From: "Matt Weaver" < >  (see also IHPVA Record Rules and Battle Mountain)

To:

Subject: Proposed/Recommended Wind Meausurement Method

Date: Thu, 12 Sep 2002 14:04:28 -0700

 

 

Something very simple occurred to me in reviewing my "clarification" on the
proposed wind rule, and a potential method of implementing it.  Consider
noting the following:

Given:

    - Two axial-vane (IHPVA type) wind meters at right angles,
    - one fixed perpendicular to the course, and
    - one fixed parallel to the course

The following could be achieved:

    - direct, unambiguous measurement of the crosswind component
    - direct, unambiguous measurement of the tailwind/headwind component
    - no need to ever "move" the wind meters as is now done.

This directly covers the entire "headwind hemisphere" as noted in the
"clarification" (copy below).  It further only requires a simple calculation
for off-axis tailwinds. (Vwind = (Vside^2 + Vtail^2)^0.5).

This approach eliminates the need to have the wind meters "follow" or be
manually directed into the wind as the current IHPVA rule is understood.
Just set them up - one parallel, on perpendicular, and leave them.  This
will work.  It will work using existing IHPVA meters.

Of course I recommend we still have smoke or streamers in use and note their
general visual orientation for the record as well.  But there is no question
of the "component" wind along the course, and perpendicular to it.

As noted before based on conversation with Paul MacCready, the off-axis
values of such "axial vane" horizontal axis turbines break down nicely into
"cosine" components.  Paul is further looking into additional data to verify
this.  If desired, this can be readily validated by simple tests "walking"
with a wind meter over a timed distance in an indoor location (gym...) while
it is inclined at various angles.

If a long "manual" recording timeframe is desired, it may be necessary to
"flip" the meters around if they can "count backwards" and effectively
negate the previous accumulation of counts.  For the sprints of primary
concern, this is hardly an issue, as the required wind reading spans only
seconds and the direction typically changes negligibly in that period.

Lastly, adding a data-logger (like the 2 megabyte wheel revolution timer
I'll have on the bike this year), can provide a continuous time-logged
history of each and every revolution/direction of using the same wind meter
setup as well, providing a seemless extension of direct "simple"
observations to full datalogs.  The meters could be "manually read" real
time, yet get data-logged as well.

That's what I propose for HPV wind reading methodology!  Actually, I think
I've already proposed this method.  However, in light of review lately of
what's known, what's important, what's doable, I like it very much!

Regards,
Matt

(COPY)
===== Clarification of Wind Rule Proposal =====
Discarding my "code" attempts, perhaps a simple re-wording of my original
wind proposal is in order.  It is as follows:

'For a straight course, divide the wind direction into two hemispheres.  One
hemisphere contains all "tailwinds," the other all "headwinds."  For the
"tailwind hemisphere", simply apply the existing IHPVA rules (wind speed
limited to 1.67 meters/sec).  For the "headwind hemisphere", simply
guarantee the sidewind component alone is limited to the existing IHPVA 1.67
meters/second windspeed.