I have an issue with combining two 3D STL prints, overprinting the first with the second.
I have created two STLs which both work fine individually on their own, but I need to remove the first print from the printer to do some manual work on it before relocating it on the print bed and overprinting with the second one.
My first print creates a 4 mm thick layer on the bed which consists of the required print contained within a surrounding frame to achieve registration for subsequent printing. The first print requires some manual work done on it prior to printing the second print on top of it. So I need to start the second print at Z=4 however I don’t seem to be able to find the necessary setting to achieve this. Obviously I don’t want to kick off the second print at Z=0 and crash into the existing first print on the bed.
Any help on achieving this on my P1S would be very much appreciated
Thanks
Why wouldn’t you simple take the two objects, place them into an assembly and then add a pause where you want to start the second model? Then apply your operation and hit resume.
You can pause the print to insert something in to a model, provided that something goes in to a pocket that’s deep enough nothing sticks up above the current top layer. But take it off the plate/modify/replace? IMO not doable. Because….
You will not be able to reposition the print accurately enough. Your jig is going to need something like 0.2mm separation from the print or you’ll never get the print off the plate while leaving the jig in place. Think “brim”. You’re printing a multi-layer brim at less then half the nozzle width. And in the time you take to rework it, the print is going to cool off and shrink, so even if you could recenter it perfectly, the further out from the center an area is, the greater the feature misregistration is going to be because of shrinkage. Not to mention that because it won’t be line-to-line fit with the jig, it is going to get moved around by the nozzle as plastic is being extruded. The mated walls will never be aligned like they had been printed contiguously. There’ll be a big discontinuity. So if you’re thinking this approach will produce a seamless result, you’re likely to be sorely disappointed.
As for getting something jacked up in to the air, use a “jack”. A really small diameter cylinder that the model stands on, that sets the height you want. The cylinder will be too small to slice, but the slicer should respect the geometry boundary anyway.
The right solution, IMO, is to design the model in pieces that can be printed separately and assembled afterwards. Done properly, this will produce better wall alignment. And if you use a plastic like ASA or PETG, you can solvent bond the parts together and the bond will be as strong as if you’d printed it in one piece.
Thank you for your very helpful comments. I don’t disagree with anything you’ve said, however I’m hoping to utilize a few tricks I’ve learned after 30 years experience in the PCB manufacturing industry to achieve my goal. It would neverthelss be extremely helpful if anyone knows of a substance that could be used as a temporary measure to re-adhere previously printed PLA to the print bed, which would release the print after cooling with any residue easily removable. Any suggestions most welcome.
If I understand correctly, you are asking whether you can remove a PLA part from the build plate, place it back in position, and resume printing on top of it.
On a typical consumer FDM printer, that will not work with any meaningful precision. Once the part is removed, you lose exact XY registration and especially Z height reference. Even very small misalignment will show up immediately in the resumed layers.
If positional accuracy is not important, for example printing on top of a disk or plate where placement can be off by a few millimeters, you could attempt to use a water-soluble PVA glue stick such as standard 3D printer glue stick or Elmer’s glue stick. Adhesion strength is usually not the limiting factor.
The real issue is positional accuracy, not bond strength. After removal, there is no reliable way to guarantee the part is back in exactly the same location and height without mechanical fixturing.
There is also a mechanical risk. If Z height is even slightly incorrect, the nozzle can strike the part and damage the nozzle, the build surface, or the motion system. If you experiment, test first with a small object and monitor the first resumed layer closely so you can stop the printer immediately if needed. In fact, I would keep my hand near the power switch because the pause button does not react as quickly, the pause will be issued and the printer will respond after the layer is completed which may be too late.
Post some pictures of what you’re trying to do.
3D printing is nothing like PCBA design and fabrication, except that PCBs can be multilayer.
If what you’re trying to do is bury a component/via inside the model, model a pocket for the component, pause the printer before the layer that bridges over the pocket, insert you object during the pause, resume the print.
Thank you for your very helpful comments
I have this afternoon thankfully achieved my objective
I used a frame to hold by model in place and provide later edge regitratoin then removed the model leaving the frame still attached to the bed. I was then able to make the manual changes to the partially printed model and subsequently replace it into the frame to continue printing.
I have to agree that it wasn’t 100% successful and rattled a bit during the second printing phase.
Although I got X Y & Z sufficiently I had not anticipated the warp factor which did affect Z and it also made the final ironing of the top layer rather ineffective.
Nevertheless I have a working model with hopefully not too much damage to the printer head.
Thanks again
Chris
Thanks for your helpful comments.
I’m pleased to report a degree of success today (see my reply to Olias)
The main problem has been the mid print modifications which had to be undertaken out of the printer making removal and subsequent repositioning a challenge.
X & Y were achieved through edge registration to an extra surrounding profile piece which remained attached to the printer bed which worked better than expected.
Z was OK ish around the perimeter but I had not anticipated warp which resulted in rather a noisy printing procedure and hopefully not too much damage !
Some appropriate adhesive for use on replacement to the bed would probably have been helpful
Nevertheless I have a functional item although the process was rather tedious.
It would be nice if there were some decent conductive filament available so that interconnection workarounds weren’t required.
Thanks again
Chris FICT
I would be interested to know what the “manual work” performed on the first part is.
Your plan of printing the first part, removing it and the putting it back before printing a second part is inherently dangerous. Printer damage, failure rates and more are all issues discussed already in detail.
Importantly, this is something so atypical, few would try it.
If we know the cause, maybe we can suggest something that means a mud print pause could be used. You later mentioned conductive filaments, I wonder if you are embedding electrical wires.
Knowing more may allow us to suggest design changes to allow the unknown factor to be melded afterwards, or creating a rig to drop the parts into the first model before resuming.
There are many other things experienced modellers could suggest if we knew the mystery factor rather than pitching what you think solves the problem.
N I C E !!!
My grandpappy used to call that “Yankee Ingenuity”, he said it’s creative solutions like that which won the war in the South Pacific during WWII.
Multilayer PCB assembly including embedded electronic components and 1 mm dia copper wires for ease of handling although 0.2 mm dia would have been sufficient. Better still 0.1 embedded in PLA filament would be great if available. Post assembly warp was the main issue.
Are you comfortable showing one or more photos of the bottom, bottom with conductive parts and fully assembled?
Could you also share Bambu Studio screenshots of the preview tab with filament view and line type view of each part?
I imagine there maybe some reticence if there is IP at stake, but it would really help us help you.
Thank you for your interest.
As stated elsewhere in this thread, the current objective was successfully achieved earlier today and consequently no further assistance is required. I have already outlined the basic principles applied in this project and I do understand that perhaps others might be interested in more intricate details of methods and results, but regrettably no further information is available at this time.
This forum functions as a community. While Bambu Lab hosts it, the value comes from members sharing their methods, results, and lessons learned so others can benefit.
Several members in this thread offered their time and technical input to help you reach your objective. The norm here is reciprocity - when someone receives meaningful assistance, they document what worked so the next person can build on it.
If there are legitimate constraints preventing you from sharing details, that is your choice. But withholding all practical information after drawing on community expertise runs counter to the “share and share alike” principle that makes this forum useful in the first place.
In due course a relevant paper will be submitted to the Institute of Circuit Technology ( see https://instct.org/ ) which will of course be made available to the forum. Until then please be patient.
Thank you for your interest
Chris FICT
