MakerGear Forum

we're restricted? because of low post count?

I have set up a thingiverse account, the Dial clamp and a few other things I've made are there, my username is Farr0wn3d

if you have any positive or constructive feedback I'd love to hear it.

Hi All,

I'm trying to install a different type of Auto bed leveling and that by using a proximity sensor, something similar to: https://www.youtube.com/watch?v=EcGFLwj0pnA or http://www.instructables.com/id/Enable- ... /?ALLSTEPS

I have all the HW and the printed parts ready but I have some issues with my Marlin Firmware.

The Auto bed leveling requires and updated Marlin version compared to what I have but when trying to upload the latest Marlin version my Rambo board stops working. I've changed the Motherboard settings to 301 but it still doesn't work.

Any other modifications in the configuration.h file other than the motherboard?

My M2 printer is from June 2013 = 12V/19V power supply based

Help is appreciated.

Guy S

Just saw a totally different approach to bed leveling, one that's compatible with a dual extruder: https://www.youtube.com/watch?v=8hDSFuC7HkQ

^^ Excellent for leveling the bed. if this was done while everything was at operating temp, then you wouldn't even have to mess with the Z-height

sprior wrote:Just saw a totally different approach to bed leveling, one that's compatible with a dual extruder: https://www.youtube.com/watch?v=8hDSFuC7HkQ

Saw this method but it looks kinda brutal on the hotend/extruder. I guess it depends on how much pressure it takes to push down the spring

i'm going to try optical myself... less hassle, flexible for surface types, and if implemented right, could be accurate within .01mm+-. which should be good enough.
in theory you could get down as low as .001mm but, i'm not going to use a laser for this.
problem with most optical approaches i've seen is they try and adapt distance sensing approaches which is hard to get this accurate at such short distances.
I have the parts on order but probably will not get to for a week or two. I'll post once done.

innkeeper wrote:i'm going to try optical myself... less hassle, flexible for surface types, and if implemented right, could be accurate within .01mm+-. which should be good enough.
in theory you could get down as low as .001mm but, i'm not going to use a laser for this.
problem with most optical approaches i've seen is they try and adapt distance sensing approaches which is hard to get this accurate at such short distances.
I have the parts on order but probably will not get to for a week or two. I'll post once done.

Hello. Have you had any success with your optical strategy? I've been thinking about doing something similar but, as far as I can tell with the minimal research I have done, the resolution of the "non-contact" versions I could try will not be good enough.

I will try to implement the mechanical strategy documented here once I get the time. I received my M2 a few months ago but I have not been able to really work on setting it up properly. The test prints I made have been unsuccessful but I know this machine is capable of more. After all the struggles I had with the Replicator 2, including a lot of upgrades, I think I can handle really fine-tuning my M2.

The optical strategy works well for me.

it is currently costly. and frankly you would likely get just as good a result using the mechanical probe for less money

the advantage i get using the optical, is that i am using it for both homing and z leveling. that lets me not have to re set the z home or z offset with a bed swap.
but the way i am doing the sharing of z home and z probe it isn't officialy supported ... but it does work for me. plus i am running smoothie board not rambo

as i get time i am still playing with my own solution. but this one works.

innkeeper wrote:the way i am doing the sharing of z home and z probe it isn't officialy supported ... but it does work for me.

I think I understand what you're doing, but can you please elaborate (because I need to do the same thing)? And perhaps we can work on Arthur Wolf to get it officially supported. . . it's much better than having to re-integrate your code every time you update the smoothieware.