Another Predator

Danal

How does the effector actually probe?

Example: The smart effector flexes the circuit board just slightly when the nozzle presses on something, and a strain gauge reads this to trigger the probe output. The strain gauge is a long "spiral" (spiral triangle) of very thin circuit traces that surround the mounting point of the hot end. Because of all of this, the Z offset for a smart effector probe is often right at -0.1

What does your hotend do? Piezo? Slight movement? Etc. What offsets have you used?

Carlo

This post is deleted!

Carlo

@Danal
Hi Danal,

It's a smart effector I've tried loads of heights and all with the same results fine in parts of the bed too high in others
currently at -0.08 trigger height.

Thanks

Carlo

@Danal

Just thinking ..first for me! is it possible as the only thing that dictates the height is the nozzle touching the bed and the end stops is this simply an end stop issue I read somewhere on these forums about a faulty optical end stop some delay in it's signal which meant false height, given the height map is flat the smart effector is working but only in the space it thinks its in, I'm claiming all is built fine but are they.. nothing else has any influence on the height? and the effects never change no matter what I change in settings or technique .
I've ordered the replacements already but if someone can say nope your wrong that would suck but I'll have learned a little more.

Cheers

Danal

If you are running a G28 G32 (which invokes bed.g) before every print, then endstop accuracy is much less important.

Two scenarios:

One, the printer homes (g28) and prints. In this case, the ONLY determinant of bed-to-nozzle spacing for the first layer is the accuracy of the home switches (and proper configuration of overall Z height).

Two, the printer homes (g28), then probes (g32), then prints. In this case, the G32 is the final arbiter of bed-to-nozzle spacing for the first layer. (and proper configuration of Z probe offset).

So, yeah, you probably did not need new endstops. IF you are doing a G32 before EVERY print.

Carlo

@Danal

I am doing G32 before every attempt.
Ah well just a thought ...about the only thing I have not changed.

Cheers

droftarts

@Carlo your bed.g should have M561 or G29 S2 at the beginning, otherwise it’s probing will be affected by any bed mesh you have loaded.

Ian

Carlo

@droftarts

I always make sure mesh is disabled as per delta cal docs when probing I'm gonna fit the new endstops and try again because nothing is making any difference, appreciate everybody helping at least I have a better understanding of reprap though if I don't get this to work not sure I'll remember it.

Thanks
C`

Carlo

Hi,

Does anyone know if I can send the board and effector back to Duet to be tested ?

Thanks

NexxCat

Not sure if it will be any help, but my full Duet2Wifi config for my predator is here:
https://github.com/nexx/Anycubic-Predator-Duet

It certainly looks like you're too far away from the bed.

On the github, you'll also find the macro's I use, including one I run prior to every first print since power-on.

I'll try and help further when I can, but hopefully this helps for now. The current github is for RRF, but if you go backwards in the history, my RRF2 config is there too.

Carlo

@NexxCat

Thanks NexxCat I started with yours and Jay_uks and your carriage adaptors Thank you for posting them.

It isn't to far away the effector is being lifted away I've tried baby stepping but it just digs up whats been printed honestly after 4 weeks constant not sure there is any combination I have not tried and I'm really not exaggerating
So today the machine is over 6mm shorter than this afternoon according to the calibration?

NexxCat

Have you checked that the probe is actually returning repeatable values?

Create a macro with the following in it:

G28
G30 P0 X0 Y0 Z-9999
G30 P1 X0 Y0 Z-9999
G30 P2 X0 Y0 Z-9999
G30 P3 X0 Y0 Z-9999
G30 P4 X0 Y0 Z-9999
G30 P5 X0 Y0 Z-9999
G30 P6 X0 Y0 Z-9999
G30 P7 X0 Y0 Z-9999
G30 P8 X0 Y0 Z-9999
G30 P9 X0 Y0 Z-9999 S-1

What does the console report after the macro runs?

Carlo

@NexxCat

I’ll try it first thing tomorrow and post back the results.

Thanks
C

Carlo

@NexxCat

Hi,

Results of the macro below.

M98 P"0:/macros/Probe test"

G32 bed probe heights: -0.078 -0.065 -0.065 -0.103 -0.090 -0.078 -0.090 -0.090 -0.090 -0.065, mean -0.082, deviation from mean 0.013

Thanks for help

NexxCat

@Carlo Your repeatability isn't as low as mine, but it still seems well within reason:

M98 P"0:/macros/ProbeRepeatability.g"
G32 bed probe heights: -0.036 -0.036 -0.042 -0.036 -0.036 -0.042 -0.036 -0.042 -0.036 -0.036, mean -0.038, deviation from mean 0.003

When you run the delta calibration, are you doing it with the bed and nozzle cold or hot? My macro for calibrating heats the bed to 60C and the nozzle to 130C before it calibrates.

droftarts

@Carlo If you're probing the bed with heaters on, turn them off by adding B1 to your M558 command. To quote @dc42 here https://forum.duet3d.com/post/97669

The reason we keep the heaters off during the recovery time is that some probes (e.g. BLTouch) need to recover from the effects of the magnetic field generated by the bed heater.

Ian

Carlo

@NexxCat
Hi NexxCat,

Hot..220c hotend 70c Bed....I'll set the heat as you have mentioned.

Thanks

B1 added Thanks droraft.

Carlo

@NexxCat

Hi,

Results at 110c and 60c

M98 P"0:/macros/Probe test"
G32 bed probe heights: 4.041 4.053 4.041 4.066 4.041 4.053 4.053 4.053 4.053 4.066, mean 4.052, deviation from mean 0.009

Thanks

Danal

@Carlo said in Another Predator:

0.009

That's pretty darn good. Some mechanics (switch and/or overall printer) may do better... but that's a very good number on its own.

NexxCat

@Carlo said in Another Predator:

@NexxCat

Hi,

Results at 110c and 60c

M98 P"0:/macros/Probe test"
G32 bed probe heights: 4.041 4.053 4.041 4.066 4.041 4.053 4.053 4.053 4.053 4.066, mean 4.052, deviation from mean 0.009

Thanks

That's an excellent result, though I'm curious as why the probe heights are now in the positive and last night they were negative. That offset would certainly explain why nothing is sticking to the bed.