Precision Levels – calibration

[NJHParticipant @njh]

[NJHParticipant @njh]

Hi All

 

A while ago, having bought a new lathe, I decided that I needed a better spirit level than the one used for putting up cupboards etc.

A trawl of our favourite internet auction site produced many contenders at widely varying prices. I finally settled on one (ex- military I guess) which is certainly far more sensitive than any others I’ve seen and not too eye-wateringly expensive. Anyway I used this to set the machine and all is well.

However, following a recent thread here, I wondered how do I ensure accuracy?

I don’t have a “known” level surface and I don’t know the state of calibration of my level. This link may be of use to any who face a similar problem.

 

Regards

 

Norman

 

[JasonBModerator @jasonb]

Put two screws in a bench or solid surface, place level on screws and adjust one screw till the level reads “level” swap it round 180deg and it should still read level.

 

J

[Nicholas FarrParticipant @nicholasfarr14254]

Hi, precision levels are very sensitive to temperature change, the higher the resolution the greater the effect temperature changes will have on it. Before adjustments are made it is recommended to have the level, and anything that you are using to calibrate it with, to be standing in an environment with a stable temperature (around 20 C usually) for a couple of hours. When using them they should also be stood in the vicinity that they are going to be used in, and out of any storage box, for at least a couple of hours, to make sure that they are of the same temperature as whatever you are trying to level. Handling them during these periods should be kept to a minimum, so as not to change the temperature from the environment due to your body heat.

 

Don’t assume that once you have made a precision level to read level, it wil stay that way, they should be checked and recalibrated from time to time.

 

Regards Nick.

[mgnbukParticipant @mgnbuk]

At my previous employment the levels were sent out annually to be checked over and adjusted as required – bases checked for flatness & burrs – box levels additionally checked for squareness – and vials checked for sensitivity. The company we used – Wharfedale Gauge in Skipton – was taken over by LS Starrett & subsequently ceased general calibration services

 

In use, the level was always checked both ways round to ensure it had not been knocked or disturbed since last use & “tweaked” as descrbed in the Level Developments leaflet you linked to.

 

Regards,

 

Nigel B.

[Clive HartlandParticipant @clivehartland94829]

It is now standard procedure to send optical levels and other levelling devices to an Instrument Workshop for calibration before they are used for survey or industrial measurement or building work.

Most of my work load is calibration and certifying that the instrument meets the manufacturers specification.

An example of cost for calibrating an Electronic Total station is about £500. If repairs are needed then that is in addition.

A small 8sec. optical level is about £45 for calibration.

We have to have extensive test equipment totaling many thousands of pounds and to have PC’s with the software to read the Electronic programming in the instruments.

In our workshop we have five people and all of us have at least 25 years experience in instrument repair and all its ancillary requirements.

Then, our test equipment has to be sent for calibration as well. Dvm’s and Frequency meters and Temp. gauges plus all our torque wrenches and dial gauges.

We also have an annual inspection from the main importer of the instruments we repair.

Looking at the cost, this is based on the time to do the calibration procedure.

 

Clive

[Les Jones 1Participant @lesjones1]

Hi,

Some time ago I thought what influences the sensitivity of a spirit level. I concluded that it was the radius of curvature along the length of the tube. (Or the radius of curvature of the surface that was at the top if the tube was barrel shaped rather than part of a large circle.) I concluded that the larger the radius the more sensitive the level would be. So I thought that starting with a straight length of glass tube with the ends clamped down and a strip of paper under the middle would give a very large radius of curvature. This assumes that the curve is within the amount a length of glass tube can be bent (Cold) without breaking. My question is are the vials in good spirit levels part of a circle or are the straight but ever so slightly barrel shaped ?

 

Les.

[Clive HartlandParticipant @clivehartland94829]

The glass vials/tube are honed out internaly by very experinced workers.

I seem to remember that the radius of curvature of the inside of the tube is some 60ft.

Tubes are filled with various mixtures but usually Ether or a mixture to slow the bubble down a bit.

The tubes come in various ‘Speeds’ from 10″ onwards. Illumination is sometimes made using Tritium tubes under the vial.

There is a move now to composite bubbles where two metal ends are glued to the tube and it is filled through a small hole in one of the metal ends and then plugged and sealed.

The ends have a propensity to snap off at the glue joint. The mounting points being the metal ends.

If you are making a spirit level for use in the workshop there is no need to have a full length contact of the base.

Two bearing surfaces of an inch or so at the ends and the center portion relieved about a

sixteenth of an inch.

 

Clive

[Les Jones 1Participant @lesjones1]

Hi Clive,

Thanks for the answer. I cannot think of how to accurately how to shape the inside of a tube to a barrel shape with a 60 ft radius of curvature of the side. On the way I was thinking of forming a curve in a length of glass tube if my calculations are correct for a 6″ length of tube the centre would need to be packed up about 12.5 tho to produce a 60 foot radius of curvature and from centre to end would show about 0.24 deg.

