Page 1 of 1
why is elevation different for intervals on the same course?
Posted: Sat Jul 19, 2008 10:53 pm
by R Mc
I did 5 crit-type intervals today involving multi-laps on the same course at a local junior college. In fact the majority of the ride took place on the same course--the major difference being that I switched directions when I started the intervals so that I could use the headwind to get some hard efforts in.
As I'm looking at the ride file in ibke2, it's easy to identify the interval portions for 2 reasons: 1. the power increases and 2. It appears as though I gained 20 feet in elevation at the beginning of every interval.
Oddly, the slope data remains unchanged.
What's going on here?
Re: why is elevation different for intervals on the same course?
Posted: Sat Jul 19, 2008 10:55 pm
by R Mc
Actually, I think I have it backwards--the question should be, why does it seem as though I descended 20 feet during each recovery period? Anyway, thanks in advance.
Re: why is elevation different for intervals on the same course?
Posted: Sun Jul 20, 2008 5:42 pm
by travispape
R Mc wrote:Actually, I think I have it backwards--the question should be, why does it seem as though I descended 20 feet during each recovery period? Anyway, thanks in advance.
The short answer:
http://en.wikipedia.org/wiki/Bernoulli%27s_principle
Look at the bottom of your unit an you'll see a little hole there that serves a couple purposes, one of them being to let the altimeter measure the atmospheric pressure. When you ride fast, air passes over the hole more quickly, which causes the pressure to go down inside the unit slightly, which causes the altimeter to measure a lower pressure, which corresponds to a higher elevation. The fact that your whole ride was pretty flat caused the verticle autoscaling to give you a zoomed in view of the elevation and make it look like a big change of elevation, but I think your results are typical.
Note that the same thing happens to any other altimeter on a bike, for example a Polar S720 does the same thing.
It is interesting that you can actually count your laps by looking at the wind data.
Travis
Re: why is elevation different for intervals on the same course?
Posted: Sun Jul 20, 2008 6:33 pm
by racerfern
I've seen this regularly on my flat intervals also. Does the air "enter" the hole or does it pass through after entering the wind port?
It the air enters, as a workaround until the Gen III comes out

, can we consider shielding this? Obviously not to block it in any way; rather to shield the wind speed.
Re: why is elevation different for intervals on the same course?
Posted: Sun Jul 20, 2008 7:10 pm
by alienator
racerfern wrote:I've seen this regularly on my flat intervals also. Does the air "enter" the hole or does it pass through after entering the wind port?
It the air enters, as a workaround until the Gen III comes out

, can we consider shielding this? Obviously not to block it in any way; rather to shield the wind speed.
As the air passes over that hole, there will be a lower dynamic pressure just outside the hole, pulling air out. One measurement the pressure sensor will make is a differential measurement, i.e. the difference between the pressure in the "intake" on the front of the unit and the pressure inside the little hole on the bottom. There would be no way to shield it so that it would only measure static pressure. If you tried to do so, you'd end up with a closed system on that side, and given its small volume, it would be very sensitive to temperature changes. In fact, it's designed to have air passing over it. Here's a simple drawing of that system:
In the picture,
b would be that little hole, and
a would be the large hole on the front of iBike/iAero. In the piccie,
hm is a fluid height difference which is a function of total pressure. In iBike/iAero, you'd replace the fluid with 2 different sensors or summat.
Re: why is elevation different for intervals on the same course?
Posted: Sun Jul 20, 2008 7:47 pm
by R Mc
It is interesting that you can actually count your laps by looking at the wind data.
kinda like tree rings . . .
Re: why is elevation different for intervals on the same course?
Posted: Sun Jul 20, 2008 8:20 pm
by racerfern
I understand the Bernoulli effect enough to get myself in trouble.
I'm still not sure if the little hole at the bottom relies on air passing by (entering the little hole) for the barometric pressure sensor or if gets its air from the air that entered the wind port (exiting out the little hole).
Re: why is elevation different for intervals on the same course?
Posted: Sun Jul 20, 2008 8:46 pm
by travispape
racerfern wrote:I understand the Bernoulli effect enough to get myself in trouble.
I'm still not sure if the little hole at the bottom relies on air passing by (entering the little hole) for the barometric pressure sensor or if gets its air from the air that entered the wind port (exiting out the little hole).
No, there is no air flow through the unit.
Re: why is elevation different for intervals on the same course?
Posted: Mon Jul 21, 2008 11:46 pm
by racerfern
No, there is no air flow through the unit.
So I'm picturing a tiny cylinder with a small screen to minimize the turbulence at the bottom side hole. Maybe something like the metal mesh of a coffee filter; allow air flow (critical) but shelter the hole from direct airflow.
Re: why is elevation different for intervals on the same course?
Posted: Tue Jul 22, 2008 12:40 am
by alienator
racerfern wrote:No, there is no air flow through the unit.
So I'm picturing a tiny cylinder with a small screen to minimize the turbulence at the bottom side hole. Maybe something like the metal mesh of a coffee filter; allow air flow (critical) but shelter the hole from direct airflow.
No, I don't think what you're thinking is right. The pic I posted is just a generic schematic. If you remove the fluid tube, in the pic, and instead place a sensor at each opening where the tube was, you'll have what I think the iBike has. I'm betting that iBike does the same think as a fluid based pitot tube by getting a differential measurement between two sensors (i.e., get two measurments from two sensors and find the difference between. Sensor 1 would be
a in the picture, while sensor 2 would be
b). There are several ways to get a system like I"m seeing.
Re: why is elevation different for intervals on the same course?
Posted: Tue Jul 22, 2008 1:28 am
by racerfern
That may be. Regardless, the goal is to slow the rush of air past the small bottom hole; that's where the pressure sensor is seeing a drop in pressure hence an increase in altitude. At least that's how I interpret what's going on. Travis?
I'm going to glue some little something around the hold with a fine mesh screen and go out and do some intervals. I'll report back when I get it done.
Re: why is elevation different for intervals on the same course?
Posted: Tue Jul 22, 2008 1:57 am
by alienator
racerfern wrote:That may be. Regardless, the goal is to slow the rush of air past the small bottom hole; that's where the pressure sensor is seeing a drop in pressure hence an increase in altitude. At least that's how I interpret what's going on. Travis?
I'm going to glue some little something around the hold with a fine mesh screen and go out and do some intervals. I'll report back when I get it done.
I don't think you want to do that. That will add an unknown variable, namely turbulence, an undefined pressure distribution, and a differential measurement that has a larger uncertainty and/or variance. Pitot tubes have always been designed this way, so they've always demonstrated the same measurement phenomena. The lower pressure in the hole in the bottom of iBike is easily corrected in software. Correcting for a new fixture, there, is not.
Also, you're just as likely to cause the pressure to decrease in that hole thus causing altitude to read even higher.