Limitations of low CG design

David_botting's picture

Please find attached a written article covering some of the less favourable aspects of low CG design. The article is in part written as a reply to the theories put forward by Royce Creasey in 'Feet First - The Physics'.

Acceleration
The average sports-tourer has nowhere near the amount of power to make wheelies a serious disadvantage. It is only over a very narrow rev range, between 500 - 1000rpm, at peak power and in first gear that such bikes are able to lift the front wheel from the ground (not the same as continuing a wheelie).

This = just a split second. Provided that a modicum of skill and throttle restraint is applied in first gear, the front wheel can easily be kept on the floor without serious loss of acceleration, and who needs to accelerate THAT hard anyway? It is wildly unlikely that a ‘looping’ situation would occur without the direct intention of a wheelie.

In the article, little account is made of weight transfer rates and their effect on grip. The statement about 85% loading and 15% slip giving maximum grip is clearly an error – as already pointed out. Also, only a single situation is considered, where the available grip is very high – the maximum possible – giving 1.5G.

On the road, particularly in the UK, a traction coefficient in the range of 1.5μ is not always available – a wet road is more likely to provide just 0.5 μ, same for gravel, tar strips, drain covers etc. The stuff road riders deal with everyday.

If the transfer rates are worked out for two bikes with wheelbases of 1470, and CG heights of 400 (as proposed by Royce) and a more typical 650, with the CGs being central between the wheelbase, we will see the limitations of low CG.

If the 400 CG bike has a weight transfer rate (k) of 400/735 = 0.54, the downward load on the rear tyre (L) is as follows:

L = mg - (mg – mak)/2

For 1G acceleration (a), the load would be 0.77mg

This means the minimum traction coefficient must be 1/0.77 = 1.3μ
If the available traction is anything less than 1.3μ, the result will be wheelspin.

The 650 CG bike transfer rate is 0.88. For the same 1G acceleration, the rear tyre load will be 0.94mg, and so any surface above 1.06μ will give grip.

The high CG bike will start to wheelie when acceleration = 1.14G, requiring traction of 1.14μ (100% transfer)

This means that the high CG bike will hold the advantage on all surfaces below 1.14μ – which, in my opinion is the more prevalent, and critical level for road riding.

Cornering
Where a low CG will be a real disadvantage, even in high grip circumstances, is in cornering.

When a bike is leant over in a bend, the centrifugal forces are using up a portion of the tyre’s available grip. There is less grip available for acceleration, and this means that a lack of weight transfer will mean that significantly less acceleration can be applied in cornering before rear spin / slide will occur.

A high-side accident is highly complex, but the start of a high-side always happens as a result of a severe rear slide / yaw. The fact that the front has less transfer unloading with a low CG, and therefore is less likely to drift in combination with the rear, means that the easily sliding rear will create exaggerated yaw (for the same wheelbase).

The fact that rear slides are more easily provoked and yaw is more pronounced means a high-side may be more likely to happen with a low CG. Arnold Wagner seems to agree with this.

The final disadvantage of a low CG for cornering is the fact that the bike will have to lean further into each bend. This is due to the way a bike rolls on its tyres. On typical tyres, a bike with a 400mm CG will have to lean 3 – 4 degrees more than a 650mm CG bike. Since ground clearance is the limiting factor for cornering speed on road bikes (at least all the road bikes I’ve ever ridden) it’s not unfair to assume that ground clearance on a low CG bike will also be a critical issue.

Deceleration
The case of rear wheel lifting under braking is also not necessarily the greatest problem for traditional CG bikes. To lift the rear wheel of the average sports touring motorcycle under high speed controlled braking is actually very hard to do. My figures suggest that -1.23G is achievable for our high CG bike allowing for some suspension compression. It is my direct experience that, even on a good, dry surface, a front lock-up is the likely ultimate result of heavy controlled braking at speed, on a sports-touring type machine. If you have ever lifted a rear wheel under braking, you will also know that there is no immediate loss of control, and it is usually when the wheel bumps back down that the lifting is noticed.
The braking performance of two tyres more evenly loaded, rather than 100% weight shift onto one is theoretically better due to a larger total contact patch. Exactly how much better has not been stated by promoters of this approach.
Also, in wet conditions, the higher contact load/mm2 (smaller contact patch) may be preferable. I’ve had difficulty finding precise data on these factors (and I’ve not seen it anywhere in defence of low CG) and suggest that it will be specific tyre sensitive. I hope to do some practical experiments soon.

The biggest problem with two-tyre braking is the fact that it will be very difficult, well-nigh impossible for the rider to precisely control both tyres to a level that he could control one. Holding the front tyre on the very point of locking requires extreme skill; doing the same thing, at the same time with the rear also is just not viable.
Don’t agree
This means that a linked system is often used. The balancing of this system is critical and requires different settings for different surface grip. Although riders are able to set the balance, the settings cannot be changed as fast as road surfaces change.

If the increased contact area gives a 5% advantage, but the balance is just 5% out, you have no advantage.

A dry surface providing 1.2μ for our low CG bike would require a balance of 82% / 18%, where a wet surface of 0.5μ would require 63% / 37%.
On a wet road with the dry setting you could only apply 76% of your potential braking power to the front, leaving only 13.7% at the rear (23% down on total braking power) – otherwise the front will lock.

The ultimate solution is ABS systems to control front and back. These are still not as efficient as a skilled rider, but getting pretty close. Don’t forget that it is more weight and cost though.

On the whole, balanced loading theory is good, but in practice, concentrating effort on the front tyre is hard to beat, all conditions considered. If you have a low CG without ABS, or without the balance being precisely set for the precise grip available, chances are you’d be better with a higher CG.

Practical use.
Toppling over at low speed or standstill is always a risk in the operation of motorcycles. This is an inescapable fact.

On a high CG bike the rider has the best possible stance to hold the bike secure. The seat is high, so his legs are straight, his arms are able to directly pull at the bars, and his bodyweight is separated from the bike.

On a low CG bike, typically the rider has to be seated. This is a very poor stance in holding the bike secure. The legs are bent at around 90 degrees, meaning that the thigh muscles take all the strain. Arms are unable to pull upward on the bars, and the bodyweight remains a part of the mass to be supported.

An easy experiment demonstrates this, please try it. Sit on a chair and put a bathroom scale under one foot. Press as hard on it as you can - 45kg is about the maximum you will achieve.

Sit on a motorcycle; place the scale under one foot. Lean the bike over as far as you can (without lifting your backside off the seat) – 100kg is easily achieved. With standing – 150kg or more is achieved (my scales wouldn’t go any further).

A rider on a high seat bike is easily able to provide 2 – 3 times the stabilising force that a low CG leg supported rider can. Even though the toppling force may be lower due to the reduced CG, it will have to be 2 – 3 times lower without any additional weight to equal a high seat bike. This is a considerable safety and practicality problem for low CG bikes. Solutions to this problem (stabilisers) are heavy and expensive.

Conclusions.

The initial impression of a low CG as put forward by Royce Creasey are very good, but further analysis shows that the advantages are not always real or achievable in practice.

Once realistic, varying and wet surfaces are considered, many of the advantages of low CG dry up.

If substantial added weight is a result of trying to get a lower CG (as it can be, and often is), it is likely that any of the proposed gains will be cancelled out and that the extra weight will have a very negative effect in other areas. A lower CG could be used to offset excess weight, like in the case of the Voyager, which has impressively low roll inertia for its large weight, but was trying to get a low CG largely responsible for the excess weight in the first place?

The above theories are limited to static analysis, and may be simplified to a point. They are intended to provide balance to the CG height debate, rather than a definitive piece of work.

CG height is just one aspect of PTW design, and I believe it is an area where on its own, little or no improvement will be made for practical, road going vehicles. Increased safety …….. Increased comfort …… Improved economy….. Improved practicality…… This is where the user focussed advancement will be made, and CG height is not central to any of these. Improved safety is certainly no.1 on the Government’s agenda. Maybe we should find the compromises WE want to make before the Government decide for us……. but that’s another debate, and these are just my opinions.

Who am I?
I’m a motorcyclist, rarely driving a car – because two wheel travel is more fun. I’ve ridden all sorts of bikes in all sorts of places over the last 22 years. I’m a person who wants to see the open minded development of two wheel road transportation. There are obvious problems (much more real and solvable than CG height) – let’s see what we can fix.

I am an apprentice trained mechanical engineer with ten years’ experience and have recently achieved a first class honours degree studying Industrial Design (BSc). My final year project was a ‘feet first’ motorcycle which I continue to develop.

I welcome any informed comment to the theories shown here.

AttachmentSize
Low CG design limitations.doc39.5 KB

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Didn't Arthur Middleton write

Didn't Arthur Middleton write a point-for-point response to this essay, or a version of it, not long after it was submitted? It would be useful and interesting to see the points raised here discussed, dispassionately, if only to save provide a summary of the debate for the less technically-minded who may not wish to have to trawl back through all the relevant Yahoo group messages.

For an open-cockpit FF design, there are practical limits on how low the seat can go, determined by how much leverage the rider can apply through his legs when the FF has to be supported when stationary. This may well vary with the leg-length of riders, but it's worth pointing out that there are people about who have been riding FFs for many years, and are of a comparable height to that of the author of the above article (well over 6', apparently).
Although not really about centre of gravity, another advantage of lowering the rider's seat is improved aerodynamics