Super simple DIY garolite G10 build plate for X1C

I’d be more concerned if the plate flexes too much, G10 is pretty rigid so just maybe… Either way, I already have the tape and the G10 wasn’t terribly expensive so I think it’s worth a try.

Olias, been there done that too, not wanting to create more work for myself but worse case I’ll toss it if it’s a nightmare trying to save an old flex plate. Hopefully it’ll work but it won’t be my first failure. :slightly_smiling_face:

For what it’s worth, when I look at the bottom of the PETG print which had been in contact with the garolite, it does indeed look both shiny/glossy and smooth, far more so than anything I’ve ever seen come off either the bambulab smooth PEI plate or the engineering build plate. I don’t know how to evaluate whether or not it’s “specular,” but it has at least that much going for it. I don’t see iridescence in it, which is something I have seen on aluminum plates which have been milled flat.

Moving on, I completed my final print test of the same object as before, but this time with no brim. This particular object has frequently failed to stick for the duration of the entire print on both the engineering plate and the bambulab textured PEI plate because it is tall and doesn’t have much contact surface with the build plate. However, on at least this occasion, it did stick to the garolite, which I find quite encouraging, even if it’s only a single datapoint:

And here are closeup shots of the same print object taken from different angles:

I don’t know what that ball of PETG is doing next to it, but maybe it arose because I didn’t allow the nozzle to wipe itself on the tab-thingy at the back of a bambulab build plate? On this print there also seems to be some other extraneous oozing splayed around the build plate, but I don’t think it would be fair to blame that on the build plate per se. It strikes me as unrelated.

Well, that’s it for my due diligence. I’m pleased enough that I think I’ll order a proper build plate, either the Lightyear one or else the G11 build plate from Australia, just to see what that might be like, given that Maker’s Muse seemed to like G11 the best (though in his youtube he didn’t give any explanation as to why).

I decided to triple my odds of failure by printing three of the same model. No problem. They all stick.

I did get some layer shifts on this batch, and if you look at the zoomed out picture you can see why: the tape had come lose, most likely from the thermal expansion of heating up the garolite.

Those of you who have opted for spring clips: good choice.

Curiously, this time I didn’t get the drooled filament everywhere, as in the last print. I used the same settings as before. Strangely random.

In case anyone was wondering: Just prior to starting on these garolite tests I had switched nozzles from 0.4mm to 0.6mm, and it seems I need to re-tune the PETG settings. i.e. Regardless of print surface, my current PETG print quality needs improvement. However, for the purpose of testing garolite PETG print adhesion and release, I’m assuming it doesn’t matter. I should have more than enough time to dial-in the new PETG settings while I wait for the professional garolite build plates to arrive.

By the way, I decided to order both Lightyear and the Australian G11 build plates. I’m skeptical I will notice any difference, but maybe a careful compare/contrast will reveal why maker’s muse prefers G11. If nothing else I can print on one while the other cools all the way down to release temperature. Also, I haven’t seen any reviews on the Australian G11 product, so it’s a total gamble as to how that will turn out, or whether it will even arrive.

One more thing:
If want to mill your own custom profile and adhere it to a spring steel plate, as some here have done, then maybe very thin garolite is all you would need for, say, PETG, since it doesn’t have much tendency to warp. In that case, you’re in luck, because the thinner the G10, the less it costs. A thin sheet of 12"x12" G10 can be had for as little as $5:

as recommended by this redit link:
https://www.reddit.com/r/BambuLab/comments/xifu77/update_on_g10_as_print_surface_for_x1/

which also gave instructions on how to make one for an X1 printer by peeling the PEI sheet off a bambulab cool plate. Worthy of note: that same user reported no problems with print scanning, unless when printing black filament on the black garolite, if you tell the printer it’s a high temperature build plate.

This, then, might be an argument for having at least two different colored garolite build plates if you want scanning to work with any color filament.

