Author Options:

I need a special tool ? Answered

Has anyone any suggestions for a tool which would be "like" a camera iris, with a hole several mm long, but which can be varied in diameter from 1 (0.040 ") to 0.2mm (0.008"). I need to set the diameter fairly carefully, to just gently pinch a bundle of 70micron (0.0029"") optical fibres passed down the centre. The hole needs to be very circular and rigid, even as its closing......

Something kind of like the mechanism of a "pin-chuck" suggests itself, but wouldn't be very round in the limit.

Any ideas ?




Best Answer 6 years ago

I think this would work. Take a drill bit or other circular metal object of diameter maximum you need. Coat it with a thin layer of grease so silicon glue will not stick to it. mount the drill bit into the center of a 1/2 inch pvc end-plug that is also coated with the non stick grease. Fill the pvc endplug with silicone glue and let it dry/harden. Remove the hardend glue from the pvc endplug and also remove the .2mm drill bit from the glue. Now you have a flexible silicone round hole .2mm wide that you can squeeze using a simple hose-clamp. The hole will squeeze shut and should remain round. This should work if you only need to gently squeeze the fibers as you said. I kept trying to think of something METAL that would squeeze and retain its shape. I think that is impossible. Must use something that can be squeezed and will flex. Silicone glue might do it.

The only thing that comes to mind is a vascular stent. I'm not sure what the specific diameters are, but built with a metal mesh, they are designed to expand/contract as needed. It probably won't help you, since it's not really a "tool" per se, but if you get a hold of one, you might have the precise and gentle pinch you're looking for...

Stents are expanded by a balloon from within.

Maybe a lasso, something like an oil-filter wrench, but very small? Instead of a chain, have a strap.

I can see how that MIGHT work, if I can find a thin enough band to wrap around 200um. Kapton film ? Presumably the film has to be << the radius of the curve

How about a braded device like the kids toy finger cuffs? That device, which is woven, works by contracting when the material inside, in the case of a child's fingers, tries to slide out. Once the material is inside the tubular accordian type band that should hold everything together perfectly. The tube could be woven to any desired diameter of contraction.

Bad drawing, but maybe the strip could be something like this? The "tines" of one end go between those of the other end, and the part in the middle makes a tiny tube.

Relax the tines to insert the fibres, pull tight to straighten them up.

strap for steve.JPG

Does the grip have to be continuous?

If you need something "like a camera iris", why not use a pair of iris mechanisms in line, like gripping the bundle with two hands? I'm sure you could machine a version with enough leaves to keep the bundle circular enough, and could link both sets so that they closed simultaneously?

200um is pretty damned titchy though, and irises aren't brilliant at zero opening.

Is there an "official" tool for the job, or are you inventing something?

If you're threading, does the order of the fibres matter? Could you use something like a funnel with a very shallow taper, down to a diameter of 201um, feed the bundle down it into whatever it is you're feeding it into?

I'm making a coherent fibre bundle to feed an astronomical spectroscope I'm building -18 guide fibres, and the measurement fibre in the centre.

I can make ONE end OK, I need to draw the fibre group back 2 metres to the other end in a coherent manner, so there is still a 1-1 correspondence from one end to the other.

I had mused on the funnel idea - do you think it would be feasible to draw a glass tube down to do it ?

I think you've got the skills/equipment to do that, but how would you keep the bundle in order during the threading? Glue, and then trim it off when the end comes out of the tip?

That's right. I made an aluminium plate, and made up a test of 19 fibres bunged down the hole and glued in with proper araldite. I heated the assembly to 40 C, and the liquified epoxy ran into the hole very satisfyingly. Now I have a 2 mm high bump to lap off, for the test piece. Next, I want to make a 2 metre coherent bundle, in a stainless sheath like this.http://www.temperaturetechnology.com/documents/catalogue/SS_Armour.pdf

I guess that if you keep each end coherent, then what happens in between isn't so important.

Maybe wrap the bundle in something soft to stop it scratching on the inside of the sheath?

It'd be a lot easier, tooling permitting, to draw the ring down the bundle to the other end, then it ends up coherent by default though.

A TEFLON funnel now.....


Line your funnel with a non stick baking sheet - Stick on with double sided tape.

Tescos are the cheapest.


6 years ago

Im sure youve considered this, but cant you get the fibres bundled as you need it? what about putting a balloon kind material around it and then applying (air) pressure?

Just throwing out ideas here:

Assuming you can lay the fibers in the required order:

pass through a silicone tube, put rigid tube out side and inflate the gap.

Heat shrink perhaps

braided memory wire - a bit like shadow muscle actuators opens and closes with application of heat usually via electrical current.

With a bundle of round objects they form an hexagonal or similar array when bunched - Not very compatible with a round hole so compressing material needs to be shaped or very conformable.

Lay the fiber down in a glue matrix at each end so each fiber is fixed in position as you lay it - let the glue set before laying the next fiber - the middle doesn't matter.

Can't scratch the fibre, or the losses will go through the roof.

How about a pre-made fibre bundle? This one's a 19 core bundle, 2m long:

They have all the SMA connectors to fit. Suitable?

It's unlikely it's the right fiber for what you're doing, but the page says:

"Custom fiber bundles are available upon request. Please contact Thorlabs' Tech Support at techsupport@thorlabs.com."

At 280 ackers for the one you posted earlier... I shudder to think what they'd charge !

Sadly incoherent by design, and WAAAY to big. The fibres ideally want to be smaller than the Airy limit for the mirror feeding the spectroscope. That's why I've used a 30um core fibre for the prototype.

What are you trying to do?

How about some kind of woven tube? As you pull it it contracts. Kind of like one of those thumb trap toys but smaller and more precise.

Alternatively you could browse the fiber optomechanics section of ThorLabs for inspiration:

Hence the stent and Chinese basket ideas. I'm trying to form a 19 fibre coherent bundle. The ESO managed it for one of their spectroscopes, I've managed to terminate 19 fibres into a plate, with a microdrilled hole, calibrated to encircle 19 optimally packed fibres, with the expectation that the little buggers have litte choice but to go into some kind of order at the end.