[TowerTalk] Thrust Bearings

Thu Dec 6 01:05:48 EST 2007

A nice discourse on this subject.

Roger's statements support the Second Law of Life;

"There are NO solutions in Life, only different sets of problems. You have
to pick the problem you want to cope with"

Tod, K0TO

> -----Original Message-----
> From: Roger (K8RI) [mailto:K8RI-on-TowerTalk at tm.net] 
> Sent: Wednesday, December 05, 2007 8:12 PM
> To: TOWERTALK at contesting.com
> Subject: [TowerTalk] Thrust Bearings
> 
>  Sorry for the length, but when I started I didn't mean to 
> type quite this much. It just kept growing.  I hope I at 
> least got the high points accurately and didn't miss too many.
> The life of a thrust bearing:
> 
> In the ham world we have a relatively limited source for 
> thrust bearings and then the use of a thrust bearing is not 
> well understood and often they are not used properly. In a 
> typical ham installation the thrust bearing's primary duty is 
> (should be) to handle thrust in the horizontal plane (side to 
> side). From what I've seen and that is admittedly limited 
> even over 50 some years, hams use the thrust bearing to 
> support loads, but often the bearings end up doing double 
> duty within both the vertical and horizontal planes. Some 
> times they are doubled up with a relatively long, heavy duty 
> mast to get more antennas higher on one tower.  Each of these 
> installations comes with it's own good and bad points, or 
> strengths and weaknesses.
> 
> The problem with thrust bearings is service.  They could be 
> constructed of much better and stronger materials. They could 
> be precision bearings such as wheel bearings. Any one with a 
> lathe could produce retainers for readily obtainable, good 
> rugged thrust bearings with many times the load capacity of 
> the typical thrust bearing, but with just two major 
> drawbacks. They need protection and service.
> 
> Even then they are at the mercy of the installer and who ever 
> takes care of the service. As several of us on here can 
> attest, wheel bearings do not survive well or long when water 
> is present.  In automotive use they survive for hundreds of 
> thousands of miles, or many years in one of the worst 
> environments possible. The average wheel bearing probably 
> never sees service such as repacking
> 
> Commenting on the TB-3 and 4 bearings the races are actually 
> designed to handle both vertical and side thrust. The 
> bearings must be removed before you can get the races apart 
> and they drain (for the most part) more or less, naturally.  
> The split(s) in the casing are at the inner top and outside 
> bottom. The design is such that the inner race is well above 
> the bottom of the casting with the outside dropping well 
> below the inner race. So both races are covered and fairly 
> well protected from the elements. You almost have to try to 
> get dirt in the things, but Nature is quite ingenious.
> 
> These particular bearings are designed to run dry. Any dirt 
> that does blow in has to be small and light. It follows that 
> they should come out just about as easily. They don't but 
> aren't usually a problem. Greasing the race and bearings 
> actually increases the point load between bearing and race. 
> It also causes the bearing to move up and down a bit (due to 
> space from
> tolerances) more so than dry. As the both races are an 
> Aluminum casting, greasing can increase the rolling off of 
> partials due to the pressure on the roller more so than when dry.
> 
> The only direction the bearings do not provide support is UP. 
>  You can put one whale of a load on them from above or the 
> side, but _don't pull_up_ as the bearings are marginal in 
> this direction at best ( This can create major problems and 
> I'll come back to this). I think many of the problems come 
> from the marriage of the hard steel balls and the soft, cast 
> casing which has a granular structure. Under some conditions 
> the hard steel balls will peel off (or roll off) the surface 
> of the race in flakes. When overloaded this can be 
> pronounced. Sometimes this overload is unwittingly built into 
> the installation. One other problem is getting the three 
> clamping/centering bolts out of the upper housing. These 
> bolts are "cad plated" 3/8" NC and used with a locking nut in 
> the units with which I am familiar. The "cad plating" is a 
> sacrificial metal which is supposed to save the Aluminum from 
> corrosion. The problem being the cad plating is usually gone 
> within a year or two at most and then those bolts almost 
> become a permanent part of the upper housing. Cad plated 
> bolts just do not hold up well out in the elements. And those 
> bearings are a bear to change when on a long mast inside a tower
> 
> I've had TB-3's up for quite a few years with SS bolts on the 
> old tower. In the case of the new tower I just did a regular 
> installation and had the usual problems getting them out. 
> (PB-Blaster to the rescue.)  In the case of the 3 
> clamping/positioning bolts Pentrox and even the Moly 