Epilog:
I chickened out on the Australian G11 as soon as I realized their shop didn’t actually show a complete photo of their product offering. What it did show was a square edge. That left me thinking it may be just another square piece of garolite. If there had been more to show, I think they would have shown it.

So it seems that Lightyear is the only game in town when it comes to a professionally profiled garolite that’s fully turn-key. Luckily, they do offer two colors (one black and one white), so I got both colors so that lidar scan will work with whatever color filament I may want to print. For PET-CF, that means their white print bed would be the better choice.

Unfortunately I can’t recommend Lightyear G10 Stellar White (I didn’t test any other of theirs so know nothing about them). Here’s the problems I have with mine:

  1. It has a QR code, but it’s misplaced 5-6 mm so printer can’t read it (I assume it’s due to the location but could also be because it’s a negative image IIRC - black and white is reversed).
  2. For some reason the steel sheet gives VERY weak magnetic adhesion. Far too weak. I can easily drag the plate around pulling at the plate tab whereas a Bambu Plate sits so hard I can almost drag the printer around. This has several consequences: The plate can actually move around and cause layer shifts (and next problem below makes it even worse), and if your model will want to warp, there is NO WAY the plate will fight it back AT ALL.
  3. It’s only 257x257 mm, Bambu plates seem to be closer to 258 mm. For full plate prints you need to align it very carefully centered between the heatbed location tabs, or prints will literally fall off the edge. I noticed recent Amazon/Ali Express ads for third party plates use to state 257.5 mm, maybe that is a better measure (cold).
  4. The plate warps with heat. You can see light under the corners. The weak magnetism doesn’t even try to hold it back. An ABL on a hot bed would cause a royal mess, as would a lidar scan.

Lightyear claims their plate can compensate for a warped bed but my G10 introduces warping to a perfectly flat heatbed, because of 2 & 4 above.

So TLDR; Damn, It could be a wonderful plate but because of all the above it’s a useless coaster for 2-3 times the cost of other plates. It does look absolutely gorgeous, white is terrific in the camera with dark filament, adhesion is fantastic and release too. But the magnetics combined with the plate’s OWN tendency to warp makes it totally useless. I got some refund but that doesn’t help me (it also doesn’t give me back any of the €40 customs charges and VAT I paid for importing it, calculated from the initial price. But that detail is not Lightyears fault).

1 Like

I have the G10 Pro doublesided plate and I’m happy with the quality, nothing like your experience with the White plate.

The Pro QR code is the same as the BL High Temp plate, never once missed a read. The plate is thick and sticks to the bed good, it is slightly less flexible compared to the normal BL plates as expected. So far I’ve never had any lifting of the plate when doing biggish flat prints. I’m kinda glad I went with the Pro and would recommend it as a main plate to go with the others.


G10/HighTemp/PEI


Pretty much the same dimensions as the HighTemp


If anything its really tight fit between the two rear guides.

2 Likes

Wow, thanks for the heads up. I hope no one else has to go through all the hassles and delay that you went through. I would have guessed that the white plates are functionally equivalent to the black plates. Thanks to you I’m now forwarned.

I’m hoping that Lightyear corrected the white plates in the new batch that they are building now. If the white plate I ordered bombs out after I receive it in the same way that yours did then then I’ll report back here and add my warning to yours. Also, should that occur, I’ll try exchanging the white Lightyear build plate for a black one.

Yes, I preordered, probably got the very first batch of Stellar White. I received it early January. I’m not trying again until I hear all problems are fixed :cry:

Reporting back:

I like the garolite so much that I didn’t feel like waiting longer for the Lightyear to deliver, so I adhered my garolite to the least expensive Bambu Lab PEI sheet chinese knockoff I could find on amazon.

The good news: 10"x10" is pretty close to a perfect fit

The not so great news is that the 468MP seemed marginally more attracted to its own backer than it was to the PEI. In a more perfect world I would have applied it to naked spring steel instead. Nonetheless, it seems to have no problem gripping the garolite, and now that it’s on, I don’t think it will be going anywhere.

