General Questions

• How do you calculate the air density from this data? http://w1.weather.gov/.../obhistory/KSAV.html
• When using a loop course how do you accurately determine the beg and end of the loop in the Aerolab interface so that you know the actual elevation is the same?

Ah, there's a "missing" variable in the air density calculator. The issue is that "official" weather stations like the one you're pointing to, typically give the barometric pressure at sea level. There's also "station" air pressure, which gives the air pressure at whatever altitude the station is at. If you're close to sea level and you have an official weather station, just use the air density calculator in GC. If you're at altitude and you're using "station baro pressure" just use the air density calculator in GC. If you're at altitude but you're using an official station's baro pressure, you need to make an altitude adjustment (which is the missing variable from the current calculator). That said, if you're looking at the *difference* between two setups (rather than the exact values of CdA and Crr) then you can pretty much ignore the altitude adjustment.

I don't usually "mark" the beginning and end of the loop -- usually, I just know the length of the loop. If the loop is, say, 1213 meters long, then I know that every 1213 meters is exactly one loop, so has to be the same elevation. It helps to run the same line through the corners. If the loop is new to you, try doing a couple of "high speed" test runs with a piece of chalk and mark the entry and exit points for your turns. If you mark the entries and exits at high speed then you'll know you can hit the same marks at low speed (that doesn't always work quite so well in the other order).

Thanks for pointing me towards the Air Density Estimator tool...I hadn't discovered it yet. Thankfully since I am in Savannah I am going to assume that I am at sea level. If we were any closer we would be under water.

Marking the corners isn't going to be needed per say. This is the loop I am using https://www.google.com/...0065,-81.2100071,17z. I was using the big loop to try and negate the slight wind that was going on, but it is a little long at ~1.1 miles. I just wish there was a little easier way to figure out my start and end point for the interval.

Along the lines of Intervals. Is there a way in Golden Cheetah to get rid of intervals that I have 'selected'?

Do you mean to delete the intervals or to unselect an interval? To delete an interval, you can right-click on it -- but I'm not sure why you'd want to delete an interval. Currently, flipping between intervals or unselecting an interval is kinda funky. If you're "zoomed-in" on an interval you can right-click on the left interval window and choose "zoom out" to display the entire ride again -- but that interval might still be selected. I can't remember how to unselect any interval -- sometimes I just click on the previous day's ride and click back on today's ride. As you can tell, I'm sorta confused about that.

I'm incredibly jealous of your testing loop.

Public road, fairly well lit at night, really limited traffic.....not a bad deal at all. The side cover is a little inconsistent and there can be some weird winds if we are being really nitpicky, but overall it's pretty good for a lot of things.

We actually hold our spring/summer crit series on the upper half of the loop. Two corner crit.

Have you tried this loop out yet? Just hit the interval timer when you start, do two or three (or more) laps at varying speed and power, then hit the interval marker at the end. Usually there are little bumps or dips that you can use to line up the laps. I could spot the dip caused by a storm drain at the side of an otherwise flat road. I've spotted a little low spot on an otherwise flat road that I hadn't noticed, but it was in a consistent spot so I went out there after a rainy day and found a light coating of mud on the road right where it was indicated.

I'm pretty familiar with it and tried a couple single lap loops the other day...mostly just goofing around. I'm pretty sure my joule gps is reading GPS for speed or my powertap has miserable speed trace data cause it is just not that clean. I asked Saris and they claim that if the powertap is connected it will use that for speed and not GPS but I am not terribly convinced. I want to try using a garmin speed/cadence sensor I have and see if that works better.

I'm just trying to figure out what I don't know and how consistent I can be.

I will def look for high spots. I usually start/stop at the intersection on the bottom of the loop. There should be a high spot from the road crown at the intersection. I will see when I get some better speed trace data.

Here's a test to make sure: hold up the rear of the bike, spin the wheel to wake up the hub, then turn on the Joule. If you do this indoors and immediately get a speed signal you know it's not coming from GPS. Make sure your rollout is correct.

A lot of people new to the test protocol think they have to hold their speed constant. You don't need to do that. In fact, on a flat course like yours, start out slow then ramp up your speed to maybe racing speed (all while holding your position) then coast (while soft pedaling) back down to maybe jogging speed, then repeat. That will get you a wide range of speed and power, and the coasting bits will help you determine how fast you decelerate under zero power.

Hi guys,

Brilliant thread, great work! I just want to ask something in case someone can help me.

The protocol recommends a course with elevation changes and changes in speed, and one of the posts was regarding testing in a velodrome when the answer was actually to do not change the line nor go up the banking, and also to keep a steady speed.

I'm confused because I've near my place a 600m outdoor track, that has only one of the corners with a banking. Because it's outdoor and quite long, I think it's more like a road loop than actually a velodrome. My question is.. should I ride up the banking (keeping the same line every lap) to change the speed/elevation, or just stick to the black line? (tho I don't know how I'd be able later to align the VE this way.. any tips?)

Cheers!!

I would think sticking to the black line would yield more consistent results than trying to ride the same banked line every time, unless you perhaps marked the curve out.

I'm planning to have a go at this soon and am trying to decide where the best location will be for testing. I have a few "halfpipe" options in neighborhoods that I could use, but I feel like it would be hard to hold mynaero position while make a sharp turn at low speed. I think my best location will be here: http://goo.gl/maps/UDHkq. It's ~half a mile and almost perfectly flat, and has almost no traffic on an early Sunday morning. My only concern is if the constant turn of the short loop would affect the data at all? If that loop is too tight I have another good option (directly northwest on that map) that is about .9 miles, but there is a stop sign I would have to run each lap, and being in a neighborhood I would think it hard to get a lot of consecutive runs without being passed by a car screwing up my data. Also the longer loop means it takes a lot more time to get each test completed. Any feedback would be appreciated. Thank you.

---

Powertap / Cycleops / Saris

Stick to the black line, though vary your speed.

Holding your position and finding a venue without traffic that is protected from the wind trumps everything else. A halfpipe lets you vary speed easily, so you'll just have to vary the speed intentionally. The turn won't affect the data as long as you can hold your position.

Hi guys,

Just 2 quick questions.. I've just started playing with the Aerolab so I'm still struggling with a few things.

1. How can you analyse different segments on the same ride, so I can select the laps that I had a different setup to get the CdA and then compare? Or do I need to reset the Garmin every time I change the setup?

2. I did a test ride with the GPS on, but later on GC I deleted the altitude profile. But now when I try to align the bottom of the VE moving the slider, the whole thing moves up/down, tho it doesn't change at all the alignment of the VE line. Is it because I had the GPS on or am I doing something else wrong? This is the printscreen of the segment of the ride where I took the bike to a outdoor track (that has a + - 1.5% gradient on each side).

You don't need to reset the Garmin each time, but you should do more than one lap in the same position. Maybe 3 or 4. Don't change position on every lap.

But for every unique position, try to match the elevation curve of successive laps. Do each position one at a time.


Move the CdA slider and it will tilt the elevation profile. If you're doing loops, you're trying to match the start elevation of the loop with the end elevation. When you've matched them, you've found the right CdA. Is that what you're doing?

---

AndyF
bike geek

I was planning on holding each position for 4 laps, so it should be fine, thanks! But how can I analyse the CdA for each 4 laps only? It seems that on GC the CdA is for the whole ride, how can I isolate only the laps that I want for the analysis?

I was trying to do it, but when moving the CdA slider the VE is not tilting, it's going up or down but keeping the same shape, so the difference between the bottoms/tops doesn't change at all. Not sure if I explained it well?

Many thanks again for your help!
Btw, nice website ;)

Rather than cycling speed within a loop interval, would it be better to "Shen" (so to speak) by slowly ramping up one's nominal power over several laps? I can't decide which would tease apart Cda/Crr better.

---

The question of who is right and who is wrong has seemed to me always too small to be worth a moment's thought, while the question of what is right and what is wrong has seemed all-important.

-Albert J. Nock

That will work, too. Both are ways to widen the range of speeds. Andy Shen's method is (to my mind) a little harder to do on the road but makes the analysis a little easier. Ramping and coasting is (to my mind) a little easier to do on the road but makes the analysis a little harder. I actually use both, at different times, depending on the venue I'm using. Use whichever makes the most sense for you.

Sounds fair; thanks!

---

The question of who is right and who is wrong has seemed to me always too small to be worth a moment's thought, while the question of what is right and what is wrong has seemed all-important.

-Albert J. Nock

Can you post the ride file somewhere? Perhaps there is something we can do in previous versions.

---

AndyF
bike geek

"It seems that on GC the CdA is for the whole ride, how can I isolate only the laps that I want for the analysis?"

I use the Split Ride tool (under the Activity menu) to split out the laps into separate files for VE analysis. I also keep a simple spreadsheet for tracking my various tests where I capture the pertinent data for each test, such as RHO, Crr, Mass, resultant CdA, etc.

As for why the VE curve isn't changing when you move the CdA slider, I have no clue. That sounds like it may be a bug in your install of GC.

Aahh...but that's one of the "beauties" of a halfpipe course, in that since you ARE going at a very slow speed, staying in an aero position is FAR less critical at the turnarounds. In fact, once I'm going very slow near the turn, I sit up a little and turn my head to look back and confirm that it's safe to make the turn, make the turn (I go slow enough that braking is unnecessary), then settle back in to the position and star pedaling again. You'll never detect the effects of that "sit up and look over your shoulder" move in the plots...as long as you're going slow enough, that is. For me, it's just barely fast enough to stay upright without doing a track stand ;-)

---

http://bikeblather.blogspot.com/

Thanks v. much guys, I went back to the track today and later on I could tilt the VE using the slider, perfect!

I just forgot to press the 'laps' when riding so I couldn't split the ride on GC, but next time I'll get it all right!

Cheers

Ok, thanks for the feedback. The loops are still easier (closer to my house, less chance of traffic), but the ones around here are almost perfectly flat so I'm not sure I will be able to see where each lap starts and ends on the VE profile. I'll have to play with it and see.

---

Powertap / Cycleops / Saris