Why did my bicycle speed increase on smooth pavement while coasting downhill?

[Tom Kunich]

Something occurred to me that doesn't seem to make any sense to the way I understand physics. Perhaps someone can explain it.

Riding a bicycle on a descending road on very rough pavement and coasting I came to a flat and they had repaved it and it was perfectly smooth. My understanding of physics tells me that if I am coasting that the bike would coast a lot further on the smooth pavement than on the rough. But it would be decreasing speed due to rolling resistance and aerodynamic drag.

But that isn't what happened - when I hit the smooth pavement the bike increased it's speed by 10%. I repeated this experiment many times during the summer and it continued to do this but as the summer wore on and the pavement got rougher this effect became less and less and then eventually disappeared.

The level road appeared to be level. My altimeter at least read 0% grade. This only means that it was less than 1%. But it did appear to be level without any visible grade. But wouldn't a 10% increase in speed need a pretty significant grade to achieve and not the "less than 1%" that my instrumentation reported?

In any case could someone offer a suggestion other than "you're crazy"?

 

[.Scott]

Was your speedometer affected by the vibration?

 

[Spinnor]

There is a bicycling website that let's you use it for free for a while. You can easily plot bicycle rides that you have gone on or plan to. Interesting to me was the apparent precision of your altitude verses your position on the map. The website is,

https://ridewithgps.com/

You might be able to use the website to determine if the smooth road goes up or down, which can be deceptive. Please let me know what you find out. My guess is that you are in fact going down hill which can be the only reasonable explanation why your speed increases while coasting.

 

[trurle]

Humans cannot detect visually grades below 2%, therefore your altimeter is more accurate tool.
From what i know about modern altimeter sensors, they have a quantization issues - not measuring accurately relief below 1m. You likely still have in "horizontal" part a slope with altitude difference under 1m and slope under 2%.
In worst case and 18km/h initial speed, this unnoticed slope may increase your speed up to 20%

 

[Tom Kunich]

Humans cannot detect visually grades below 2%, therefore your altimeter is more accurate tool.
From what i know about modern altimeter sensors, they have a quantization issues - not measuring accurately relief below 1m. You likely still have in "horizontal" part a slope with altitude difference under 1m and slope under 2%.
In worst case and 18km/h initial speed, this unnoticed slope may increase your speed up to 20%

Well, I most definitely do not agree with your assessment of being able to tell grades less than 2%. And this isn't just the gradient measurement. It shows perhaps a change in altitude of a meter. The quantization of an altimeter has to do with the change in altitude over several meters and the distance was about 100 meters so it would easily show anything of 1% or more.

To answer Scott - the speedo/altimeter is totally unaffected by vibration. It simply measures forward speed by counting the rotations of the wheels and making the calculations based on the circumference of the tire OD. The change in rotational speed by vibrations would have to overcome the inertia of the wheel/tire/tube. This is high enough that becoming completely airborne doesn't effect it enough to be detectable. That is - the distance going down the decline is the same as going up the opposing incline at a much lower speed.

I would write this off as being my imagination if I didn't do this so many times that convince myself that it was actually occurring.

 

[berkeman]

when I hit the smooth pavement the bike increased it's speed by 10%.

That's pretty obtuse. Your speed increased from 3mph to 3.3mph? or 40kph to 44kph?

I repeated this experiment many times during the summer and it continued to do this but as the summer wore on and the pavement got rougher this effect became less and less and then eventually disappeared.

Can you post pictures of how rough this pavement is now?

Also, are you on a road bike or MTB? What was your tire pressure for all of these rides? Do you check your tire pressure before each ride and top your tires off (I do)?

 

[nitsuj]

Well, I most definitely do not agree with your assessment of being able to tell grades less than 2%. And this isn't just the gradient measurement. It shows perhaps a change in altitude of a meter. The quantization of an altimeter has to do with the change in altitude over several meters and the distance was about 100 meters so it would easily show anything of 1% or more.

