Need recommendation for heated bed attachment method.

So I have a new project to build a new 3D printer for myself. My Makerbot is close to it's last days and I wanted a bigger/better replacement. Here it is.
Print envelope is x=24" y=12" z=12". Most of the design and fabrication is complete except for the heated bed.

https://www.cnczone.com/forums/attac...d=447252&stc=1

The problem I am having is dealing with the heated bed expansion based on temperature changes. I haven't even designed the bed because I am struggling with how to attach the bed to the 3 screw/bearing mounts. I have examined other printers that are a little smaller. Most of them put a slot in the bed in the direction of the expansion and then loosely bolt it down. Since my bed will be twice the size of theirs the expansion joint will have to allow about .030" expansion in the X axis, and about .015" in Y.

The second issue is deciding how to attach the bed loosely enough that if the steppers go haywire the bed will not be bent if the 3 steppers are at drastically different heights.

So that is my issue. I expect to use 6061 aluminum .25" thickness bed with a heater attached to the underside. Depending on what we come up with for an attachment there may need to be a minimal frame attached to the bottom of the bed plate.

Here is a close up of the screw/bearing carriage. These can come off an have more holes drilled if needed.

https://www.cnczone.com/forums/attac...d=447254&stc=1

Any of your idea or recommendations are welcome.

I'm working on my own industrial quality 3d printer design at the moment....but am not an expert on 3d printing.

IMO, you would want a frame underneath the bed, perhaps made up from 1"x1" T-slot or something similar. I would think that it will be easier to get the entire bed flat as a whole with a frame under it.

As far as accidentally bending the bed goes

I'm guessing that the 3 motors are some of the highest torque Nema 17 steppers? 5mm lead ballscrews? Sure, you could get quite a large amount of force on there that could bend things. I'm not sure what the best solution is, but is it possible that you could make some easily replaceable attachment points designed to break away before the point of damaging your bed occurs?

Are you using a duet board and trinamic drivers?

https://duet3d.dozuki.com/Wiki/Stall...sorless_homing

There might be a way to reduce the available motor torque while homing, combined with a stiff enough structure under the bed, assuming that during a homing sequence is the only time this accident might occur.....

Quote:

---

Originally Posted by Duet Wiki

If you are using stall detection for homing, reduce motor current while homing (see the M913 command)

---

Sounds possible perhaps.

For the bed

I've been thinking about using 120V silicone heat pads with a PID temperature controller. Haven't sorted out the details. It might be nice to have two sections on the bed for only heating half of it on smaller prints.

The way I have envisioned it is underframe, aluminum plate, insulation, silicone pad, glass, and just a couple of clips on the four corners to hold the insulation, silicone heating pad, and glass on there.

I wonder if I could get away with underframe, phenolic, insulation, silicone pad, glass (or buildtak?) instead.

I have been thinking about using unheated phenolic for printing Nylon. In that case I could just take off the top layer. But I need to investigate the properties of this material first to see what temperatures it can be exposed to and what it's coefficient of thermal expansion is. Plus finesse the concept a bit.

What are your thoughts on all of this? What print surface are you planning to use?

I need to look up the coefficients of thermal expansion for aluminum, phenolic, and glass to get started with a better final design for myself.

NIC 77. Thanks for the feedback.

Yes, Duet 3 main board is the controller. At the moment, there is a problem with the board, Duet says that their fab house sent a batch out with bad diodes. I got one of those so I am in the process of getting it replaced. I love the board though.

The Z axis steppers are small NEMA 17's but they are geared 3:1 with a belt drive. So they could generate some force. The question with regard to this is should the bed be strong enough to overcome the force and make the motors stall, or should the attachment mechanism accommodate the motors getting out of wack. The two methodologies are diametrically opposed, I just have to pick one and also accommodate for the thermal expansion.

Initially I had planned to simply create the bed from a 1/4" thick piece of 6061 aluminum plate with slots in the front corners for attachment. The plate would sit on the bearing brackets, 2 at the front one in the center rear, with the center rear attachment being fixed and not floating. The two front attachment points would use a shoulder bolt through a slot in the plate to allow for about .030" of expansion. Then I got to thinking about what if!!! Maybe I should just go with a 1/2" plate an say F it.

For now the plan is to use a AC silicone heater pad attached to the underside of the plate. The electronics are all setup where I am using an output of the Duet board to control a solid state relay to turn on the heater and a temperature probe input to the Duet for feedback. The Duet runs the PID loop to control the bed temperature. It does the same for the hotend. With such a large heated bed plate it will need some insulation on the bottom side to keep it hot but I haven't selected a material yet for that. I suspect that I will go for a glass wool of some type. The print surface will probably be different materials. With my Makerbot I print ABS on Borosilicate Glass. I will probably do the same for this printer as it works really well. I am going to try other surfaces too. There are several flexible surfaces that I have been looking at that are more forgiving than glass. All of these surfaces will use magnets to hold them in place and maybe some clips in the front to eliminate shifting.

Like you I thought about a multi zone heated bed. The problem with it, however, is bed warping. When you take a metal plate and heat part of it the thermal expansion will cause it to bend. Half the plate will grow and the other half will not. The thermal expansion has to go somewhere. So for this reason, it is best to heat the whole plate as evenly as possible.

The rest of the printer is based on a CoreXY motion system driven by NEMA 23 steppers. The hotend is water cooled so I should be able to achieve any temperature that a future material will need. I can use a Bowden extruder but it is currently setup as a direct drive geared head. I will admit the printer is WAY OVERBUILT. I am using 15mm rails on all axis, 1204 ball screws on Z, 10mm belts driven with 27oz/inch NEMA 23 steppers. Endstops are magnetic sensors. Filament is dispensed from a sealed/dehumidified container and never touches air until it comes out the extruder. The extruder is using dual Bondtech gears on an 8mm shaft, driven by a NEMA 17 stepper with 14:1 gearing. Currently the extruder motor is mounted on the head, but it is really heavy so acceleration may be an issue. The machine frame is 1" Aluminum extrusion with CNC machined side panels that keep it absolutely square. All the brackets and such are machined from 6061 billet Aluminum.

For your viewing pleasure:
https://www.cnczone.com/forums/attac...d=447272&stc=1
https://www.cnczone.com/forums/attac...d=447274&stc=1
https://www.cnczone.com/forums/attac...d=447276&stc=1
https://www.cnczone.com/forums/attac...d=447278&stc=1
https://www.cnczone.com/forums/attac...d=447280&stc=1

I am shooting for speed, hot fast, move fast, extrude fast. We will see how well it works.

So I think I have decided to just us a shoulder bolt in the front corners to attach the bed. Shoulder bolt through an oversize hole in the bed with a spring washer on the bottom side to raise the bed off the bracket to minimize heat transfer into the bracket.
https://www.cnczone.com/forums/attac...d=447298&stc=1

Final photos.
https://www.cnczone.com/forums/attac...d=451332&stc=1https://www.cnczone.com/forums/attac...d=451334&stc=1https://www.cnczone.com/forums/attac...d=451336&stc=1

I like your work...well done!!