X tower lift with HE280?

[DerStig]

I have a V2 that upgraded to the HE280. I'm getting everything dialed in and have my calibration box printing at 20.005x20.008x10.008 so I'm pretty happy with that. Now I'm printing some FSR mounts for my MaxMetal build and I'm getting X tower lift. I thought with the auto cal of the HE280 that should be gone.

Anyone with ideas of what to look for or how to fix it?

[mhackney]

Read the forums here! A few of us are experiencing a similar issue with thinness of the print near the X tower and thick opposite. No resolution yet but in my case, removing the FlabLam and cleaning and reinstalling all the bed components nearly eliminated the problem. I posted a large single layer test donut on another thread here you can use to test and post a photo of your problem to make sure we are dealing with the same thing. There are some other recommendations there too, do those as well.

I can't repeat this enough: auto calibration does not magically fix geometry (and other) problems. And the V3 auto calibration is only a partial auto calibration (it does not address tower lean and rotation for instance). That said, if your build is pretty good (some instructions in my other thread on testing) then it works nicely. Auto compensation will fix problems and compensate for mechanical problems. But that is not supported in RAMBo for deltas.

The other thing to keep in mind is that on a delta, resolution is not constant across the build plate. It is best at the center and decreases as you approach the perimeter of the plate. You can do the trigonometry to verify this. This does affect the calculations that position the nozzle in X, Y AND Z (cartesian coordinates) near the perimeter of the build plate. In fact, near the tower, the carriage motion steps/mm for that tower predominates. On an RMax (post early V1s like mine) that is 80 steps/mm (alpha_, beta_ and gamma_ steps_per_mm parameters in EEPROM). So near the tower, eight steps result in .1 mm of movement. That is a HUGE percentage (50%) of a .2mm layer thickness. And keep in mind, those steps are the result of stepper motor micro-stepping, a not-so-precise mechanism to "improve" stepper resolution. It all adds up. That's why prints near the center can be spectacular and those near the periphery not so good.

But don't lose heart (or faith)! One can dial things in well enough to do large prints out to the edges of the universe (or, actually, build plate!) with a little work. The issue you and a few (8 at last count) are experiencing with the V3 (and many others are not) I am convinced is the result of some slight movement that happens somewhere during probing. It could be in the bed or bed mount, the effector/joints/arms, or the belts or other drive components. The fact that my V3 went from bad (showing this problem) to good after removing the surface laminate (a side effect I'm convinced) has me sniffing on the bed/bed mount track.

[DerStig]

mhackney, first off thanks for all your posts you are a wealth of knowledge.

So I did more research last night and looked at my EEPROM settings I think there is a problem. First off after doing a auto calibration with the HE280 ran the end stop macros and then did a paper test. X tower 1.2mm Y tower .82mm and Z Tower 1.1mm Center of the Bed was .18mm So even with the HE280 doing it's thing we are not level. then I looked at the EEPROM and got these numbers.

Tower X offset 0
Tower Y offset 761
Tower Z offset 205

what is going on?

also should I not have the same X,Y,Z arm lengths?

any ideas?

Thanks again for the help here, without this forum a lot of us would be lost

[mhackney]

X/Y/Z "arm" lengths - are you referring to "X max length [mm]"? If so, that is correct and are the X/Y/Z Cartesian coordinate for the hypothetical cube for the build volume. Which, the firmware knows is actually circular so the 250/250 for X and Y are a circle 250mm diameter.

Did you clear the EEPROM before flashing the new Repetier firmware? If not, do that and recheck. Otherwise, read on...

Your tower offsets are horrible! That is almost 10mm on Y. Can you physically see an issue? Can you get a square and check the towers like I show in my build thread? When the printer homes, you must see the Y carriage drop down that nearly 10mm!

[DerStig]

The towers are square to each other and the build surface, measured with a square and a digital angle meter. The travel is the same tower to tower when homing BUT when you start a print and it homes the print head moves to the Y tower and then back to the center it doesn't drop

I'm going to clear the EEPROM again and reload the firmware again.

[DerStig]

Nevermind, with the help mhackney we figured out that it was a carriage offset issue. Little bit of work and its printing beautifully

[morgandc]

Can you elaborate for the rest of us?

[mhackney]

He meant to say "endstop offset". His stop screws at the top of the carriages were very far off. So he manually adjusted them as per the procedure in my build thread.