Hunter Allen interviewed.

An interesting interview with Hunter Allen over at BikeRadar...

In response the ever popular what are the benefits to a power meter question:

Top three, then...

Number one is the planning. I can actually plan my training around the response I want to get; this is called the dose and response system. Pacing is also a very, very big part. On-the-bike pacing in an event is very important, whether that’s a time trial, a criterium, a road race or even a breakaway. We often lose sight of the fact that this sport really is a sport of pacing. So we’ve got pre, during and then the post side, the actual analysis side; figuring out what the data means – did I improve, and how much can I handle?

And some more generic info:

Has coaching changed because of the use of power meters?

Now, I have a very clear understanding of what needs to be done to elicit a certain response. Before, if we wanted to improve your ability to go hard for a short period of time and recover quickly, we knew to do short intervals. Now, there are wattage numbers that we are trying to hit, a goal effort. It’s changed the prescription side of the coaching.

I actually think this is the primary benefit of training with a power meter. It allows you to optimize your time on the bike:

Does volume still count for what it did before the popularity of the power meter?

That has changed. There are a lot of these old myths that have been propagated throughout the years – base training for two months, riding really slow, or whatever. Those things work well for pros, but are just silly for the rest of us. If you’ve got eight to 10 hours a week to train, you can’t afford to ride slow. It does mean that you’ve got to keep a higher level of fitness throughout the season and take more of these smaller rest periods.

And finally a bone for us data analysis nerds:

What's the biggest mistake you see people make when training with a power meter?

They don’t value the cumulative power of all of the data. It gives you your training stress and tells you how much you can handle over three months, four months, six months... It allows you to look at it in that periodic type of way. What you really need to understand is how it all builds together. It can tell you your chronic training load [what you do every day], but you have to account for that. If you just ride, collect information and look at it in the performance manger chart in our WK0 software it can tell you a whole lot about what you can handle.

I have been trying for years to get Hunter to put this into a one-pager (well two pages double sided, and we can use a small font). Then I could print it out and put it into my track bag to hand out to people.

Notes from the blogosphere....

Robert Chung uploaded this interesting graphic to the wattage list. It shows an estimated wattage to CdA to speed relationship.

And in another post Alex Simmons has posted an update to his famous Seven Deadly Sins called The Sins of Sins (Testing FTP #2).

BVC Structured Training - Motorpacing, 500mTT practice

BVC Structured Workout and some more 500m TT drills.

It was Motorpace night. We did two sets of take-a-lap. First solo and then as a small group of three. Finished with sprint drills, from back of line to front.

Afterwords, I did four 500m TT practice drills. Two each with 51x14 and 50x14 gearing (95.6 and 93.8 as measured.) Results where only slightly conclusive. Looking at the best two, one each was with each gearing. Both times the second test with that gear...

This shows the power profiles. The first 51x14(1) was the only one that seems drastically different. I think I didn't start hard enough. That effort finishes best (power wise). The other efforts are more comparable. With the 50x14(2) ramping the fastest and ending the lowest.

This shows the actual distance travelled by time. Showing that the 50x14(2) effort was (slightly) better than the 51x14(2) effort.

This shows the WKO workout graph. From this perspective it was very difficult to tell which was best. Even comparing the average watts didn't help.

1. 51x14(1) - 493, fourth
2. 51x14(2) - 438, second
3. 50x14(1) - 500, third
4. 50x14(2) - 470, first

The tentative conclusions are:

• go out hard
• try a slightly smaller gear (49x14?)

N.B. There where other riders on the track. And at least one effort required some slightly sub-optimal track riding (up track). I think it was the 50x14(1) effort. But can't remember for sure.

N.B. The power and distance graphs where generated by exporting the numbers from WKO as an SRM text file. This was imported into Excel and the relevant portions cut and pasted to another spreadsheet to generate the graphs.

BikeRadar on Aero savings

BikeRadar has a great article up on Aero Savings.

Key points from the article:

• Position: Going from drops to aerobars will save you ~30 watts at 40km/h
• Bike: A time trial bike will save you ~20 watts over a road bike with clip-ons at 40km/h
• Helmet: A time trial helmet will save you ~10 watts over a road helmet at 40km/h

I'm not sure about how rigourous the tsting protocols where, they mention changes in wheelsets and clothing between the indoor and outdoor tests. And while they give a price/watts number for wheelset changes they don't mention actual watts savings for wheelset changes elsewhere...

But still an interesting read and probably a good first approximation of savings.

Method for Calibrating Ergomo Power Meter using Computrainer

This describes how to calibrate an Ergomo Power Meter using a Lab Computrainer.

It is hard to verify that the Computrainer is more or less accurate than any particular Ergomo. But using a simple ramp test we can confirm two things. First how the two devices diverge. And second get a reasonable idea on how reproducible the two report.

Finally, if you have more than one Power Meter for different bikes, then using this method, you can re-calibrate them so that they agree. And that they agree with your trainer (assuming you are using a CT for your trainer.) 

