We ran ceramic bearings on all four corners of the Formula Car. May have been a slight advantage due to less friction but in my opinion not worth the big expense. Metal bearings are a lot more forgiving and last longer. Was a nice thought but no real lap time advantage.
Now if you are going for land speed record at Bonneville over 300 mph or racing a car over 175 mph for hours, then the ceramic bearing will make sense. Less thermal growth than steel (see table). This means the inner and outer bearing race and the ball bearing itself will GROW with temperature. So if the bearing components will grow 0.0001" in size for every 72° F increase, things will get real tight real quick. We have all seen the right front hub of round track cars getting red hot on a long green run. Nascar right front wheels are 1200° F! Brakes play a huge part of this so we have cooling hoses blowing on the brake rotors.
A number of things can alter the radial play during the fitting process. A tight shaft fit where the shaft is slightly larger than the bearing inner ring (often called an interference fit or a press fit) will stretch the inner ring so making it bigger. This reduces radial play by up to 80% of the interference fit. A similar thing occurs if the outer ring is a tight fit in the housing. A difference between the shaft and housing temperatures can also be a problem. If a bearing inner ring gets hotter than the outer ring, it will expand more and reduce radial play. This can be calculated as follows:
Chrome Steel: 0.0000125 x (inner ring temp - outer ring temp °C) x outer ring raceway diameter* in mm.
440 Stainless Steel: 0.0000103 x (inner ring temp - outer ring temp °C) x outer ring raceway diameter* in mm.
*
The outer ring raceway diameter can be roughly calculated as: 0.2 x (d + 4D) where d is the bore in mm and D is the outer diameter in mm.
There can also be problems where, for example, the shaft is made of different material to the bearing and housing and expands more due to a different expansion coefficient. In such a case, a bearing with a looser radial play may be needed.
A standard radial play is usually suitable and these bearings are more readily available but, sometimes, a non-standard clearance is recommended. A tight radial play is better for low noise, greater rigidity and running accuracy if the load is purely radial. A loose radial play is preferable for high axial loads as it increases the bearing's axial load capacity. It will also better accommodate misalignment between the shaft and housing.
An 80 mm diameter "loose " bearing ( high load applications) has 0.0007" play built in. Now you can see effect of thermal growth of inner , outer races and the bearings themselves.
nominal bore
|
C2 (tight)
|
CN (normal)
|
C3 (loose)
|
C4 (looser)
| | | | | |
|---|
Over
|
Incl.
|
Metric .001mm
|
Inch .0001”
|
Metric .001mm
|
Inch .0001”
|
Metric .001mm
|
Inch .0001”
|
Metric .001mm
|
Inch .0001”
|
_
|
10
|
0-7
|
0-3
|
2-13
|
1-5
|
8-23
|
3-9
|
14-29
|
6-11
|
10
|
18
|
0-9
|
0-3.5
|
3-18
|
1-7
|
11-25
|
4-10
|
18-33
|
7-13
|
18
|
24
|
0-10
|
0-4
|
5-20
|
2-8
|
13-28
|
5--11
|
20-36
|
8-14
|
24
|
30
|
1-11
|
0-4.5
|
5-20
|
2-8
|
13-28
|
5-11
|
23-41
|
9-16
|
30
|
40
|
1-11
|
0-4.5
|
6-20
|
2-8
|
15-33
|
6-13
|
28-46
|
11-18
|
40
|
50
|
1-11
|
0-4.5
|
6-23
|
2-9
|
18-36
|
7-14
|
30-51
|
12-20
|
50
|
65
|
1-15
|
0-6
|
8-28
|
3-11
|
23-43
|
9-17
|
38-61
|
15-24
|
65
|
80
|
1-15
|
0-6
|
10-30
|
4-12
|
25-51
|
10-20
|
46-71
|
18-28
|
80
|
100
|
1-18
|
0-7
|
12-36
|
5-14
|
30-58
|
12-23
|
53-84
|
21-34
|
Less temperature disfussion ( runs cooler ) , lighter weight per bearing.
Make sense?