Given that I am the kind of person who can be motivated by numbers, I decided to pick up a bike computer today – the simplest waterproof model available at MEC. After installing it, I wanted to make sure I had selected the correct wheel size (I think it’s 2174mm on the 700x32c wheels of my Trek 7.3FX). A few kilometres of cycling allowed me to confirm both its measure of velocity and distance traveled against my GPS receiver (a marine unit too big and cumbersome for cycling).
Unfortunately, it also confirmed that the little rare earth magnet that the sensor detects shifts around quite easily on my spoke, and it needs to be very carefully aligned to work. First, I tried gaffer tape, but it really wasn’t right for the job. Then, I tried the blogosphere, which suggested superglue. Glued in place, I hope that magnet isn’t going anywhere while I rack up the kilometres over the coming months.
For those keeping track, the trip out to get the computer, return home with it, and calibrate it amounted to 17.8km.
What GPS do you use?
I have a Garmin 76CSx, which also happens to be a floatable marine unit (great for hiking and canoeing), that I also mount to my bike. It is a bit larger but not all that out of place if you mount it to the handlebar stem rather than out on the handle bar. Of course finding something to mount it to the stem was the hard part. The mount I ended up going with is the RAM RAP-274-1 mount and the corresponding RAM-HOL-GA14 Cradle. So far its rock solid (RAM Mounts are known for their toughness) and with the easy on off feature quite handy as I bought two of the mounting assemblies so I can switch my GPS between the mountain and road bike (and bring it inside at work when commuting).
Here is a pic of it mounted to my road bike to give an idea of size:
http://farm4.static.flickr.com/3536/3470011392_1325bc875f_b.jpg
I looked into a couple of cycle computers, but I would miss the tracklog my GPS keeps. OTOH a computer is good for an indoor training stand where the GPS is obviously of no help, which was why I was looking at them and if you get a fancier one it can do cadence, power, etc. The perfect solution would be the Garmin Edge 705 released last year but I couldn’t justify the expense when I already had a perfectly good GPS. Actually for touring I prefer my GPS as it uses easily field replacable standard AAs vs. the Edge’s custom Li-Ion battery which poses a charging conundrum when camping out on a tour.
I also have a Garmin GPS76. I got it after I realized the patch antennas in their smaller devices are useless under trees and in mountain valleys.
The 76 is much better. I suppose I could have mounted it on my bike, but it’s really too complex a thing to operate while cycling, and it’s a bit of a battery hog.
All told, I don’t think the $13 (plus tax) for the bike-specific device was a waste of money.
I cannot imagine going as long as you did without a bike computer. I am addicted to them as I enjoy keeping track of what they offer.
Its definitely a sensitive GPS as it uses the Sirif III chipset. I can get accuracy down to 3m, even under canopy.
I don’t find it hard to use on the bike, I set it to the trip computer page where it has all of your stats in easy to read boxes. (Speed, Trp Odom, moving time, etc… which can all be customized with quite a few user selectable data items) and leave it there, but what I like is it records my ride log in the background so I can upload and analyze it later. Which is a plus for actually mapping off road trails on the MTB. I’ve mapped quite a bit of the greenbelt in the Stoney Swamp area where the NCC maps don’t even cover half the actual trails. Not to mention if you are riding somewhere you don’t know, having a map at your fingertips is invaluabe, saves getting a map out of your pack.
I actually get a good 12 hours out of 2 rechargeable AAs. About 16 if I use alkaline. I can go about 2 weeks commuting before I have to recharge them. Definitely not as friendly as a watch battery that lasts 2+ years in a regular cycle comp though.
YOU can be motivated by NUMBERS???
Come on now. Stop messing with me.
… sandwich economics …
Its definitely a sensitive GPS as it uses the Sirif III chipset. I can get accuracy down to 3m, even under canopy.
From what I understand, it’s the antenna that really makes the difference. Hiking in North Vancouver, under trees and between mountains, I found that the newer and smaller Garmin units were useless, but the 76 often worked even in a backpack.
I don’t find it hard to use on the bike, I set it to the trip computer page where it has all of your stats in easy to read boxes.
You may have better eyesight than I do. During my bike computer calibration run, I really had to struggle to read the trip computer page while in motion. That said, it might have been easier with a mounted device.
Which is a plus for actually mapping off road trails on the MTB.
The 76 is useful for mapping (though perhaps less so than a fancier device that includes a built-in map). I used it during countryside and canalside trips in Oxford, to give me a better idea of how to get back into town.
I actually get a good 12 hours out of 2 rechargeable AAs. About 16 if I use alkaline. I can go about 2 weeks commuting before I have to recharge them.
Part of the problem for me is that I tend to use the 76 infrequently, so it suffers a lot from self discharge. As such, it always tends to be almost dead if it has rechargeables in it. I switched to alkalines, which should make things less annoying.
From my research it was the Sirif III chipset that made the difference… That would be the “x” in the model number 76CSx. There were older versions of these GPSs that had the Sirif II chipset (76CS) but otherwise identical hardware (minus an SD card slot) so same antenna, casing, etc. This review of the 60CS vs. 60CSx shows the chipset difference with all other factors being equal.
http://www.patrick-roeder.de/reviews/garmin_gpsmap_60CSx.htm
It would be the same for the 76CS vs. 76CSx as it uses the same components as the 60 just in a different casing.
That said the 60 and 76 models do supposedly have a better helix antenna in them as well compared to the smaller eTrex Legend, Vista, etc…
Though I do apologize, I feel a bit like Lisa Simpson, the answer to the question nobody asked… but it is Friday and I would much rather be riding my bike then sitting in my cube talking about it, especially given the weather today!