> Disulfide grease has kept corrosion to a minimum where it's 
> never been a problem for me. The ones with the cad plated 
> (sacrificial plating) rust and for all essential purposes can 
> turn a $125 bearing into a single use item. If you figure the 
> time required to save the ones where the bolts need to be 
> drilled out and re tapped it takes little to exceed the cost 
> of a new bearing.
> 
> Now going back to the weakness in the upper direction: 
> Normally a thrust bearing never sees thrust in the UP 
> direction, but there is an installation in which they may, or 
> are likely to do so.  This happens when multiple thrust 
> bearings are used. If the mast has a different thermal 
> expansion than the tower, one of them is going to get longer 
> than the other when the temps go up and the other will get 
> longer when the temps go down. Under some conditions this can 
> produce more force in the vertical axis than the antennas and 
> mast we can put up there. Getting warmer will reduce or even 
> reverse the load on one of the bearings while increasing it 
> on the other. 
> Getting cooler will reverse the process and put the stresses 
> on the opposite bearings. In many of the cases I've seen the 
> installation would have been better using that sloppy, crude 
> sleeve bearing at the top of the tapered tower and letting 
> the rotator support the weight instead of a thrust bearing. 
> Most rotators are designed to be loaded from the top when in use. 
> Some even need that weight. Eliminating, or even using a negative load
> (lifting) can adversely affect a rotator's life. Many of 
> todays rotators are designed to support as much or more 
> weight than we'd have even with a large array above it, but 
> they are not designed for a negative load. OTOH many rotators 
> are not designed to handle the large antennas they are 
> struggling to hold or turn.
> 
> A mast or shaft is required to get from the rotator out 
> through the top of the tower and to support the antenna. This 
> greatly reduces the forces on the rotator and greatly 
> enhances its ability to handle larger antennas, but only 
> within reason. Now all of us can end up putting the logic 
> together and follow the train of thought that if we gain 
> strength by placing the rotator in the tower compared to on 
> top and we have to run a mast up to the antenna we should be 
> able to go with a longer shaft and put up another antenna 
> without exceeding the previous rating which might have been 
> adequate. We all know these towers are built with a safety 
> factor so we should be able to operate in that safety factor 
> zone without much of a risk. Right? Ask the manufacturer why 
> they put that safety factor there.
> 
> Well, we've gone from the logic of an extended mast and 
> another antenna and we've all seen large arrays supported 
> like this. It then seems to follow that if we do away with 
> that pipe at the top of the tapered tower top section and 
> replace it with a thrust bearing and another on the mast down 
> to the rotator we should be able to handle more and/or larger 
> antennas and reduce the stress on the tower and rotator.
> 
> Without going too deep and already having covered some of the 
> pitfalls of multiple thrust bearings there are several 
> problems that may show up when mounting the rotator down in 
> the tower, or near the bottom and using a longer mast between 
> the antenna(s) and rotator along with the required thrust bearings.
> 
> That mast between the antennas and rotator may be of a very 
> strong steel alloy, but it is still flexible from side to 
> side and radially as a torsion bar. That makes it a spring 
> and springs have resonances in two directions and multiple modes.
> 
> That means thrust bearings need to be properly placed within 
> the tower in relation to these resonances AND guy placement 
> on the tower which also has resonances of its own. Where and 
> how? It differers with every installation.
> 
> The longer the mast the more rigid it must be and that 
> usually means a larger diameter and that usually means more 
> mass which means more inertia. 
> In the end that changes the resonant points higher except one 
> due to inertia where period goes up with mass. It's all a 
> compromise and the farther we get from standard the farther 
> we are into experimenting and uncharted territory. 
> BTW, a rotator mounted in a tower at or near the bottom 
> exchanges torsional movement of the tower into shear at the tower legs
> 
> It pays to follow good engineering practice, but most hams 
> aren't engineers, or at least of the right type. No mater 
> what designs we look at we are likely to find some singing 
> the praises while others tell us that may be a bad way to go. 
> Some installations survive for years in spite of them selves 
> while well engineered systems fail in a relatively short 
> time. However I'll stick with the odds that favor the 
> engineered approach.
> 
> 
> 
> 
> Roger (K8RI)
> 
> _______________________________________________
> 
> 
> 
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
> 
> 