Marlin and delta setup question

Hi guys,
for financial reasons, I´ll start running my kossel800 delta using marlin on a ramps/Mega/LCD2004 combo.
I started looking into the configuration.h for deltas and was able to identify most of the required parameters.

Only the z-height is a mystery to me.
The endstops will home at the top, but where do I tell Marlin what distance to the printbed this is?

All I found was this:

//Manual homing switch locations:

#define MANUAL_HOME_POSITIONS  // MANUAL_*_HOME_POS below will be used
// For deltabots this means top and center of the Cartesian print volume.
#define MANUAL_X_HOME_POS 0
#define MANUAL_Y_HOME_POS 0
#define MANUAL_Z_HOME_POS 250 // For delta: Distance between nozzle and print surface after homing.

But what for do I have a microswitch on the effector? How do I tell marlin to set z-zero to that level?
-Olaf

Edited 1 time(s). Last edit at 08/13/2015 03:06AM by o_lampe.

That's it right there . . . . Deltas home to X=0, Y=0, Z=max . . . Z=0 is the bed, so you effectively tell it where max (at the endstops) is. Don't confuse tower X,Y,Z with cartesian X,Y,Z. . .

- Tim

Then the effector switch is used for ABL only?
I downloaded Marlin 1.1 today and see if I can make it work. My 1.02 delta version had no ABL-section....
-Olaf
Is there a special branch of marlin, that works best for deltas? I´ve also seen Rich Catells Marlin branch but the files were 2 years old?!

You have Johann's fork of Marlin which used to track the main Marlin development tree, but did not incorporate Rich Cattell's autocalibration.

Then I believe Johann's code was completely merged into Marlin (I am not 100% sure about that, though). Unfortunately as far as I know Rich Cattell's branch was never merged into mainline Marlin.

I started working on merging Rich's fork with Johann's fork and have a github repository for that work (just check under AndrewBCN on GitHub). It needs more testing but I haven't had time to work on it lately.

The idea is to merge this fork into mainline Marlin and Marlin4Due (32 bits) some time in the future... when I have time.

Edited 1 time(s). Last edit at 08/14/2015 12:48PM by AndrewBCN.

Yes, the effector switch is ABL only, and in that case, the fixed parameters you quoted get it close, and probing 'dials it in'.
This works with 1.0.3 . . . I never bothered with 1.0.2 myself . . .

- Tim

Quote
o_lampe
Then the effector switch is used for ABL only?
I downloaded Marlin 1.1 today and see if I can make it work. My 1.02 delta version had no ABL-section....
-Olaf
Is there a special branch of marlin, that works best for deltas? I´ve also seen Rich Catells Marlin branch but the files were 2 years old?!

Edited 1 time(s). Last edit at 08/14/2015 01:49PM by tadawson.

Quote
tadawson
Yes, the effector switch is ABL only, and in that case, the fixed parameters you quoted get it close, and probing 'dials it in'.
This works with 1.0.3 . . . I never bothered with 1.0.2 myself . . .

- Tim

If I send a "G0 Fxxxx Z0 S1 //move with endstop-check" the effector switch wouldn´t stop the nozzle from cracking the glassbed?
I better start with a manual z-height way above the real distance, then...
...and use a cardboard bed for first tests, too
-Olaf

Edited 1 time(s). Last edit at 08/15/2015 02:40AM by o_lampe.

Quote
AndrewBCN
You have Johann's fork of Marlin which used to track the main Marlin development tree, but did not incorporate Rich Cattell's autocalibration.

Then I believe Johann's code was completely merged into Marlin (I am not 100% sure about that, though). Unfortunately as far as I know Rich Cattell's branch was never merged into mainline Marlin.

I started working on merging Rich's fork with Johann's fork and have a github repository for that work (just check under AndrewBCN on GitHub). It needs more testing but I haven't had time to work on it lately.

The idea is to merge this fork into mainline Marlin and Marlin4Due (32 bits) some time in the future... when I have time.

They are currently busy with debugging the ABL code, I hope your work isn´t obsolete then?
I´d like to try Marlin4 Due one day, especially when it would work on one of the cheap all_in_one due boards ( replikeo, MKS SBase )
-Olaf

Nope. You need to set the overall height correctly, and then if you want to prevent travel t onegative Z, enable soft endstops. By definition, the probe *must* be lower than the nozzle in order be able to probe, so Z=0 will push the probe through the bed. And when not probing, it should be retracted to it doesn't snag the print, rendering it useless anyhoo . . .

- Tim

Then I´m in trouble, because my nozzle is the probe.
I`m using this effector, which is a clean and easy solution without servo and allen key hack. ( If I can make it work with marlin )
I tested the "S1" option and the steppers started to stutter a bit, when I triggered the endstop, but they didn´t stop ( Marlin 1.03 dev )
I hoped this would be a clever way to measure the nozzle/bed distance after G28-homing.
I`ll try to issue a fake G29 ABL and see, if the probe switch is setup correctly.
-Olaf

#define DELTA_SEGMENTS_PER_SECOND 200

How many segments can I actually ask from a Mega2560? I tried it with 160, but flipping through the LCD2004 menue was almost impossible then.
Or are there other parameters I should reduce to keep the default 200 segments?
What happens, when I print very fast? Will marlin reduce segments or reduce speed?
-Olaf

Edited 1 time(s). Last edit at 08/17/2015 12:51PM by o_lampe.

Quote
o_lampe
Then I´m in trouble, because my nozzle is the probe.
I`m using this effector, which is a clean and easy solution without servo and allen key hack. ( If I can make it work with marlin )
I tested the "S1" option and the steppers started to stutter a bit, when I triggered the endstop, but they didn´t stop ( Marlin 1.03 dev )
I hoped this would be a clever way to measure the nozzle/bed distance after G28-homing.
I`ll try to issue a fake G29 ABL and see, if the probe switch is setup correctly.
-Olaf

That should work . . . I was more talking to deploy/retract probes that require clearance . . .

- Tim