Analyzing Wind Post Ride vs Real Time data

Post Reply
turbomentor
Posts: 85
Joined: Sun Dec 07, 2008 6:51 pm

Analyzing Wind Post Ride vs Real Time data

Post by turbomentor »

I am playing with a PT/iAero combination and comparing ride files. Quite interesting.

My first ride the PT data showed significantly higher wattage than my GIII iAero. I emailed Velocomp and was advised to adjust my ambient wind to 0 which caused the iAero data to marry up pretty well with the PT data. All well and good.

This is great to have the tools to do that in the IB3 software. I have read on other posts about using wunderground to get historical wind to plug in to use to adjust your iBike data. My question is: I would like to improve my real time accuracy while riding! A lot of my riding currently is doing intervals. Obviously, I'm trying to hit specific target numbers. If I am hitting the numbers on the iBike only to discover post ride that I was too low or too high (after I adjust for wind) it seems that the value of the iBike is somewhat diminished.

I did a very careful cal ride with 6 coast downs. I do a wind cal before every ride. I do NOT do a tilt cal before every ride as I understand this is somewhat detrimental.

What else can I do to improve my real time data accuracy?
rruff
Posts: 445
Joined: Wed Apr 23, 2008 10:48 am

Re: Analyzing Wind Post Ride vs Real Time data

Post by rruff »

turbomentor wrote:My question is: I would like to improve my real time accuracy while riding!
Your calibration is off... the wind scaling number. This is determined during your out-back ride.

It isn't difficult to adjust. Simply do another out-back ride, preferably on a reasonably calm day, or at least when the wind is consistent. Afterwards iBike3 should show a net wind of zero. If not, IIRC take the avg speed from that graph, divide by the avg wind speed, and take the square of this number. Multiply this times your wind scaling factor to get the new one. Do another out-back ride to see if it is correct.

For instance, say your current wind scaling factor is 1.0, and the "analyze wind" graph shows an avg bike speed of 17 mph and an avg wind speed of 15 mph. So (17/15)^2= 1.28 x 1.0= 1.28.
turbomentor
Posts: 85
Joined: Sun Dec 07, 2008 6:51 pm

Re: Analyzing Wind Post Ride vs Real Time data

Post by turbomentor »

How do you know the wind scaling factor is off? It could be the cRR or the cDA right?

I guess on a reasonably wind consistent day (or ideally a NO WIND day) an out and back ride should yield a net ambient wind of zero, correct? So if I can look at my cal ride and see if it's the case. If not, I will do a recal ride.

It would be very helpful if we could compile a list of ah-ha's and gotchas that might prevent a good setup/cal for new IB users such as myself. I am an engineer and a pilot so these principles aren't difficult to understand, but this particular fix would have been nice to know all along! I may have missed reading where the wind speed and actual speed should be the same in the installation guide as well...
rruff
Posts: 445
Joined: Wed Apr 23, 2008 10:48 am

Re: Analyzing Wind Post Ride vs Real Time data

Post by rruff »

Wind scaling is the most likely thing since adjusting the wind in iBike3 fixed the problem. Of course, it would be nice to hear from him on this...

Yes, you should have zero net wind on an out and back. You probably don't need to do a recal, though. The out and back is completely different from your coast downs, which determine CdA and Crr. The out-back is only for determining the wind scaling and it is easy to tweak this after the fact.

The calibration usually works well, and obviously Velocomp does not want to give the impression that you need a degree in physics to get a good cal and understand what is going on. A "checklist" might be a good idea... but on the other hand, I'm impressed with how responsive they are to questions and problems.
turbomentor
Posts: 85
Joined: Sun Dec 07, 2008 6:51 pm

Re: Analyzing Wind Post Ride vs Real Time data

Post by turbomentor »

rruff wrote:Wind scaling is the most likely thing since adjusting the wind in iBike3 fixed the problem. Of course, it would be nice to hear from him on this...

Yes, you should have zero net wind on an out and back. You probably don't need to do a recal, though. The out and back is completely different from your coast downs, which determine CdA and Crr. The out-back is only for determining the wind scaling and it is easy to tweak this after the fact.

The calibration usually works well, and obviously Velocomp does not want to give the impression that you need a degree in physics to get a good cal and understand what is going on. A "checklist" might be a good idea... but on the other hand, I'm impressed with how responsive they are to questions and problems.
Thanks for your directions on how to fix this.

I went in and did the calcs you specified. Rode yesterday and ambient winds = 0! Beauty. Unfortunately, today (same course, same temp, same ride) ambient winds = -1.3! Dunno how to explain that. A bit frustrating that something is happening to cause this. Or maybe that's just the nature of the IB.

