Figure1—Differential hub cracking on high load test due to wear of material from insufficient press fit.

Welcome back to Part 2 of our inner ring and creep discussion. We left off with our creep calculation resulting in a 10.5 µm minimum inner ring fit to avoid creep. For the sake of making clean dimensions, let’s call it 10 µm on the lower end and the upper end is simply whatever your manufacturer can hold. For the 6205, something around 20–25 µm tolerance range would be expected. Often this can be accomplished with hard turning alone. Surface finish is not too critical here. Most bearing manufacturers will recommend a 0.8 Ra for small bearings and 1.6 Ra for large bearings, without clearly defining what a large bearing is. The main consideration for surface roughness is that if the surface is too rough, the peaks of the material will plow off during the bearing press resulting in less of a press fit than intended.

Figure 2—NSK Rolling Bearings E1102k c.19.

Now we consider our shaft tolerance range of 10–20 µm. If we review the standard shaft fit chart in Figure 2, we see that this custom range falls between m6 and n6.

If we review our standard fit table in Figure 3, you will notice on the load condition column that a Heavy Load is defined as > 13 percent of the dynamic load rating Cr. We have designed our creep loads for 50 percent of Cr, so falling near the n6 range makes perfect sense. Applying these fits is very easy. The 6205 inner ring bearing tolerance is 25 0/-0.010 mm. Bearings are usually written in negative unilateral tolerances like this. The nominal diameter (what we call it by), 25 mm is the largest diameter. We size our shaft by taking the bearing nominal bearing diameter and applying the fit tolerances. Our shaft diameter is simply 25.01–25.03 mm.

Now let’s see how we use this value for our internal clearance. For this example, we will look a solid and hollow shaft. Most of the large bearing manufacturers have a fit tool on their website. I like Koyo’s, it is easy to use and customize, they include both 3 sigma and min-max methods.

The 3-sigma method can be useful if you have confidence that your parts will follow a normal distribution. If you expect to see parts at the limits, you may prefer the min-max method which uses the true limits of the parts. You can usually rely on bearings falling within the center 50 percent of the tolerance range. If we say that we want to the 3-sigma method to stay above zero, our solid shaft falls a little off by 2.7 µm. This is an easy enough fix by dropping the shaft diameter by ~ 3 µm, but recall, that number was set for a reason, creep. This is the game we have to play. Revisit your creep calculations and see what that does for your operating load. Perhaps it only affects a very small portion of your cycle, or it may cut into a high load test that you need to avoid creep on. Effective clearance includes temperature conditions which we aren’t worrying about right this minute.

Figure 3—NSK Rolling Bearings E1102k Fit Table A85.

Figure 4—koyo.jtekt.co.jp/en/calculation/operating_clearance/?pno=6205

Figure 5—Internal Clearance Results for solid shaft.

Now I take this exact same calculation and add a 15 mm through hole to the shaft (5 mm wall thickness if you prefer) and look how dramatically the clearance changes. I get almost 7 µm of internal clearance back due to the shaft compliance. Recall in the previous article, I mentioned that I often assume a 1 class higher fit for hollow shafts—this is the reason.

Figure 6—Internal Clearance Results for hollow shaft.

Ok, I’m going to leave you with that to stew on for this article. I have a blast with shaft and housing fits; I find this an enjoyable part of the job. It is often challenging, there is often not a perfect answer. You will find yourself with simultaneous problems of not hitting your creep numbers and having high hoop stress and pushing your clearance to the limits. I hope I can talk you into enjoying this process rather than losing sleep over it. Think of this like being a meteorologist and learn to confidently say that we have a 50 percent chance of rain. Until next time, enjoy!