One quick aside about spam filtering:
In WordPress, there seems to be a significant trade-off between the speed at which a site runs and how good it is at catching spam while not blocking legitimate comments. That is because the best spam blockers don’t seem to work well with WP Super Cache, which is necessary for making a site run on cheap hosting, like my current GoDaddy setup.
My apologies if your real comments get eaten. I generally did through my spam every couple of days and will restore them on sight. Otherwise, send me an email to alert me more quickly of the error.
Another 20.2km today: along the canal to downtown, then out along Sussex past the PMs house, farther into the very fancy neighbourhood beyond, then back, across the Cartier Bridge, along the Quebec side of the river, across the Confederation bridge, and home.
Perhaps I will aim to do 50-70km per week, as a start.
I will log future distances on this free site.
In my experience I’ve found the most accurate way to set the wheel diameter is to measure how far the bike travels forward for one revolution of the wheel, using a spoke as a marker. I’ve found this is a lot more accurate that the pi x diameter method. Your result can change by a few centimeters which can be a few percent difference, particularly in the distance displayed.
I agree with Matt, although there is no reason to limit yourself to one wheel revolution. When I set up my last bike computer, I think I used 3 or 4 revolutions and divided, using a long tape measure. Accurate to the millimeter!
If you wanted a more accurate velocity gauge, you could put more than one magnet, then divide the wheel size number appropriately. That way, the data would be more frequently updated.
If you really want to appreciate the energy penalty associated with higher velocity, I accidentally invented an effective demonstration:
Choose a street that is (a) normally quite busy (b) studded with stoplights every block and (c) largely abandoned at the time of the trial. Somerset in Ottawa after midnight on a weekday works well.
Once on the street, race every car that comes up beside you for as far as you are physically capable of doing so. Doing this for 5.59km was exhausting: far more so than cycling 40km at a natural feeling speed. At the end, I was out of breath and sweating profusely, though I hope I will be able to handle it much better after a few more weeks of dedicated cycling.
Over the course of the 17 minutes, I wasn’t even averaging a dramatically higher speed than normal, but it was enormously more tiring. On one downhill section, I did manage to hit 51.02km/h, however.
You just described the concept of interval training!
If you want another intense workout, try hill training. Riding from the Ottawa side to the Champlain lookout in the Gats and back really humbles you (and the back portion is mostly downhill so its fun!) but just getting up to the lookout can be tough.
50km/h is seriously fast on a bike. There are no good places to hit high speeds in Toronto – the best I managed was 65km/h downhill on a busy street, passing at speed through an intersection. That’s faster than one really should go when conditions are not ideal.
I managed over 90km/h when I was young and stupid… but my favourite record at the time was a 71km/h on a slight uphill. I was in much better shape then.
Since I mostly enjoy travelling by bike (instead of commuting or going for afternoon rides) I now count by villages (i.e. 3 villages to go before lunch). In the St-Lawrence and Lower Ottawa valleys, this generally means one village per 10-15 km. This is mostly due to the fact that my last 3 cycle-computers died prematuraly and that I commited not to buy cheap ones again.
That being said, I barely use my bike anymore… work and familly are taking a toll.
Don’t Let Drag Interfere With Your Cycling Training
Cycling is a sport where forces are constantly working against you. Gravity tries to drag you back down every incline, from a short hill to a long climb. Just as importantly, aerodynamic drag makes every mile per hour harder and harder as you pass 15mph. Over this critical speed, aerodynamic drag exceeds rolling resistance caused by the tyres and mechanical resistance in the drive train.
…
Drag increases as the square of velocity; however, the power required increases as a cube of the velocity. For example, recent research data show that to travel at 21mph requires around 190 watts; however, it requires another 110 watts just to go 4mph faster.
To convert some of that to metric:
“Just as importantly, aerodynamic drag makes every kilometre per hour harder and harder as you pass 24 km/h. Over this critical speed, aerodynamic drag exceeds rolling resistance caused by the tyres and mechanical resistance in the drive train.
At 32km/h a rider will displace over 454kg of air per minute, and the wake it causes requires a lot of work to overcome.
Drag increases as the square of velocity; however, the power required increases as a cube of the velocity. For example, recent research data show that to travel at 33.8 km/h requires around 190 watts; however, it requires another 110 watts just to go 6.43 km/h faster.
If you want to ride fast, you need a lot of power!”
“If you want to ride fast, you need a lot of power!”
Fortunately the cyclist does not always need to supply this power thanks to inclines and back winds.
I am up to 189.46 km since I got the computer.
I did another late night, interval training-type run up Somerset, then along the canal and Ottawa River pathway tonight. They are quite exhilarating, and tolerably safe when done at very low-traffic times.
What kind of bike light do you use later at night, if you use one, or are the canal and river pathways reasonably well lit (I know Somerset isn’t a problem)? Getting one is my next purchase.
It doesn’t seem that bright out behind the Hill and the war museum, though I took in a show at the cheap seats at Bluesfest last year without any problems, but there were tons of stage lights, etc…
I have some moderately powerful Cat Eye bike lights, but last night I was also using my 4-LED headlamp and a bright red blinker clipped to my back pocket.
Actually, my original Cat Eye front light failed shortly after I got it. The shop replaced it with a Sigma Sport TRILED. It’s a good light, but not powerful enough to be your sole front light on a dark path or trail.
Since getting the computer, I have cycled 325km. That averages out to 13.54km per day.
So far, I have cycled for just under 22 hours: about 3.82% of the total time elapsed since I got the computer, and about 5.73% of my waking hours.