I have three different Ergomo BB's (road, TT and track) and train indoors on the CT. Prior to developing this procedure I was getting fairly divergent numbers. Large enough to make a difference to CTL/ATL/TSB numbers in WKO+. Getting them all to agree was the reason for doing this.

1. Setup

Setup is simple. You just need to get your bike with PM installed and setup on the Computrainer. 

Tire pressure should be about 110PSI. 

The roller on the CT should be cleaned.

Do the initial re-calibrate for the CT. Typically I try for about 3.00 at this point. Adjust the roller appropriately.

2. Warm up

Do a simple warmup. About 100-200 watts. Do a re-calibrate every 3-4 minutes. If the number falls below 2.5, increase the roller clamp pressure.

Once you have a stable number over 2.5, seen in two or more re-calibrations a couple of minutes apart proceed to step 3.

3. Ramp Test

Start the coaching software, using the following erg file:

[COURSE HEADER]

VERSION = 2

UNITS = ENGLISH

DESCRIPTION = Power Meter Calibration 60w per min

FILE NAME = PowerMeterCalibration20.erg

MINUTES     WATTS

[END COURSE HEADER]

[COURSE DATA]

0.0     100

8.0    580

[END COURSE DATA]

This is a simple ramp test, starting at 100W and increasing 60W every minute. 

4. Test

To ensure that the CT didn't cool down, do another few minutes of warmup with the coaching software running, the test loaded, but not started.

Then start the test, ensure that your Power Meter is also recording.

Try to maintain a fairly constant and comfortable cadence. If possible over 80RPM. 

The first few minutes are easy... the last minute is hell... hang in as long as you can.

5. Results

Copy the CDF.txt file produced from the Coaching file into WKO+. Also import your PM data.

Inspect both files and cut out the parts that do not pertain to the test. You should end up with about 4-6 minutes of data from both the CT and your PM.

Use the Export facility to save these into a CSV file.

6. Excel

Open the two CSV files in Excel, and also create a new empty workbook. Cut and paste the Minutes and Watts columns from the CT and PM CSV files into the workbook. You can use Excel tools to compare your results.

7. Sample one - Soloist Ergomo with K-factor 184

This is from a Soloist with Ergomo BB and the factory K-Factor setting.

The two time series are plotted as an X/Y scatter plot (as lines) with trendlines added. Finally a difference column was added and plotted.  

This shows that there is a substantial difference, with the Ergomo reading much higher. The difference does not quite scale linearly. It increases slightly as can be seen in the Diff plot (which uses the right axis).

8. Sample two- Soloist Ergomo with K-factor 164

This test is for the same bike with the K-factor adjusted down from 184 to 164. 

Note that while not an exact match, it is reasonably close now.  

The Diff line is still not exactly linear and does show that the Ergomo is more variable (again Diff is show using right axis.)

9. Conclusions

As can be seen, it is reasonably easy to get the Ergomo to match the Computrainer by adjusting the K-Factor appropriately. 

While this does not guarantee accuracy, it does provide a way to ensure that multiple devices are all setup to produce similiar results.

The (presumably) best results for this procedure would be with a reasonably new Lab Computrainer. But it is possible that acceptable results could be obtained with the non Lab version or possibly a Tacx or other trainer that can control the watts in "Erg" mode (i.e. set the load level absolutely, regardless of tire speed, rider cadence etc.) and can record the results.

News from the tubular testing front, they are fast if you over glue!

A very interesting post over at Bike Tech Review....

It appears that the old track axiom of over glueing may have some basis. Adding an extra two layers of glue (3 for rim, 2 for tubular, instead of the suggested 2 for rim and 1 for tubular), at least using Mastik, may result in a lower CRR...

Not recommended for Tri-Guys though if you want to be able to fix a flat in a race... The tester reports that it took him a day to remove a tubular glued in this fashion...

Glued in this fashion a Vitorria Crono dropped down to 0.00235.. a substantial improvement over the previously reported 0.00276 when lightly glued. And possibly gettting it close to the "fastest" tire in the AFM list.

I am just half way through glueing a new Pista on a Zipp 404 front wheel for the track racing season... So it's out to put another coat of glue on... instead of just the final mount... I should have mounted it yesterday, but got busy... then stumbled across this tonight..

Lennard Zinn - Rolling resistance

As pointed out by samk on the wattage list, here is a summary of tire Crr from Leonard Zinn.

iBike - too cold, bad position

The numbers crunched, but not too well.

Two problems I think. First I need to get the iBike moved from above the aero-bars to below (using for example the Topeak Bar xtender.)

Second, I think that the cold caused the battery level to go too low.

So, at this point inconclusive results.

Neaves Road - iBike testing

Out to Neaves Road to try out the new iBike2 firmware.... I will (hopefully) provide some CRR and CdA numbers.

Ran three sets of tests on my P2C:

1. non-aero with Bontrager Aeoleus wheels
2. aero with Bontrager Aeoleus wheels
3. aero with Bontrager X-Lite Aero

At the end of the day I also did an iBike torture test, the section of Neeves Road just to the north of my normal TT and wheel test area has some of the worst chipseal and cracked pavement...

