I was a tester for the Bambu Lab H3C

I wrote this for a reddit post, but I figured some of you might be interested as well.

Disclaimer: I have been very kindly provided a testing unit of the H3C by Bambu Lab.

Bambu Lab has not specified what they’d like me to say about it. All the following observations are my own after a few weeks of use.

For my first project with this printer I chose to make a cosplay weapon inspired by roman gladius swords. With this project I wanted to focus on the ability to print single pieces with different nozzle sizes. I’m using 0.4mm nozzles for the larger features and a 0.2mm nozzle for the fine gold details, both in walls (grip) and on top layers (pommel). I’m still learning how to best design things for multiple nozzle sizes. Designing top surface details for a smaller nozzle is trivial, mixing different nozzle sizes in side walls can take a bit of planning and experience. I think I have room for improvement there.

I will upload this model and print profile to my Makerworld profile soon.

The Printer:

  • This is a big boy! If you have it on a regular table and you’re on the short side, you may want to keep a step handy to load filaments into the rear slots.
  • I’m hugely impressed with the speed of the tool changes. Since nozzles get pre-heated before it’s their turn to be fetched, print-to-print times are extremely short. Bambu Lab specifies the tool change time at six seconds, I hand-timed it at five. The pre-heating of the hotends is very well timed, they reach their print temperatures the moment they’re getting fetched. Not too early, not too late. The tool changes have been completely reliable, no malfunctions at all. I pushed it with some prints, one requiring nearly 3000 tool changes. The H3C didn’t miss a beat.
  • Despite the tool changer mechanics, the toolhead doesn’t seem especially heavy and the gantry is up to the job. I’ve noticed no issues with ringing.
  • One downside I see compared to my H2S is the ease of nozzle swaps. They don’t just clamp in, they are fixed to their carriers with 3 small screws. But since there are nine hotends, you likely won’t have to swap nozzles too often.
  • Print quality has been great for me so far. Comparing it to my H2S, it easily matches and often exceeds that machine.
  • I found it very easy and intuitive to set up a print with different nozzle sizes in Bambu Studio. The slicer combines the different line widths automatically, I’ve had to do very little tweaking to get the exact outcomes I wanted.
  • There’s no poop on tool changes. There is still a prime tower to be printed though. I did run a multicolor print without a prime tower just to see what happens. The print quality takes quite a hit, I would not recommend that. I’m really hoping to see the “no sparse layers” feature for prime towers in Bambu Studio as well, that could further cut down on waste significantly.
  • There are 2.85mm filaments now. I got three spools of 2.85mm TPU - the really squishy stuff, 75A. The large diameter filaments need their own hotends, one of which came pre-installed with the AMS. A TPU this soft in 1.75mm often leads to extruder gear issues, this larger size makes the filament less stretchy and easier to grip. I had no issues printing it. I think this is a smart solution.
  • There are some interesting smaller improvements over the H2 series. The poop chute now has a little flap that flips backwards when the nozzle moves away from it, making sure bits of filament don’t land inside the print volume or get stuck on the rim of the poop chute. The nozzle wiping pad also has seen a redesign, now the nozzle can rotate around the edge of it to wipe the outsides of the nozzle tip. Unlike on my H2S, I’ve seen no material buildup at all around the nozzle tips. They look pretty much as pristine as they came out of the box after many hours of printing. One common gripe with the H2 series was the placement of the USB port at the top of the printer. It has now been moved to the side of the touchscreen mount, a much more sensible place in my opinion.

The AMS 3H:

The AMS 3H deserves some special consideration, it is quite different from the AMS 2. It is an integral part of the printer now, not an “add-on”. The 3H has no active rollers, which means it can’t retract filament by itself. To unload a filament, the printer heats up the nozzle shortly, extrudes a bit of filament and then cuts it before prompting you to retract the filament by hand. It’s not a big deal as the PTFE tubes are not that long, but it is a bit more involved to unload a filament. Something to be aware of for serial spool switchers. Loading a spool is a breeze. Pop it in, the AMS reads the RFID tag without rotating it at all, insert the filament end into the feeder, done.

Two spool slots have a dryer function for fancier multi-material jobs or just regular filament drying. The other seven slots have no dryer functionality. Personally I’m fine with that, that’s plenty for me. Maybe you find this too restrictive for your use cases.

The AMS 3H is screwed onto the printer. With the removable magnetic side panels getting to the hotends is easy enough, but come maintenance time the attached AMS might make things a little trickier than with the open-top H2 series. The 3H can be removed as a whole or just the top part holding the spools, with four screws each.