I also find charting the amount of cycling on a computer enjoyable. The increased daylight hours allow for more of this.
Another fun event is to participate in a critical mass ride. Vancouver has one on the last Friday of each month beginning at 6:00 pm on the north side of the Art Gallery. This is a fun group event. I had some qualms in obstructing the flow of motor vehicle traffic and alienating the motorists with whom I share the road . However, I overcame that.
Perhaps Ottawa has something similar. I believe Toronto does and I participated in one during Bike Week.
Ottawa has critical mass rides… or they did. I participated in one last July… I am not sure what the status of it is now though, I haven’t heard of one this year.
http://www.punkottawa.com/community/criticalmass/index.shtml
With critical mass I find the ride itself enjoyable but I think the message of cycling advocacy is lost. I readily agree with the chant “We aren’t blocking traffic, we ARE traffic!”. It’s a great slogan. However if one is expected to be taken seriously one should obey the laws, which means not going through reds, not being a clown and taking up all lanes on a multi-lane road so cars can pass, etc. I do understand it is meant as an act of civil disobedience but it seems a counter productive way of promoting cycling, by being a deliberately bad cyclist.
Now on the ride I was on most riders stopped at reds, however “corkers” would block traffic on through streets so the entire mass could get through and not split up, sometimes resulting in the back of the pack going through the red. Obviously the two abreast rule was broken, as it was the point to block traffic. All in all our ride was very reasonable, however I think our route hit a snag when it went down Colonel By Dr. for a long stretch which is a single lane road, and likely backed up traffic and angered many motorists… I blame that on the (deliberate) disorganization of the movement. No one leader, no preplanned route… even the ride itself is organized flash mob style… I think they do it so blame cannot be pinned on one individual.
As it turns out we were busted by the police. OK busted is the wrong word, about 3/4s of the way through the ride about 6 cruisers stopped the mass in Chinatown just off of Somerset (which was not much more than a bulge at that point – about 30 riders from about 60 at the start) and after talking with one of the guys who stepped up into a leadership role the police agreed to escort us into the Market… Two cruisers, one at the front and one at the back with flashers on (no sirens though) lead us down Wellington, past Parliament and into the Market. It was kinda neat. They don’t often get enough kudos, but good on the cops for dealing with situation in the appropriate manner without malice.
All that said the message is good, I just don’t know if this is the best way to deliver it.
I have attended a few different Critical Mass rides, including both Ottawa and Oxford in 2007.
I may well attend some others here this summer.
I had some qualms in obstructing the flow of motor vehicle traffic and alienating the motorists with whom I share the road . However, I overcame that.
That sentence made me laugh.
For example, recent research data show that to travel at 21mph requires around 190 watts; however, it requires another 110 watts just to go 4mph faster.
In reading this, I hope people don’t think these numbers are absolute, because they would be very heavily dependent on the aerodynamics of the bike and rider combo. With the numbers above 300 watts yields 21mph. Only a highly trained athlete could sustain 300 watts of power, but I’m sure that the same highly trained athlete could sustain easily above 25mph (on flat ground) with a good bike setup. Similarly the enclosed bikes used for record breaking attempts would absolutely fly with 300 watts of power.
^
Correction to the above:
With the numbers above 300 watts yields 25mph.
Matt,
Can you think of a practical way for an ordinary cyclist to test their coefficient of drag?
I am guessing you could do so by riding at a constant speed, testing power output somehow, and using the drag equation.
You would just need to know the mass density of air under the conditions where you are riding, and your area.
Fall 96 – Considerations for Data Quality and Uncertainty in Testing of Bicycle Aerodynamics
By: Michael J. Flanagan, Ph.D.
Reproduced from: Cycling Science Fall ’96
For this test effort, it was possible to simulate the full-scale environment. As the test article was full-scale and the wind speed was matched precisely, duplicating the real-life flowfield was simple. With no post-test manipulation of the data necessary, it was believed that highly accurate drag measurements would be possible. Little did we know that, in our attempt to measure highly accurate drag levels associated with bicycle aerodynamics, we would uncover a Pandora’s box of data quality issues related to the very nature of cycling dynamics. The nature of this testing did bring to light a peculiar point of view in the age -old quest of accuracy versus precision. Simply put, even though a test can be organized and run utilizing state-of-the-art equipment giving precise answers to 0.05 lb, the flowfield can be so complex given such variables as wheel rotation, pedaling speed (cadence), and rider posture that accurate measurements are impossible. Confused? Don’t be! This is the age-old question of accuracy versus precision. The answer you get is very precise – unfortunately, it is precisely the wrong answer!
For the person who wants to experiment with aerodynamic modifications, there is still another another quite inexpensive method that gives only relative predictions, but supplies no absolute values (9). With constant boundary conditions, heart rate is a very good approximation of the rider’s applied power. Modern heart rate monitors with chest electrodes can measure (partly) one heart beat per minute exactly.
With the heart rate monitor, the rider can keep his power nearly constant and repeat a time trial with different equipment, thus detecting possible differences by the different elapsed times. The method with the heart rate monitor is suitable even for the evaluation of different drive systems (e.g. round chain rings versus a drivetrain modulating the angular velocity).
This method can be simplified somewhat by using rear wheel hubs which measure drive torque – like the ones by Look and Balboa Instruments (The Power Pacer) to be produced shortly.
Off the top of my head I can’t think of a practical way to measure rider power, which would be the first difficult step in trying to even ballpark your drag.
I think that a professional setup probably uses strain gauges on the cranks to get torque along with cadence (crank RPM) to get power.
Perhaps an easier method would be to install a motor with a set power level in as aerodynamic way as possible, then have it drive the bike forward while the person pretends to pedal.
That’s an interesting idea. You could also measure wheel power with a dynometer which would allow you to subtract drivetrain and rolling resistance losses (at least for your rear tire) from the power equation. Definitely some assumptions would have to be made, but I think you could come up with something reasonable given access to the right equipment.
A pedaling robot with a suitable shape and surface could also work. It could be programmed to keep applying power until it reached a certain constant speed. I presume it would be easy to measure its power output, by looking at how its motors were using energy.
In thinking about this about further, wheel power is actually the number you are after, because it is what is driving the bike forward. We don’t care about the drivetrain losses, etc. This would be easy enough to measure with a dynometer, and you could figure out what your sustainable pedal power was if you averaged a few trials. After, you could measure your maximum sustainable speed with a bike computer or GPS (hopefully on a no-wind day). You could roughly figure out from this your drag. It wouldn’t necessarily be accurate to 10 decimal places (far from it), but it could give you some ‘good enough’ type numbers.
What I think we are most interested in is how the drag force varies with speed, both in the context of cycling and other modes of transportation.
That might be even easier to measure, since absolute values aren’t required.
If all you are concerned about is how drag varies with velocity, couldn’t you use the drag equation linked above, cancel out most of the terms, and get a solution without doing any experiments?
Well, I get that, but the coefficient of drag is going to vary enormously between bikes and riders, and even the same rider in a different riding position, which is what my initial point was: the post saying that a 4mph increase at 21MPH initial speed requires 110 watts more power is completely specific to one coefficient of drag. My feeling is that a bicycle is going to have a CdA (coefficient of drag X frontal area) that’s bigger than a medium to small car. (Which is doesn’t mean the car’s the more efficient choice, you lurkers!)
Anyway, it’s a good post from an illustrative point of view, but my worry is that someone’s going to read that and quote it as though it’s absolutely absolute for all situations.
Matt,
One more way to estimate the drag force at different speeds would be to get a rider who is in good shape and has a good understanding of how their body works, have them cycle at a set speed, then make them cycle on an unmoving machine that measures power – with them trying to match their power output from before.
Perhaps the most accurate approach (and this would all be less accurate than some of the other options) would just be to have them pedal all-out on the measuring machine, then do so on a real bike, measure their velocity on the bike, and work out drag from that.
I agree that it is misleading to think the specific figures cited apply to any bicycle. That being said, the general concept that increases in velocity cause disproportionate increases in drag seems to be a valid one.
R.K.,
I don’t think that would work, though I cannot explain why mathematically. Could it be because that equation is just an approximation?
Could it be because that equation is just an approximation?
The equation is completely accurate for low speeds. Nearer to the speed of sound the compressibility of air will distort it. I’m not sure what R.K. was asking to be honest (to construct a physical proof of the equation?) but it is known to be valid.
Perhaps the most accurate approach (and this would all be less accurate than some of the other options) would just be to have them pedal all-out on the measuring machine, then do so on a real bike, measure their velocity on the bike, and work out drag from that.
This is what I was getting at in the post 5 or so above this one.
I see. I thought you were proposing installing a dynometer between the pedals and the rear wheel, when I first read that comment.
I’m not sure what R.K. was asking to be honest (to construct a physical proof of the equation?)
I think the idea was that if velocity and drag are mathematically related through the drag equation, perhaps it should be possible to derive their relationship from the equation itself, by cancelling out the other terms.
We want to know is how drag will change with velocity for a particular vehicle.
In that case, the mass density of the fluid, reference area, and drag coefficient would all be constant. That would leave force and velocity as the only variables.
For a cyclist, it seems plausible that the drag coefficient would not be constant at all speeds. Those in normal clothes would have loose fabric flapping at high speeds, posture would change, etc.
We want to know is how drag will change with velocity for a particular vehicle.
Oh. Well, for a vehicle of fixed geometry ( a car, train) drag increases with the square of speed through air and power to overcome said drag increases with the cube of the speed through air (I think this was mentioned above). The reason that power goes up by the cube and not sqaure is because, in this case, Power = Force * Speed. If you substitute the force of drag which already includes a speed squared term, you get a speed cubed term.
It’s very easy to measure the drag of any vehicle, so long as one is willing to lump together aerodynamic and mechanical drag. What one needs is a speedometer, a clock, and a recording device. It’s called a “Coastdown” test.
Accelerate to whatever speed you want to determine the drag from, begin recording, and stop pedaling. When you get home, look at the data, and graph the deceleration compared to the zero-drag deceleration (constant speed). The negative acceleration with respect to constant speed will allow the relevant figures to be calculated.
For instance, if it turns out the deceleration is 2 meters per second per second, and KE=1/2MV^2, you just need to know mass, figure KE at, say, 10m/s and 8m/s, subtract will give you delta KE, and change in KE time over is power. That power is the drag. To get coeficient of drag, you need to compare actual decelerative force with the idealized situation of something with the same frontal area but entirely flat, and the ratio between those two figures is the CD. Once you have the CD, multiply it back by your frontal area to get theoretical loads at high speeds. Of course it’s not perfect because I’ve forgotten about mechanical drag, which is linear whereas aerodynamic drag is logarithmic.
To be much more simple- all you need to do is accelerate to 40km/h, and measure how long it takes to coast down to 35, 30, 25, 20. That and your weight (including the bike), will be enough data to do some very precise figuring.
