Is it important for a shaft key to be ground flat?

  • Thread starter Thread starter DTM
  • Start date Start date
  • Tags Tags
    Shaft
Click For Summary
SUMMARY

This discussion centers on the importance of flatness in shaft keys, specifically regarding the manufacturing tolerances and surface waviness. The company has historically produced keys with a flatness of .0004 and is considering keys with a tolerance of .4990 +/- .0005. While the keys meet width specifications, the presence of waviness, measured at .0015", raises concerns about uniform load distribution on the key seat. The consensus is that while the waviness may not significantly impact shear strength, it could affect load bearing uniformity.

PREREQUISITES
  • Understanding of shaft key manufacturing processes
  • Familiarity with measurement tools such as calipers, micrometers, and dial indicators
  • Knowledge of tolerances and fits in mechanical engineering
  • Basic principles of load distribution and shear strength in mechanical components
NEXT STEPS
  • Research the effects of surface flatness on mechanical key performance
  • Learn about different types of fits and tolerances in mechanical assemblies
  • Investigate the role of keyways and keys in load-bearing applications
  • Explore alternative key designs, such as splines, for dynamic load situations
USEFUL FOR

Mechanical engineers, manufacturing specialists, and anyone involved in the design and production of shaft keys and related components will benefit from this discussion.

DTM
Messages
75
Reaction score
9
TL;DR
How important is it for a parallel shaft key to be ground flat? If key measures the right size but the surface isn't especially flat, will there be any negative affects?
Our company has made our own shaft keys for many years, grinding them on the 4 loaded surfaces.
For example, let's say the pulley and shaft key seats are .5005 +/- .0005 wide.
We've made ground keys that are .4995+/- .0002 with a flatness of .0004.
We now are considering purchasing keys with a tolerance of .4990+/- .0005
The keys are measuring this width in tolerance, as long as you use a calipers or a micrometer with reasonably large anvil. However, if you put a dial indicator across it, with the key on a surface plate, The needle moves about .0015". So there is waviness to the key surface, but the high points measure in spec.

My thought is that this variation has minimal affect on the shear area, so there should be little to no difference in strength or performance. However, the key will not be bearing load onto the key seat as uniformly as if they were ground keys.

Can anyone see any issue with these slightly wavey surface keys?

Pictures showing examples of key seat in a shaft, in a gear, and in an assembly.
1737153343817.png
 
Engineering news on Phys.org
If the slot width is 0.5000...0.5010, and the key width is 0.4993...0.4997, resulting in a gap fit. If we add 0.0015 to the width of 0.4990, we get 0.5005, that is, a normal fit or with a weak tension. If the highs are within the specification, it is unclear why the readings are 0.0015. Maybe +0.0005 on both sides and the error of the device? Personally, it seems to me that there is no problem. But in general, for such mating, as far as I know, a low-tension landing is used in order to eliminate backlash and impacts during launch
 
Only two sides of the key should matter.
The attached drawing shows a key which top surface must have a gap.
Are the four sides showing the measured waviness?
 
Reply
  • Like
Likes   Reactions: berkeman
A key is used to fix the phase angle of a hub on a shaft. Initially, the key will contact the keyway wherever, both will deform, until a static situation is reached. Some form of clamp or taper must be provided, to lock the hub securely on the shaft.

A key should not be in a dynamic situation, that is the job of a spline. Where the sides of the key is carrying a dynamic load, it will flog out, damaging the key and the keyway.
 
Reply
  • Like
Likes   Reactions: Rive, gmax137 and Tom.G