Measuring straightness

[geoffthehammer]

I have a straight edge approximately 38mm long by 20mm wide and 1mm thick. The straight edge is bevelled both sides and is believed to be better than +/- 3 microns for straightness (overall).

Can anyone suggest how to measure the straightness to substantially better than 1 micron both for the overall length and over shorter distances of approximately 3mm?

 

[FrenchCAD]

3D metrology I suppose

Cyril Guichard
Mechanical Engineer Consultant
France

 

[geoffthehammer]

In my experience a 3D metrology machine will not be anywhere near accurate enough although the principle is probably correct.

 

[Ashereng]

Perhaps try contacting the "standards bureau" of your country, perhaps also called "measurements bureau"?

These are the agencies responsible for maintaining the "standard" units for length, weights, volumnes, etc.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[IRstuff]

You might be able to configure a surface profilometer to run down the edge with a modified probe. Something like a Dektak:

http://www.veeco.com/appnotes/AN525 _Dektak_Surface.pdf

Don't remember if Wyko makes a physical contact profilometer, though.

TTFN

 

[RCarman]

There are coordinate measuring machines that have sufficient accuracy to do what you want. There are commercial measuring services, so you don't have to buy a machine. Probably Mitutoyo, Zeiss, Leitz, and SIP at a minimum would make machines that could do it.

A distance-measuring interfereometer (DMI) with straightness optics could perhaps suffice, something like a Renishaw ML10. You'd have to work out a way to guide it.

If you're using it as a straightness reference for metrology in a horizontal plane, you can always use straightedge reversal to null the straightedge error, so it would already be good enough

 

[JJPellin]

I am having a hard time picturing the part. If there is a flat face at the edge with a highly polished finish, it might be possible to check the flatness using an optical flat and a monochromatic light of the type used to evaluate the flatness of mechanical seal faces. That would give a visual indication of flatness within 1 helium light band or about 0.3 microns. Of course, if it comes to a knife edge, this will not work. Otherwise, mechanical seal manufacturers use coordiate mapping equipment that can provide a seal face profile accurate to a fraction of a helium light band. If you contacted a major mechanical seal manufacturer, they might be able to help direct you to a suitable machine.

 

[MikeHalloran]

I doubt you can hold that ruler straight to one micron.

Mike Halloran
Pembroke Pines, FL, USA

 

[IRstuff]

The Dektak profilometer can measure surface roughness over several inches

TTFN

 

[Cockroach]

Laser, it follows a true straight path. Dial it in with a scope on the recieving end.

Well, true being straight means in the absence of high gravitational field, which I think you would qualify.

Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada