gross surface finish quantification

[Tmoose]

We just received three 8000 lb cast steel hubs from a supplier.
Each hub has a 28" long machined Ø760mm/Ø29.92" bearing bore.
Our incoming inspection determined the bores' sizes and roundness are acceptable using large inside micrometers ( "stick mikes" ).

As the hubs were being moved by crane to the assembly area, folks noticed that, when the hubs were back lit by the shop lights, the bore machined surfaces looked mighty suspicious.

What looked like Small divots were visible. There were even some entire areas with a rippled uneven appearance.
It became pretty obvious that the bores had received a great deal of hand finishing after machining to make a uniform looking surface.

We are trying to come up with a DIY method of quantifying the surface defects to a useful degree.
The shop has determined that spraying a paint "guide coat" and then gentle block sanding to reveal high and low spots is out of the question.

Any kool ideas ?

thanks,

Dan T

 

[IRstuff]

Much depends on what your contract/specification says for surface finish, and whether, and how, the actual surface deviates from the documented requirements. Seems like your incoming inspection didn't have a surface finish requirement, or was the surface in-spec according to their inspection? Were their instruments inadequate for the measurements required?

Typical surface finish requirements are statistical in nature, i.e., some sort of root-mean-square (rms) deviations are allowable. In your case, you'd need to specify either some sort of minimum area where the rms finish must be applied, i.e., "rms of X over any 1 inch square area" or some sort of peak/valley specification.

There are numerous, and expensive tools for characterizing surfaces, starting with simple mechanical instruments all the way to laser/optical

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Gr8blu]

TMoose Is the inside bore meant to be a true "bearing" surface (which likely means there is a clearance to the other part), or is it just the matching surface to a machined shaft (which likely means an interference fit) - like putting a spider assembly on a thru shaft? Regardless of which it is, is the assembled hub-and-shaft meant to come apart again later?

My first thought on looking at the image was If the bore looks this porous, what is the rest of the casting like?. If the porosity of the casting might compromise the intent of the assembly, you have bigger issues to chase than surface finish. It would be useful to at least try to measure the depth of some of those pits/pores.

For some of our larger constructions over the years, we've used traditional "blueing" to determine fit/finish. In some cases we used the actual part that goes inside. In a few cases we juried up a fixture - basically two sheets of 20mm thick plywood glued together to make a 40mm thickness with a thin steel surface made from 0.25mm shim stock formed to match the circumference. I believe our largest diameter was around 1120 mm (44 inch), with a "fit length" of roughly 1400mm (55 inch).

Converting energy to motion for more than half a century

 

[Tmoose]

Sorry for the confusion

bearing bore = bearing seat for the OD of a couple of large roller bearings.

https://us.misumi-ec.com/linked/material/mech/IKO1/PHOTO/221302297064_001.JPG?$product_main$

https://upload.wikimedia.org/wikipe...l-roller-bearing_double-row_din635-t2_120.png

≈ .001" to .003" interference (diametral) fit
diameter tolerance ~ ±.0015"
surface finish ~ 63 µin RA

Probing with a feeler gage or wire under a straight edge held against the bore seems a logical thing to try first.
Historically our specialists are rarely keen on my science experiments, and the hubs are about 3000miles away.

 

[IRstuff]

OK, so there was a surface finish requirement, so how did incoming inspection pass the parts?

Are the surface irregularities an issue with fit and function?

Irrespective of casting quality in general, etc., one could change the spec to "surface finish ~ 63 µin RA, over any 1 inch square areas" which would at least get the supplier to sit up and pay attention.


TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Tmoose]

Several eye witness confirmed that when the hubs are on the floor the ID surfaces looked acceptably smooth.
Some of those same folks were the ones that noted the visual bore "defects" when the hubs were airborne and viewed from their nether regions.
It seemed pretty obvious that the bores had received a great deal of hand finishing after machining. We can only presume that was to make uniform looking surfaces regardless their actual geometry.

Now the trick is to determine -

If there are high spots that will deform the race geometry enough to cause the bearings distress when in service.​

If the low spots will leave significant portions of the outer race unsupported and possibly cause the bearings distress when in service.​

We won't hesitate to weld the bores on these brand new hubs and re-machine if necessary.
We rebuild assemblies like these after years of service all the time.

 

[Tmoose]

thanks all.

Dan T

 

[IRstuff]

Several eye witness confirmed that when the hubs are on the floor the ID surfaces looked acceptably smooth.

That's troublesome, given a quantitative roughness specification; did these "eye" witnesses have calibrated eyeballs?

Some of those same folks were the ones that noted the visual bore "defects" when the hubs were airborne and viewed from their nether regions.
It seemed pretty obvious that the bores had received a great deal of hand finishing after machining. We can only presume that was to make uniform looking surfaces regardless their actual geometry.

Did they actually pass incoming inspection? How were they measured? If not measured, how did incoming inspection pass the material?

Now the trick is to determine -
If there are high spots that will deform the race geometry enough to cause the bearings distress when in service.

That should be easily measured with a mechanical profilometer; but that should have been encompassed in the roughness specification, modified to include small areas

If the low spots will leave significant portions of the outer race unsupported and possibly cause the bearings distress when in service.

That would seem to me to be less of an issue, assuming the thickness of the outer race is more than 3x as thick as the largest dimension of the divots.


TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[MintJulep]

Butter the surface with "plastic clay"

https://en.m.wikipedia.org/wiki/Polymer_clay,

than screed with a straight edge.

That should leave the low spots filled, and be reasonably easy to clean up.