Comet suspension

[hadronuk]

Has everyone seen this TT accident?

http://www.youtube.com/watch?v=LZ1srcQMa_0

This is one of the incidents that was analysed in the Cranfield and Imperial study.

It looks to me as if at one point he nearly regains control? It certainly looks as if fighting the bars may be making it worse.

The slowish incident I once had on an old (and well worn) Triumph T100 was over before I could react at all, which perhaps is why the bike recovered on its own?

Although the low speed is probably what saved me:-

http://libra.msra.cn/Publication/6239374/on-steering-wobble-oscillations-of-motorcycles

"Results show that steering wobble oscillations grow more vigorously as their amplitude increases beyond a few degrees of steering and that the stabilizing influence of the rider's tensing his/her muscles in response to a growing wobble problem is small. The work supports the idea that any machine which has a very lightly damped wobble mode at some operating condition may be made unstable by an unusual set of initial conditions and that the natural response of the rider to the problem will be largely ineffective. This idea is closely aligned with anecdotal accounts from general motorcycle usage."


Another study:

http://ieeexplore.ieee.org/xpl/logi...re.ieee.org/xpls/abs_all.jsp?arnumber=5717690

"Burst oscillations occurring at high speed and under firm acceleration are suppressed with a mechanical steering compensator. Burst instabilities in the subject racing motorcycle are the result of interactions between the wobble and weave modes under high-speed cornering and firm-acceleration conditions. Under accelerating conditions the wobble-mode frequency decreases, while the weave mode frequency increases so that destabilizing interactions occur. The design analysis is based on a time-separation principle, which assumes that bursting occurs on time scales over which speed variations can be neglected. Therefore, under braking and acceleration conditions linear time-invariant models corresponding to constant-speed operation can be utilized in the design process. The inertial influences of braking and acceleration are modelled using d'Alembert-type forces that are applied at the mass centres of each of the model's constituent bodies. The resulting steering compensator is a simple mechanical network that comprises a conventional steering damper in series with a linear spring. This network is a mechanical lag compensator."

Wonder what it means in english?
And where can I buy a mechanical lag compensator?

 

[hadronuk]

I think there are 3 arguments that it is mainly the spring preloading that makes Girdraulics vulnerable to tankslappers:-

1. David Dunfeys extensive experience that stiff springs with minimal preload are safer.
2. The problem appears in modern motorcycles mainly under acceleration. In Vincents, it can occur under braking, but as this can cause the forks to extend, it is creating the conditions present in the modern bike when accelerating.
3. Other bikes fitted with Girder forks are less prone, but (to the best of my knowledge) none have preloaded springs.

 

[greg brillus]

I think you have hit the nail on the head. The only downfall i see with using shorter springs to lower the links to a "Safer Possition" and also softening up the front end is that on a race bike you have effectively lost valuable ground clearance.....not quite such a big issue with a road bike though.....I have these springs in my Rapide and the front end feels much safer, the ride is softer but not too soft, and the brakes work very well, as the front (head stock) is not raising up as it may have been when the stock springs were in place.....I think a hydraulic steering damper would be the final icing on the cake in helping this issue as well....I would like to see and perhaps purchase one of the kit form dampers that Big Sid mentioned earlier.......Greg.

 

[bmetcalf]

If you want to use the Kawasaki damper, here are two sources:
http://www.partsnmore.com/parts/kawasaki/h1/?filters[category]=chassis&filters[page]=2
http://www.tga.co.uk/catalog/product_info.php?products_id=143

They are in the UK and US, though. Maybe EBay in Oz has some?

 

[hadronuk]

I think you have hit the nail on the head. The only downfall i see with using shorter springs to lower the links to a "Safer Possition" and also softening up the front end is that on a race bike you have effectively lost valuable ground clearance.....not quite such a big issue with a road bike though.....I have these springs in my Rapide and the front end feels much safer, the ride is softer but not too soft, and the brakes work very well, as the front (head stock) is not raising up as it may have been when the stock springs were in place.....I think a hydraulic steering damper would be the final icing on the cake in helping this issue as well....I would like to see and perhaps purchase one of the kit form dampers that Big Sid mentioned earlier.......Greg.