And for the most part, again ignoring very high and very low speeds, you’d be correct to take all other variables as constant provided we are ignoring rolling resitance.
Sorry Tristan, that was meant to go with the one above yours. Good post.
My math, might be wrong by the way. But, I’m pretty certain the correct math is possible from the measurements I specified.
Ok, so I’ve reckoned that if the deceleration at 10m/s is 0.5m/s^2, that would mean the air was decelerating you (specifically you if the combined weight of you and your bike were 85kg), would be 380 watts.
I got this by calculating the different kinetic energies at 8 and 10m/s, their difference, that difference over the time it took to change the speeds is the power, and the time is the rate of change divided by the change.
What one needs is a speedometer, a clock, and a recording device. It’s called a “Coastdown” test.
This link mentions this mode of testing:
“To date, a significant portion of the known experimental data comes from a technique referred to as “coast-down testing.”1 This method involves the derivation of bicycle (and rider) drag from the velocity achieved at the end of a ramp of known geometry. Typical drag levels at 30 mph are of the order of 6.5 to 8.0 lb for a rider on a racing bicycle in a racing posture. 2 This particular speed, bicycle, and posture conditions are chosen to provide a degree of consistency for presenting test results. A drag reduction on the order of 10 percent would save an estimated 2 minutes total time during a typical 60-minute race. As races are won by mere seconds, serious cyclists would gain an advantage with drag reductions on the order of 1 percent, or 0.08 lb. Proponents of the coast-down technique criticize the lack of adequate simulation during static testing of bicycles in wind tunnels. Coast-down testing inherently produces time-averaged results of the bicycle trajectory (the same type of results that would result from static testing within a wind tunnel). The final velocities derived from coast-down runs are affected by rider position, pedaling technique, and the mechanical condition of the bicycle, as well as environmental factors. Despite these influences, the uncertainty of the drag levels associated with these tests have been quoted to within approximately 0.02 lb.”
I think the math is a bit wonky, because as you slow down both the drag is going to be decreasing due to less wind resistance. You will slow down faster the faster you are going, and on something as small as a bicycle I think even a small change in wind resistance would skew the data enough to make it meaningless. Still, the idea is a good one. You could probably come up with an equation that would allow for an input starting and ending speed and time for deceleration, and spit out an approximate Cd, but that math is beyond my abilities, at least.
Calorie usage per hour is indicative of drag. Some estimates:
Bicycling, <10mph, leisure (16.1 km/h) – 345
Bicycling, 10-11.9mph, light effort (16.1 – 19.2 km/h) – 518
Bicycling, 12-13.9mph, moderate effort (19.2 – 22.4 km/h) – 690
Bicycling, 14-15.9mph, vigorous effort (22.4 – 25.6 km/h) – 863
Bicycling, 16-19mph, very fast, racing (25.6 – 30.6) – 1035
Bicycling, >20mph, racing (32.2km/h) – 1380
Chart
I find it dubious that 20mph is considered racing. I don’t have a speedometer on my bike right now, but in the past I’ve definitely sustained 35km/h for long periods of time. I’ve also seen cycling enthusiasts talk about things like their top speed and their pursuit speed, in the 35-40mph range. I don’t trust these numbers.
I will get a cycling computer and do my own measurements. The funnest way, I think, would be if you could find a very long consistently sloped hill – your coasting terminal velocity, and the slope of the hill, would be enough to calculate CD.
The winner’s average speed for the Tour de France has increased steadile from about 25km/h in 1903 to over 40km/h in 2008.
I wonder how much of that is improved bike technology, how much is better training, and how much is drugs.
* increased steadily
It’s mostly training, and drugs. The improvement in the last 30 years in bike technology looks fancy, but it’s really quite incremental.
It doesn’t seem overly unfair to call 20 mph (32.2 km/h) ‘racing,’ given that keeping it up for an absurd distance would have been adequate to win the Tour de France up until the end of the 1930s.
That said, a 100m cycle sprinter could be expected to go much faster.
As for the data above, it is the relationship between velocity and calorie consumption that is being highlighted, rather than the descriptive terms used for each velocity range.
Apparently, the world record for fastest speed achieved by a bicycle is 82.33 mph (132.5 km/h), set by Sam Whittingham in 2008.
How fast can you go on a bicycle?
It’s well known that a human being on a bicycle is the most efficient of all ways of translating energy into motion. Over short distances the cheetah is the fastest land animal, its top speed usually quoted at around 70 mph, whereas a human being on foot is much slower: Olympic sprinters peak at about 30 mph and they’re already slowing down when they hit the 100m mark.
By comparison, any fit cyclist on a good-quality road-bike should be able to get up to 30 mph (48.3 km/h), at least briefly, and top riders can sustain it for an hour or more under time-trial conditions. In short-distance events on the track riders will get close to 40 mph; in the recent World Track Championships, Chris Hoy’s winning time for the kilometre time trial was 1.00.999. The absolute world record, set at altitude in Bolivia, is 58.875 secs by Arnaud Tourant of France (Chris Hoy is currently planning an assault on this record). Remember, this is from a standing start.
I am up to 513km of cycling since I got the computer, about five weeks ago.
Today, I did my fastest long ride so far: 19.34km at an average speed of 25.17km/h – after cycling 31.33km earlier in the day.
Another enjoyable , and for me most enjoyable, cycle experience is bicycle touring . I especially enjoy a tour over lightly travelled roads in the mountains for a week or so in the company of good friends. Last year we road from North Vancouver to Banff in six days in the “Tour for No Good Purpose”. This year we are planning a seven day trip in British Columbia and Washington.
