Separate names with a comma.
Discussion in 'Hacks' started by Get Back, May 25, 2009.
Ok, so my KZ has a wheelbase of 57" so my lead should be 8" or so.. Me thinks...
So, going back to the original post, it appears absolutely necessary to have an accurate measurement of the lineal offset to use this formula. What I don't see is a practical way to arrive at this on a real bike, assembled, with fender,wheel and tire in place. How the hell do I do this?
Dog gone it there is more to it that just doing some math. The rigs we run are not rigid all around and supensions vary a lot. A ridgd sidecar suspension with a heavy bike like a harley may be able to do fine with less lead than a stock sprung Oil Head GS with a havy sidecar on it.
Simply put .. Suspensions move and cause things to happen. Light bikes unload the rear end with aheavy sidecar in turns away from the sidecar easier than heavy bikes.
I would love for one of the enginner types to be able to come up with an easy way to establish roll centers on each corner of a sidecar rig and from that be able to calcuslate the roll couple under various circumstances. To treat a sidecar rig as it it had no suspenaion may get one intop the basic ball park but to not be able to think things out related to how a suspension works on a given rig is a good way to put it 'out of the ballpark'.
A few years ago this deal bothered me and I spoke to some folks who were susposed to be in the know as far as suspensions go. I knew that if we could raise the roll center we coudl have decent results and create better cornering on a flatter plane. Lowering the center of gravity can only be done to a point on many rigs and off road or dual sport rigs are kinda stuck with what they have. The suspension guys basically did not really desire to get into it as far as something as weird as a sidecar goes I suppose.
So we built a k bike rig with a dual torrsion bar rear suspension on it. The torsion bars were set very high in hopes of by doing that the roll center woudl be raised. There were two bars so it coudl have a more progessive spring rate. Anyhow once the rig was done and run hard we knew we were onto something. I actually spun the thing out twice in turns toward the sidecar with no body on the sidecar frame. But, the thing was disassembled and sold before we coudl get into it more. Maybe some of the ones here like RichardNL or Phyllis would have the time to experiment with such stuff. trust me it does work but time and money are always an issue when you are walking so far out of the box.
Food for thought anyhow
Sorry for rambling and I HOPE my typing wasn't too bad
Maybe I'm missing something here Claude:huh
I just read that last bit a couple of times to make sure I had read it correctly.
With race cars, the aim is to get the roll centres as low as possible, generally speaking. There is always room for adjustment to slightly change handling, i.e. to induce more oversteer/understeer for a particular track or situation. But when building the car, much thought is put into building low down weight, and when ballast is added to bring the car up to a competition weight limit, it's always placed low in the frame.
So why, when building an outfit, would you need to raise the rollcentre.
I'm currently in the early stages of building a dual purpose rig and have been planning to put a 38 litre auxilliary tank low down, inside the triangle formed by the three wheels.
Should I be re-thinking where I put the tank?
Input from Richard, Phyllis etc welcome too
I only wish J.R. was still here. Can you imgine him chiming in on this thread...LOL. He seemed opinionated to some folks but they just didn't under stand that most of his opinions were indeed facts from a lot of trial and error.
Many confuse the 'roll center' with 'center of gravity' (CoG). The center of gravity to put it as simple as possible rotates around the roll center .
For example When watching a nascar race you will may hear them say " they came in and raised the track bar'. In essence they raised to roll center.
With a low roll center and a high center of gravity the weight transfer in a turn is in a large arc. This translates into body roll. In most cases it is desrable to lower the center of gravity but there are definatly limits. If we are stuck on the possiblity to lowering the CoG then raising the roll center is a means to get rid of some body roll and allow the weight transfer to be in more of a lateral motion than 'up and over' so to speak.
When we determine the roll center locations at each suspension point then from there we can establish how they work together. This, in basic terms, is called 'roll couple'.
Clancy you wote:
>>I'm currently in the early stages of building a dual purpose rig and have been planning to put a 38 litre auxilliary tank low down, inside the triangle formed by the three wheels.<<
Yes, by all means do so. You will be lowering the center of gravity which is a good thing. This is different than the roll center.
Hope that made some sense.:eek1
I have to admit I DO confuse the two myself sometimes
But in the NASCAR situation, when the track bar is raised or lowered, it only moves the roll centre miniscule amounts in order to slightly change a handling situation. In the'build' of the car, the aim is always to get the roll centre (and CoG) low in the frame.
But I think I've got my head around what you're saying about roll couples.
Most will use the term 'ceneter of gravity' and 'center of mass' as being the same. Pretty much true.....but both are different than the 'roll center'.
If the roll center and the center of gravity were on the same plane laterally the weight transfer woudl be in a lateral motion(little body roll).
When the roll center is low (the location of the roll center is determined by suspension geomtry and can be many times below ground level) we get more body roll. The higher the cog is above the roll center the more body roll is induced in a turn.
Again, the center of mass or center of gravity if we want to use the terms together rotate around the roll center. They are not one and the same.
Keepin weigth inside the wheels of the vehicle is usually a good step toward be able to manuver well. weigth bias front to rear and side to side is important and can be a good or bad thing depending upon the situation. This is really getting into a whole different thing though.
I have been doing some research into the whole “roll centre” – “centre of gravity” thing, and have come up with the following thoughts, which I now share with you for discussion/character assassination:
All the information I can find about roll centres has to do with cars, which have suspensions completely different to the suspensions seen on almost all sidecars.
By different suspensions, I refer to the arc that the suspension allows the wheel to travel through.
On cars, the movement is up and inwards, necessary to ensure that the flat car tyre tread remains flat to the road.
Even very simple car suspension, the model T type solid front axle, does this.
Motorcycle suspension works on a different plane, as it needs to, being after all a single-track vehicle, combined with rounded profile motorcycle tyres it isn’t needed to move in any other plane.
That plane is purely vertical with no lateral movement relative to the chassis at all, or at least this is what the designer had in mind, the “hinge in the middle” Japanese chassis of the 70’s being an example of result not matching design intention.
I am forced to interpret this major difference to mean that the volumes of work done on car suspensions is of little use when applied to sidecars, made even less relevant if flat treaded sidecar tyres are used, as the contact patch area is influenced by the angle of the wheel to the road surface.
Some of the points made with regard to car suspensions hold true, such as the use of “roll bars” to ensure the correct tyre contact patch relative to the road surface, but is done for different reasons.
On a typical car system, the virtual centre of pivot of the essentially side acting suspension system ensures that the tyres remain correctly oriented to the road surface, and so the “roll bar” is just that, it controls body roll, which at small lateral force situations is a comfort driven system.
The typical motorcycle system, being longitudinally based, means that the body roll of a sidecar also causes the tyre contact patch to roll relative to the road surface, less of a problem with sidecars still fitted with motorcycle tyres.
But reducing the contact patch area, and hence its ability to resist the side forces being applied to them, on sidecars fitted with flat treaded tyres.
In this situation, the “roll bar” is essential from a road holding point of view, not just comfort.
If you take my point that there is no “roll centre” with sidecars, due to the suspension orientation, then the argument that it is the “roll centre” to “Centre of Gravity” distance that is driving the cornering dynamics of a sidecar is no longer valid.
It is driven more by the distance from the road surface contact with the tyres at the three extremes to the “centre of Gravity” that is important.
Roll bars are useful to maximise the tyre contact and so resistance to lateral forces.
Off road sidecars have a higher “centre of gravity” than road sidecars due to the greater under body clearances required and will be more susceptible to “tipping” than a low slung road outfit, and there is little that can be done to help the situation other than clever loading to move the C of G to its most advantageous position between the three tyre contact patches.
EDIT, just found this site here, http://www.cyclesidecar.com/Guides/Advanced-Cornering/index.html
Thank you for taking the time to write this wonderful response, I found it informative and helpful.
If this stuff interests you keep reseaching.
NOTE TWO THINGS:
1) Any vehicle that has more than a single track and a supension will have a roll center.
2) Sidecar outfits are no longer motorcycles.
Too many times there is the assumption that sidecar rigs have no suspension on them. Even some of the books written have been guilty of this
The following is a statement taken from a website devoted to race car suspensions. I hope it will shed light on what we have discussed here.
"It is clear that low roll centre give little geometric weight transfer and most of the weight transfer goes through the springs (elastic weight transfer), and is therefore delayed by the time it takes for the vehicle to take a set. Conversely, with high roll centre most of the weight transfer precedes the body roll, leaving a smaller amount of weight transfer to go through the springs".
