are not ideal for this task since they're basically the standard V6 shoehorned into the HT role - but they get the job done.
   
   So, all I have to do is spend some money, and maybe figure out a mounting solution. No big deal.
2. 200C Bed.
   
   There are two ways you could go with this one. One, take the easy and expensive way out and buy an E3D HT bed. Two, spend a bit less and build your own heated bed also capable of these temps.
   
   Being tight on money, I initially aimed to build my own bed, but it's useful to know that the commercial option was available as a fallback to me.
3. 100C ambient.
   
   Here we go.
   
   100C ambients are...not easy. Hell, 100C is enough to boil water!
   
   To even reach 100C ambients in the first place without overloading a US 15A household circuit, you're going to need a small and well insulated enclosure. That probably means enclosures for bedslinger styled machines - Prusas, Ender 3s, etc - are out, because they by necessity have to be large in order to clear the bed, and usually to encapsulate the whole printer as well. So, ideally, this would be a box-framed printer where you could enclose just the printed volume.
   
   At 100C, lots of other problems start popping up. PLA and PETG printed parts are out, and even ABS is going to have trouble (especially if there are local hotspots). Genuine Gates belts are only rated to 80C. Most fans are only rated to 70C. Many electrical connectors are only good to 70-80C, and bearings with plastic retainers could start suffering above 80C as well.
   
   And finally, 100C is practically out of the question without some sort of active heater for the enclosure. Bed heat will only get you so far.
   
   We'll, uh, cross that bridge when we get there?
4. Bed adhesion.
   
   If you look at some of Intamsys' or Vision Miner's videos on the subject, their initial solution was to apply PVP glue to a glass plate, bake it in the printer for a few minutes, then scrape it with a scraper, and repeat that process a few times. They also go on to say that this method has been deprecated in favor of this stuff:
   They also make some rather bold claims about this stuff - that it makes printing PEEK and PEI and such practically easy, that it works with almost every filament in existence, that it eliminates warping, and that it can cure certain terminal diseases. Okay, maybe not that last part, but color me skeptical.
   
   They also price it at $50 per bottle. Yeah...no thanks.
   
   PVP glue and a scraper it is.(As we later find out this decision was a big mistake, but we'll get to that.)

Okay, now that we know what we're after, let's jump in, shall we?

Printer selection

I could say that I planned this all out ahead of time and drew up a CAD, a budget, a timeline, and all of the other things for a formal project before approaching this. Honestly, though, it all started from a meme and sorta spiraled.

At this time, I had been saving money for what was my dream printer for years: an Ultimaker 2. At the same time, I had found this neat little printer on Monoprice called the Maker Ultimate, also known as the Wanhao D6. It was pretty cool: it seemed to have an all metal frame, Ultimaker/Zortrax mechanics, and lots of stuff you'd usually find in much more expensive printers. However, as interested as I was, I couldn't buy it because I was saving money for an Ultimaker!

Then, the open box version went on sale for $320. Then, a certain user in a 3D printing discord I'm part of sent me this meme.

And just like that, I had hit the order button. I need to work on impulse control.

It came in, I did some tests, printed some PLA with it, and decided it was a pretty neat printer. I stuck it on a shelf to add to my growing collection of printers, and that was that.

Until...some weeks later I started giving some serious thought to printing high temp stuff. I realized that this printer was almost ideal to be the base for a high temp printer for several reasons:

• It was mostly metal, from the frame to the bed stage to the sliderblocks to the print carriage. There were a few plastic parts that wouldn't survive the high temps, but I wouldn't have to replace anything major.
• It was a small, compact printer with a box frame. It would be easy to enclose and eventually heat - in fact, it has both first- and third-party enclosure options that screw straight into holes already pre-drilled in the frame.
• Its electronics and power supply were already reasonably well isolated from the main print chamber, instead being located in the bottom of the printer under a panel.

There were still several significant challenges, though:

• It did not look easy to isolate the belts and motors from the main chamber. Bellows, as would usually have been used, would have been very difficult to install.
• The hotend was only rated to 265C, and more frustratingly, it was a proprietary hotend in a proprietary mount. It would be difficult to get any high temp hotend into this machine.
• There seemed to be very little space to install any sort of a chamber heater that will not interfere with the motion of the bed.

We'll tackle those when we get to it. Thinking about things too much tends to discourage me from even starting a project, so the most important thing for me right now was to jump in head first. So, I ordered the enclosure.

Parts purchased: 

Printer ($320)

PEI build sheet ($20)

Total cost of printer: $340

Total project cost: $340

Baby steps: ABS first?

Now, to this date I had never actually even printed ABS well. I had done small things, but medium sized prints always came out warping and splitting. I wanted to see for myself how well ABS could print if I did have some way to trap some heat. The enclosure was on its way, but I got impatient.

Yeah, I know, this is very much not a good idea. Flammable paper and plastic sheets near a 250C heater block is a recipe for fire. Check out the result, though!

I had never produced an ABS print that clean before. Okay, there was a little bit of vertical banding and moire, but overall geometrical accuracy was great and there was zero warping!

So, the enclosure arrived, and having committed to this high temp thing I immediately decided to insulate it right away. Now this printer is starting to look pretty serious.

Parts purchased:

First-party enclosure kit ($110)

Foam insulation ($25)

Total cost of printer: $475

Total project cost: $475

Part 2: Aiming for Ultem