Machinist level with a perfume vial

Would it be possible to turn a sample perfume vial into a precision machinist level? I've given it a try and had some interesting findings as you will see in this video.
Precision machinist's levels are very sensitive and can detect very small slopes. The one I've made has a sensitivity of 0.01 millimeter over 200 millimeter or 0.4 thou over 7.9 inches.
Spirit levels are made using curved vials, and to make a level one should make a bended vial in the first place, not an easy thing to do. I also experimented capacitive sensors, but reverted back to the traditional technology of the bubble trapped into a vial.

[@quirty864]

You win the internet today! Bravo!

[@SciFiMind]

Nicely done. I've been contemplating the exact same problem for a few months now. Never considered trying to bend the glass tube with a shim. The idea I settled on was making a bidirectional level by lapping glass like a telescope mirror and then using that to make a level like oxtools repaired. Bending the vial seems easier.

[@EnriqueVetere]

This sounds like a pretty good idea. I wonder if you pursued it.

[@_Caedwyn]

dude the precision level of the metal working is so sexy lol great job

[@tomsuica8731]

Very labored English but very clear and not bad! Keep working hard!

[@thestupidquestionguy2223]

I clicked so fast when I saw you had a new video. Keep up the good work it inspires me to keep working on my lathe project

[@couldhaveseenit]

I love this idea! props for getting it working too, looks great :)

[@peppem94marsala]

Cool! I saw some people using liquid gas ( gpl for lighter) inside the "vile"

[@alphacapo]

I'll just buy one. A ton of work went into this. Credit given

[@gamemeister27]

Great stuff. I also love your accent. Wonderful combination of interesting content and great narration

[@AccidentalScience]

Thank you Brian, I really appreciated your comment.

[@sodster68]

Excellent work! Science and art in a beautiful tool! Thanks for sharing!

[@yagwaw]

Amazing work, thanks for sharing!

[@das250250]

Very interesting to apply a very small pressure to the vile works ..i thought it would crack before being operatable

[@AlmostMachining]

Most awesome. Fantastic job!

[@bombero34fr]

Nice idea and fine work too. congratulations.

[@HopeInProvidence]

You can buy a precision level vial from leveldevelopments for $20. I applaud the curiosity.

[@contemporiser]

I was considering it myself many times. Thnx for tips. I think copper is not the best material for this, since it's heat expansion rate is huge.

[@AccidentalScience]

Actually brass is even worse!

[@contemporiser]

@@AccidentalScience I meant brass, sorry, my brain is slowly shutting down, tough week.

[@somebodyelse6673]

Maybe this could be built on a sine bar base, to make calibration simple?

[@KravchenkoAudioPerth]

Very interesting!

[@glenking3337]

Awesome build buddy. What's next? A dial indicator from sewing machine and fishing real parts?

[@AccidentalScience]

That would be a cool idea but I'm afraid my wife would kick me out if I'd try to dismantle her sewing machine :)

[@glenking3337]

@@AccidentalScience yeah mine too but it cant hurt to think it thru.

[@Bob_Adkins]

I made one some time ago. Bought a vial online for us 15 that has around 9000:1 resolution, or ~ .001" over 9", or 1mm over 9m. Works great, but the 40 TPI adjustment screw is far too coarse. I'm going to rebuild with a 10:1 lever, making my 40TPI threads more like 400TPI. Yours is beautiful, mine is aluminum.

[@AccidentalScience]

Despite the (small) leverage even my screw is too coarse, on the other hand I find it useful to compare surfaces that are not even close to level. Thanks for commenting.

[@nathanhays1746]

I did exactly this with a shim. Wasn't perfectly stable though. Temperature changed the flex shape.

[@AccidentalScience]

Thank you for sharing your experience.

[@jeffevarts8757]

Nice video. Was this also for the lathe? Looking forward to part 5

[@AccidentalScience]

Yes.

[@piotrlenarczyk5803]

Thank you for video.

[@ScrapYardDog64]

Fantastic... Funny as well 👍

[@lennym1273]

Love the challenge, but man that is way too much work, lol...

[@AccidentalScience]

Yeah :D

[@GraphicManInnovations]

nice, so about 0.05mm over 1 M ? what thread pitch did you use for balancing?

[@AccidentalScience]

0.5mm at a approx. 70mm radius from the trunnion ...it is quite sensitive, it would have been better a finer pitch, but it is usable.

[@GraphicManInnovations]

@@AccidentalScience let me ask it in another way how many arc sec per division? the thread pitch is 0.5mm? thanks

[@AccidentalScience]

So my memory didn't served me well, the actual pitch is 0.7mm (not 0.5) and the distance from the trunnion is 103mm (not 70). Therefore one revolution of the adjusting knob causes a tilt of 0.39° or 1401 arcsec (since you like this unit :) ). The sensitivity is 0.01mm / 200mm per division, equivalent to 0.05mm per metre, thus an angle of 0.0029° (or 10arcsec) per div. (run by rise). So the adjustment knob causes a correction of a full division per each 2.6° or knob rotation, which is pretty touchy as I said in the previous comment, yet usable with some care. Because the knob has a diameter of 18mm this means that with a rotation of ~0.4mm at its circumference makes a full division tilt. Using a finer thread would have been better, but I hadn't any finer screw handy and even less the tapping tool.

[@GraphicManInnovations]

@@AccidentalScience yes that makes sense, like I said in the first comment 0.05/M i did many calculations for the thread pitch also and find it has to be around 0.25mm and even on that pitch it is very tricky coz i am using a 0.02/M vial too sensitive