Origin of bearing fits 1

[spigor]

The hole fits in ball bearings are not in line with the ISO fit system, beacause the tolerance of the hole is always pointed outside of the material. What is the reason for that?
Thank you in advance for your input.

 

[Tmoose]

"HOLE FIT" = bearing housing tolerance?

 

[spigor]

No, I mean the fit between the shaft and the inner ring of the bearing. The table 15.1 (link below) shows the tolerances of the inner ring hole and it is just the other way the ISO fit system is. What could be the reason for that, historical?

http://www.fltpolska.pl/pl/produkty/26-flt_tolerancje_i_luzy_lozysk

 

[Tmoose]

The shaft to bearing fit depends on the arrangement, starting with if the shaft rotates, the rpm, direction of load, and more. For most rotating shaft arrangements the race must not be loose on the shaft. "Line to line" (zero clearance) up to a few thousandths of an inch or dozens of microns interference may be required.

ISO does cover all the possibilities.

http://www.skf.com/binary/109-124478/Fig-17_tcm_12-124478.png

 

[spigor]

If we take a bearing with the inner hole dia. 30 mm then we can find the hole tolerance would be e.g. 30_-0.010
On the other hand, ISO tolerance would typicaly be in minus material, e.g. 30H5=30^+0.009
If we now examine different ISO tolerances trying to express the bearing's tolerance with ISO, we may find:
30Js5=30^+0.0045_-0.0045
30J5- does not exist
30K5=30^+0.001_-0.008
30M5=30^-0.005_-0.014

So, the best match would be 30K5, but it is not possible to express the bearing tolerance with ISO tolerance exactly. That leads to a conclusion, that some other tolerance system or rule must have been used. The question is, what was that?
Hope it is more clear now.

 

[Tmoose]

http://www.ntnamericas.com/en/pdf/2200/tolrance.pdf

 

[Tmoose]

http://www.astbearings.com/bearing-tolerances-precision-levels.html

http://www.iso.org/iso/iso_catalogue/catalogue_tc/catalogue_detail.htm?csnumber=60356

I think a slight broadening of the definition of "ISO tolerances" might help.

regards,

Dan T

 

[Tmoose]

And note that "bearing fits" are superimposed on the bearing tolerance.

Metric bearing IDs typically are toleranced "opposite" inch bearings, but the desired resulting fits will be the same.

 

[spigor]

By "ISO tolerances" I mean the system of standardized tolerances called International Tolerance grades by ISO 286-1:2010.
This system is apparently not used for bearing bore standarization, as shown above. Other system(s) must have been used instead. The question is- why?

 

[Tmoose]

Hi Spigor,

"Other system(s) must have been used instead. The question is- why? "

I think ISO 492 is where you should start. Not ISO 286. Or ISO 5775.

As to "why", I have no clues, let alone actual knowledge.
I find it is difficult many times just to correctly identify truly "what is" without even considering "why it is."

Per-Haps because the ABMA was founded 2 score years before ISO was even invented?

http://www.americanbearings.org/?page=bearing_timeline

1917 – U.S. Bearing Manufacturers create an informal group to aid bearing manufacturing for World War I, which led to the founding of ABMA.

"

http://www.iso.org/iso/about/the_iso_story.htm

In London, in 1946, 65 delegates from 25 countries meet to discuss the future of International Standardization. In 1947, ISO officially comes into existence with 67 technical committees (groups of experts focusing on a specific subject)."

And, ISO has aligned itself with the US and Japanese bearing manufacturers.

 

[spigor]

Thank you Tmoose for this information.
I agree it's not very important why they do it this way as long as they end up with the same numbers, but I just find it interesting.
You said inch bearing bores are toleranced opposite the metric bearing bores. So the inch specs would be "correct", and the metric specs "opposite". That and the historical timeline might serve for an explonation of the inverted metrical system.
That, or it's all just an accident...
Thanks everybody for your input!

 

[Tmoose]

"Per-Haps because the ABMA was founded 2 score years before ISO was even invented?"

1947 - 1917 = 30 not 40.
So 1.5 score, not 2 score.