 

Les.

[Les Jones 1Participant @lesjones1]

Hi Rob,

The idea for the suggested method came from an article I remembered in a copy of “Electronics &Wireless World” (Jan 1988) I was a design for a tilt meter that could detect the movement of the earth’s surface caused by the moons gravity. It consisted of two shallow dishes containing mercury (90 mm dia) connected by a thin tube. The dishes were space 750 mm apart. Each dish had a metal plate just above the surface of the mercury. When the unit tilted the distance between the mercury and the plate increased one end and decreased the other end. If Gray has not shown you the article and you would like to see it PM me with an email address and I will send you a copy.

 

Les.

[blowlampParticipant @blowlamp]

Posted by Les Jones 1 on 23/12/2011 15:06:22:

…if my calculations are correct for a 6″ length of tube the centre would need to be packed up about 12.5 tho to produce a 60 foot radius of curvature and from centre to end would show about 0.24 deg.

 

Les.

 

 

I wonder if we could be making our own precision levels?

Those figures sound like they might be within the bending tolerance of a piece of glass tube.

So it might be worthwhile for someone to set some tube on a frame and jack it up in the middle a few thou’ to see if this could work.

 

Martin.

[Billy MillsParticipant @billymills]

Perhaps you could mount the tube to be honed against a bar with the centre of the tube shimmed out and the ends clamped. Hone the tube then fill and use. It is easy to bend a hard glass tube by the required amount without damage. Once the honed straight tube is released from the clamo it will then have the honed bore.

 

Merry Xmas All, May Santa get you some nice tools.

 

Billy.

[Billy MillsParticipant @billymills]

Justfollowed Bog’s link then found a reference to MEW 10, 33, 36 and 77. Had a look at issue 77 ( page 6 archive Oct 2001) in the digital archive, the artice tells you about making a honing jig and a small angle calibrating device, all from stuff we might have in the workshop so it’s already been done. Vial sealed with brass lug and epoxy, filling by hole in the plug sealed with a screw and washer. Nice article.

 

Billy -on the level!

[Les Jones 1Participant @lesjones1]

I’ve just realised I made a mistake working out the thickness of the packing under the centre of the tube. It should be about 6.25 thou. Billy, I like your method of honing out the bore of a length of tube.

 

Les.

[Clive HartlandParticipant @clivehartland94829]

A lot of Vial tubes are used in two positions,standard level and inverted!

There are ‘Turn over level’s used for drainage layout and Optical plumbs are sometimes inverted for Nadir setting.

Honing the Vial tube depends on the sensitivity you require. The shallower the arc the faster the bubble will travel and the more sensitive it is.

There is the fact that a commercial Vial will be accurate whereas a home made one will be literally any reading!

I think I would rather buy one of a known sensitivity.

 

Clive

[Sub MandrelParticipant @submandrel]

Somewhere in my piles of old MEs, MEWs and a few EIMs there is indeed an article on making a precision level that depends on gently bending a piece of glass tube.

 

You obviously need an accurately sized tube, but on the other hand the less the bend, the more sensitive the level.

 

You also need flourisceine(?) to colour the spirit – I have some cheap ‘bubbles’ and all the colour has bleached out of them over less than a decade. I used two in a bit of 1/4″ brass to make an LP turntable level – two slots at right angles, vials fitted with slow set epoxy and aligned by hand using the ‘turn around test’ – but then my kind of hi-fi doesn’t need pinpoint accuracy.

 

Neil

[Billy MillsParticipant @billymills]

Neil,

You can colour the liquid- which could be IPA ( the alcohol not the beer ) or use a coloured card underneath the tube. The MEW 77 article is a great deal better than has been given credit for. The method is based on one developed by the NPL in the 1950’s and is most definitly not Mickey Mouse. The glass tube is internally honed on a steel rod which has been flexed into an arc set on a bending jig. After honing the tube will have a barrel shape, if the honing is done with reasnoble care then there is NO reason why it should not be BETTER than a commercialy made vial. The honing makes sure that the bore is round, the barrel shape makes the sensitivity independant of axial rotation.

The article also describes an easily made calibration rig which is very simular to a sine bar. The final section has a short section on calibration.

 

My comments on amateur accuracy are based upon experience of telescope mirror making, many home made mirrors are very much better than commercial mirrors because of the time element in perfecting the figure. If you want to get obsessive about shaping the bore then you can test/rework/retest untill you have as good as you want or need.

 

The most expensive telescope mirror in the world was figured by experts at very great cost and all of the high tech available. They were so sure of their abilities that they never tested the figure, had they done that they would not have left out the bit that every home mirror maker understands. The mirror is now corrected by a lens, the telescope is called Hubble.

 

Billy.

[Author]

Posts