Rolling tests

[Mick F]

Thanks.
I do have to work harder on the Moulton than the Mercian. No doubt about it. My average speeds are down as per my stats.
Having ridden Moulton for some time and getting used to it, it's amazing how fast Mercian is.

The hard work with a Moulton, has been discussed on here before. They're heavier, poorer aerodynamically, and the suspension system absorbs the pedalling action. As for rolling along, that is what I'm curious about.

Yes, they lack acceleration, they are slower ........ despite what the Moulton blurb might say about small wheels vs big wheels.
They are heavier for climbing up hills too, and the suspension absorption makes it even worse.
Flying down the hills, they are slower due to the space frame causing drag.

Although you can't get something for nothing, I reckon (and I would like to find out one way or another) that when Moulton is freewheeling on level ground, it is just as good as Mercian. It feels the same. The decay feels the same.

If you take the argument to extreme, if one bike has stiff bearings and draggy tyres, it would slow down almost suddenly compared to a free-hubbed bike with hard easy rolling tyres. I reckon that Moulton and Mercian slow down at the same rate ............. it's just that Moulton needs more of a push to get it up to speed due to the poor acceleration.

Willing to be proved wrong here of course. I often am.
I'm not convinced that any given bike is symmetrical when it comes to acceleration and deceleration by gravity alone plus rolling resistance.
Also, could it be that when I lift my foot off the kerb, the suspension dips a bit before the whole machine starts to roll properly?
Could it be that the suspension takes away the initial "push" from the slope? It might only be a second, but Mercian flies off instantly, and Moulton sort of pauses?

That's what it feels like.

[Mick F]

Out on a short chilly ride this morning on Moulton, and called it a the station on my way back and did a couple of individually recorded runs.
Both seem very similar indeed, so I've included only one or the pair as a screenshot of the data and the track as analysed by Ascent. I do, of course, have the original GPX files.

Click the pic to get a bigger view.

The highlighted band is where the bottom of the slope is, and you can see the position on the satellite map.

It seems that the speed at that point is 6.9mph

Ascent can produce a graph of speed against distance, but not against activity time. That is the graph that needs to be drawn, but I don't know how to do it on a computer. Perhaps someone could use the data above, to produce a Decay Graph? It would need to be smoothed, because there are speed jitters there.

Any road up, I'll need to go back up there with Mercian and try a few different start positions, recording as I go, and find which position gives me 6.9mph at the bottom of the slope. It needs to wait until the weekend, as I'm busy for the rest of the day as well as tomorrow.

[MartinC]

Why don't you work out where you need to start the Mercian rolling so that you can match exactly the speed that the Moulton enters the 'flat' part then you can compare the distances they roll directly?

[rfryer]

Or do a series of runs on the Mercian each starting 1m further down the slope. When you analyse them at home, you'll probably find one of them to be close enough to the Moulton run.

[Mick F]

Yep.
That's my idea.

One issue, is that Garmin GPS devices show their data on-screen a couple of seconds late. Must be something to do with the processing time involved to calculate then display it.

A good experiment to show this anomaly, is to ride along at a gentle but constant speed - 10mph? - then stop abruptly whilst watching the speed readout. It deceases by increments down to zero, but speed is shown when you have already stopped. The zero speed is nearly two seconds late after you've stopped.

Therefore, I can't look at the screen to judge the speed, and have to do a series of checks to see which one is the correct.

I'm actually not at home now, but drove off to Daughter 1's place. On the way, I turned in at the station and stopped at the lamp post briefly, and then found out how a Fiat500 rolls with respect to a Moulton!

The car rolled off very slowly. Any bike would have left me for dead, but the car kept going - albeit slowly. It eventually rolled to a halt six kerb stones prior to the base position - three kerb stones short of Moulton. One kerb stone is 600mm maybe? I reckon it did very well indeed.

[ChrisF]

Surely you need a longer slope to compare the two bikes. Over 20 yards or so all the figures/data you collect will be too dependent on inaccuracies. With a longer slope you can more easily measure (capture) the speed at a certain point.
And living where you are you have a decent choice of slopes, don't you ?
Even with the data you have from above, you could plot time against speed for both bikes (using Excel or similar) and see the shapes of the resulting curves. But it would be better with a longer slope: you'd then see any difference between the starting acceleration (which, if your feeling is correct, will be the determining factor here) and the more steady acceleration once you 'get going'.
Are you 100% sure the flat section is flat? I would have thought that both bikes should go more than 20 or 30 yards when starting at 7mph.

[later]. Done a graph for the Moulton, have you got similar data for the Mercian?

[Brucey]

IIRC Mick's test is a runout test, i.e. the bike is allowed to roll out on a near-flat course until it stops. This kind of test can (if the air is still) give a distance result that consistently reflects the rolling resistance of the tyres and is less dependant on the aerodynamic properties of the bike + rider etc.