To answer Scott - the speedo/altimeter is totally unaffected by vibration. It simply measures forward speed by counting the rotations of the wheels and making the calculations based on the circumference of the tire OD. The change in rotational speed by vibrations would have to overcome the inertia of the wheel/tire/tube. This is high enough that becoming completely airborne doesn't effect it enough to be detectable. That is - the distance going down the decline is the same as going up the opposing incline at a much lower speed.

I would write this off as being my imagination if I didn't do this so many times that convince myself that it was actually occurring.

! For such claims I would call such a speed measurement imprecise, particularly when used to compare rough pavement vs smooth...

So far we know the speed measuring device's measurement of speed changes depending on pavement.

 

[.Scott]

To answer Scott - the speedo/altimeter is totally unaffected by vibration. It simply measures forward speed by counting the rotations of the wheels and making the calculations based on the circumference of the tire OD. The change in rotational speed by vibrations would have to overcome the inertia of the wheel/tire/tube. This is high enough that becoming completely airborne doesn't effect it enough to be detectable. That is - the distance going down the decline is the same as going up the opposing incline at a much lower speed.

The device you describe gets the distance measurement. For speed, you need something that takes time into consideration. If this is an electronic speedometer, then there is no chance for a 10% error - it will be crystal controlled and the time precision will be much better than that.
But if it it mechanical, then it may depend on friction, centrifugal force, or something like that. Those components can be very subject to vibration.

 

[Trawlerman]

To accelerate there would have to be downhill slope. Many times surrounding terrain can fool the eye. Do a search on "Mystery Hill". They exist all over the country. Some of the sites do an amazing job of optical illusion making the participant feel they are actually heading uphill.

 

[PeroK]

Something occurred to me that doesn't seem to make any sense to the way I understand physics. Perhaps someone can explain it.

Riding a bicycle on a descending road on very rough pavement and coasting I came to a flat and they had repaved it and it was perfectly smooth. My understanding of physics tells me that if I am coasting that the bike would coast a lot further on the smooth pavement than on the rough. But it would be decreasing speed due to rolling resistance and aerodynamic drag.

But that isn't what happened - when I hit the smooth pavement the bike increased it's speed by 10%. I repeated this experiment many times during the summer and it continued to do this but as the summer wore on and the pavement got rougher this effect became less and less and then eventually disappeared.

The level road appeared to be level. My altimeter at least read 0% grade. This only means that it was less than 1%. But it did appear to be level without any visible grade. But wouldn't a 10% increase in speed need a pretty significant grade to achieve and not the "less than 1%" that my instrumentation reported?

In any case could someone offer a suggestion other than "you're crazy"?

Assuming you are descending the rough track and forces are balanced, then when your front wheel reaches the smooth surface the overall resistance will decrease. As you and the rest of the bike are still on the downslope there ought to be an acceleration during this transition phase, until the rear wheel is on the flat.

If, further, the smooth surface begins slightly before the end of the downslope, then you may get an additional acceleration for a short time before the smooth surface is truly flat.

 

[Tom Kunich]

That's pretty obtuse. Your speed increased from 3mph to 3.3mph? or 40kph to 44kph?

Can you post pictures of how rough this pavement is now?

Also, are you on a road bike or MTB? What was your tire pressure for all of these rides? Do you check your tire pressure before each ride and top your tires off (I do)?

The downhill is about 4% and I enter it at about 15 mph and coast down and at the bottom where the smooth part was, I was doing about 20. Then my speed would increase to 22. This was particularly odd because at that speed there is enough aerodynamic drag that you sure wouldn't expect that to be. And that increase in speed occurred almost immediately.

I have been thinking that this might be the speedometer catching up there. But I haven't noticed that other places. But in general on other rides I didn't have that abrupt transition from very rough to very smooth so perhaps I'm simply not aware of the lag time.

