Automatic Bed leveling

Re: Automatic Bed leveling

Postby NeoTheFox » Fri Apr 18, 2014 5:56 pm

@Nicholas Seward, what about kossel? It uses Z probe, and you stated before GUS only needs a sign change to run rostock firmware?
NeoTheFox
 
Posts: 92
Joined: Tue Jan 28, 2014 5:54 pm
Location: Moscow

Re: Automatic Bed leveling

Postby Nicholas Seward » Fri Apr 18, 2014 6:09 pm

@NeoTheFox: The math doesn't scare me but having an offset probe introduces many problems. Now you have to measure the xyz offset of the probe relative to the arm that rotates with the effector. That can be done. You also have to have a probe retract mechanism. That can be done.

Now the bad news, there is no quick way to make sure the xyz offset hasn't changed since you calibrated last. You also have another point of failure with the retract mechanism.

With a hinge mount for the effector and one FSR, you remove all need for calibrating the xyz offset and you don't need a retract mechanism.

All that said, a servo actuated z probe is very cheap, well documented, and easy to understand. I think it is a great solution. I personally want a cleaner solution because I obsess over little details (that are sometimes not important).

EDIT: If you implemented a z probe on GUS and shared it, that would be fantastic. Assuming you have a flat surface so you can ignore offsets altogether, this is a great drop in solution.
Nicholas Seward
 
Posts: 738
Joined: Mon Nov 25, 2013 10:41 pm

Re: Automatic Bed leveling

Postby see3d » Sat Apr 19, 2014 1:05 am

Here is a variation on a theme using a load cell from a kitchen scale:

https://groups.google.com/forum/#!topic/deltabot/2fjnBXNVrew
see3d
 
Posts: 67
Joined: Tue Dec 03, 2013 1:24 am

Re: Automatic Bed leveling

Postby Nicholas Seward » Sat Apr 19, 2014 4:44 am

http://www.robotshop.com/en/micro-load-cell-0-78-kg.html

I like this load cell. I can imagine mounting that into the GUS effector. Too bad you have to do some custom electronics to use it.

That is the beauty of the FSR. You only need a comparator and a variable resistor to get it to act like a normal endstop.
Nicholas Seward
 
Posts: 738
Joined: Mon Nov 25, 2013 10:41 pm

Re: Automatic Bed leveling

Postby see3d » Sat Apr 19, 2014 2:54 pm

If I am reading the data sheet correctly, that 780g load cell would output about 1V differential full scale with some unknown DC offset voltage in the 1.5V-3.5V range. It seems that it could be connected to an analog uP input much like a FSR. It could go into the main controller, or a tiny Arduino board (<$10) to operate like a limit switch. No custom circuit board would be needed in that case. It should be worth an experiment if the load cell is mechanically ideal for GUS or Wally.
see3d
 
Posts: 67
Joined: Tue Dec 03, 2013 1:24 am

Re: Automatic Bed leveling

Postby tbcarlson » Sat Apr 19, 2014 10:43 pm

I don't understand the desire to use load cells or FSRs and potentially additional electronics instead of a micro switch. I know someone else on this forum mentioned a hinged hot end that is used directly as a z-probe, but I couldn't find the post. Here is a video showing someone that implemented the idea for a delta.

Is there an advantage in sensing a variable level of force rather than the off/on you get from a switch?
tbcarlson
 
Posts: 6
Joined: Fri Mar 14, 2014 6:46 am
Location: Colorado

Re: Automatic Bed leveling

Postby Nicholas Seward » Sat Apr 19, 2014 11:31 pm

It is a trick to get a setup that is nice and rigid for printing and sensitive to a light touch with a limit switch.

I like the FSR or load cell because they are essentially a solid state solution. I can have a very rigid effector while it still sensitive to the lightest touch. With a simple turn of a pot we can get it to respond to light touches. If we have an ADC on the controller we can do all sorts of extra fanciness but that is not required. I would also contemplate using this for a digitizer
Nicholas Seward
 
Posts: 738
Joined: Mon Nov 25, 2013 10:41 pm

Re: Automatic Bed leveling

Postby daftscience » Thu Apr 24, 2014 7:26 am

Have you seen anything like this? http://www.thingiverse.com/thing:161753
[youtube]UPeNtwEEcww[/youtube]
I'm using it on my prusa, it's a hall-o sensor that uses two magnets.


Code: Select all

[ ]     <-magnet one
----
 *      <-- point where magnet 1 will retract the probe

 *      <-- point where magnet two is detected

[ ]     <-magnet two
 |
 |
 |      <--steel probe
 |



That has a fixed offset, that shouldn't change. It will auto retract automatically when the nozzle goes to z=0. To deploy it you just need to get the second magnet to touch a stationary magnetic something while the nozzle moves up a bit.
daftscience
 
Posts: 4
Joined: Fri Mar 28, 2014 4:47 pm

Previous

Return to GUS Simpson

Who is online

Users browsing this forum: No registered users and 1 guest

cron