[TowerTalk] wind load vs Rohn specs

[Bill Coleman AA4LR]

On 9/7/00 3:28 PM, Kurt Andress at K7NV@contesting.com wrote:

>
>Bill Coleman AA4LR wrote:
>
>> This discussion of projected area has always made by head spin. Seems to
>> me it doesn't account for antenna members that could add significantly to
>> the drag of the installation.
>>
>> If we take a small tribander and optically project a shadow of the
>> antenna vertically on the ground, then measure the area of that shadow,
>> we fail to take into account say, the boom to mast plate. Since the plate
>> sits vertically, it casts a small shadow, yet it presents a big area to
>> the horizontal wind.
>>
>> Similarly, if we optically project a shadow horizontally against a wall,
>> and measure the area of that shadow, we'll take into account the boom to
>> mast plate, provided we have the antenna oriented correctly. However,
>> we're likely to miss either the boom or one or more of the elements,
>> whose shadows overlap one another.
>>
>> So, which is right? The vertical projection or the horizontal projection?
>> Neither seems to be well correlated to the actual drag of the antenna.
>
>The vertical projection technique would be useful if you plan to mount your
>antenna with the elements vertically aligned or if all you get for wind 
>events
>are microbursts directly over the tower. In these exposures the wind will see
>both boom and element areas simultaneously.

Now you are toying with me. The odd thing is, the way that these 
projected area values are calculated appears to be by using the vertical 
projection.

>For normal installations, the horizontal exposure is the appropriate one to
>chose.

Right! Which is why the horizontal projection would appear to be more 
appropriate -- except that it isn't the one used in the "projected" area.

> But trying to do it with a some-kinda projection against a wall, while
>providing at least a six-pack of entertainment, is useless for determining 
>how the antenna will develop loads on a tower.

My point exactly.

>Antenna area measurements should be made with a calculator. Simply summing up
>the Length x diameter of the members is more accurate.

If you do this sum, then you have computed the area of the vertical 
projection of the antenna. (Assuming a standard yagi-type antenna) Just a 
few paragraphs before, you decried this projection as being inaccurate. 
Now you say it is the way to calculate them. Again, I don't understand.

>The peak area (FPA or EPA) and hence load applied to a tower by a 
>horizontally
>polarized yagi type antenna will occur at either 0 deg or 90 deg 
>orientation to
>the wind. In between these two orientations the area (hence, load) will be 
>less.

Doesn't this depend on the configuration of the antenna members?

I know enough about aerodynamics to understand that a tubing member 
receiving clear, undisturbed area is going to have a different amount of 
drag than one receiving the turbulent air that has already passed over 
another antenna member. The effects are very non-linear and difficult to 
predict with simple algebra. The drag effects will vary with the speed of 
the wind, and also with the density (altitude, pressure, humidity) of the 
air.

But I also know that you can't just ignore the area of the additional 
elements, even in turbulent air. The horizontal projection would appear 
to ignore that area completely, since the shadows of the elements (and 
possibly the boom) overlap one another.

>There is no longer (never was) a single peak peak load orientation caused 
>by the
>combined element and boom areas as used to be common thought. 

Isn't there? I can think of two extreme cases. One is a very long 
wavelength antenna on a very short boom. Most of the drag will be 
contributed by the antenna elements. Virtually none by the boom. An 
alternative is a long boom antenna with very short elements (say a 70 
foot 1296 MHz beam). The elements contribute little drag to the antenna, 
virtually all by the boom. (Indeed, due to the turbulent effects of the 
elements, the antenna may offer less drag than the boom by itself)

>That notion was
>effectively challenged in 1993 when K5IU dragged some old long standing
>aerodynamic principles, kicking and screaming into the communications 
>arena. I
>think there have been several extensive posts regarding this on this 
>reflector
>and other places.

Could you dig up some references?

>Your concern over the mast/boom plate is unwarranted because you are thinking
>about it in the old obsolete reference frame.

If someone puts a big mongo mast / boom plate on a very small antenna, it 
certainly would be a big contributor to the total drag on an antenna. 

>For a TH7, the proper inclusion of the mast attachment or not makes a 2.2%
>change in the effective area with the boom broadside to the wind, but who 
>cares
>when the area of the antenna pointed into the wind is 2.8 times greater than
>broadside.

Since we can't control the wind direction, the appropriate computation 
would be for the "peak" drag on the antenna. If the structure can 
withstand the wind loads at the peak drag, then the loads at lessor drag 
orientations seem immaterial.

But computing peak drag seems dependant on the specific configuration of 
the antenna. A low aspect-ratio yagi (1296 long boom) has peak drag 90 
degress to a high aspect-ratio yagi.

So far we've talked about yagi's. What about other configurations,  like 
quads? How do we compute their "projected area"? 

I also know that many old biplanes got most of their drag, not from the 
whetted area of the wings or fuselage, but from round flying wires, which 
generate a ton of drag from the turbulent airflow they produce. 
Streamlined flying wires were introduced as far back as 1915, and were 
later eliminated entirely with cantilevered and monocoque construction.

>> Seems to me a better measure was the old "flat plate equivalent" -- which
>> is the drag of the antenna expressed as equivalent to a flat plate of a
>> given area. This FPA and EPA for round and square antenna members just
>> leads to a lot of confusion.
>>
>
>Well, while intrigued by your proposal, it leaves me wanting to know more 
>about
>how it would actually work in practice, and how it would make anything 
>better.

You'd have one number, normalised to a given unit. You wouldn't have to 
deal with the wildly varying units and computations for this "projected 
area".

>Things to address:
>
>There are more than one specification used to design towers. They are all
>configured to accept Flat Projected Area as an input.

I have yet to see a concise definition of this Flat Projected Area. If I 
have a given antenna design -- how do I compute this FPA? You laughed at 
my vertical and horizontal projections. This leaves me more confused than 
before. 

Let's demystify this whole process by providing proper definitions.

>How will the users (and the P.E's we hire) of the new "Flat plate equivalent"
>value, generated in a similar unknown fashion, know how to work their way 
>back
>to the fundamental projected area to then commence evaluating their tower
>according to its required design specification?

I would assume they wouldn't use projected area at all, but use flat 
plate equivalent instead.

>There are no universally accepted antenna design specifications. Nor are the
>ones used for any particular antenna clearly defined. Hence, confusion 
>reigns!

So is it OK to confuse people further?

>Seems likely that we would have the same confusion over a different number.

Not if it is one precisely defined number to a standardised unit.

>I think we'll find that the current M.O. works just fine when the proper 
>values
>and methods are used.

Except that no one seems to understand it well enough to explain it.



Bill Coleman, AA4LR, PP-ASEL        Mail: aa4lr@radio.org
Quote: "Boot, you transistorized tormentor! Boot!"
            -- Archibald Asparagus, VeggieTales


--
FAQ on WWW:               http://www.contesting.com/FAQ/towertalk
Submissions:              towertalk@contesting.com
Administrative requests:  towertalk-REQUEST@contesting.com
Problems:                 owner-towertalk@contesting.com