Aero Magic

three_mbt_sensors front_view2 acrylic_probe5

Chung-On-A-Stick Update

Due to the unexpected delays in getting tunnel time, the team was able to regroup and redesign the aero probes from scratch.  The result is a whole new set of probes that have been CFD-designed to optimize yaw sensitivity.

Here is the whole story in photos:

[View with PicLens]

[...]

Win32 Builds for Golden Cheetah (And Aerolab!)

Thanks to Gareth Coco, Windows users now have a constant stream of fresh win32 builds for Golden Cheetah.

Get Golden Cheetah 1.4 Dev with all the latest goodies including Aerolab) here!    It’s just a dev release, but you’ll find this version of Golden Cheetah is chock-full of terrific new features, including a comprehensive Metrics module, [...]

The Chung Method In Windy Conditions

How does the windless Chung Method fare in the presence of  windy ride data?   What constitutes a favourable field testing day? How sensitive are CdA and Crr estimates when data is polluted with wind that we don’t know anything about?  What accuracy do we need to adequately measure wind with Chung-On-A-Stick?   How [...]

Validating Aerolab

So, how do we know Aerolab actually works?    Oh, sure, it seems to give the right answers.   But how do we build confidence that Aerolab and all its future versions will be the best they can be?

Enter  toy_ride, a perl script to generate test files in iBike format with known answers. Toy [...]

Aerolab: A Sneak Peak at GoldenCheetah's Upcoming Virtual Wind Tunnel

As GoldenCheetah prepares for its upcoming release, I wanted to keep you all posted about the development of the virtual elevation module, Aerolab.  Aerolab is not currently in Golden Cheetah, but I’m hoping it may eventually find its way there.   There’s tons of work still to be done, but at least we [...]

An Effective Rotational Mass Database for Wheels

Thanks to Adrien Gontier of Roues Artisanales and David Linke, I am able to provide you with a unique database of moments-of-inertia and effective rotational masses of a large number of wheelsets. The data allows greater accuracy for VE, a virtual elevation perl script I wrote.

The big surprise is that [...]

Virtual Elevation: Does Rotational Inertia Play a Role?

The method of Virtual Elevation and all its variants are derived from Newton’s 2nd Law for linear motion, neglecting rotational inertia terms.   Do those additional terms play a significant role in elevation-matching? Can we accurately measure the moment of inertia (MOI)? In this post, I’ll describe the terms and provide a [...]

VE: A Test of the Small-Angle Approximation

A recent discussion on the Wattage Training Forum got me to hack the VE code to quantify the difference between the exact and small-angle version of the VE calculation.

The columns of the resulting CSV file are as follows:

SAMPLE #: row index
SLOPE_EXACT,SLOPE_SA: exact and small-angle slope
DELTA_EXACT,DELTA_SA: delta-elevation, exact and small-angle
ELEV_EXACT,ELEV_SA: cumulative elevation, exact and small-angle

The final [...]

VE Update: Improved Stability and Convergence

I was looking at the convergence of the first version of VE and thinking, “this can’t be right…”. When it’s coded right, Newton-Raphson typically converges extremely well, or not at all.

So looking at the code I realized I made a beginner mistake and used a one-sided numerical differentiation scheme:

sub fy
{

my $x [...]

VE - A Perl Script for Virtual Elevation Calculations

In trying to create a prototype for Golden Cheetah, I put together this perl script that I submit for your perusal.   The script comes with the sample data used by Robert Chung in his recent post to the Wattage Forum, so it may look familiar to many of you.

My aim was to match a known [...]