Important to point out is that to get to your hotends you need access to the sides of the printer, which could be a problem if you have a cramped print space. Moving it a lot is not something you want to do, it’s heavy and has very grippy rubber feet. Unfortunately the front cover of the AMS base is not easily removable. For print-farm operators, this means the H3C should likely be placed on a sliding tray.

Apart from holding nine spools, the biggest upgrade over the AMS2 in my opinion is that you can now print soft TPUs and all manner of other “unfriendly” filaments through it. Seven spool slots have a bypass hole at the bottom, allowing you to route filaments straight to the hotends without going through the AMS gears and PTFE tubes. There’s much fewer restrictions on what you can throw into a multi-material print. A nice touch: the hotends have a holder for the PTFE tube so you don’t need to remove it or figure out how to secure it.

Summary:

This is an impressive machine! The tool changer feels like a real generational step up for Bambu printers, and I was very positively surprised how mature the system felt even early on in my testing time, when some firmware and slicer kinks were still being ironed out.

I used to avoid prints that require filament changes and spent a lot of time designing my parts around them, with the H3C I don’t feel like I have to limit myself at all. The low print-time penalty and the number of hotends massively lowers the hurdle. Not just for color; combining different materials is trivial. Combining different nozzle sizes also turned out to be a lot easier and straight forward than I thought.

So far I’ve printed PLA (Basic, Matte, Palette, Silk, Translucent, Glow, CF), PETG (HF, Translucent, CF), TPU (forAMS, 85A, 75A), ABS, ASA, ASA-Aero and PA-GF with the H3C, all without issue.

There are some small ergonomic compromises, chief of which being the requirement to have enough space for access to the sides of the printer to work on the hotends. In my personal opinion a small price to pay for all the upsides.

On a side note, I’m never printing supports without using non-bonding materials for interfaces again. OMG it’s so clean! I’ll just keep a spool of PETG permanently loaded in the AMS for all my PLA prints.

I will add a few more projects I did during my testing phase tomorrow.

I must admit, I’m a little jealous :slight_smile:

It was a surprise, lol. Bambu Lab contacted me on reddit out of the blue and asked if I wanted to test a new product. Hell yeah! …turns out it was the H3C! :scream::scream:

I own a P1S with two AMS HT units and I’m planning to upgrade to the X2D. That’s about the limit of what I can afford…

I wonder what the price point will be on the soon to be released P3C. I think it’s a 7 tool changer. Could be very competitive, as the H3C honestly was lower priced than I would have thought.

The main problem is that in my country, that’s a lot of money. :slight_smile: People could live for half a year, or even longer, on what an H3C costs.

Yeah, I get that..

Prusa must be concerned.

I’d have to sell out everything and still couldn’t. :unamused_face:

I’ll have to see what Sovol comes out with. $1K less for 7 heads and others to be hot on their heels. As it is Bambu Slicer is a bit obtuse in AMS mat handling with more than one printer and AMS design has left me a :pinching_hand: bit underwhelmed, sure it’s inexpensive, until it become problematic. Also the issue of Slicer security, when I cannot print via my PC on my own network easily/ seamlessly if the internet goes down I get a bit perturbed. And what’s up with putting a roll of filament in the AMS, it loads it, but the slicer does not pop up and say “Hey! I see you put a new roll in, tell us what it is and we’ll get that set up for you?”.

As it is, the slicer thinks the 2 AMS units on the X1C are also on the X2D. The AMS2 on the X2D is no where in sight. Head scratcher. In filament management “No data” even though I have 4 spools of Bambu filament in it.

I am not sure I’d buy another Bambu as the tech is maturing and others will be catching up,

Not sure what to think of the drying limits, especially after the AMS2 not being sent with the wall transformer in the X2D bundle, finding out later “Why can’t I dry while printing like they said it could???” it lacks function without it, not mentioned in the fine print. Do I need drying in all slots? Probably not. But I could if I need to. Cheaper than running a dehumidifer 24/7 in my average %70 humidity location.

Bambu cheaps out in some of the wrong places. Having to stand on a ladder to roll the filament back up? Well I have not seen the inner workings so I’ll reserve judgement. Not putting spool motors in like the AMS seems a bit dubious.

What Bambu has going for it is great quality prints with appliance-like experience with a little frustration here and there in slicer integration.

What gets me is they are almost there. But seem to stop a little short. I had to sell stuff to get the X2D and still need to sell the X1C. Won’t work the for H3, Too much. But would I? What is 6 months down the road other than Bambu?