calculating bearing radial load 2

[nnjunger]

I am having a bad day evidently. I have forgotten how to run calculations for radial load on a bearing. Does any one have a good reference for this type of work or guidance along the subject? Thanks in advance.

 

[enirwin]

Most bearing catalogs have pretty good information in their catalogs. Timken has general engineering info online at:

http://www.timken.com/industries/torrington/catalog/

INA has very good information in their catalogs, but I'm not sure if its available online.

 

[diamondjim]

What kind of calculations are you
trying to run, static or dynamic
capacities? AFBMA has formulas
to calculate these. Or are you
asking how do you calculate the
radial forces for your application?

 

[nnjunger]

I am looking for both. At this location, as well as others that I have worked at, bearing application has been a sloppy undertaking to say the least. There is a lot of trust put into salesmen to aid in specifying the proper bearing, size, etc., without having the engineering (load calculations, etc.) behind it. In addition, a lot of what I have seen go on is the "we have always done it that way", or "that should be good enough" or lastly "bigger is better". A particular example that I have currently involves a vertical shaft mixer, with a rotor above the fixe bearing. This rotor spins, causing our mixing action in a slurry. The machine came in, and has been running, but we have experienced bearing failures, we do not have the engineering information and assumptions that were made during design, but I am not sure that there was much true "engineering" since it would not be required. The vendor could sell pretty much what the wanted, since it is not for public use, and is just a machine. I want to verify the loads actually experienced on the bearing, but my skills have gotten rusty after so many years of doing project management and other tasks.

 

[diamondjim]

The hardest part of any application is
determining the true loads and duty cycles
that the manufacturer needs to select the
proper bearing. This plus the size restraints
determine whether to use a ball bearing,
cross roller bearing or a combination
bearing. You should contact several bearing
manufacturer's. If you have the correct
application data ie loads, duty cycles, etc,
then they can give you the expected life as
well as the static safety factors. Try,
INA, KAYDON, PSL, ROTEK, GEAR PRODUCTS, and
check for other manufacturers of heavy
industrial bearing suppliers. They do the
load calculation for free. If you know the
bearing manufacturer of the bearing you are
using, they should be able to give you free
information. If you do not feel that they
are reliable, then check with the others.
Also check the application guidelines as to
how to mount the bearing and also the lubrication
guidelines to make certain that you are in
compliance with their recommendations.
Not an easy task. Sometimes simply adding
a pilot or spigot support can help keep the
bearing loads more uniformly supported.

 

[vanstoja]

Turbomachine bearing loads depend on the force generating mechanisms acting at shaft rotating components, the distances between force-generating components and the number and spacing of all the bearings supporting the shaft. For your vertical fluid mixer application supported by a single fluid film (or rolling element??) bearing, if the driver is a motor then there are probably two more REBs in the motor giving a 3-bearing supported coupled rotor shaft having hydraulic loads at the mixer impeller, motor electromagnetic forces in the motor airgap and possibly shaft unbalance forces acting at either or both the impeller and motor ends of the shaft. It becomes a beam loading problem once the hydraulic, electromagnetic and shaft unbalance forces are defined in magnitude and location. Motor magnetic force calculations need motor parameters like airgap flux density and gap dimensional data. Hydraulic forces on the mixer impeller depend on impeller geometry which is likely to be of axial flow configuration. Calculation of impeller hydraulic forces (both radial and axial forces need to be considered) is not a trivial task nor is it certainly amenable to hand calculations with reliable formulas. Computational Fluid Dynamic (CFD) calculations are probably best for impeller hydraulic forces. A method of integrating CFD hydraulic force calculations with a shaft rotordynamic analysis is given by Berger,T. etal, 2003, "Fluid-Structure Interaction of Stirrers in Mixing Vessels", ASME J.Pressure Vessel Technology, Vol.125, November, pp.440-445.

 

[nnjunger]

The rotor is supported by the clamping forces of the bearing (flange bearings), mounted to the shaft and then bolted to the frame. The driving mechanism is belted. There are two bearings on the shaft.

 

[Hippo41]

Sounds like your vertical shaft is supported by 2 4 bolt flange mounted ball bearing units of the UCF style. That would be the norm in my experience for this type of machine.

These are too often used as they are cheap and easy to fit.Your shaft support also dependant on the grubscrews in the inner rings and nothing else.

You don't appear to mention the type of failure, but contamination would be a consideration, as well as the axial and radial loads (of course). And you don't offer much detail with regard to size of impeller, shaft size, speeds etc. So what I say may be slightly off course.

Once you have either estimated or calculated the forces involved, consider using flanged bearing units that have either small double row self aliging ball bearings or even small spherical roller bearings in (use sealed ones even inside the sealed housing if contamination is a significant problem); step the shaft to ensure proper transfer of the axial forces to the bearings and ensure for adequate lubricant (to keep contaminats out more than anything). Retain the bearings in place either by proper interference fits (with stepped shafts) or use adapter sleeves.

If your impeller is really big you may have to use a spherical thrust bearing to support the shaft - but that might be 'over the top' from what I understand from your postings.

Hope this helps.



Lester Milton
Telford, Shropshire, UK

 

[nnjunger]

Let's see, I apologize for not providing all the information on the first post. The rotor diameter is 45", the shaft is 4" at the top, stepped to 3 15/16 at the driven end (don't ask me why, they probably saved a few bucks on the sheave and bottom bearing). We do have a "hub" on the top that is machined into the shaft, so it acts like a step to prevent the shaft from falling. We have tried a couple of bearings to see if we could keep them from failing. RPM on this machine is around 285. The mode of failure seems to be twofold. One is a slight drop in the bearing, which then causes our rotor to rub since we are targetting 0.009 to 0.012" clearance under it to start with. One of our plants also has contamination issues, but I haven't personnally experienced that too much. The other failure mode for us just seems to be "looseness" within the bearing, and shortened life 6-9 months. We are sending one of the bearings to the OEM for analysis, so when I get a report I will share it with everyone. We have experienced problems with the set screw issues, and have even milled flat spots on the shaft for these to seat on. The latest bearings are spherical roller bearings, and they have an adapter to take the weight of the shaft.

 

[Hippo41]

Reducing the shaft to 3.15/16" is as near as 100mm as anything - maybe to clean up the shaft surface, maybe to make metric bearings useable - maybe more available.

Impeller size and speed don't concern me at this stage, but your need to control the clearance would make me move away from ball bearings and spherical roller bearings (which will inherently have some axial clearance) and look at either tapered roller units or angular contact ball bearings that I could set up in a preloaded system and control the clearance. This could be done in the 'top' housing (drive end) with a simple NU cylindrical roller at the bottom for radial load/ guide purposes only.

If that clearance is critical I don't see any other way around it.

4" tapers would handle whatever load you have - I wouldn't waste too much time with huge calculations.


Lester Milton
Telford, Shropshire, UK

 

[nnjunger]

Thanks for the insight. I have also been in contact with the bearing vendors to see their recommendations, but it is always nice to have a way to check their work and make sure all the assumptions that they make are reasonable.