The problem with this test is that the starting conditions into the flat section ought to be equal. In most cases the test is run to compare various (basically similar) tyres of the same(ish) size. In such cases simply starting the run up at the same point on a downhill run is usually deemed to be good enough, but in this case Mick is concerned that this won't give identical conditions between different bikes so that the entry speed into the flat section isn't close to being the same.

I think this is a valid concern, although some would say that it may indicate that one machine is inherently slower than the other anyway.

My take on it is that there are known reasons why one machine ought to accelerate more quickly than the other, so that if the results are in accordance with that, this reason can't be dismissed. However if they aren't, it is a fair bet that something else is going on.

cheers

[Mick F]

Hi guys.
At home now, and a sunny day.

Later this afternoon, I'll go for a spin with Mercian with the Garmin Montana mounted, with the same settings as previously on Moulton. I'll do a series of runs ............. a couple from the correct spot, then from further down the slope - maybe try two positions and two runs from each. I'll use kerb stones as reference points.

The carpark at the station is a level as it can be. It's parallel with the station platform.
We're not short of slopes round here, but finding somewhere with a gentle slope followed by a level section that is both straight and smooth, is a rare thing indeed. Coupled with that fact that it's a quiet carpark not far from home and fairly sheltered from the weather and winds, it makes an ideal runout track.

Thanks Chris, for your graph. I'll post the results from today's tests later.
Brucey, wise words again.

Hopefully, after the testing the afternoon, we'll know more. (fingers crossed!)

[Mick F]

Sadly, yesterday's runs were a bit problematical.
Wind. Head wind too.
I had a go, but I could feel the differences in the rolling.

It's even windier today. Shipping Forecast says that there's a gale warning in the Plymouth area.

Unless I can get the runs in in very little breeze, it would make them invalid. By all accounts, the wind and weather should be ok by Tuesday.

[fausto99]

...Yes, they lack acceleration, they are slower .....

I have to disagree with this somewhat. Starting from standstill, I find (purely subjectively) the initial acceleration much higher on a Moulton than any of my big wheel bikes. I put this down to the lower moment of inertia of the smaller wheels. I found this on both my gaspipe framed 60s Moultons as well as a spaceframe AM7 I used to own. I'm not sure at what speed the feel of the smaller wheel ceases to be the most significant factor, but at start off from stationary, the effect is undeniable.

[Mick F]

Try a rollout test yourself. See at what speed you get to at a particular point and compare it to a "standard" bike.
The weather is forecast to be good tomorrow, so I'll be carrying on with my testing.

The gales have gone, but it's chucking it down and still a bit breezy.

[Mick F]

I picked a window in the weather as the wind has gone and left us with occasional showers, I rode up to the station on Mercian.

I think it's back to the drawing board with these tests as far as measuring speed with a Garmin GPX uploaded onto a computer is concerned. It needs something more accurate I reckon.

I did nine runs.
Three each from the normal start point, one kerbstone down, and two kerbstones down. If I look at the speeds, they are rather inconsistent. Yes, the higher speeds are from the higher points, but the traces and the scatter are too much and some of the speeds overlap. I must have been lucky the other day.

Any road up, I found that by dropping down one kerbstone - 3ft down the slope? - Mercian was coming to a stop where Moulton comes to a stop. Two kerbstones down was too much and Mercian didn't roll as far as Moulton.

I suppose it needs a note pad and pencil and using a stopwatch from the start point to a known point at the bottom of the slope as has been already been suggested by Tim B. I may be off on a nice ride tomorrow as the weather forecast is good, so I'll take my stopwatch and a pen and paper and call in at the station on my way home.

[Mr Evil]

...I think it's back to the drawing board with these tests as far as measuring speed with a Garmin GPX uploaded onto a computer is concerned. It needs something more accurate I reckon...

Try a 'phone accelerometer. They typically have good accuracy, and are fast enough to provide multiple readings per second. You'll have to do integration if you want to convert that to velocity though - how's your calculus?.

[Mick F]

........... how's your calculus?.

It was ok back in 1972 as an apprentice, but I have difficulty with adds, tecks and guzintuz these days.
I was up in the loft the other day, and I saw my old course books. Gordon only knows why I keep all this stuff.

My Garmin Montana has a stopwatch app. It'll be easy to tap the screen to start and stop.
http://garminmontanagpsr.wikispaces.com ... +Stopwatch

[nigelnightmare]

All other variables being = the greater the mass the slower the acceleration and deceleration.
This is shown by the fact that the Moulton is Slower at the bottom of the slope.

If they both had the same rolling resistance then the Moulton would roll further due to the extra mass( it doesn't).

Therefore the Moulton MUST have a greater rolling resistance if it stops quicker.

What else are you trying to discover with this experiment?