accurately determining the amount of radial play in a ball bearing

Intlwaters

Help Support Intlwaters:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Jim Allen

Well-Known Member
Joined
Jun 7, 2005
Messages
1,622
I will post photos of the tool, with explanations, that we use to accurately determine the radial play of ball bearings used in our racing engines. We purchase WIB Swiss made bearings that are made to our specs. Since the bearings are pressed onto the crankshaft & shrunk into the crankcase a typical C-3 bearing would not work. This type system works as well as crankshafts that are mechanically fastened to the inner race of their ball bearings & crankshafts that have the inner race of the main bearing ground on the crankshaft.

Jim Allen

PB140013.JPG
 
The simple fixture in the photo accurately determines the bearings radial clearance amount. It also shows the concentricity of the outer & inner races to each other. The "V" shaped pressure pad assembly is held away from the bearings outer race with the red locking screws. The "V" shaped pressure pad assembly, which is spring loaded, is mounted with the four stripper bolts. Notice the indicator reading of .0000" has been set simply by positioning the bearings mount beneath the "V" shaped pressure pad. Also notice the red mark on the bearings outer race, at the six o'clock position, that will be needed for orientation.

More to come................................

JA

normal_Deep-Groove Ball Bearing 004.jpg
 
I can't exactly figure out what you are talking about but I sure appreciate the level of skill and knowledge.

I've been told that the bearing clearances used in marine and car model race engines are kind of loose to account for heat caused by high RPM/friction. Also that crank balancing can add considerable life to the bearings.

We have had several issues the last 2-3 years in the Pacific Northwest of getting bad fuel and boy is that hard on bearings. I had some fuel that was supposed to be 60% and hydroed over 90%. When the motor wouldn't start It cost me any chance of winning the high points year before last in UNW. The only time I ever grenaded a CMB 67 was on some of this junk fuel. If you hydroed a case it was rare to get 4 gallons that tested out the same. Some under and some over. It makes me wonder how many times I chased boat problems when the culprit was bad fuel.

The photos on your gallery page look like pieces of art for a gearhead like me
default_smile.png
. Awesome stuff!
 
Last edited by a moderator:
jim do you have hot and cold readings for your radial runout readings on bearings? also do you take measurements with the bearing pre loaded or free wheeling.
 
No & No. There are several reasons to explain this. Any aluminum crankcase engine must have a sufficient radial play amount in the cold bearings to provide enough end play or axial play amount (.005" to .006"), after the bearings are mounted in the case & on the crankshaft. This compensates for the greater growth of the aluminum front end compared to the steel crankshaft as the engine heats up. Failure to have this cold end play amount will cause bearing failure! After many precisely controlled tests we discovered that bearings with to little radial clearance failed rapidly & bearings with large amounts of radial play continued to run. Types or amounts of lubricants used had little effect on this. Gimmicks such as special cages (phenolic machined types, riveted steel types, crimped brass or steel types), do not help when the cold radial clearance is insufficient!!

All steel front ends cannot be compared to aluminum front ends with a steel crankshaft because all the parts involved in the steel front ends are made of steel! In some all steel designs the inner race of both front bearings are mechanically fastened together to the crankshaft. This allows positive control of the cold end play amount (axial end play amount), which remains the same, because all of the parts involved are made of steel!

Jim Allen
 
Last edited by a moderator:
"I can't exactly figure out what you are talking about but I sure appreciate the level of skill and knowledge."

Maybe the next group of photos will help explain better. When the red locking screws are released the radial clearance can be read on the indicator; .0018". The "V" shaped pressure pad is applying spring pressure only to the outer race's side opposite the indicator. The total amount of clearance between the inner race, the outer race & the size of the balls used is read on the indicator. Will the indicator read the same amount when the bearings outer race is rotated to the 90*, 180*, 270* positions?

Jim Allen

normal_Deep-Groove Ball Bearing 005.jpg
 
After rotating the bearings outer race 90*, while the anvil is in its locked position, another reading is taken when the anvil is released. Notice the red mark on the outer race is now at the three o'clock position. Also notice that the indicator is now reading .0008" while the "V" shaped pressure pad is applying the same pressure as in the .0018" reading. What does this mean? Does this fact & the additional readings taken at 90* intervals tell us anything about the bearing.

Jim Allen

normal_Deep-Groove Ball Bearing 006.jpg
 
I totally understand the measurements at 90 degree intervals is this telling you your bearing races are out of round or are the ball bearings are not all the same size? pre loading a bearing weather it be spring loaded or mechanically loaded eliminates most of the radial runout also, on a preloaded bearing the ball bearings roll on the race which is what you want. on a freewheeling bearing the balls tend to slide or not roll on the race. any thoughts?
 
Daryl,

In the case of the ABEC #3 bearing being tested the metal balls have been replaced with class 3 ceramic balls. These balls are round within .000020". Therefore the bearing races themselves must be out of round or the races are not concentric with their OD's or ID's. In our testing of various failed ball bearings, we discovered that the part of the bearing that fails is not the inner or outer race, but the balls themselves, which were no longer round. Replacing these out of round balls with round balls made the failed bearings good. It is also true that the balls in bearings used in our miniature, very high RPM, two cycle engines spend some time sliding & some time rolling on their races. This one fact requires bearings with sufficient cold radial clearance so that there is enough end play (axial clearance) after front end assembly. I used the micrometers shown to make comparative ball roundness measurements.

The next photo shows the reading when the bearing is rotated another 90*. The red mark is now at the 6 o'clock position. Notice that the indicator reading is .0017".

Jim Allen

normal_Deep-Groove Ball Bearing 007.jpg

001.JPG

normal_Deep-Groove Ball Bearing 009.jpg
 
Last edited by a moderator:
A final indicator reading of .001" is taken at the nine o'clock position with the same amount of "V' shaped anvil pressure. The bearing turns smoothly & it appears to be good. This is a manufactured ABEC #3 bearing that was purchased with a "special radial clearance" of .0014". However, is it defective? If the four indicator readings are averaged together, the total radial clearance amount is .001325". I have measured bearings where the four indicator readings did not vary more than .0002". Does a perfectly round bearing run faster? My testing showed that this variation does not matter, providing the bearing has sufficient radial clearance for the application. My testing definitely showed that there is no "finger feeling test" that can be used to determine the radial clearance amount of any ball bearing. My testing also showed that many of the bearings sold to modelers DO NOT have sufficient radial clearance & the only positive method that can be used to determine what you paid for is, TO MEASURE!

Jim Allen

normal_Deep-Groove Ball Bearing 008.jpg
 
Last edited by a moderator:
Excellent technical information jim, i will have to dust off my dial indicators and micrometers .
 
It is a fact, that radial ball bearing inner & outer races are "NOT" designed to support themselves. Without the press fit of the steel crankshaft into the inner race & the shrink fit of the aluminum housing on the outer race, bearing life will be greatly compromised. Doing both things will require bearings with a cold radial amount that is greater than a typical C-3 (.0002" to .0005") or a C-4 (.0005" to .0008") radial clearance amount. A C-5 (.0008" to .0011") radial clearance amount may suffice in some applications. For a hardened steel front end that uses a #6002 bearing, a special radial clearance amount of .0014" to .0017" will be necessary. A #6000 bearing in the same hardened steel front end has a special radial clearance amount of .0011" to .0015".

Jim Allen
 
Last edited by a moderator:
Back
Top