Together, my bike, panniers and I have a mass of just under 100kg. That means that at a velocity of 30km/h, I have 3472 Joules (J) worth of kinetic energy. By comparison, a .45 calibre handgun bullet (with a mass of 0.015kg and a muzzle velocity of 288 m/s) has a kinetic energy of 619 J.
At 40km/h, I have 6173 J of kinetic energy, and a momentum of 1111 kilogram meters per second.
I just went for a pretty great ride:
The main road through the Gatineau Park has some pretty good hills and drops. During the latter, I was averaging around 55km/h. At the bottom of the biggest drop, I hit 63.22km/h, with two non-aerodynamic panniers loaded with photo gear.
All told, it was 39.18km.
June 24, 2009, 12:26 pm
Can You Get Fit in Six Minutes a Week?
By Gretchen Reynolds
A few years ago, researchers at the National Institute of Health and Nutrition in Japan put rats through a series of swim tests with surprising results. They had one group of rodents paddle in a small pool for six hours, this long workout broken into two sessions of three hours each. A second group of rats were made to stroke furiously through short, intense bouts of swimming, while carrying ballast to increase their workload. After 20 seconds, the weighted rats were scooped out of the water and allowed to rest for 10 seconds, before being placed back in the pool for another 20 seconds of exertion. The scientists had the rats repeat these brief, strenuous swims 14 times, for a total of about four-and-a-half minutes of swimming. Afterward, the researchers tested each rat’s muscle fibers and found that, as expected, the rats that had gone for the six-hour swim showed preliminary molecular changes that would increase endurance. But the second rodent group, which exercised for less than five minutes also showed the same molecular changes.
According to a table in Mackay’s book, an approximate drag coefficient for a cyclist is 0.9.
That compares to 0.26 for a Prius, 0.36 for a Honda Civic, 0.425 for an intercity bus, 0.027 for a Cessna, 0.031 for a Boeing 747, and 0.8 for a typical car.
On the same page, it explains that if you assume:
then the energy used per 100 km by the cyclist is 3% of that used by a car doing 105 km/h on a highway. The efficiency compared to a faster-moving car would be even better.
First, the coefficient of drag is the ratio between how much energy it takes to move the car along, compared to the amount of energy it would take to move a flat piece of plywood with the same frontal area as the car along.
Second, an average car is nothing like 0.8 – that would mean, on average, a car was only a fifth more aerodynamic than a flat piece of plywood. It seems much more believable that a cyclist is only a tenth more aerodynamic than a flat piece of plywood.
An average car prior to 1985 might have been about 0.4, but now it’s much closer to 0.3. The Civic you cite as 0.36 is the 2001 Civic. I can’t find the rating for the 2010 conventional civic, but the civic hybrid for 2010 has a 0.28 rating. The Subaru Forester my parents had for a while had a truly awful CD at 0.42 – about the same as most cars from the 40s before anyone had invented the wind tunnel (or at least use it in automotive design).
I should probably look into making some aerodynamic modifications to the Van before I set off, as any improvements could make a large absolute difference over 5000 km.
An intercity bus has a drag coefficient lower than a typical car? The former is a big rectangular box on wheels. I dunno, I’m skeptical.
I remember commenting before that McKay’s book didn’t impress me that much (I’ve not nearly read all of it, though). I’ve somewhat changed my stance on this because I see his goal, he’s trying to illustrate for people with things they are familiar relative climate impact. My problem, however, comes from the fact I often find the the numbers hard to believe.
“The efficiency compared to a faster-moving car would be even better.”
That’s true – but 105km/h is already quite fast moving. The difference in economy in the Taurus between 75 and 105km/h is over 2L/100km – and that was a reasonably economical car to start with. I notice when I drive the Suburban (which might have a CD closer to 1 as it is basically a piece of plywood driving down the road) that I can hugely increase my economy if I drive at 80 or below compared to 100 and above. That kind of economy difference would be similar on an intercity bus, at least until they start making them more aerodynamic.
Reducing the national speed limit to 90km/h would significantly reduce the carbon emissions associated with travel.
This Wikipedia entry has some independent estimates of drag coefficients. It notes that the drag coefficient is “not an absolute constant for a given body shape. It varies with the speed of airflow (or more generally with Reynolds number). A smooth sphere, for example, has a Cd that varies from about 0.47 for laminar (slow) flow to 0.1 for turbulent (faster) flow.”
The numbers are comparable to MacKay’s (and he is not cited as the source of them):
0.25 for a Prius
0.9 for a cyclist
0.031 for a 747
The Wikipedia page doesn’t have an estimate for a ‘typical’ vehicle.
You can plug the Prius number into an equation to get a Prius-to-cyclist comparison:
[ Drag coefficient of bike * Area of bike * Velocity of bike squared ] / [ Drag coefficient of car * Area of car * Velocity of car squared ]
Assume, again, that the area of the bike is 1/4 of the area of the car and that the bike is going 1/5 as fast.
The ratio of the drag coefficients is about 0.9/0.25 = 3.6
[energy-per-distance of bike / energy-per-distance of car] = [ Ratio of bike to car drag coefficients * Ratio of bike to car areas] * ( Ratio of bike to car velocities ) ^ 2
= (3.6 * 1/4) * (1/5) ^ 2
= 0.036 = 3.6/100
Therefore, the cyclist uses about 3.6% of the energy of the Prius, to travel the same distance in five times as much time.
“My problem, however, comes from the fact I often find the the numbers hard to believe.”
