Shop made Machinist Level

[tornitore45]

Machining levels are expensive and hard to justify for a hobby shop.
The high sensitivity is due to the fact that the vial has a very large curvature radius.

I wander if one could be built by flexing a glass or plastic tube.
Clamping down the ends while shimming up the center a few thousand should achieve a serviceable curvature.
Any thought on this?
Any data on how flexible a glass tube of the proper dimensions (3/8 dia 6" long) can be?

[John Hasler]

It certainly should be possible to bend glass tube that much, but you'll need precision tube.

Physical properties of glass: http://uk.saint-gobain-glass.com/appcom ... 42/711/378

[WesHowe]

You might also explore using a plumb bob to level with instead. I don't have a picture, but here is a video of that being done:
https://www.youtube.com/watch?v=2qIdsnl5vpg&t=16s
The needed skills set to make such a rig are probably already in your "kit".

- Wes

[SteveM]

You might want to contact Starrett's parts department and ask how much the vial from the series 98 level costs. They aren't all that expensive. The 98 has a ground vial and the 97 has one that is molded, you want the ground vial.

You can make your own tube for the vial and a base to put it on. Just make sure the tube is adjustable for height on one end.

Alternately, you can sometimes get one for cheap on bay. If you find one with a broken vial, you can buy the vial and install it yourself - they are just bedded in plaster.

Steve

[tornitore45]

Appreciate all response.
John Hasler, why precision tube? May be I can answer my own question.
The curvature is so small that any variation in in the inside diameter is going to affect and may be negate the expected arc.

[John Hasler]

Appreciate all response.
John Hasler, why precision tube? May be I can answer my own question.
The curvature is so small that any variation in in the inside diameter is going to affect and may be negate the expected arc.

Yes, exactly. I wonder if it might be possible to grind a tube, though. The exact diameter is not important, nor is concentricity with the exterior.

Or even select a piece that happens to be straight enough.

[tornitore45]

I wonder if it might be possible to grind a tube, though.

I may try reaming a Lucite tube. Just an experiment for those winter day when is too cold to touch steel.
I did some math and a 6" tube bowed (lifted) 0.002" in the middle would deviate the bubble 0.2" by lifting one end 1/2 of one thousand.

Anyway a good discussion, raising practical points I did not consider.

[pete]

The best vials are ground on that large curvature and to some impressive numbers. That's what makes them expensive. Doing some Google searches for Building a Kingway Tool or a few variations on that should get you some links to a few manufacturers of precision vials. My actual knowledge about precision glass and the exact requirements for an accurate enough level is pretty limited though. If bending the tube would work well enough I'm assuming the high end vial manufacturers would do so. A long time ago I worked at a coal testing lab so I know ground labratory grade glass tube is available and it's not very expensive. Finding a supplier that would deal with a us might be tough but I've never tried. I bought a Mit. 960 series level and your sure right, they are expensive for the amount there used.

Today and if I had to do it over again I think I'd do as Steve mentioned and find a Starrett, Mitutoyo or even one of the high end European machinist levels on Ebay that's got a broken vial and pick that up if you know replacement vials are available and then order a new vial. About the cheapest way of getting a very good level for a whole lot less. I also think a proper box level is a lot more useful than the machinist levels are and new there much more money than what most of the machinist levels sell for since it's about 4 times as hard to make an accurate one. Used they seem to sell for less most times than the more usual levels do because many don't know everything they can be used for. If you ever think you might do even some minor machine tool rebuilding then a box level would be at the top of my list. Anything with name Starrett and Mitutoyo on it gets higher prices on the used market because people know the name. Hilger & Watts is likely even better and not all that many know that so they usually sell for less. Unless you have surface grinding available in your shop or can do high precision hand scraping then obviously the condition of a used levels working surfaces are extremely important. Making one from scratch then the same would apply. (maybe)

I've been more than impressed with the changes I've seen on a well leveled lathe verses one that's not. But in a home shop I now think the accuracy limits of the level are a bit less important than what those high end levels have. There still well above a carpenters level of course. On my machines after spending hours and getting as close to absolute perfection as I can the final test cuts have always needed some minor tweaking on the adjustment screws to get the last adjustments done. Your idea of building one at home might well get you close enough so those last minor adjustments can then be done. But that's a lathe, bigger cnc bed mills etc do need leveling to some pretty fine limits before the machine even gets used. I guess it all depends on all the uses you want a level for.

[John Hasler]

I thnk that grinding the vial on a precise curve is probably necessary to make a precision NIST traceable measuring instrument but we don't need that. We don't need to know the angle to within +-10" over the entire calibrated range of the instrument. It doesn't matter if 1mm at one extreme of the range represents twice the angle that 1mm does at the other or that the sensitivity varies a bit with temperature. We just need repeatibility and adequate sensitivity so that we can be confident that when the bubble is centered the level is not tilted more than, say, +-30". It's a nulling instrument, not a measuring instrument.

[tornitore45]

It's a nulling instrument, not a measuring instrument.

I think that last statement just clinches it.

I have a brass desk top level like this one. I have no illusion of being adequate, is more like a desktop conversation piece that a precision instrument. For curiosity I center the bubble with screw I tapped into the base and then inserted a 0.002" feeler leaf and noticed an appreciable movement of the bubble.

https://www.google.com/search?q=brass+d ... nIsx_VL6LM: