Tightest Angular Tolerance that can be Specified & Measured?

[TorsionalStress]

I would like to know, what is the tightest angular tolerance that can be specified on a drawing which can actually be measured? We can specify 0.001” on a drawing & actually measure it, what would be the case for angles?

I’m seeing some drawings with a plus/minus .003 degrees

Any response will be greatly appreciated!

 

[NomLaser]

.003° = 0°0'10.8"

So it really depends on how accurate your measuring tool is and how accurate you need the part. I have seen most nominal angular dimensions in the +/- 1/2 degree but if you need tighter then you need to figure out what you need and if you have the tools to measure it. Then determine if it can be machined with the equipment that you have.

 

[TorsionalStress]

You can put anything on a drawing, but if you can’t measure it, what’s the point?

My question is, are there devices used in the shop that can measure a difference of plus/minus .003 degrees?

Any response will be greatly appreciated!

 

[MikeHalloran]

See 'sine bar'.


Mike Halloran
Pembroke Pines, FL, USA

 

[unclesyd]

What is the envelope size of the part that you want measured?

Depending on the size, aside the above mentioned sine bar there are several possibilities, laser, tool makers microscope, or a comparator

 

[KENAT]

Agree, without giving some idea of the size of the part it's difficult to say. On a very large part you might expect to be able to measure a very small angle by measureing linear offset and then doing the trig.

As to what you put on the drawing, first is it functionally required. If yes then you look at if you can make & inspect it. If you can't make it then who cares if you can inspect it.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[TorsionalStress]

I have to machine a locating hole to fit guide pins, on a bolt pattern at a precise location, but the larger the bolt circle gets, the larger the error, therefore the tighter the tolerance needs to be specified.

Any response will be greatly appreciated!

 

[BobM3]

Do you need an angular dimension for the holes? Why not just x,y dimensions?

 

[KENAT]

Or, you could look at a more functional way of dimensioning it. You don't really care about the angle, you care about the final location of the holes. This can be expressed in polar (angle & diameter) or in cartesian (X & Y). If you do the math what is the equivalent cartesian tolerance, +-.001, .002 ...?

Have you ever heard of position tolerance, it seems this would be a situation where it is appropriate & perhaps benefficial.

http://en.wikipedia.org/wiki/Position_tolerance

If you are worried about specifying what you need rather than actually making or measureing it and want help with how to specify this then forum1103 may be a good place to look.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[TorsionalStress]

Personally, when it comes to tight tolerances on a bolt circle, I like putting ordinate dimensions. It’s cleaner, less chance of error & you can measure.

Regarding the angle, it was something that I was curious about. What do you do when tight tolerances are involved?

Any response will be greatly appreciated!

 

[KENAT]

USE POSITION TOLERANCING

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[TorsionalStress]

Even with position tolerance, you still need a theoretical center.

Any response will be greatly appreciated!

 

[NomLaser]

"Even with position tolerance, you still need a theoretical center."

Which you can measure and calculate. I would suggest talking with your machinist and find out what tools they have available to measure and inspect parts with. Everything can be measured, given enough money to spend on the tools.

 

[TorsionalStress]

I have spoken with one machine shop, they suggested plus/minus .25 degrees.

Any response will be greatly appreciated!

 

[KENAT]

"Even with position tolerance, you still need a theoretical center." yeah but the dimension locating it are 'basic', ie no tolerance. You then apply the tolerance directly to the hole.

This avoids a tolerance on the angle as such.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[mfgenggear]

yes positional tolerance
hold .001 true position
this is generaly jig bore/grind tolerance.
some cnc mills can hold it.

if you need it apply it, leave it up to the
make or buy (sub contract)person
it's typical to sub contract work out if necessary.

yes a discription of the part would be better
& fit form & function.

 

[BobM3]

I'd guess most machine shops would choose (if they had one) to measure the x,y position of the holes with a CMM. Whether you specify in degrees or not, the operator and the the machine itself measure the x and y coordinates of features with touch probes. The software in the CMM can then translate to degrees and bolt circle dimensions if asked to.

A good bench type sized machine with a good operator could measure relative locations within a few .0001"s.

 

[mfgenggear]

.001 true position = +/-.00021
angular can be specifed as a basic so can the
x, y coordinates., it seems to me x,y basic would be easier.

 

[mfgenggear]

oops math was off, .001 true position = +/-.0004
which is close enough.

 

[racookpe1978]

Wrong question.

8<)

It's not how tight you can demand.
Nor how tight you can specify. Or "draw".
Nor how tight he can measure.
Nor how tight he can reproduce that measurement.

But how tight do you need it?
(Looking the other way, how loose can you afford to let him get it?)

Low cost, fast production, high quality.
Pick two of the three.