Silvio 1.5 vs 2.0 - Power vs Speed

[BBL]

Just for giggles, I plotted some average power vs. average speed data for the Silvio 1.5 and the Silvio 2.0. I think you will find the data interesting.

The vertical axis is average power in watts. The horizontal axis is average speed in miles per hour. All data points were collected from course loops in calm wind conditions. Although I have more Silvio 1.5 data, the eleven data points are all I have for the Silvio 2.0. As I get more, I'll fill in the chart, particularly at the higher power levels. But it's painful for this little old man to crank out much above 150 watts for any length of time.

There are complicating factors -- aren't there always! The tires on the Silvio 1.5 are 28mm Continental 4 Season (115 psi). On the 2.0 they are 25mm Rubino Pro III (125psi). The Silvio 1.5 has a standard SRAM GXP bottom bracket bearing. The 2.0 has an SRAM ceramic bearing. The Silvio 2.0 is probably a little lighter but nothing too significant here. Other than that, they are similar builds. Temperature was probably within about 10 F between all rides.

The form of the curves is not exactly what I expected for the Silvio 2.0, for which I anticipated the 18mph data point to begin turning the corner a little harder. Clearly, more data at higher speeds is needed.

Admittedly, this kind of data is crude and subject to the obvious errors. Still, it is valuable and paints a reasonably realistic picture of the kinds of results one may expect from both bikes.

 

[Charles.Plager]

I think you've just

I think you've just discovered aerodynamics...

 

[John Tolhurst]

Quite a bit faster for the

Quite a bit faster for the same power. Over 1 mph at 20 mph. hmm. That would have put Maria Parker's first 12 hour record from 240 to 264 miles. Hmm - don't tell her!!!!!!!!!

 

[flashburn]

I can tell you from anecdotal

I can tell you from anecdotal experience (the worst kind of evidence, I'm sure) that when I ride my lowracer with a 17 deg seat it feels like the upturn in the power curve comes somewhere around 20 - 22 mph.

So the linear trend you seem to have for the silvio 2.0 north of 18 mph seems to put it almost in the same ballpark for aerodynamics. A pretty impressive feat.

Here's some related charts:
http://www.m5-ligfietsen.nl/site/EN/Why_an_M5/Measuring_results

 

[pugwash]

Drag area

More interesting is the plot as drag area. Assumptions: Cr=.003, mass bike rider and gear = 100kg

if the assumptions are correct , the lines should be straight and horizontal. Whatever the absolute values, the CDA of the Silvio 2 is about 20% less that the 1.5

 

[iow]

thanks for the post - very

thanks for the post - very interesting

 

[Kavman]

Switch over wheels with tires?

It's a shame that the tires are different. Could you just switch over the wheels and tires from one bike to the other, just to see whether the results shift left and right?

Kevin

 

[BBL]

Drag Area

pugwash ....

Thank you for adding the drag area computations. The 100 KG assumption is accurate. I agree with your conclusion.

Kavman suggests switching wheels, but because of the way data is collected here, I doubt we will be able to see the difference. This does not mean there isn't any, just that the inherent inaccuracies in the manner I am taking data will not show it. I think it is remarkable how horizontal the CdA data are for the two bikes using Cr=.003 for both, further indicating that we won't be able to measure the effects of tires between the two bikes.

Bottom Line: The aerodynamic difference between Silvio 1.5 and Silvio 2.0 is large and might be as much as 20%.

 

[Doug Burton]

This is good work...

Hi BBL,

I see you're in Bumpass; I'm in Glen Allen, not too far from you.

What equipment are you using to gather raw data?

It looks like you're using MS Excel to plot your data. if you use the function "add trendline" some useful results can be obtained.

I do this type of work all the time in my "day job". If you're willing to share your raw data spreadsheets, I can port them to Minitab. I think there is a lot of additional info here that a little more powerful statistical analysis can tease out.

I'm really happy to see you taking a quantitative analytical approach to this. Not many folks are able to do this this well.

Cheers,

Doug

 

[BBL]

More Analysis

Doug .....

Thanks for your help. I am using the SRAM Quark (Riken) power crank. Picture below.

I chose it because it allows me to use wheels of my choice. On the Silvio 1.5, I mounted the magnet that came attached to a ring between the GXP bearing and the bottom bracket face. Since the Silvio 1.5 uses a split boom clamp and is inboard of the bottom bracket faces, it was an easy installation. Although I removed the power crank from my Silvio 1.5, I left the magnet in place. A picture is shown below.

Magnet installation on the Silvio 2.0 took a little more doing. I machined a pocket into the boom clamp and press fit the Quark magnet into position. Pictures are shown below.

On the bike, before everything gets assembled and you can't see anything, the boom clamp assembly looks like this. See picture below.

Had I to do over, and I still can, I would imbed a much smaller, but very powerful, rare earth magnet into the pocket, cover it with an encapsulant, and machine it back to the original contour. After touchup paint, no one would ever know there was a magnet installed. If I buy a Vendetta, this is what I plan to do.

OK, so that answers the question about the equipment I am using to measure power.

I am using a Garmin Edge 800 to record the data. It is set to average zeros and record on one second intervals. It is also set to stop recording at speeds below 3 mph (when I am stopped, which I rarely am). I calibrate the power crank using the Garmin Edge 800 prior to each ride, which typically last less than two hours. The rides are in the morning, usually completed before 9:00 AM.

I have had some time to look at the data this weekend (rainy Sunday). Working forward from what pugwash provided, I have attempted to characterize power vs. speed for the Silvio 1.5 and 2.0 bikes. I used pugwash's suggested Cr of .003 for the Silvio 2.0, used air density of for 30 in Hg at 73F, Rider plus bike weight of 203 lbs, then tried to reduce the high and low spreads of deviations in data and zero the sum of the errors across the range of data. I think I could play with this until the cows come home in Bumpass and still not converge on a solution! The numbers I settled on for the Silvio 2.0 were Cr=.003 (pugwash's suggested value) and CdA of 4.26 (best fit?). For Silvio 1.5, I used a slightly larger Cr = .0035. This is a WAG in an attempt to account for the presumably lower rolling resistant tires and ceramic bearings on the Silvio 2.0 bike. I came up with a CdA of 5.38 (best fit?) for the Silvio 1.5. The figure below shows how the characterization fits for the Silvio 1.5 and 2.0 with my measured data.

The Cr assumptions (or arbitrary selections within a rather wide range of acceptable choices) are a bit weak. Maybe you can do something with the data. Yes, I have the data on an Excel spreadsheet. I would be glad to e-mail it to you.

Maybe the Silvio 1.5 fit looks a little high? Anyway, good luck Doug!

 

[Ivan]

Ha ha! BBL, I don't suppose

Ha ha! BBL, I don't suppose that machining equipment is in your garage! Very cool stuff...

 

[Doug Burton]

Wow, you have the toys!

Very impressive work. Cool machining toys are no good without the skillz, which you obviously have in spades.

I would very much like to take a cut at your data. The inference is that this is a really efficient bike (which anyone whose ridden one knows) but you are the only person I know who has such a cleanly comparable data set for both frame configurations. Solid quantitative measurements are so hard to come by when we evaluate our design progress.

I checked out your other thread on this bike, it is a really nice technical magnum opus. You've really maximized opportunities Di2 provides for a clean-looking bike.

As John would say, "Good Onya!"

Cheers,

Doug

 

[COBRA GT]

Speed differences / Old Silvio versus new

I would like to hear from old Silvio 1.0/1.5 owners the differences with their new 2.0. Sprinting , climbing etc../O.P.? Did you have chain stay ext. on both bikes?

 

[BBL]

Chain Stay Differences

GT ....
I did not have the 100mm chain stay extension on my Silvio 1.5. In my opinion elevating your feet and legs by adding the 100mm chain stay extension will tend to lower the drag.

Since the time of posting I have played more with the data. My best fit (by eyeball) is Silvio 1.5: Cr=.0055, CdA=4.45 sq.ft. Silvio 2.0: Cr=.0050, CdA=3.75 sq.ft. Fit curve using these coefficients is shown below, for what it is worth.