Electronics > Projects, Designs, and Technical Stuff
Cutting open tiny filament bulbs
(1/4) > >>
ChristofferB:
Hi! Figured it's been a while since I've made any headway in the "DIY-gas-chromatograph" department, but i've run into a very simple solution to a complex problem:

I need to mount tungsten filaments in 1/8" OD compression fittings as sensor elements. Winding 4 identical filaments, making an insulating yet gas-tight and temperature resistant mount in 1/8" space is a terrifying prospect.

But fortune struck! They make tiny cylindrical filament light bulbs with 1/8" envelope! cutting the top of one of those and mounting it in a PTFE ferrule would sole ALL the problems.
Here's the one: https://uk.rs-online.com/web/p/filament-indicator-lamps/6559198/

Issue is, cutting the top off. What's the best approach? I'll buy a whole bag of them and experiment, but I'd like to know if someone had tried similar.

I've thought of two methods: 1) score around with carbide tool, tap against sharp steel edge.
                                           2) dremel with a diamond wheel or similar, cutting it off.

with both methods, the question is if i should drill a tiny hole with a suitable tool in the round end of the bulb to relieve the vacuum.

Also, i'm not certain the glass bulb with top removed will survive the compression of the ferrule, but that's impossible to know before it's tried. maybe i'll cut back the ferrule so it compresses on the solid glass base.

Whatcha think?

Thanks for the interest and merry xmas :)

--Chris
duak:
I would try to mount the bulb first, score it and then attach a small handle to the piece to remove with cyanoacrylate adhesive.

An old glass blowers trick is to put a drop of water on the crack before breaking.  The water molecules force themselves into the crack and help make the fractured end square.  My chemistry teacher taught me this and it works.  He would use a drop of saliva as it holds itself in place.

Good luck, Merry Xmas & Happy New Year.
ChristofferB:
that is a good point, getting some leverage on the offcut end of the bulb would be good. being able to snap it over like a piece of tubing would be ideal.

Scoring around and pulling the two pieces straight in either direction, then putting the cut on a hot NiCr wire might also do the trick.

The spit trick is indeed very good, the difference between a wet and a dry snap is unbelievable.

I'm not sure how snapping a glass tube under vacuum will differ though.

Thanks!
SeanB:
Get an industrial carbide tool insert, or a mounted industrial diamond scribe. Mount the lamps in a tube with a drop of epoxy and long leads, and fill from the rear of the tube. Cure epoxy and then superglue a small length of bamboo skewer on the end to act as a lever, and score with the carbide, then break off. You will break a number of them doing so, but regard the first few as "alignment of the process line" instead.

Tip for better sensitivity is use the 28V versions of those lamps, thinner filament, faster response and they make a very good replacement for RA53 thermistors in Wein oscillator bridges if you design it to run them at around 3v across the lamp.

Got a whole little box of those lamps around, but sadly they are a mixed bag, 5V and 28v versions mixed, I was just a little careful in checking each one on 3V coin cells before installing them. glow and it is 5V, no glow but still is a resistor probably a 28V version. Front panels that would variously use a 5v one, and another string in the same panel would use 28V lamps, and they were both soldered in, and buried in acrylic resin.
ChristofferB:
That would be ideal, yeah, unfortunately the idea is that I really need the OD to be 3.1mm to crimp the ferrule around, so glueing in a tube would make them too large. Making a jig to hold them steady while rotating and having a fixed carbide scribe would really help, maybe if i temporarily shrink wrap them and then mount carefully in a drill chuck on a (very high precision) drill press, mounting the scribe sorta like a lathe tool and rotating slowly by hand while pulling at the end.

I fully anticipate to break more than I make, but considering commercial tungsten TCD filaments cost around 250 usd a piece, I think i can live with it.

Would the 28V versions also run at 12V? I'm really aiming for a 12V filament supply.

Navigation
Message Index
Next page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod