How did we delude ourselves about 23mm?

For discussions about bikes and equipment.
User avatar
Mick F
Spambuster
Posts: 42066
Joined: 7 Jan 2007, 11:24am
Location: Tamar Valley, Cornwall

Re: How did we delude ourselves about 23mm?

Postby Mick F » 10 May 2018, 2:14pm

You see, it's those words again.
Bumps
Road Buzz

Some of us don't have them.
Mick F. Cornwall

Samuel D
Posts: 2223
Joined: 8 Mar 2015, 11:05pm
Location: Paris

Re: How did we delude ourselves about 23mm?

Postby Samuel D » 10 May 2018, 2:48pm

pwa wrote:I'm a bit old school and I've not ridden a carbon frame. But I wonder if race frames are now stiffer than they used to be in the days of 753, and the suspension now has to be transferred more to the tyres. Just a thought. Any truth in this?

Until a few years ago: yes. However, very recently some carbon frames have been developed that have more give than typical steel frames, some approaching the spring rate of the tyres they’re shod with! This is explained with supporting data here.

As already said, tyre construction and materials matter more than section width for both comfort (low-amplitude vibration if not big hits) and speed. Unfortunately, fast and comfortable tyres tend to be expensive in any width.

A distinction should be made between strict rolling resistance (hysteretic losses in the tyre caused by sluggish rebound and incomplete energy recovery after deformation at the contact patch) and losses in the cyclist arising from the body being vibrated. Vibration is (1) certainly in extreme cases and perhaps always resisted by wasteful muscular force and (2) attenuated in the body until it dies out, heating the body. The energy for that heating comes from your leg muscles propelling the bicycle, so this slows you down directly, quite aside from the discomfort it produces.

Vibration is the main reason why sitting in a car for eight hours is more exhausting than watching TV on the sofa for the same time. When you’re next in a car, observe the constant vibration of your flesh. This happens on a bicycle too – even more so as you can see by glancing at your arms on even a good road – and tires even the fittest riders.

Drum tests are the best method we have of ranking tyres for rolling resistance but tell us nothing about suspension losses.

Here’s a few factors to consider when choosing your ideal section width:

  • IDEAL TYRE IS WIDER < > NARROWER
  • rough road surface < > smooth road surface
  • you prefer low pressures for comfort < > less interested in comfort
  • you crash into objects and risk pinch flats < > you ride lightly and observantly
  • fat cyclist < > muscular cyclist (I postulate that taut muscles are more elastic than fat and so suffer less hysteretic losses when vibrated)
  • heavy cyclist < > light cyclist (because (1) heavier cyclists spend proportionally more of their power on rolling resistance and less on aerodynamic drag and (2) large cyclists, having more mass, suffer proportionally greater absolute suspension losses)
  • slow tyres < > fast tyres (because fast tyres incur a smaller penalty of rolling resistance at lower pressures and/or narrower section widths)
  • low speeds < > high speeds (strong correlation due to drag increasing with the square of speed while rolling resistance increases about linearly)
  • aerodynamic rims < > box-section rims (because rims with aero features reduce the drag penalty of wider tyres)
  • small-diameter wheels < > large wheels (because small wheels have less aerodynamic drag, higher rolling resistance, and poorer averaging of road irregularities)
  • frame is rigid < > frame has suspension whether nominally or via elastic deformation of key parts
  • long rides < > short rides (because (1) longer rides tend to be slower rides with the attendant drag-versus-rolling-resistance implications and (2) vibration eventually exhausts us even if it has a negligible tiring effect on short rides).
Tom Anhalt has done better work than anyone to help us estimate our ideal width. Explaining all his assumptions would fill another chapter but the main points are that he is using a figure for rolling resistance multiplied by an empirically established correction factor to model typical suspension losses; that he is assuming a 38 kg load; that he is testing at a speed of 35 km/h; and that he is using state-of-the-art aerodynamic rims.

With these assumptions, the fastest of the 22, 24, 26, and 28 mm S-Works Turbo tyres is the 24 mm.

On the other hand, I suspect many of the 23 mm defenders in this thread are using comparatively slow tyres, riding at speeds much below 35 km/h, loading the rear tyre (and maybe the front!) beyond 38 kg, etc. In these cases there is little doubt that wider tyres at appropriate pressures would be faster as well as more comfortable.