The problem is not that his numbers are hard to believe, but that they are sometimes brutally false. He is off on the average car CD by more than an order of magnitude. The responsibility here needs to not only be on him but also his publisher – who should be considered less reputable for allowing such a garbage number to get through.
An intercity bus has a drag coefficient lower than a typical car? The former is a big rectangular box on wheels. I dunno, I’m skeptical.
If you look at the table, you will see that the error is mine. The units change partway down.
0.8 is the area of a typical car, in square metres. The listed drag coefficients for cars range between 0.25 (Honda Insight) and 0.51 (Citroën 2CV). The figure given for a long-distance coach is 0.425.
Incidentally, the Citroën 2CV seems like a case study in how to make a non-aerodynamic vehicle.
You can basically learn what the typical CD of a car is by looking at the 1986 Ford Taurus. When it came out, it had a CD of 0.34, which was excellent. Now, it has a CD of 0.34, which is pretty average-to-bad.
Although, a lot can also be seen by the 2010 Camaro, which has a “Slick” CD of 0.37. Compare that to the average CD of the previous Camaro of 0.34 – which was considerably worse than the 3rd Generation Camaro at under 0.3 (that 3rd Gen Camaro was incidentally the first car designed for slick aerodynamics by GM in a wind tunnel with computers).
What this shows is that there is progress, but styling sometimes takes precedence over progress. When we saw the return of grills on cars in the early 2000s – those had been eliminated for a reason: they are not aerodynamic.
Also, engine size is not unrelated to aerodynamics. A big engine requires a big radiator which must be supplied with air hitting it directly (not flowing around it). Sometimes the bigger engine version of the same car has a CD of .2 or .3 worse than the small engined version because of the design of the front with respect to getting airflow into the engine. One of the reasons hybrids have such good CDs is precisely that they have small engines and small radiators (and also, of course, they are optimized in other ways – pankake wheels and flat underbodies).
If one of you has academic access to this journal, this paper might include a good drag coefficient number for a bus: Numerical Study of the Flow Around a Bus-Shaped Body.
Motor Trend confirms the Prius number:
“The Prius boasts a drag coefficient of just 0.26, making it one of the most aerodynamic production cars ever to hit the road.”
He is off on the average car CD by more than an order of magnitude.
Firstly, there was the error I made.
Secondly, even an estimate of 0.8 isn’t off by an order of magnitude. The average of the ten cars he lists is 0.331. If I had estimated a typical drag coefficient of 3.3 or 0.03, I would have been off by an order of magnitude. An estimate of 0.8 is off by much less.
Since CD compares a car’s aerodynamics with a sheet of plywood with the same frontal area – cars with smaller frontal areas will take less energy to move through the air and this is not reflected in their CD. So, we should probably just stop using CD altogether. Why not use CDA instead? CDA simply multiplies the CD by the frontal area – giving in a sense a more honest value for the car’s aerodynamics.
http://en.wikipedia.org/wiki/Automobile_drag_coefficients#CdA
This also allows direct comparison to bicycles – “The drag area of a bicycle is also in the range of 6.5-7.5 ft²”. Compare that to something like a 2004 Prius at 6.24. But, at least a bicycle is more drag efficient than a Hummer! 26.5! Although, the Hummer seats five people, so in fairness it has a per person CDA of only about 5.5, still better than a bicycle…
Now all we need is a car that runs on Weetabix.
Reducing the national speed limit to 90km/h would significantly reduce the carbon emissions associated with travel.
Doing this would be enough to incite revolution.
If you look at the table, you will see that the error is mine.
Apologies to McKay.
Did the reduced speed limit incite revolution during the oil crisis?
Are the reasons for needing to reduce oil consumption now greater or lesser than they were during the oil crisis??
During the oil crisis, gasoline was both expensive and obviously in short supply (lines at gas stations, etc). There was also a convenient outside enemy to blame and try to foil, in the form of the OPEC countries.
The reasons for reduced speed limits now are arguably much stronger, but they are less acute. People don’t see scarcity or feel the effects.
None of those conditions are present now, though I would personally support reduced speed limits.
I also think a fuel price floor is a good idea. It could reduce volatility, while improving the financial security of any state that introduced one.
Did the reduced speed limit incite revolution during the oil crisis?
Are the reasons for needing to reduce oil consumption now greater or lesser than they were during the oil crisis??
Listen, I’m not disagreeing with your reasoning. I’m just saying that touching the speed limits would be political suicide. Besides climate change, lower speed limits would save a lot of lives every year, too. But people are willing to sacrifice a few fellow citizens, and increase their own risk to be able to drive fast. I think that says a lot about the speed issue.
I don’t think lower speed limits on highways would save many lives. Reducing limits in cities and towns from 50km/h to 35km/h would save hundreds of pedestrian lives per year, however. Also, it would make cycling in traffic much safer. And, it might save some fuel since people would not be breaking as much.
There must be statistics online somewhere for deaths from different kinds of highway collisions. In some cases, speed may not be a factor, but I think it would definitely be a factor in some.
For instance, you are probably a lot more likely to die if you fall asleep and drift off the road at 120 km/h rather than 90 km/h. The same goes for avoiding a moose that suddenly appears in front of you. If you drift over into the other lane of traffic, you are also likely to kill more people at 120 km/h than at 90 km/h.
This was released in February of 2007:
Canadian Motor Vehicle Traffic Collision Statistics: 2005
Here is the 2006 version.
In 2006, there were 1,625 fatal collisions on rural roads (either: primary or secondary highways, local roads, or roads with speed limits over 60 km/h).
There were 954 fatal collisions on urban roads (metropolitan roads and streets in urban areas, or streets with speed limits below 60km/h).