It's fairly easy to make your own steering damper set up. I used a small pillar drill, the recommended Kawasaki damper, some ally plate and a bit of angle iron. Make the brackets oversize, drill holes for the damper after a trial fit, then cut away the uneeded metal. (Original sketch by Big Ed -thanks)

(In the photos the steering is on full right lock, so the rod end is nearly touching the top of the damper body.)
Not sure I got the most optimum position, but it works. (The mucky state of the bike is down to winter road tests and lots of WD40)

Greg, I don't understand your comment "the ride is softer but not too soft"? I would have thought it was firmer?
David’s springs have a rate appreciably higher than the standard springs, and if you shorten a standard spring by lopping a bit off, it will also be a bit stiffer.

 

[Big Sid]

Sid . The plate shown is very much like Lords but the damper body lies pointing towards the left across and beneath , and bolts to a bracket attached to the head nuts on the left side . The adj knob is right there accessible to ones left hand . Once set there seems no need to readjust nor is a low speed weave much felt . Works great . Sid .

 

[Big Sid]

A note to Jim Burgess , please E mail me directly for book prices , details . Thanks , Sid .

 

[Howard]

I think there are 3 arguments that it is mainly the spring preloading that makes Girdraulics vulnerable to tankslappers:-

1. David Dunfeys extensive experience that stiff springs with minimal preload are safer.
2. The problem appears in modern motorcycles mainly under acceleration. In Vincents, it can occur under braking, but as this can cause the forks to extend, it is creating the conditions present in the modern bike when accelerating.
3. Other bikes fitted with Girder forks are less prone, but (to the best of my knowledge) none have preloaded springs.

I'm a bit confused (about normal for me) what you mean by "preloaded". I spent some time in the 70s sorting suspension on a couple of specials I built, unfortunately I've forgotten most of what I learned, but I'm sure the idea was to set the spring "preload" so that, with the rider on the bike, the suspension was about 25% compressed.
The theory being that bumps go down (potholes) as well as up, and the springs keep the tyre in contact as the road falls away. If this wasn't the case, the front end is likely to become light and light front end plus bumps seem to be the prerequisites for a tankslappper.
I'm happy with 1 and 2, it's point 3 that I don't understand, my knowledge of girders is about the same as my knowledge of the dark side of the moon, but surely the springs become preloaded when they take the weight of the bike.

H

 

[timetraveller]

I have just heard from the spring manufacturers that they have made the batch of davidd type springs [if only all engineering companies were so efficient] so I will be collecting them on my trip to the north for Xmas. This will save the considerable delivery costs. I have had 10 of each strength made so that is enough for five Comets, five twins and five twins with chairs. As the cost per pair is going to be £30 or less it is going to be cheap enough to try them out. Where possible I will arrange personal delivery to save postage cost. I should be able to start and deliver them just after New Year so by, say, February we can start and get some feedback as to how people find these on the road. Please note that more than half the first batch are spoken for so please be patient and let us see what people think of them.

 

[hadronuk]

I'm a bit confused (about normal for me) what you mean by "preloaded". I spent some time in the 70s sorting suspension on a couple of specials I built, unfortunately I've forgotten most of what I learned, but I'm sure the idea was to set the spring "preload" so that, with the rider on the bike, the suspension was about 25% compressed.
The theory being that bumps go down (potholes) as well as up, and the springs keep the tyre in contact as the road falls away. If this wasn't the case, the front end is likely to become light and light front end plus bumps seem to be the prerequisites for a tankslappper.
I'm happy with 1 and 2, it's point 3 that I don't understand, my knowledge of girders is about the same as my knowledge of the dark side of the moon, but surely the springs become preloaded when they take the weight of the bike.

H

Hi Howard. As I understand it, preload is the force the springs are exerting when the suspension is fully extended. As you say, the theory is that as the forks reach the limit of travel, the force between the tyre and road drops abruptly from whatever the preload is, to zero and this causes instability.

Say you have total suspension movement of 3 inches and want to achieve 33% of suspension travel ride height when the static load on the suspension spring is 100 lbs.

You could have a spring rated at 25 lbs/inch that was 3 inches longer than the available space. When fitted, it is exerting a force of 75 lbs with the suspension full extended. That’s the preload. Compress the suspension by 1 inch, and you have the 100 lbs to support the static weight and give the required ride height.

Alternatively, if you were to use a 100 lb/inch spring with a length that exactly matched the available space, this would have zero preload, but when compressed by 1 inch would produce the 100 lbs required for 33% ride height.

So the same ride height can be achieved by any combination of spring stiffness (rate) and preload.