Need advice detect small objects in a tube.

3 possible approaches

> backpressure -- something akin to the glaucoma pressure test pulsed pressure
> laser reflectance -- shine laser down the tube, look for reflection
> photoelectric transmission -- shine light from other end of tube, look for absence.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! faq731-376 forum1529 Entire Forum list

I was more thinking along the lines of drilling a small hole or thru hole and using a fiber optic sensor. Triggering when light is visible on the other side, but having no experience in this I need advice on components and mounting, etc.

That's basically my 3rd approach.

That's certainly doable, but you'd need:
> light shielding to prevent stray light from the detector
> possibly fiberoptic on both source and detector
> if no other optics, then tube ends need to be within 0.2" to minimize stray light

How are the tubes going to positioned laterally?
Are the tubes stopping at all?
What tells the sensor when a tube will be directly aligned with the source?

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! faq731-376 forum1529 Entire Forum list

I'm thinking thru holes perpendicular to the tube. I will probably place sensors both at the bottom and top of the tube. The top fiber switch can turn on the feed mechanism (absence) and the bottom sensor can stop the production line if the tube is near empty (absence). I assume the switch can have a built in time delay for instance the top sensor ignoring a falling primer so there is not an instant off/on caused by the falling primers past the top sensor before the tube is full.

I have seen thousands of fiber sensors, etc. yet having no experience in this I need some help with exactly what is necessary to purchase. Can you make recommendations on exact model/mfg and components necessary and perhaps how to mount. I have complete design latitude. Thanks for your input! I'm trying to decide if I want to go with PLC or solid state.

Let's answer this one first: "to go with PLC or solid state." they're both "solid-state," so there is a terminology confusion here. The only things that might be considered non-"solid-state" would be vacuum tube, hydraulic/pneumatic, or possibly, optical. But, there are no optical processors commercially available, and hydraulic/pneumatic can't really do much math.

as for the rest, sorry, I do have a day job. You should probably find a job shopper

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! faq731-376 forum1529 Entire Forum list

Cross tube fiber would work fine here. Find a local supplier and discuss details with them include stray light concerns and any other you might have.

You may be able to do this with a simple one sided retro-reflector model reducing it to only one hole in the side of the tube.

Keith Cress
kcress -

Keyence would be my friend because they have beaucoups of little tiny fiber optic heads.
I might also try a trick from the bowl feeder folks. Instead of a perpendicular through hole I would put bigger holes or slots on either side in order to make the beam axis more parallel with the tube axis. That way my sensor would see more than one object.

TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering

Oh, I didn't realize you wanted to look perpendicular to the tube axis. That makes no sense. You look DOWN the tube. How many primers or where they are in the tube is irrelevant, only their presence is relevant. That means you look to see if ANYTHING is blocking the tube, even if it's not even a primer.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! faq731-376 forum1529 Entire Forum list