Smart Effector issue

dc42

@dc42 said in Smart Effector issue:

@tecno please answer the first question in my previous post. Also tell us what M558 command you are using.

M558 P8 R0.4 C"probe+servo0" H15 F800 T1500

G29 and tiggers on point where it should go down. Idle = NO

Try reducing the probing Z acceleration, using M201.1 if you are running 3.4.0RC firmware, or M201 if not. High Z acceleration combined with a high mass hot end create the same strain in the effector at the start of the probing move as nozzle contact does.

tecno

@dc42
M201 X1000.00 Y1000.00 Z400.00 E1000.00

Slightly worse = more jump over points

dc42

@tecno try even lower, e.g. 100.

Also, if the green LED flashes at the end of the XY travel move rather than at the start of the probing move, increase the M558 R parameter.

tecno

@dc42

R1 or R2 Still same problem

tecno

No problems with fan shroud taken off.

Phaedrux

What does the fan shroud look like when on?

tecno

@phaedrux
https://youtu.be/6YHxmoWgLI8

Phaedrux

Ah, so quite large then.

tecno

@phaedrux

Jay has been using this model without problems, so go figure.

Phaedrux

Interesting. I wonder if it's a sensitivity problem then.

I assume it's been disassembled and reassembled already?

tecno

@phaedrux

Still as in latest picture = no duct.
Total weight is 84grams with 3 fans, printed in PETG.

jay_s_uk

@phaedrux that duct design (which is mine) works fine for me on my smart effector v3 so i doubt thats the issue. I still think its a mechanical issue with effector tilt

tecno

Have found my main issue = the printed effector adapter is way too small = too short rods.

Back to the drawing board!

tecno

A bit larger adapter

tecno

One violent move and this is the result.

tecno

So some days has gone and I am now trying to get the Smart Effector to work = no go!

New non 3D printed effector adapter (original in alu with slight modifications)

Area of bed 9 to 3 down to 6 can be probed G30 and 0,0 towards back about 50mm works
Out of that area the smart effector is not sensitive enough as the effector needs to tilt a lot before it triggers.
I find this a bit odd as the trigger LED on head moves. All set to default values.

bed.g

config.g

Phaedrux

Your dive speed seems to be half the recommended value. M558 F600 vs M558 F1200

https://docs.duet3d.com/en/Duet3D_hardware/Accessories/Smart_Effector#configuring-firmware

M558 P8 R0.8 C"probe+servo0" H10 F600 T2000

M558 P8 R0.4 F1200

So perhaps it's not striking with enough force to reliably trigger.

tecno

@phaedrux

And looks like that, had to go all way up to 2500 to get reliable trig.

Where do I set hotend offset and where to set effector offset?
Looking from front X0/Y0 effector is to left and the hotend is a bit left too.

Phaedrux

@tecno said in Smart Effector issue:

Where do I set hotend offset and where to set effector offset?
Looking from front X0/Y0 effector is to left and the hotend is a bit left too.

I'm not sure what you mean.

dc42

FWIW we've identified a modification to the Smart Effector version 3 that improves the signal-to-noise ratio, which in turn allows you to use a higher sensitivity unless fan noise etc. prevents that. The modification is to remove capacitor C8. This can be done using a soldering iron with a bit large enough to cover both pads. Alternatively, you could use a small twist drill bit to remove a thin sliver of copper around the adjacent via that connects one end of the capacitor to the ground plane.

This image is as viewed from the top of the PCB, so you need to invert it when looking at the bottom.