I don’t share amediasatex’s enthusiasm for road testing, because noise real and imagined drowns out the tiny differences we’re seeking to detect. This is especially true for speed, which is famously hard to estimate when vibration varies. My experience tells me I’m as good as anyone at observing small changes (for example, I can just about feel the drag when I turn on my dynamo lights), but I don’t have any illusions about reliably detecting even significant changes in rolling resistance by going for a ride.

User avatar
Mick F
Spambuster
Posts: 42066
Joined: 7 Jan 2007, 11:24am
Location: Tamar Valley, Cornwall

Re: How did we delude ourselves about 23mm?

Postby Mick F » 10 May 2018, 3:05pm

Samuel D wrote:On the other hand, I suspect many of the 23 mm defenders in this thread are using comparatively slow tyres, riding at speeds much below 35 km/h, loading the rear tyre (and maybe the front!) beyond 38 kg, etc. In these cases there is little doubt that wider tyres at appropriate pressures would be faster as well as more comfortable.
I'm not a 23mm defender, but a 23mm on my bike defender.

Vittoria Rubino Pro tyres and latterly slicks.
Fast, supple, and comfortable ............... on my bike.
I weigh 12st and a bit. Let's say 79kg.
120psi in the rear, and 85psi front.

PS:
I regularly ride down hills here at 40mph+.
Superb. Faster the better.
Mick F. Cornwall

amediasatex
Posts: 482
Joined: 2 Nov 2015, 12:51pm
Location: Sunny Devon! just East of the Moor

Re: How did we delude ourselves about 23mm?

Postby amediasatex » 10 May 2018, 3:09pm

I don’t share amediasatex’s enthusiasm for road testing,


Not everyone does, but I love to try stuff out for myself and also enjoy the stats :-) but....

because noise real and imagined drowns out the tiny differences we’re seeking to detect. This is especially true for speed, which is famously hard to estimate when vibration varies. My experience tells me I’m as good as anyone at observing small changes (for example, I can just about feel the drag when I turn on my dynamo lights), but I don’t have any illusions about reliably detecting even significant changes in rolling resistance by going for a ride.


...I think you might have slightly got the wrong end of the stick about my road testing comments. I'm not suggesting your bore yourself silly with striclty controolled back to back same-day style testing in an attempt to identify tiny transient differences that will be lost in the noise. I'm on about trying things over extended periods, weeks, months, hundreds of miles and then evaluating that with both an objective 'how did that feel' way, but also paired with any data collected and averaged out over the time. I think it's the only way you get a proper feel for what works for you, whether you're basing the 'works' bit on actual measurables or just on how you feel, or a combination of both.

Looking at both the subjective 'feel' and data will allow you to decide where you want to draw your compromise line. Although it can be very hard to work out how much of what you feel is real Vs imagined, which is why I think a long term approach coupled with data analysed in retrospect (rather than at the same time) is a good way to get real answers.

Individual short-term tests normally require an extreme variation, like the deliberately under/over inflating by a large margin and seeing if (a) you can feel any difference and (b) if there is any actual measurable difference, and they're only useful for illustrating extremes, but I think they're still worthwhile from time to time.

PS:
I regularly ride down hills here at 40mph+.
Superb. Faster the better.


PPS. I hit the magic 100kph on one descent on an audax recently, the bike I was riding is the only one I'm happy at that speed on, I have another bike that rolls on exactly the same tyre combo (28s in case it matters), and it's a nervous mess above 80kph, I don't think a change of tyres would change that much, the bike and geo plays as big a part as tyres do at that point.
Last edited by amediasatex on 10 May 2018, 3:14pm, edited 1 time in total.

User avatar
Si
Moderator
Posts: 14784
Joined: 5 Jan 2007, 7:37pm

Re: How did we delude ourselves about 23mm?

Postby Si » 10 May 2018, 3:11pm

Mick F wrote:You see, it's those words again.
Bumps
Road Buzz

Some of us don't have them.


Are you sure....or could you be so used to it that you just don't notice it any more? OR those HP 23mm tyres have killed off your nerve endings :lol: I do find it hard to believe that someone could do a reasonably long ride on the average British roads and not feel any kind of vibration, shock, buzz or whatever you want to call it through the bars and saddle....I can discern the nature of the road surface even on a fat tyred full susser!

reohn2
Posts: 30529
Joined: 26 Jun 2009, 8:21pm

Re: How did we delude ourselves about 23mm?

Postby reohn2 » 10 May 2018, 3:16pm

The utility cyclist wrote:Aero, even chris boardman states that aero counts even at slower speeds going uphill, he stated this last at the commonwealth games, maybe you should have a word with him and the secret squirrel club and enlighten them as to where they are going wrong :roll:
As per my previous post,I've ridden plenty of wide tyres, they aren't faster IME

Which secret squirrel club's that then :?
And where did I claim wide tyres were faster other than on bad roads?

Re the testing, go do some reading, it won't take a few minutes to figure out what the testing was and it's constant for all tyres
.
I've done he reading and I've done the riding too,we come to different conclusions,it's why now I only own bike with clearances for big tyres.
Show me the data on Mike Burrows roll down tests, was it outside so wind vairables which even with a small changhe have a big difference in measuring rolling resistance, was there other variables like body position (that will be a yes btw), were the tyres pumped up to out the ordinary for the wider tyres? There is no consistancy with testing and I've not seen what tyres or pressures or the environment he tested them in.

I'll need some time to dig out the tests he did and whilst I agree there may be variations, IIRC this point was not to go off roll down tests hence my posting about it.
EDIT,I've had a dig on the web and can't find the Burrows article but I'm almost sure I read it in Velovision some years ago.
Last edited by reohn2 on 10 May 2018, 4:15pm, edited 3 times in total.
-----------------------------------------------------------
I cycle therefore I am.

amediasatex
Posts: 482
Joined: 2 Nov 2015, 12:51pm
Location: Sunny Devon! just East of the Moor

Re: How did we delude ourselves about 23mm?

Postby amediasatex » 10 May 2018, 3:16pm

Si wrote:
Mick F wrote:You see, it's those words again.
Bumps
Road Buzz

Some of us don't have them.


Are you sure....or could you be so used to it that you just don't notice it any more? I do find it hard to believe that someone could do a reasonably long ride on the average British roads and not feel any kind of vibration, shock, buzz or whatever through the bars and saddle.


I was debating whether to reply to that comment from Mick or not, as I live on the other side of the moor to him and I'm sure we both often ride some of the same roads, and I think Dartmoor is a terribly bumpy, buzzy place to ride (great fun though!), despite me normally running slightly bigger tyres and at lower pressure than Mick.

And just to throw a spanner into a potential explanation, namely the bike, well....one of my bikes is a 531C Mercian :lol:

I'm sure neither of us are wrong, we just have different experiences, so maybe it is just a perception and personal tolerance thing.
Last edited by amediasatex on 10 May 2018, 3:21pm, edited 2 times in total.

reohn2
Posts: 30529
Joined: 26 Jun 2009, 8:21pm

Re: How did we delude ourselves about 23mm?

Postby reohn2 » 10 May 2018, 3:19pm

pwa wrote:If my regular routes involved long sections of cobbled surface I would already be riding a MTB, without any prompting from pro race mechanics. But my regular road surfaces are just chip n seal and a few cracks. Potholes to be avoided, naturally. And I find 25mm broadly okay on that, and better than okay on good surfaces. 28mm I find a bit slower for very little extra cushioning, which would put me off that size on a long ride. My current 35mm Hypers are good for not having to worry about preserving the wheel rims, but they definitely take more pushing around my regular routes than my 25mm GP4 seasons. I'd not use the 35mm tyres for an audax.

My experience is different.
-----------------------------------------------------------
I cycle therefore I am.

Scunnered
Posts: 144
Joined: 11 Apr 2014, 11:23am

Re: How did we delude ourselves about 23mm?

Postby Scunnered » 10 May 2018, 3:26pm

Raleigh Steve wrote:Interesting article here on the topic of tyre testing.

https://janheine.wordpress.com/2018/05/ ... isnt-easy/

Addresses the issues with testing on a drum, and the difficulty substituting a rider with a solid weight that doesn’t generate friction when jiggled.

The BQ testing methods seem more convincing to me than roller tests. This is subject to the limitation that they were conducted at moderate speed and so the aerodynamic impact of wide tyres may be understated, and don’t address the behaviour of wide tyres when climbing out of the saddle etc.