I hope I can get this issue worked out!
coachboyd
Posts: 527
Joined: Thu Mar 27, 2008 7:01 pm
Contact:

Re: Analyzing Wind Post Ride vs Real Time data

Post by coachboyd »

Having an average wind of -1.3 on a loop is perfectly fine. Think about most of the loops we do, they go all over the place. You are going in one direction for a while, another for longer, and the winds can be sheltered by things like hills, trees, buildings. There are so many variables that would make it so your average wind speed compared to wheel speed is not 0mph.

With the cal ride you are supposed to go out two miles and back two miles on a traffic free (relatively) road. In this situation when you are not changing directions and for that small of a sample time, you can expect 0.0mph relative wheel speed. In fact, this is how wind scaling is calculated.
Boyd Johnson
http://www.boydcycling.com - high performance carbon wheels and accessories
rruff
Posts: 445
Joined: Wed Apr 23, 2008 10:48 am

Re: Analyzing Wind Post Ride vs Real Time data

Post by rruff »

What Boyd said. With variable winds, drafting, etc there is no reason to expect a "0" result. -1.3 is a fairly large deviation, but my last two rides (each about 50mi) were about 2 mph from zero. This was believable though, since it was windy and the wind speed and direction varied during the ride. The best "check" is to ride back and forth on a course (1/2-2mi say) with little or no traffic, and ideally either little or no wind or no obstructions to the wind (ie flat and open). You should expect close to zero wind in these conditions.
coachboyd
Posts: 527
Joined: Thu Mar 27, 2008 7:01 pm
Contact:

Re: Analyzing Wind Post Ride vs Real Time data

Post by coachboyd »

Exactly, sometimes I have people do that if I see a ride with suspect wind readings. . .like the entire ride with a tailwind. I'll have them do a two mile out and two mile back ride in the normal screen (the lap feature is great for this) and I look at their wind readings for just that part.
Boyd Johnson
http://www.boydcycling.com - high performance carbon wheels and accessories
djconnel
Posts: 34
Joined: Tue Jan 13, 2009 11:33 am

Re: Analyzing Wind Post Ride vs Real Time data

Post by djconnel »

rruff wrote:Velocomp does not want to give the impression that you need a degree in physics to get a good cal.
I have a degree in physics and I have to admit to finding all of this pretty scary! It's enough of a challenge for me to make sure my tire pressure is good before a ride; allowing a computer to come into thermal equilibrium with the ambient, doing a wind-zero test, then doing out-and-back rides in sufficiently low wind conditions.... add in having to weigh myself and my bike, make sure the batteries aren't ready for their bimonthly replacement.... I've not bought an iBike yet but was considering getting one as an alternate to replacing a malfunctioning Powertap head unit. I'm starting to think the extra data promised by an iAero might not be worth it! I admit the CdA extraction, wind speed determination, and route profiles with gradient are attractive, though.
coachboyd
Posts: 527
Joined: Thu Mar 27, 2008 7:01 pm
Contact:

Re: Analyzing Wind Post Ride vs Real Time data

Post by coachboyd »

It may seem like a big list at first. . .and the results you get are dependent on getting a good calibration (which could possibly take a couple tries). But once you start to learn the device, it really is a snap to get going. Every ride, I just put my iBike outside while I am getting dressed for the ride. Then when I get outside I put it on the bike, check the wind offset (take about 5 seconds), and go. I usually go two days at a time between pumping up my tires, I haven't noticed any change with that. I did notice a change the day I headed out and my wheels were at 70psi.

One of the things we pride ourselves on is the customer support. When you get your unit and do the coastdowns and calibration ride, you can send me the file and I will take a look at it for you.

Every device has it's learning curves, from mp3 players to computers to power meters. The first couple rides you may seem a little lost about all the features we offer, but usually within a week everybody learns all the features and setup processes.
Boyd Johnson
http://www.boydcycling.com - high performance carbon wheels and accessories
turbomentor
Posts: 85
Joined: Sun Dec 07, 2008 6:51 pm

Re: Analyzing Wind Post Ride vs Real Time data

Post by turbomentor »

I have found the iBike to be very easy to use. Very fun to play around with. Customer support has been second to none, not to mention the great feedback from the forums.

In my posts I am just trying to get to know the device, what it's capabilities and limitations are and how to best use it, and what to expect out of it.

I just got the ANT+ sport hub updater from Saris and am going to work on that over the weekend. Next project is real time continuous Cda measurements. Looking forward to that. That is one reason I'm trying to dial in my iAero so precisely.
Post Reply