Heater plate warping

Is it common for heater plates to warp or be distorted by the leveling process??? - I can "clock in " my Mendel 90 to within .005mm on the four corners, hot. But I am left with a diagonal distortion, almost exactly at 45 deg. to the platen front. This of course gives serious temperature variations across the diagonal void left by the bow under the glass.Please forgive any ignorance as I only have a short acquaintance with 3D printing, but there appears to be no provision for thermal expansion in the design (?) - My solution may be to reinforce the underside of the heater plate with brown insulation board, float the corners between "O" rings and cut diagonal slots at the four corners. If this is just a newbie over reaction or if there is a common solution of which I am unaware please be kind enough to advise.

For the PCB type, they only heat on one side and that bows the board. Mounting that PCB against an AL plate helps with the heat distribution.

I use a silicon type heater that is glued to a aluminum plate with a three point mounting and if you use 1/4" AL MIC-6 tooling plate, you won't need the glass.

If it's 4-point leveling (AKA all the corners) that is the cause. Changing one corner affects the diagonal, and having them all anchored will cause either the bed or the support carriage to flex when heated.

3-point is the way to go. You may have to make a change to the way it's mounted. But for the bed you'd level 2 corners. And then you'd level a 3rd point on the opposite side of the bed, Hopefully at the middle edge to make a triangle.

Agreed. I had so many issues with 4 point leveling that I just switched to a 3 point system. Haven't tried it yet but I have high hopes.

Four corner leveling has to be the dumbest common error in 3D printer designs. Any dope who has ever stayed awake in a high school geometry class knows that 3 points define a plane, not 4. With 4 screw "leveling", when you try tighten one screw that corner tries to move down, the plate wants to pivot on the diagonal line between the two adjacent screws, and the diagonally opposite corner wants to lift but can't because of the screw holding it down. So what happens? The corner of the plate bends downward and the undercarriage bends upward. How can bending a plate make it level?

Using a glass plate is a poor "solution" to the bent plate problem- clamp a piece of glass to the plate and it won't flex so much, so the undercarriage flexes more. With a slightly bent plate under the glass, as you have discovered, the heating is uneven. That makes it hard to get prints to stick without using glue or some other slop on the glass.

This cascade of band-aids on top of band-aids is necessary simply because the kit makers are too cheap/dumb to use a piece of flat, thermally conductive, cast aluminum tooling plate mounted on a three point leveling system that actually levels the bed plate without bending either it or the undercarriage.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

In defense of multi-point bed leveling, On my 350x350mm printer, I'm using a 1/8" aluminum build plate with a silicon heating pad (no extra glass). It is much lighter and less expensive than 1/4" cast plate, but it is not quite flat, and it also warps a little as it is heated. It is thin enough, though, that with multiple leveling points (in the 4 corners and one in the center), I can bring the bed up to temperature and tweak the plate into sufficient flatness. This is a different approach to building a printer, where instead of making everything massive and rigid and precise, I keep things as light weight as is feasible and add adjustability where required. If things are designed right, once everything is adjusted, there is very little fussing.

I have mixed feelings about the concept of 3 point vs 4 point leveling.

On one hand, it is certainly dead-simple for 3 points to define a plane, but that doesn't mean that any other number of points can't also define a plane; the plane just needs to be precisely flat to begin with. When using cast aluminum tool plate, the flatness is there, so really 4 point leveling is fine as long as no distortion is introduced. Really, the level should be set with 3 points, then the plate SUPPORTED by more than this. This is how machine tools are leveled. Because while 3 points defines a plane, it also allows the corners to sag more than 4 points. If the unsupported corners of your bed sag more than the error you would find by using 4 points, then is anything improved?

Think of a milling machine. You clamp a "flat" surface (the workpiece) to the "flat" base, and it is supported at many more than 3 points.

Edited 1 time(s). Last edit at 08/11/2016 01:37PM by n8bot.