Steve


Having actually measured rolling resistance myself, I wholeheartably agree - it is NOT an easy thing to do.

I would be very interested to learn how all posters who have expressed an opinion on which tyres they consider to be fastest have actually measured data to support their viewpoint.

Nb. the plural of opinion is not data

reohn2
Posts: 30529
Joined: 26 Jun 2009, 8:21pm

Re: How did we delude ourselves about 23mm?

Postby reohn2 » 10 May 2018, 3:28pm

Mick F wrote:
reohn2 wrote:
Mick F wrote:Trouble is, I don't think a bike that would take 32mm wouldn't be getting the best out of 23mm.

If by that you mean that it isn't possible to build such a bike,I think you're mistaken.
I can't see any reason why that can't be achieved with a leettle more clearance and say canti or V brakes if the frame is made from the right material.
Sorry, been outside doing some chain maintenance on Moulton and Mercian, plus measuring the chains against a brand new one. All ok.

I digress eh? :wink:

I'm sure a frame and fork set could be designed to take both 23mm and 32mm, but that frame wouldn't be ideal for either.
This is what I'm discussing. My Mercian was designed from the outset to take narrow 700c tyres and have close clearances (but still allow mudguards - just) and take shallow drop brakes too.

I had the choice of 700c or 27" back then. I went for 700c because I knew that 27" was on its way out. Had I gone for 27" clearances, no doubt by now, I'd have used 700c wheels - and therefore I could fit wider 700c tyres and maybe up to 32mm but braking would be compromised due to long drop brakes. The bike wouldn't look as good, and I'm fairly sure wouldn't ride as well. You could argue that the frame would have been even springier due to the longer stays and fork blades, but I'm not sure.

Your frame and fork made from the same steel with 10 or even 15mm more clearance(brakes and looks apart)wouldn't ride perceptively any differently than it does now would it?
-----------------------------------------------------------
I cycle therefore I am.

reohn2
Posts: 30529
Joined: 26 Jun 2009, 8:21pm

Re: How did we delude ourselves about 23mm?

Postby reohn2 » 10 May 2018, 3:39pm

Si wrote:Im going to come out with my usual responce...,.,,,,depends on what kind of 'fast' you want. Over a relitively short distance my hp 23mm tyres was fastest....i could putup with the bumps for a short while. Over a longer distance the fat tyres were faster as i didnt get worn down by road buzz. Also, up hill the 23mm tyres were quicker, down hill the fat tyres were quicker. Different tyres are suited to different uses.

I agree and as I ride bad roads and some gravel/dirt roads,wider tyres(Hypers)are the best overall IME.Descending is far faster,upto 10mph on technical rougher tarmac roads.
-----------------------------------------------------------
I cycle therefore I am.

amediasatex
Posts: 482
Joined: 2 Nov 2015, 12:51pm
Location: Sunny Devon! just East of the Moor

Re: How did we delude ourselves about 23mm?

Postby amediasatex » 10 May 2018, 3:59pm

Scunnered wrote:Having actually measured rolling resistance myself, I wholeheartably agree - it is NOT an easy thing to do.

I would be very interested to learn how all posters who have expressed an opinion on which tyres they consider to be fastest have actually measured data to support their viewpoint.

Nb. the plural of opinion is not data


Precisely because it is so difficult is why I gave up and approached it from the other side. I have tried, and admittedly failed, to come up with a decent reliable way of measuring RR that I can easily do and repeat reliably. Instead I've gone for the data-volume approach and looked at a combination of power output and speeds over a several thousand miles on various setups.

It doesn't give me any definitive numbers for comparing RR, but it does give me indicative overall differences, on average, about certain setups.

eg: if over X thousand miles of mixed riding I was generally 1-2kph faster on tyre X than on Y on the same types of ride, then it's a good indication. When or if that also coincides with lower required power outputs for the same or better speeds I can be pretty sure that tyre X is 'faster' for my use case.

Pairing that with subjective things like how tired I feel, how much I notice vibration and how comfortable I feel, helps me make decisions. It's not as easy as looking at a chart, reading a RR figure and proclaiming a tyre as 'fast'. But it has lead to some surprising results for me.

I prefer this averaging over real miles approach as it essentially nullifies any testing methodology issues, and with enough time and miles it smooths out the spikes and anomalous results from a 'good ride' or 'bad ride' and tells me what works well for me and what doesn't even if I can't quantify it with a RR figure.

