Motor/Gen Shaft Keyway/Keyseat Tolerance Discrepancy Between Ansi B17.1-1967 and Nema MG 1 1

[Papabyrd]

Hey,

Just curious if anyone had run across Table 6 in B17.1 says the width tolerance of square keyways/keyseat for 0.25-0.75" wide was +0 -.002/ 0.75-1.25" wide was +0 -.003". Therefor the keys would be shrunk. There is no direction as to what class of Key is required,Class 1 or 2 on a motor shaft.

While MG-2009

4.9.2 Keyseat Width
3/16-3/4", +.002 -0
3/4-1.5, +.003 -0.

Again no reference to required Key class.

It makes sense that back in the day shrink the keys in would be better for shipping and not loosing the key. Why the discrepancy...?
Double posted in Gear/Pully forum.

Thanks,

Papa

 

[dArsonval]

Okay, I'll take a stab at this one.

The term [discrepancy] in the post caught my attention.
Merriam-Webster uses the word [Variance] in describing the word Discrepancy.

There are too many variables related to keys that would make a rigid specification for them impractical.

In a subtle way, the Key remains one of the last strongholds of actual handwork still performed by mechanics around the globe in an era of increasingly rigid specifications.

A key and key-way on an electric motor shaft serves more as a guide in positioning or receiving a "fit".
The more important shaft dimensions are the diameters of all the other mating components.

Yes, the "key" is an important element for the successful transmission of torque.

Yes, if the key is used in reciprocating or reversing loads and if sudden shock or torque is
known to exist, a tighter key is required. In this instance, a class 2 ANSI fit which requires the
key to be in immaculate condition with no burrs... would still require the key to be lightly driven in place.
There is no spec for "lightly driven".

Performing a quick Google search there is a copy of EASA's AR100-2010 posted on line.
It will remain on-line until EASA becomes aware of it. The document sums up the answer for the OP
in a sentence having only 11 words. Scroll to section 2 page 1 and refer to 2.1.3.

"... Keyseats should be true and accommodate keys to a tap fit."

URL to PDF below:

http://www.kellerelectrical.com/pdf/EASA Standard Recommended Practice.pdf

There's actually quite more about this topic that could be discussed.

An analogy may also be useful here. There are all kinds of dimensions related to the kind of seats we sit in.
Typically, there are no dimensions specified for the person doing the sitting.

John

 

[electricpete]

Table 6 in B17.1 says the width tolerance of square keyways/keyseat for 0.25-0.75" wide was +0 -.002/ 0.75-1.25" wide was +0 -.003". Therefor the keys would be shrunk.

Are you sure about that?

This link purports to show ANSI B17-1 keyway tolerances and the tolerance allows clearance (not interference).
For example 1/2 shaft, keyway width is listed as .125 / .127. That is nominal one eight, +0.002/-0


=====================================
(2B)+(2B)' ?

 

[Papabyrd]

 

[electricpete]

Well, that's just bizarre.

I looked in the standard you referenced and that table is included just the way you posted it. But it is not referenced anywhere in the text though. Reviewing class 1 and class 2 that you mentioned, it states:

ANSI B17.1-67 (reaffirmed 2013) states:
“This standard recognizes that there are two classes of stock for parallel keys presently used by industry. One is a broad negative tolerance bar stock and the other is a close plus tolerance keystock. Each is combined with appropriate keyseat tolerances to establish assemblies, respectively designated as Classes 1 and 2.”

Table 4 gives class 1 fit. Table 5 gives a class 2 fit. Neither of these have a negative tolerance on the keyseat width. Considering combination of both the keyseat width tolerance and key width tolerance, class 1 always gives 0 or clearance, class 2 may give either interference or clearance.

The keyway width tolerance numbers from EASA AR100 and NEMA MG-1 and ANSI/AGMA 9002-C14 all seem to match table 4 = class 1 fit. That also agrees with my experience (motor keys are not generally interference fit to the keyway).

I think maybe you're already aware of all this. Personally the best I can recommend is to stick with the tolerances from the modern standards NEMA MG-1 and EASA AR100 and AGMA 9002-C14. If you go back to B17.1-67, then use table Table 4 rather than Table 6.

I have no idea where table 6 came from.


=====================================
(2B)+(2B)' ?

 

[Papabyrd]

Hey ElectricPete,

I thought exactly as you'v noted it that there would be some back story on this. It's odd that ANSI B17.1 doesn't give the key class for motor shafts either.