Thursday, April 02, 2009


Alternative to Opto Endstops

I have been in the process of trying to build a repstrap for about 2 months now. At the beginning of the project I decided on two main goals: 1. The whole thing had to be cheap. Being a college student, I set a budget of $300 for the entire project. 2. Knowing the original goals of the Reprap project, I wanted to try to make the entire machine from parts that you could pick up at your local Lowes or Home Depot. This would really help bring the project into the hands of your average hobbyist. Now there will of course probably be a few electronic parts that you will have to order (Arduino, motor driver, etc), but those can usually be ordered easily and shipping is much lower than the mechanical parts.

So moving onto the reason for this post, I am now nearing the end of my build. I finished the mechanics about a week ago, and the electronics are completely done short of one item: the endstops. I went ahead and ordered 3 opto endstops v2.1 from the RRRF store. When I got them in I soldered them together and tested them out after hooking them up to my electronic breadboard. They all worked as expected. So next, I went about attaching them to the machine. I'm not sure it I maybe but a little too much stress on the boards or if they are really just that finicky, but after attaching them, 2 of the 3 had stopped working. I had made sure that none of the contacts on the back of the PCB were shorting against anything metal on the machine, but I couldn't get them to work. I took them back off and resoldered the connections to the photointerrupter thinking that was the culprit. On one of the endstops I was able to get is working intermittently, but still not reliably.

With respect to those designing the generation 2 and 3 electronics, I very well may have gotten a finicky board, or just been a little too rough with my PCB, but either way I decided I wanted something a little more reliable. Being a mechanical engineer by trade I was much more comfortable with something mechanical interacting with the machine. I looked through my huge tub of spare electronic parts and came across a lever switch I had picked up from Radio Shack a while back. For reference the switch is part # 275-0016 SPDT Submini Lever Switch. The nice thing about this lever switch is that even once tripped, the lever has about another millimeter that it can travel before it contacts the switch housing. Knowing that the carriages on the machine probably wouldn't stop instantly once the switch was tripped due to the effects of inertia, this extra travel distance is a requirement if you don't want your platform crashing into the switch and throwing the alignment off. Thankfully, the switches also had 2 convenient mounting holes to I ran a twisty-tie through these holes and used that to secure it to the machine. This worked fine in testing and the switches didn't move when contacted, but I eventually added a little hot glue once I had the positioning perfect for some extra support. Here's a few pictures of the switches on various axes.

Z axis

Y axis

X axis

During testing, the switches worked fine and the platforms were able to stop before contacting the switch housing. I added a simple LED readout similar to the circuit on the opto endstop and a 1k resistor for current control. The LEDs give a nice visual indication that the switch is in fact working. I feel much more comfortable with this method after by questionable experience with the opto endstops and for $2.99 a switch, this represents an easy to find, and relatively cheap alternative to the photointerrupters.

If you are interested in my other progress see my personal project blog:

I'm using the same kind of switches on my repstrap for the same reason. I don't think you're using them the best way, though. The switches don't always actuate instantly, and a fast moving x or y axis can strike the body of the switch before stopping. Mine are also from Radio Shack, but have rollers on the end of their levers and are mounted so that the carriage of the axis strikes the roller and pushes the switch down without risking impact with the switch.
Good point Corwin. Any chance you could post a link to some pics? I am thinking that one of the first parts that I print will be a more proper way to mount these switches and I will look into positioning them in this manner so that the platform would just keep sliding over the top of the switch after it is actuated.
Can you post a pic of your x or y axis? What's in the U channel that runs on the rod?
Very nice. (I'm relieved to see someone else using hot glue as much as me).

Does anybody know if there is an advantage with the Opto-Couplers as to the repeatability of the switch position? From memory I think micro switches use an internal buckling mechanism, and I'm not sure how repeatable it is.

Could you try some homing experiments and see if you see 'missing steps' when you re-home?

Also, can I ask if you have limits on both the Upper Z and the Lower Z? From your photo it looks as though you home by moving the table upwards. It seems that if you have the limit on the lower Z then you can home at any time without the possibility of a collision with the workpiece. (I don't currently have a RepRap though, so I may be missing something).
I am using the RRRF Opto-devices for Home detection where repeatability & accuracy is required.
For my XY&Z max sensing I am using Micro Switches to prevent end stop crash. As micro switches have zero cost when you have a few old ones lying around.

Note to self must make use of glue gun.. for picture opportunity.. Integrity of the name Bodgeit must be maintained.
Pardon my ignorance but... How do you connect these switches to the 3 pins on stepper boards 1.x ?
Post a Comment

Links to this post:

Create a Link

<< Home

This page is powered by Blogger. Isn't yours?

Subscribe to
Posts [Atom]