Kreuzotter Calculator - How Accurate?

Porkyboy

Hi

Does anyone have any idea how accurate or inaccurate the above might be, has anyone done any field tests with a power measuring device? I've seen the calculator mentioned quite a few times. The reason I ask is because I've just plugged in my weight, height, bike weight etc. and done a comparative calculation to see the power saving by simply shifting from the tops (top of handlebars) to the drops (bottom of handlebars) and changing nothing else.

According to the calculator if I ride on the tops and produce 160W I'll move along the road at 16.5mph. If I then just shift to the drops and keep riding at 16.5mph I'll only need to produce 121W.

How accurate does this sound to the experts out there? Sounds like a heck of a drop in power requirement for the same speed to me

Thanks.

PB

__________________
What do you mean your legs are hurting? Give it some welly man!
Visit My Training Weblog

squidwranglr

I think the "golden standard" (as accepted on these forums) for web sites providing analytical calculations involving cycling is Analytic Cycling. I tend to prefer the Kreuzotter calculator because the interface is far easier to use and haven't seen its answers to be dramatically different than those of Analytic Cycling.

The thing to keep in mind with either of these calculators is the principle of "garbage in, garbage out." Their accuracy depends solely on the accuracy with which you can provide the input measurements or conditions and depending on how the various parameters are used mathematically in the calculations, single-digit percentage inaccuracies in the measurement of these inputs can be amplified into far greater inaccuracies in the results.

Berend

daveryanwyoming

No argument to your post, but one key difference is that analyticcyling doesn't give you a canned estimate for CdA based on tops or drops. You've got to plug in your frontal area which is what riding high or low changes so you can't really compare the numbers he got to what you might get from analyticcyling.

But in terms of the OP's question. That doesn't seem totally out of line for riding on the drops vs. tops on a windless day(have you actually found many of those?). You're getting more aerodynamic by lowering your frontal area in the drops, 16.5 mph isn't really fast, but as you go faster the power savings you get by dropping frontal area becomes even more dramatic which is the whole point of aerobars.

I think the 40 watt savings at that speed might be a bit optomistic but not totally out of line depending of course on how your bike is fitted and how aero you actually get when riding in the drops. I know I can add about 2 mph for the same power when riding in my aero bars on a TT bike and that's at ~ 25mph. I'm saving a lot more than 40 watts at that speed by getting aero but it's a lot more streamlined than standard drop bars and I'm talking about faster speeds where the savings is a lot more signifigant.

-Dave

RChung

The reason the "interface is far easier to use" is because it makes far more assumptions about the amounts the resistance. For example, the Kreuzotter calculator uses your height, weight, and a rough indication of position ("hands on bartops" or "hands in drops") to estimate drag area using a hidden formula. It also presumes that rolling resistance is only a function of tire pressure, and the amount of loss due to drivetrain inefficiency is unspecified.

However, position can have a huge effect on aero drag -- and whether one has a higher stem or narrower handlebars or deeper drop bars or floppy wind jacket or long torso and short legs can have much more effect than simply choosing whether your hands are on the tops or in the drops.

Likewise, the amount of rolling drag from a Tufo Elite Jet Tubular vs. a FMB tubular is huge, even at the same pressure.

Kreuzotter makes assumptions about all those things and is easier to use. Analyticcycling makes none of those assumptions and puts the burden on you to enter the right values. If you don't know what those values are you'll have to use Kreuzotter but don't mislead yourself into thinking the answers you get will be more "accurate" than if you actually knew them.

dkrenik

Another option is from MachineHead Software. They have a power calculator that kind of falls in between Kreuzotter and AnalyticCycling. It's an app that you download to your PC and I've found it easy to play around with a few parameters to get it to agree with my SRM.

Dave

lanierb

I once got sick of using those web interfaces so I made an Excel spreadsheet that will compute speed given power, CdA, gradient etc and will also let you break up your route into sections if you want to plan a time trial, for example. Would you guys be interested in seeing it? For me, anyway, it's much easier to use than either Kreuzotter or analytic cycling.

Alex Simmons

Sure, why not?

__________________
http://alex-cycle.blogspot.com/
http://www.cyclecoach.com/

lanierb

Yeah, I figured that was a stupid question. I put a copy here:

http://www.stanford.edu/~lanierb/Pacing.xls

Let me know if you have any comments.

Roadie_scum

Any chance of this calculator:

Enter course segments (gradient, length, wind), rider characteristics (mass, crr, bike mass, CDA, etc), then power for each segment, then solves for speed/segment time?

If it then used all the segment powers and the time at each power to calculate a normalized power... it would be a real blast...

Anyone know of something like this?

Squint

This might be what you're looking for:

http://www.whitemountainwheels.com/Speed_Power_v2.xls

Roadie_scum

Glorious! I am gonna write the NP bit myself...

acoggan

Now here comes the tricky part: how do you deal with the 30 s smoothing when the time required to transit each segment is variable?

Roadie_scum

I have minimal excel skills... so maybe this is harder than I think?

But couldn't you use some kind of function to generate a list on a separate sheet with a datapoint every second for each segment? Then do the calculation on that?

Eg... for a segment that takes 45s generate 45 datapoints at the ascribed wattage. Then combine all these generated points into a single column, then smooth, then raise to the fourth power... etc...

Is this really hard? Maybe I will find out today...

tigermilk

I wrote my own program that would

1) interface with TopoUSA to get course data (road direction, elevation)
2) allow wind and random gusts
3) constant or variable power input

and could be easily modified for variable CdA, Crr, etc. If you just use numerical integration, it's not difficult to do all this.