Some thoughts on TPU multi material printing

Bambu Lab H2D
1

One of the new H2D features that had me most excited was the ability to mix AMS-incompatible materials like 95a TPU into multi-material prints. I can confirm that it does in fact work, with a few major headaches.

  1. Setting up the TPU pass through sucks. Freeing the bowden tube from the buffer’s pneumatic lock is a pain since that buffer is at the back of the large enclosure and the lock is hidden under a spring mechanism. The good news is that once you’ve taken the bowden tube out of the buffer, there is no reason to reconnect it if you only have 1 AMS. However…

  2. TPU only works on the right nozzle. During setup, the manual specifies that you can connect the AMS to either nozzle. So if you’re like me and connect the AMS to the right nozzle and then try to use the left nozzle for TPU on the external spool holder, you’re in for a rough time when you find out you have to fiddle with the bowden/buffer situation AGAIN to swap the AMS to the left nozzle.

  3. TPU nozzle swaps are slow. Even though there is no cutting/purging, the H2D has a lengthy ramming/priming routine it performs before each TPU nozzle swap to reduce oozing. Nozzle swaps were taking near as long filament swaps on the X1C. I tried to speed up the ramming routine, but even slight changes resulted in extruder jams.

  4. Extruder jams are horrific. This is more an issue when switching from high temp materials like ASA to TPU/PLA. The tool head has enough thermal mass that it stays warm long after the enclosure has cooled down. If you try to load TPU or PLA after using the chamber heater without giving it at least 30+ minutes to cool down, it’s very likely the filament will soften and jam. The intricate nozzle swapping mechanism makes extruder disassembly into a technical and involved process. The wiki lists this repair as a 60 minute task, it took me closer to 90 because I had to use a heat gun to slowly work out bits of filament from several of the enclosed passages.

  5. TPU/PLA mixed prints require a release agent like glue stick. I wasn’t able to find a happy middle ground bed temp where PLA would stick during printing and TPU would release after cooling. As much as I hate using glue stick, it’s a necessity on the included pei sheet.

  6. It’s a real shame that the AMS can’t dry while printing. In a humid southern environment, TPU starts to bubble and ooze after just a few hours out. While I can print from an external dry box, it’d be nice to be able to dry dessicant and backup rolls of TPU during printing.

All in all, I’m loving the H2D. It isn’t quite a full replacement for tool-changer printers, but the 2nd nozzle definitely adds some new utility that I plan to use in my projects (such as) . The firmware is rough around the edges right now. Hopefully it improves over time. Mainly, I think they need to lock the user out of loading PLA or TPU filament for a specified period of time after using the chamber heater, as well as adding a low-heat setting (~40C) for the AMS that can be used during printing. It definitely doesn’t have the same kind of plug-and-play simplicity of the P1/X1/A1 printers, but for the price tag I think most people would prefer added features over simplicity.

5 Likes
2

I hate how annoying the TPU set up is.

But mixing materials in a single AMS is no different than what it was on the X1 and P1’s of the past. So, the jams make total sense. Different materials have to be on the other nozzle. Or maybe its better to state, only one material per nozzle in a print. Sure… you can break that rule, but you’re on your own there. Different colors and same material is fine, but different materials on the same nozzle is a recipe for problems.

If TPU is on the bed… yeah, bed adhesion will be problematic. But that’s the same for mixing PETG and PLA. Any different materials on the first first layer that have a significant difference in bed temps are going to be tricky. TPU sticks too well at 70C (when mated with PETG) to really chance not using glue stick. Especially, with prices of replacement parts bumping up.

***Updated with quotes to clarify

3

Can’t put 95a TPU in an AMS. This is all talking about running TPU from an external spool holder to a dedicated nozzle. The AMS is only used for PLA on the left nozzle.

Even with a dedicated nozzle, the H2D goes through a ramming/priming routine to reduce oozing for TPU. The jams occur if you try and adjust that routine.

4

I believe one of the early access YouTube reviewers said Bambu is aware and working on a solution. It was not specified if this is related to a firmware update, or if they’re going to release an actual hardware kit.

5

I understand the limits, I was referring you referencing the increase of jams.

In reference to that, it would only be an issue if you are using the same nozzle. From your text, I would guess you meant ASA/PLA. But please, if I have misunderstood you, correct me.

6

Oh gotcha, my mistake. That makes sense that ASA in one nozzle and PLA in the other would be problematic. I wonder if the software locks you out of using PLA in one nozzle while also heating the chamber.

My worst clog occurred after doing a single material ASA print. I started a single material PLA print about 10 minutes after the ASA print finished. The residual heat in the tool head was enough to make a mess of things.

7

Probably more my fault. I do have a problem of thinking of things and not spelling them all the way out. It definitely leads to this type of comment here and there.