I'll run the analysis later and post results..

Yesterday was heavy rain, so spent an hour and a half on the Computrainer, doing the Athens Triathalon course... 5 8km loops with one medium and steep climb (.5km at 7-8%, 300m at 15%) and another shorter climb. Mostly done as L2/L3.

ibke2 - new firmware and post-analysis software

There is a new version of the iBike software (aka ibike2) available.

ibikepro download

Both firmware and new application (Mac and Windows). Along with new procedures for coast down, and doing a "Four Mile" ride, the new iBike2 software will do post-analysis on your coast down(s) and ride to get a much better set of numbers (CdA and CRR) to allow the iBike to report more accurate numbers.

The post-analysis software can also eliminate many of the old problems (tilt problems, wind changes, vibration problems) by filtering the ride files.

I'll be trying this out soon. And I put together a quick checklist of the procedures for getting the initial profiles (tilt, wind, coast down, 4-mile ride) etc, done correctly.

iBike Checklist...

The new iBike2 software produces CdA and CRR numbers as part of the new process of creating a "profile". This involves multiple coast downs, then doing a 4-mile ride under reasonably controlled conditions, then doing post-analysis.

Hopefully the produced numbers are reproducible and can be used as the "poor mans wind-tunnel" to evaluate aero-gear, wheel sets, bike position etc.

New Teamate - good aero position

We're always interested in getting our teammates outfitted in the latest gear so they can be as fast as possible.

Pez reviews Bontrager Aeolus 5.0

I know I like my Aeolus 6.5's, here is what Pez says about the 5.0.

My defining moment with aero wheels came unexpectedly, quickly, and at surprisingly low speed. It happened at the Amstel Gold cyclosportif ride in April, as I was coasting along chatting to some other riders. It made a believer outta me – both in aero wheels – and these Bontrager Aeolus 5.0 carbon clinchers…

Go here for the full review PezCycling

BVC Testing - 2k

BVC Coach Jeremie Storie was doing testing at the track this morning as part of the masters structured workouts.

Just a 2k (10 lap) effort at VO2Max.

2k test:

Duration: 2:49
Work: 61 kJ
TSS: 6.3 (intensity factor 1.156)
Norm Power: n/a
VI: n/a
Distance: 2.067 km
Min Max Avg
Power: 246 486 358 watts
Heart rate: 0 0 0 bpm
Cadence: 96 110 104 rpm
Speed: 41.1 46.8 44.0 kph
Pace 1:17 1:28 1:22 min/km
Crank Torque: 24.5 44.7 32.9 N-m

I'm still surprised at how hard I find it to maintain this power level over this short a distance on the track compared to (e.g. Ryder) TT for 8-10k distances. Not sure if it is bike positioning or fit, or just fixed gear versus normal etc.

Also did a warm up and then a 30 minute tempo workout.

And of course because I got up early and went to the track the predicted 1" of rain never materialized...

CdA testing

CdA testing at Neaves Road... But probably a bit too much wind to get really accurate results.

Two tests. Each test was 2km out and back at target speed (kph) of:

- 26
- 32
- 38
- 44

The first set was at my current TT setup:

- Cervelo P2C
- seat in rear position
- bars set high (stem up)
- Bontrager Aeolus 6.5 wheels
- Veloflex Record Tubulars
- short sleeve skinsuit
- full finger gloves
- Rocket aero helmet
- Campanogla shoe covers
- bottle cage, no bottle

For the second test I move the seat to the forward position and reversed the stem to lower the bars (but moved up on the stack, probably about 1.5" lower.)

Results:

Test 1 Test 2
CdA 0.261 0.259
CRR 0.004 0.004

So perhaps a very slight difference in the setups with a preference to the seat forward position. But not outside the error range for the test conditions.

Probably the best I can say is that current CdA is no higher than about 0.260, and CRR no higher than 0.004. Good, but certainly not great.

The forward position with lowered bars was not very comfortable. I'm going to pickup a flat stem so I can try some intermediate positions. Forward but with more comfortable positioning of the bars.

Spread sheet

Wheel testing - Neaves Road

The last three TT's have been very slow. Even though I have been producing good power numbers, the times have been poor. Barely better (if) than last year. And people I should be matching have been getting times minutes(!) better.

For example at Aldergrove Bill Riley was 23:04 and Gerry van Ganz was 23:10. With me back at 24:21.

So I decided to look at the bike and see if there where any potential problems. And decided it could be the Zipp 404's with Tufo S33 tubulars.

To test I went out to Neaves road and did two tests each to compare the Zipps against my tried and true (and cheap) Bontrager Lite Aero (aluminum) wheels. Each test was 2.5 k out and back (5k total). Tests at 36kph and 42kph.

The results seem to be unequivocal, every Zipp test had higher average power for lower average speed.

Possibly by as much as 20 - 30 watts.