Do I need any equipment to use Run123?

You need a magnet-based ANT speedometer logged by a bike computer. A powermeter and aero sensor (e.g. Notio, AeroSensor) are optional.

How does it work?

The Run123 uses 2 runs in different directions to separate rolling resistance from road slope, and 2 runs with different velocities to separate rolling resistance from aero drag. So a minimum of 3 runs are needed to calculate CdA and Crr. The precision is very high!

What's a good test venue?

You have to ride the EXACT same path, so I recommend a flat straight bike path about 250m long. A slight slope of 0 to 1% is good. Some slight dips and bumps are ok. For every run you should strive to ride in the (imaginary) wheel tracks of previous runs. I divide the path into thirds drawing A-B-C-D lines with chalk, where the middle third is the test section. Each section has exactly the same length (e.g. 80-100 meters).

How do I get repeatable results?

Run123 is extremely precise. If your results are varying and you do not see a single cluster for AA testing then there are 4 possible error sources: (a) wind gusts, (b) your CdA is varying between runs, (c) your path is varying in the test section, and/or (d) your powermeter is not accurate.

• Test when it is not windy (typically mornings). Ideally the bike path has foliage on both sides which helps shelter you from gusts.
• Strive for a comfortable and consistent riding position. Before starting each run remind yourself where your hands, elbows, torso and head are supposed to be.
• You have to ride the EXACT same path, it may help to make a X-shaped chalk marks every few meters. Ride in a straight line -- no turning or sinusoidal path since this increases tire forces and distance traveled.
• Review the velocity curves of your Fast Downhill, Slow Downhill, Slow Uphill runs. Are all (e.g. Fast Downhill) similar? If so then select additional similar runs which have very similar velocity curves for analysis which will result in a single cluster. More similar data points also tightens the solver bounds.
  
  However if dissimilar speeds, then practice generating similar speed curves. Enter the test section at the same speeds. For example for Fast Downhill aim to enter at 10 m/s (36 km/h), for Slow Downhill aim to enter at 3 m/s (10 km/h), for Slow Uphill aim to enter so you can coast to the end of the test section -- for me that is about 6 m/s (22 km/h) for a slight uphill on a ~85 meter test section.
• Coast through the test section for all runs. When coasting your power is exactly zero.

Testing is a skill, so practice until you can create consistent clusters of (CdA,Crr) points. Only then you are ready for AB testsing!

Will a TT bike have cleaner results than a Road bike?

Yes. A TT bike with aerobars 'locks' in the position of the rider much more than a road bike. On a road bike the rider can't really 'lock' the elbow angles leading to different aero drag resulting in a larger CdA variations. A road bike with aero bars will have more consistent CdA results than one without.

Can I measure which configuration is more Aero?

Yes. The Fast Downhill runs dominate the aero drag. Ride several Fast Downhill with configuration A, then several with configuration B. The combinations of all selected runs will show up as 2 separate clusters on the CdA vs Crr graph. CdA resolution is typically about 0.002 or better.

Can I measure which configuration has lower rolling resistance?

Yes. The Slow runs dominate the rolling resistance contribution. Ride several Slow Downhill and Slow Uphill runs with configuration A, then several with configuration B. These runs will show up as 2 separate clusters on the CdA vs Crr graph. Crr resolution is typically about 0.00010 or better.

If I am changing tires or tire pressure during testing do I need to do a rollout?

Technically yes, but you can elect to skip it. The test course is measured in wheel revolutions. On a run with a different circumference the start and end of the test section will NOT match for a Downhill and Uphill run -- most of the test section will be the same, but not the ends. Also the total distance traveled will not be accurate which affects the absolute magnitude of the Crr. If you are ok with these approximations then you will know which configuration has a lower Crr and you don't need to do a rollout and re-chalk the test course A-B-C-D markings. If however you want to publish the measured Crr values then you need to do a roll-out and re-chalk the course A-B-C-D markings.

What happens if I brake?

Braking is a dissipative force that is not modeled. If you need to abort a run, brake to a stop. The Run123 automatic categorization will discard all runs that are too short.

How do I document my testing?

All Run123 graphs have a 'Camera' icon which adds the current graph to a report list. On the 'Report' page select each item and add markdown-formatted annotations. When done adding all report items press the 'PDF' button to save the report as a PDF file.

You may also consider using a 360 camera on the bike to record all testing. This captures voice annotations (e.g. configuration description), and records your position for every run.