[TowerTalk] FW: FW: Thrust Bearing Installation

[Grant Saviers]

Matt,

Got the image, thanks and nice work.  Very informative about some of the 
issues.  The three bearing study would also help settle a lot of 
debates.  Were real bearing models developed?  The design that was 
developed for radial float (done in some machine tooling to float 
reamers) of the rotator would be interesting.

My thinking re mast length is most towers have the largest yagi close to 
the top to minimize its load on the tower.  Given the usual 20ft mast, 
that leaves a lot to add VHF/UHF, etc.  A recent TT post asked "what 
should I do with the extra N feet?"

Your analysis makes it more interesting that UST made the tube mount on 
the HDX really stiff with the double top plates.

I was involved in FEA way back for designing high performance disk 
drives.  Autodesk Fusion360 CAD/CAM now includes it, so I should try 
some modeling.  A lot easier with all the assistance now in the FEA 
software.  Ridiculously cheap to run on a PC rather than a Cray 
supercomputer back then.

73 and HNY,

Grant KZ1W

On 12/27/2020 03:42, maflukey at gmail.com wrote:
> Hi Grant & Happy Holidays.    Sorry to mislead anyone.   I lowered the 
> mast in my example because I'm not sure I ever met a ham that welded 2' 
> more mast on the end to compensate for a thrust bearing installation - 
> hi hi.   The original point I was making was relative to radial and 
> axial loads which I think has been pretty well ventilated at this 
> point.   Beyond statics 101 I have run a lot of finite element stress 
> analysis for complete tower installations and even more analysis on mast 
> installations.    Typically there just isn't enough load to create 
> significant statically indeterminate stresses of significance at that 
> location in the tower...   below is a FE model zoomed in on the top of a 
> crank-up beyond tower leg failure...  I’m not sure if the reflector will 
> handle HTML image.    Peak stress at the leg connection of the rotor 
> plate is only about 9000 psi at tower failure loading (which occurs 
> lower in the tower).   Displacement of the rotor plate relative to the 
> top plate is less than 10 mils.   It’s all typically pretty stiff here 
> relative to typical peak loads if that is what you are trying to get a 
> handle on.
> 
> Regarding a 3^rd mast support, I have done that scenario too on a 
> project a few years back and I know we ran a FE model of it.   I’ll see 
> if I can find the file.   I recall that the key to getting it to work 
> was to float the rotor in resilient bushings while restricting it’s 
> rotation (obviously).   I think you made a point about the alignment of 
> 3 constraint locations being problematical and I agree with you.
> 
> Thanks for your insight and response.
> 
> Best 73
> 
> Matt
> 
> KM5VI
> 
> -----Original Message-----
> From: TowerTalk <towertalk-bounces at contesting.com> On Behalf Of Grant 
> Saviers
> Sent: Saturday, December 26, 2020 9:08 AM
> To: 'Tower Talk' <TowerTalk at contesting.com>
> Subject: Re: [TowerTalk] FW: FW: Thrust Bearing Installation
> 
> Matt,
> 
> Your statics 101 analysis is misleading.  You lowered the load above the
> 
>    tower by 2 ft, At 10ft load height, the mast would still have 
> 1000ft-lb bending moment at the thrust bearing in the analysis, and no 
> reduction as claimed.  Others pointed this out.
> 
> While instructive, it's also bit misleading and more complex since mast 
> support bearings have resistance to bending forces - they aren't a 
> friction-less point pivot of simple mechanics. The resistance is higher 
> when there is axial preload on an their commercial angular contact 
> bearing designs. The mast support bearing really isn't an unconstrained 
> pivot and is limiting the moment below when the mast is loaded by the 
> wind forces.
> 
> For a tower top tube radial bearing ala UST crank ups and Rohn tube tops 
> there is much higher resistance, depending on the plate thickness, 
> welding, and tube properties. The two separated tube mounting plates on 
> an HDX589 make that resistance to bending very high. Thus, the moment 
> that is transferred to anything below will be substantially lowered or 
> close to zero depending on the design.  In this case the stiffness of 
> the tower matters as I posted.  If the tower bends (it does, how much?) 
> then the result is different.
> 
> For top bearings with pillow block ball bearings in spherical (self
> 
> aligning) or cylindrical blocks or thick polymer etc. radial bearings, 
> then there are different properties.
> 
> Except for the tower bending situation, I think what happens below the 
> top support will have very little influence on the mast bending moment 
> just above a real support.  However, the forces on the tower and rotator 
> can be redistributed.
> 
> The thread start mentioned a concern about the benefit of a third 
> support of the mast inside the tower.  So, that question hasn't been 
> tackled.  I think that is a big challenge for a number of reasons and it 
> doesn't fit into a statics 101 tool bucket.
> 
> Grant KZ1W
>