Can anyone recommend a sensor for detecting a paper clip on a GT2 belt so it acts as an endstop? Yes, I know mechanical switches are the best and that inductive sensor can do the job, but they’re too big. I’m looking for something small like a photosensor. Has to be accurate as well of course.
You’d use an optical slot encoder. Which is what’s on the optical endstop boards. I’m sure smaller sensors can be found.
Otherwise, you could do a electrical contact like how you probe right now.
It needs to detect the metal that is wound around the GT2 belt. Not sure how an optical slot encoder would be set up for this?
Ah, wound. Thought you’d have the end sticking up like a finger which would trip it.
Trying to set up something like how SeemeCNC’s Eris is. They have a plastic piece clipped onto the belt and the switch on the bottom. But our Delta Go doesn’t have space for a plastic clip to pass through. So wrapping something like a paper clip would work well. But need a way to detect it when it passes under the bed.
What about a piece of reflective tape or paint on the belt? Like how those optical tachometers work? I have no idea how you’d detect it but maybe it’s another idea (and a larger target than a paperclip wire).
Tape would fall off. Paint is a bit more expensive for mass production and wouldn’t be accurate to manufacture I imagine. Needs to detect something solid.
@Whosa_whatsis any ideas? 
Obvious choices are mechanical, optical, our Hall effect. How tight are your space requirements?
missing/deleted image from Google+
I was thinking putting a rivet into the belt might be an idea… but the likelihood of it stretching and ripping apart…
Magnet with S/N direction in the same direction as belt. Hall-effect sensor detect wher S/N switches. Thist point is very precise and easy to detect. Its totaly uneffected by light and dust. The magnet can be weaker by time but the point for S/N will allways be the same.
Anoter advantage is that ist contactfree and will not effect the movement. The magnet can ge very small. Actually no mounting is needed. Some superglu on the belt and the mounting is done.
Here is a small european Hall-sensor…
http://www.ebay.com/itm/Precise-magnetic-HALL-endstop-limit-switch-/122405486122?hash=item1c7fef7e2a:g:pW0AAOSwsW9YzyOQ
This cab be alot smaller because the bulky part is the socket and the PCB. The sensor itself is in the size of a small transistor. If you build the sensor yourself, choose a sensor that have everything built-in. The more advanvced Hall-sensors are able to feel magnetic force and/or create a digital signal,
I have tested this version and replaced all microswitches.
remember to make the cord twisted or use shielded cables. A good choice for shielded, thin and flexible cable for endstops is to use the cord from broken inear-speakers.
Have you already thought of a reflective photosensor? White mark on your belt.
Mount a small magnet to the belt. Superglue or stick it on, doesnt matter. Then proceed like @Johnny_Linden suggested. Hall-Sensors are easy to handle…
I agree with mounting a small magnet on the back of the belt. Check the RADDS hall endstop boards and the tiny magnet they use.
@Eddy_Castro Unfortunately those are expensive 
They are worth every penny AND versatile. You can use them for filament tracking, too: https://www.youtube.com/watch?v=F1QcBbc12zc
I was thinking of using something like this around the belt since it’s readily available and cheap: https://www.aliexpress.com/store/product/Socks-aluminum-clip-General-socks-packing-clip-1000pcs-lot-free-shipping/1092398_32627249565.html but need to figure out a way to detect it using a cheap sensor that can be mass produced for under $0.50 for it to be worth it.
Simplest might just be to use your clip as a contact completing the circuit, wired up as you would an end stop switch, but without the switch.
@Shai_Schechter Under $0.50? Hard to see how that’s going to happen aside from a microswitch.