What they are talking about is that with a high roll center the mass of the vehicle weight is transferred around the low roll center and doirectly onto the outside springs. With a higher roll center the weight transfer is in more of a lateral motion which induces less body roll.
Check out the site here:
As mentioned earlier regarding the track bar adjustments we hear of during a NASCAR race what they are doing is moving the roll center up or down at the rear to either unhook the rear or hook it up more.
When we were still racing the sprints and midgets we had a bar on the front axle that worked in similar fashion ( at the time most called it a swaybar even though that was totally wrong nomenclature). By raising or lowering this bar you moved the roll center which either put more down force on the right front or less.
In the rear there was also a bar called a 'jacobs ladder' which also was adjustabel up and down for the same purposes. Lower the jacobs ladder( some call it a 'w' link because that is basically what it is shaped like) and it would plant the right rear....because the roll center was lowered and the weigth woudl transfer down on the right rear suspension. It also induced more body roll. Raise the jacobs ladder and it woudl unhook the right rear as weight was then tranferred in more of a lateral motion.
This is basic suspension stuff and is really oversimplified here.
Look back at my earlier post about the torsion bar suspension on the rear of the k bike rig. The results were pretty amazing to me. No, it was not calculated on paper but the results were what we wanted to see.
We can quote all kinds of stuff off the net and get a mindset that we are right or wrong but the proof is in the doing.
In fact (don't run me out of town on this one) think of what would happen if we coudl get the roll center above the center of mass....
Oh, by the way one of , if not THE, biggest influences of CoG location on these sidecar rigs is the operator. He or she can only be lowered so far.
But how do you calculate the roll centre on a system that moves the way a motorcycle and sidecar swinging arm system moves, EVERY bit of information I can find assumes that the suspension pivots inwards, not the way a bike system works.
Saying it has one isn't showing how it is calculated.
If you can't calculate it, does it exist in any meaningful way at all.
It will be from the tyre contact patch, that is the only consistant bit of information I have found, but to where is the line extended.
Is it along the ground?
If so then it can be ignored as it has the same vertical distance to the C of G as the tyre contact patches.
They are still motorcycles as far as the planes the suspension works in are concerned, not talking about steering dynamics here, but about suspension interaction, and mostly back and sidecar suspension.
Trikes have a roll centre that can be calculated, because they have symetrical car type suspension, but we aren't talking trikes here.
I am not winding you up specifically Claude, this isn't in any way a personal attack, just trying to understand how it works, and I am more and more coming to the conclusion that it can't be calculated, doesn't exist in a useful form, and so treating it empirically is the only way to go.
This is how all the "Manuals" for sidecar set up go, Phil Irving said as much in the quotes I have seen so far.
So either through ignorance or knowledge the "old" systems are the ones to use.
No offense taken.
The delima you expresed is exactly why I was bummed out when talking to some suspension gurus left over from our racing days as mentioned in my earlier post. That frustration led to building teh twin torsion bar rig and the results were interesting.
Any suspended machine has a roll center and a center of gravity.If the supension moves when weigth is trnaferred in a corner there is a roll center. How to calculated it on a sidecar rig? I dunno. Read my first post carefully. We raised the supenaion figuring what the heck maybe that will raise th eroll center. It obviously did. I still do not knwo how to calcualte it and was hoping someone here woudl have an idea.
Hey we're all in this together right?
Keep at it Boys, we'll get there yet.
Andy is almost completely right! Claude too!
Trailing arm suspension systems do have a roll centre, and it is situated at ground level!
The centre of gravity of the average motorcyle-sidecar combination (rider included) is very high off the ground, and, as in the german textbook, it is biased toward the motorcycle, not centered within the "tipover lines"(the contact points of the tyres joined by three lines to form a very skewed triangle) Trikes and the new Can Am Spyder have a centrally located C.of G. within a symetrical triangle, and are equally stable turning left or right. Sidecars are, as we all know, prone to roll over, rather than slide, especially when turning towards the chair.This a consequence of a narrow track (relatively speaking) and a high C.of G. close to the tip over line under the bike.
Anti-roll bars (to apply their tecnical name....sway bars to the general public) do just that, they resist roll, or the tendency to load the suspension on the outside of a turn, and unload the suspension on the inside of a turn,or in other words to oppose the "weight transfer" to the outside of a turn. If over done this is like having no suspension, so any bump will cause the wheel in question to lose grip and slide, and if it grips again could cause a "highside".(Is that what you meant Claude?)