Worthy of note: From the X1C control panel you can disable the checking of the build plate. I did, and now I don’t get any annoying errors about it complaining the expected build plate isn’t there, especially since the knock-off build plate didn’t have any of the 2d markers on it. I suppose I could have copied the 2d marker from a real bambu labs build plate, but this obviates the need for that.

I just now completed my first print with the new configuration, and no problems to report.

2 Likes

I’ve got 2 G10 sheets, both 1mm. One I attached to a high temp adhesive magnetic sheet but unfortunately it has a warp in the sheet which doesn’t completely flatten out when put on the bed. I thought the force of 2 magnets together would be strong enough to hold it down but the warp isn’t going away. The other G10 I just attached to one side of one of the gold aliexpress PEI with about 10 strips of double sided tape and it’s been perfect. This is the cheap G10, not higher temp G11 but it’s been fine with 280°C so far. It leaves a nice smooth glossy finish on the part.
The only thing I’d like to change is the hotend nozzle wipe where it rubs across the strip at the back. I cut the G10 the same as the original sheet with the slit and it does still flex but I’m not sure on the wear. I may have to look through the g-code and just remove that part of the sequence.

1 Like

The cleaning strip on BL’s High Temp plate is the metal plate, the Hot surface doesn’t extend to it. It has a step down on that area so I’d be inclined to leave it bare (running hardened nozzle of course). Edit, just checked my Cool Plate and its the same.

1 Like

Good eye! So in actuality not having the Garolite cut to a perfectly identical profile is a feature, not a bug!

Yes, I know but if the garolite is in attition to the plate thickness on top of the bed, then there will be a much larger step down before the nozzle will touch it (depending on your G10 thickness, in my case 1mm extra). I guess I could cut off a little more of the G10 on the bed side of that area so there is enough room for the nozzle. But I was worried when it does it’s wipe in between, that the nozzle might hit the side of the G10

With my dodgy verier I get about 0.4mm thickness with the Cold Plate step, could be a bent nozzle if it does clip eek.

Mine is 2mm thick. No collision that I’m aware of. I think it may know where the garolite top is, because doesn’t it first home to the center before doing all of this?

BTW, my earlier posts were testing the garolite with PETG. For unrelated reasons, I’ve switched to PLA, and this was my first test with that. It worked perfectly. It adhered when the build plate was hot, and it totally released 100% on its own when it cooled off. Honestly, it’s all so perfect I’m surprised that garolite isn’t the gold standard.

2 Likes

Good to hear with regards to the nozzle wipe. I might have to get out the dremel and take off those last few mm’s to break the piece off and give it a go. I print everything on mine also. PLA, the parts just slide off the plate after it’s cooled. I have poked the G10 after heating to 250 and it’s got some stickiness to it. It’s really the only thing to use for printing nylon.

1 Like

Also, this PLA I’m printing is black. And my garolite is black. The lidar scanning that it does after the first layer is not reporting any problems. I had read on redit about the experience of some others where they reported it as a problem with black on black, but at least so far it’s not rejecting the scan results or complaining.

1 Like

This is false. I just now watched it more carefully, and it cleans its nozzle first. However, it does drop the build surface by more than an inch before moving over to do its homing immediately afterward. So, it seems we’re all lucky for that, as I do not see any risk of collision.

Thinking about it now, it makes perfect sense that it would want to clean its nozzle first to remove any leftover filament from the tip so that it doesn’t interfere with either homing or the bed leveling measurements. Presumably that’s the entire reason it does it at all.

1 Like

I notice that I’m able to crank up the heatbed temperature to quite high in order to get even better bed adhesion, and yet the prints still disengage after cooldown. Very nice. With PEI this risks “welding” the print to the build surface, but I’ve so far had no welding to garolite, and I’m not sure if it’s even possible.