Bed compensation, which type?

I put my auto compensation co-ordinates into config.g as suggested in the manual, using the M557 P1, 2,3 command.
I also have different co-ordinates in my setbed.g file for manual stored readings.

Upon prepping for print I run setbed.g but thought I would try an automatic run so issued a G32 from pronterface.

Instead of moving to the M557 coordinates it went instead to the manual ones and proceeded to mash the head into the bed as there are no white squares at those positions.

My question is why did it not go to the right coordinates?
Should you not run setbed.g if you want to use auto compensation?

My main query is that if you don't run setbed.g, you don't get the orthogonal compensation M556 line as that is in the file too.

I know auto comp is a bit iffy but hopefully that will be fixed soon, I'm just trying to get a handle on the correct startup procedure.

---

Another RS Ormerod Mk1 meets the world

Retired now but I used to make....
CNC Machined Mk1 aluminium bed support plates for the Ormerod
CNC machined X-plates and ribs for Mk1 & Mk2 Ormerods
CNC machined bed support arms for the Mk2 Ormerod.
Dual Hot-End heatsink blocks.

Once you set up *any* bed compensation once, it is difficult to do manually with the release firmware, because the Duet will apply compensation to the reported Z positions and you won't know what the real values are. So it's not a good idea to have it in config.g Unless you have carried out one of the Z sensor modifications, I would not trust the automatic compensation or the automatic Z homing. I do my bed compensation and Z homing manually before every print because I have found that to be the only way that is reliable on an unmodified machine.

Edited 1 time(s). Last edit at 03/07/2014 05:10PM by dmould.

The Duet firmware only stores a single set of coordinates for bed compensation. So running setbed.g which did the manual compensation overrode the ones you set up earlier in config.g. I'll take a look at storing them separately in a future version of the firmware.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

One driving force behind me trying out my alternative framework to house the workings of the Ormerod (other than noise abatement) is so that I can have stably orthogonal axes (and to make the orthogonality easily testable) - it's possible for the Z pillar to tilt towards the Y motor or away from it, and for the X axis to droop (the axis, not the carriage) if the Z pillar tilts toward or away from the bed, or if there's slackness in the Z carriage, or if there's torsion about the X axis itself. All of these tilts and twists would register on a bed probing as if the bed itself had tilted (of course the bed itself CAN tilt too).

In the best-case real-world scenario you run bed compensation before a print, and if X,Z and the Y frame are orthogonal, but the bed slopes and you apply compensation, then the nozzle is made to conform to the slope of the bed - you get vertical prints with 90° between X and Y, but the top and bottom of the print slopes relative to Z in X and Y to the same extent as the bed does (but at least you first layer sticks).

In the absolutely best case you run ormaxis.g before every print then run a bed compensation (or is it the other way round?)...

So, I'm building a frame I can level using a height gauge on a flat surface, with a bed I can level using a height gauge and an X axis ditto, whose corners I can true with a try square and whose Z column(s) can be adjusted with a try square, so I SHOULD get orthogonal prints (I just hope my first layer sticks )

Ray

Edited 1 time(s). Last edit at 03/07/2014 05:40PM by rayhicks.

Wow, that's some impressive effort

Would it then still be an ormerod

Nice ideas though, I'll probably settle for a new non-droopy bed mount when I get them made and manually the level bed to the mount then hope to not need compensation.

---

Another RS Ormerod Mk1 meets the world

Retired now but I used to make....
CNC Machined Mk1 aluminium bed support plates for the Ormerod
CNC machined X-plates and ribs for Mk1 & Mk2 Ormerods
CNC machined bed support arms for the Mk2 Ormerod.
Dual Hot-End heatsink blocks.

Whoah! I haven't actually got there yet just got a trial base going so far - but that's the plan, to make it like setting up a mill, you clock it then lock it down, though with the lower forces on the printer, and rigid joints, hopefully it'll need reclocking less often. I've got a scrap optical breadboard that I'll probably use as a stand for it, then I can run a height gauge around on that for truing - the disadvantage of clocking using a gauge in the extruder position is it's not clear whether the bed is out of true or one or more of the axes (in my mind anyhow as mentioned above). You're right a stiff, stable and adjustable bed mount is probably crucial

Ray