The problem is it's both time consuming and entirely limited to my use, I don't think it can transfer to others, as our riding styles, speeds and terrain will be different, but as a personal data gathering exercise it works for me. This is also why I really think people should try stuff out and see what works for them, and why I have so far steered clear of mentioning specific sizes, brands or models of tyre in any of my comments as my choices are exactly that, mine, and based on my data and my riding.

What I would love to do, were it possible, is be able to stick my tyres in a magic machine, get a 'fastness' number out, and then compare that to how they performed in the real world, and see how the two match up. And if they don't to work out why.


* As a side note this approach also shows up interesting seasonal differences!
I ride pretty much the same year round, I don't do 'seasonal riding', and I don't really change my bikes much through the year either. But there is noticeable difference in average speeds and effort levels required through the year which aren't attributable to changes in fitness, bike/tyres etc. It really is surprising how much of an impact the weather, temperature, air density etc. have on performance and I've posited before that a lot of the 'my winter/mudguarded/bigger tyred bike is slower' assertions might have more to do with the environmental conditions those bikes are used in than anything else.
Last edited by amediasatex on 10 May 2018, 4:09pm, edited 1 time in total.

User avatar
Mick F
Spambuster
Posts: 42066
Joined: 7 Jan 2007, 11:24am
Location: Tamar Valley, Cornwall

Re: How did we delude ourselves about 23mm?

Postby Mick F » 10 May 2018, 4:08pm

Si wrote:
Mick F wrote:You see, it's those words again.
Bumps
Road Buzz

Some of us don't have them.


Are you sure....or could you be so used to it that you just don't notice it any more? OR those HP 23mm tyres have killed off your nerve endings :lol: I do find it hard to believe that someone could do a reasonably long ride on the average British roads and not feel any kind of vibration, shock, buzz or whatever you want to call it through the bars and saddle....I can discern the nature of the road surface even on a fat tyred full susser!
Well, there you go then. You're taking what I said and analysing it too much.

Even the Moulton can feel the road on the 'bars. I'm not immune to it and I doubt any bike would be. You'd need a hovercraft of a bike not to feel the road.

It's that I don't feel any discomfort with bumps and road buzz. Bumps, depending on what they are of course!
Perhaps its the way I ride?

I ride with the majority of my weight on the pedals. My bum sits lightly on the saddle and my hands lightly on the 'bars. If you lean on the 'bars and grip tightly, your arms will ache. Sit hard on the saddle, and your bum will ache.
Mick F. Cornwall

reohn2
Posts: 30529
Joined: 26 Jun 2009, 8:21pm

Re: How did we delude ourselves about 23mm?

Postby reohn2 » 10 May 2018, 4:22pm

Mick F wrote:.....It's that I don't feel any discomfort with bumps and road buzz. Bumps, depending on what they are of course!.....

Ah,so you do feel road buzz(HFV).
What I was saying was that it was greatly reduced if not eliminated entirely when riding big supple rubber at the correct PSI for load.
IME such buzz/vibration also gives the impression of speed and once eliminated the bike feels slow but,IME the numbers measured on a 70 mile ride on my bike computer don't lie
-----------------------------------------------------------
I cycle therefore I am.

Scunnered
Posts: 144
Joined: 11 Apr 2014, 11:23am

Re: How did we delude ourselves about 23mm?

Postby Scunnered » 10 May 2018, 4:31pm

amediasatex wrote:
Scunnered wrote:Having actually measured rolling resistance myself, I wholeheartably agree - it is NOT an easy thing to do.

I would be very interested to learn how all posters who have expressed an opinion on which tyres they consider to be fastest have actually measured data to support their viewpoint.

Nb. the plural of opinion is not data


Precisely because it is so difficult is why I gave up and approached it from the other side. I have tried, and admittedly failed, to come up with a decent reliable way of measuring RR that I can easily do and repeat reliably. Instead I've gone for the data-volume approach and looked at a combination of power output and speeds over a several thousand miles on various setups.


Fair enough, although too time consuming to compare many combinations of tyres and pressures.
Since you presumably have a power meter, have you tried Robert Chungs method?
http://anonymous.coward.free.fr/wattage/cda/indirect-cda.pdf