Making a machinist level with a perfume vial

[Claudio HG]

Thanks Johan. About the fluid I've tested many fluids.
- Ethilene 90% + blue methilene (as dye): slow.
- Perfume (yes, the original perfume): quite good. I have no idea of the composition.
- Acetone 100%: excellent but it would eat up the seal in the long run,.
- Ethil-methilene 94% (ethilene+10% methilene, up to reach 94%): good enough.
For the seal I've used the original cap of the perfume vial.
About the bubble. I've noticed that the smaller the bubble the slowest the response. There is a physical explanation for this but I didn't digged much into it to check it out. So for a faster response let the bubble be generous as much as possible.
For the fluid my final choice was ethil-methilene. In short it is industrial "alcohol" with the classic pink dye.
Be careful to add a dye because it could change the viscosity of the fluid. In general aniline dyes, in very little amount, should not modify the viscosity. Blue methilene should be a good choice but I noticed that the one I bought is "already diluted" in water! That's probably why in my experiment the bubble moved so slow, I think.
An other point to keep in mind is the possible reaction of the vial's material with the fluid, i.e. if the fluid is polar. That should be not the case with glass, though.
About the seal. Sodium silicate should bond quite well with glass, but I have no experience.
Epoxy won't work at all. Not only you could experiment snags in making it to cure, but also it could even decompose in the long run. Also, epoxy in general requires almost 15°C to cure, if you put it in freezer it won't cure at all.
If the sodium silicate works to seal the nib, then epoxy could be used to externally keep in place and reinforce the sealing.