It doesn’t include figures on the total passenger kilometres traveled on either kind of road.
Reducing limits in cities and towns from 50km/h to 35km/h would save hundreds of pedestrian lives per year, however.
Between 2001 and 2005, there were an average of 358 passenger fatalities in Canada, representing 12.72% of total vehicle fatalities.
Cutting that by 200 would be a 56% reduction. Do you really think a 15 km/h reduction would have such an effect? Someone must have done an experiment of this kind somewhere in the world at some point.
And, it might save some fuel since people would not be breaking as much.
True. You can calculate what the average distance between stops must be before energy losses associated with air resistance exceed those associated with acceleration and braking:
Assuming cars have a frontal area of 3 m^2, a drag coefficient of 0.33, and a mass of 1000kg, it works out to about 750m. If you prefer different numbers, you can recalculate it yourself.
Consumer’s Union website has a report that analysed two speed limit increases in the US. The report shows that both speed limit increases led to statistically significant fatality increases.
It certainly doesn’t surprise me that it proved statistically significant, but I would be interested in seeing just how big an effect there is, as well as what sorts of accidents are most affected by speed, in terms of how often they produce fatal outcomes.
This page seems to refer to the study you linked:
“A recent study examined the impact of higher travel speeds on US rural interstates after the repeal in November 1995 of the national speed limit. Researchers found states that had increased their speed limits to 75 mph (120 km/h) experienced a shocking 38 per cent increase in deaths per million vehicle miles than expected, compared to deaths in those states that did not change their speed limits. States that increased speed limits to 70 mph (112 km/h) showed a 35 per cent increase in fatalities.”
That seems like a bit of a misleading way of describing the study, though.
Based on the rural and urban tables, it seems the speed increases increased the number of fatal accidents (not deaths) on rural roads by 15-16%. On urban roads, it was about 37%.
I doubt people increased their speeds to 75 mph (120 km/h) on the urban roads where most of the extra deaths arose. I wonder how much the limits rose by in urban areas.
One issue with the combination of drag coefficient and area (as mentioned by Tristan, but originally eaten by the spam filters – one risk of using fake email addresses) is that it is far more dynamic for a bicycle than for a car. Things like posture and cadence have an effect, as mentioned earlier.
A car, by contrast, presents pretty much the same profile to the air at all times.
Pedestrians – a child hit by a car at 30km/h is something like 15% likely to be killed, compared to 80% likely at 50km/h.
Do you have some actual sourced stats? I don’t doubt that this could be true, but it would be nice to have some data to examine.
The claims are corroborated by this site: http://www.whatstherush.ca/\
The site, however, provides no links to statistics.
In Peter Norton’s Fighting Traffic: The Dawn of the Motor Age in the American City it is argued that the large number of child fatalaties associated with cars was one of their biggest early barriers to adoption. In response, car companies invented the crime of jaywalking, established urban playgrounds as a way to help make roads exclusively for cars, and effectively shifted the blame for many child deaths to carelessness on the part of the child, rather than the fundamental dangerousness of cars.
For quite a while, cities were erecting large concrete monuments engraved with the names of children ‘murdered’ by motorists.
Before the advent of the automobile, users of city streets were diverse and included children at play and pedestrians at large. By 1930, most streets were primarily motor thoroughfares where children did not belong and where pedestrians were condemned as “jaywalkers.” In Fighting Traffic, Peter Norton argues that to accommodate automobiles, the American city required not only a physical change but also a social one: before the city could be reconstructed for the sake of motorists, its streets had to be socially reconstructed as places where motorists belonged. It was not an evolution, he writes, but a bloody and sometimes violent revolution.
Norton describes how street users struggled to define and redefine what streets were for. He examines developments in the crucial transitional years from the 1910s to the 1930s, uncovering a broad anti-automobile campaign that reviled motorists as “road hogs” or “speed demons” and cars as “juggernauts” or “death cars.” He considers the perspectives of all users–pedestrians, police (who had to become “traffic cops”), street railways, downtown businesses, traffic engineers (who often saw cars as the problem, not the solution), and automobile promoters. He finds that pedestrians and parents campaigned in moral terms, fighting for “justice.” Cities and downtown businesses tried to regulate traffic in the name of “efficiency.” Automotive interest groups, meanwhile, legitimized their claim to the streets by invoking “freedom”–a rhetorical stance of particular power in the United States.
That there was a conflict between motordom and traditional street users should come as no surprise. According to the Millennial Edition of the Historical Statistics of the United States, between 1909 and 1923, the number of automobiles registered in the U.S. grew by a factor of 43. The same source indicates that the number of traffic fatalities grew by a factor of 16 over the same period. Initially, when pedestrians were killed by motorists those deaths were cast as murder. The presumption was that the child or adult walking in the street had the right to be there; the motorcar was the trespasser. Moreover, the “overwhelming majority” (p. 29) of accident victims were children and a large proportion of the rest were young women. Cities throughout the country began erecting monuments to memorialize the deaths of innocent children. Norton argues that the deaths of so many women and children gave the traffic safety movement a “feminine” face. Evidence for this can be seen in the many posters showing mothers grieving over children lost to automobile accidents.
Apparently, the Tesla Model S has a drag coefficient of around 0.27.
Halfway through this video there is some explanation of why. For instance, the flat bottom.
Speaking of bike-riding robots: Joules robot rides shotgun, helps pedal on two-person bicycle.
Race: Yogurt (bicycle) versus Gasoline (motorcycle) in NYC
http://www.youtube.com/watch?v=cwT2k9tE_SE&feature=player_embedded