The Roll Axis of a car is a line joining a car's forward roll centre (usually above the ground with modern types)with the rear roll centre, and the inclination of this axis determines how a car will behave in a corner, understeer and oversteer are the terms usually applied.
The car blokes don't know how to deal with us because we have a two track device with only one wheel on the "other" track(the right, for europe & USA, left for OZ & UK)
We still roll, just like a car, but there is no "other" front wheel to support the front corner, so if we over do it turning away from the sidecar, we are in danger of draging the nose of the chair on the ground just like McCardigan's BMW above. [Which looks to have the sidecar axle lined up with the fwd edge of the rear wheel rim (about 9" perhaps?). A bit more lead wouldn't have hurt. IMHO.]
It's getting late, must go, more sermons later. phyllis
FOR PHYLLIS' NEXT 'SERMON' WHEN HE WAKES UP...LOL:
>>Trailing arm suspension systems do have a roll centre, and it is situated at ground level! <<
If this is true can the roll center be raised? If not, how does that explain what we did with the k bike. Not being difficult just curious what your input may be.
The pic below is really bad but you can see the torsion bars and how high they were mounted. As stated I had spun it out in a turn toward the sidecar two different times. Once with the swaybar hooked up and once without. Both times there was no body on the sidecar frame. We figured it was due to a higher roll center being acheived by the placement of the bars.
Oh, in the picture it had a 205 tire on the rear and a 195 on the front and the sidecar. Track width was close to 60 inches and lead was around 14". No lean out. The spin outs took place with 135s all around.
Prior to doing the high mounted torsion bars there was no way it coudl have been spun out the same way under the same conditions with no body on the sidecar frame.
Dang it are you all going to make me build another one of these things?
Can you draw us a diagram of this set up?
I don't have any real figures on it Vernon. The torsion bars were mounted quite high. The torsion arm rested on a roller above the rear end housing. The roller is barely visable in the picture.
That rig had quite a few other weird ideas built into it but that is another story.
I assume that the lower tube and short lightened arm in the photo is the front torsion bar termination.
And that there is a similar termination of the top torsion bar at the front.
Interesting set up, but I can't see how it can have any influence on roll centre, as it is the suspension pivot points that determine this, not the springing arrangement, and the back pivot points appear to be standard K bike, I assume the front is a leading link of some sort.
The tyre offset outwards would have much more of an affect on stability than the torsion bars.
I don't want figures, just a clearer idea of how it all goes together
I'm sorry to change the subject somewhat, but I want to throw this out there and see what you all think of this idea of mine. I've been thinking of this for years, since the days when we used to play with stock cars. We were always looking for an advantage that hadn't been outlawed yet, and I never had the chance to try this one out.
I've seen alot of you hack purists making a big thing over sway bars. I thought of a way to do the same thing with pneumatics and make it fully adjustable, even so far as to enable left or right turn preload. I think this would be easily applied to hack rigs, and may make roll control much more easily installed than with a steel bar.
Picture 2 hydraulic cylinders installed like shocks. The lines attached independantly from the bottom port on one cylinder to the top port on the other and vice-versa. Then charge these separate lines with air or high pressure gas. As long as the pressure remains the same in both lines, the rig would remain level. But if the rig tries to lean, it will load the top of the outside cylinder which will transfer pressure to the bottom of the inside cylinder....pulling that side down.
If large diameter lines were used so gas flow was not restricted no effect would be felt when both wheels encounter the same bumps, because as one side's top was compressed it would feed the other side's bottom as it retracted. This setup would only operate when there is a difference from one side to the other. Increasing the gas pressure would have the same effect as increasing the thickness/stiffness of a sway bar. And in the case of oval track racing, one side could be given higher pressure to preload for cornering.
I don't know if this has ever been tried, but I have no doubt that it would work. My knowledge of hydraulics tells me that the best choice of cylinders would be ones with the smallest possible rods in comparison to cylinder diameter, to keep the differential as small as possible. I don't have the $$ or the facilities to build a setup to test this, but I bet Richard or some of you other "well connected" FF's can find a willing and worthy test bed for this. I want a ride report, that might be a billion dollar patent I just gave away