Would you suppose that what I was seeing was a lag in the speedo response and then as the road got rougher it had a high enough rolling resistance that the speed reduced fast enough that by the time the speedo would have caught up that I happened to be going the speed I though I was going on the entrance to that section?

This seems more likely wouldn't you think?

 

[berkeman]

I don't think the speedo is "catching up", but perhaps the pickup is not adjusted well, and it is missing some of the impulses on the rough ground (which would cause it to indicate a slower speed on rough ground).

Check the alignment of the magnet on your spoke with the pickup on your front fork to be sure the alignment is good. Maybe also consider riding with a buddy through that section, to see if their bike computer/speedo reads the same things...?

 

[Tom Kunich]

To accelerate there would have to be downhill slope. Many times surrounding terrain can fool the eye. Do a search on "Mystery Hill". They exist all over the country. Some of the sites do an amazing job of optical illusion making the participant feel they are actually heading uphill.

Well, remember that I had instrumentation. While the grade is a somewhat coarse means of grade measurement the actual changes in altitude are pretty accurate. I have done rides over and over with over 3300 feet of climbing and the total climbing hasn't varied more than about 50 feet and that is easily excused for the variations in rollers from one side of the road to the other. So on a total climb of only 600 feet we can assume that the instrumentation is accurate.

 

[Tom Kunich]

Assuming you are descending the rough track and forces are balanced, then when your front wheel reaches the smooth surface the overall resistance will decrease. As you and the rest of the bike are still on the downslope there ought to be an acceleration during this transition phase, until the rear wheel is on the flat.

If, further, the smooth surface begins slightly before the end of the downslope, then you may get an additional acceleration for a short time before the smooth surface is truly flat.

The change from the descent to the flats isn't that abrupt.

 

[PeroK]

The change from the descent to the flats isn't that abrupt.

So, you're saying that you are going along a flat rough road, then the road gets smooth and you accelerate?

Or, are you saying that you are going down a rough downslope that gets smooth?

It must be one of the two, assuming that the transition from rough to smooth is abrupt.

 

[Tom Kunich]

I don't think the speedo is "catching up", but perhaps the pickup is not adjusted well, and it is missing some of the impulses on the rough ground (which would cause it to indicate a slower speed on rough ground).

Check the alignment of the magnet on your spoke with the pickup on your front fork to be sure the alignment is good. Maybe also consider riding with a buddy through that section, to see if their bike computer/speedo reads the same things...?

It Is a VDO so It is pretty insensitive to the magnetic field distance. Having actually written this out now and looked at it with the knowledge that conservation of momentum is a law that simply can't be broken that has to be the only answer. I also think that I've seen a delay off of stop signs and lights as well now that I think of it. Especially if there is a slight decline in the road so that I can accelerate pretty hard.

I thank everyone here for helping me to get a better idea of what occurred.

 

[Tom Kunich]

So, you're saying that you are going along a flat rough road, then the road gets smooth and you accelerate?

Or, are you saying that you are going down a rough downslope that gets smooth?

It must be one of the two, assuming that the transition from rough to smooth is abrupt.

The rough downhill ends and there was a couple of bike lengths before it got onto the smooth pavement. This is what was throwing me - there didn't seem to be any easy answers.

 

[berkeman]

It Is a VDO so It is pretty insensitive to the magnetic field distance

Sorry, what is a VDO?

 

[Tom Kunich]

Sorry, what is a VDO?

Pardon me, it is a German manufacturer that is pretty well known for accuracy. But it never occurred to me before this discussion that the accuracy is offset by the response time.

 

[trurle]

The rough downhill ends and there was a couple of bike lengths before it got onto the smooth pavement. This is what was throwing me - there didn't seem to be any easy answers.

This discussion seems to be nearly fruitless and may require additional data. Efficient way to check for light slopes is to pour a cup or better bucket of water to pavement. This method have a sensitivity down to 0.5%, and not as sensitive to roughness as carpenter`s "spirit level" device.