smart51
|
| posted on 4/6/13 at 09:37 PM |
|
|
Trike roll oversteer
I've recently built a trike with 2 front wheels and one rear. The rear being a whole scooter rear end. It weaved at speed and I guessed the
scooter trailing arm wasn't up to the job. So I made a new one, swapping from this...
 
Piaggio swing arm
to this...
 
My replacement swing arm
The result is more different than an improvement. It is less twitchy but more dramatic. It still wants to go one way or the other instead of
straight ahead, though less so. It's worse when turning.
When turning into a corner, the trike initially turns in from the front. As the trike rolls, the rear starts to pull out and the rate of this
increases with increasing lean. In long fast corners the rate of change becomes quite rapid and you have to steer out of the corner to hold the line.
Lift off and the trike darts the other way. I'm guessing it is roll oversteer. Is this how a trike normally handles?
The whole thing weighs under 240kg and has a Fiat Cinquecento front ARB which I thought would be more than enough. However, if I put my feet against
the rear wheel, hold on to the roll over hoop and lean out, the trike leans over by about 10° or 15°. I don't know what to do to fix it. Any
ideas?
|
|
|
|
|
ali f27
|
| posted on 4/6/13 at 09:45 PM |
|
|
sounds to me like a front end prob at a geuss to much roll on front need bigger wieght springs to take out roll and stop rock and roll steering sounds
like steering through wieght transfer but need to think on it bit more
|
|
|
v8kid
|
| posted on 5/6/13 at 02:21 AM |
|
|
My first reaction was its rear end steer. I'm not sure how the rear suspension works but assume that the big rubber doughnut will allow the rear
wheel to steer- a very small movement is enough can you test this with long levers?
Cheers!
You'd be surprised how quickly the sales people at B&Q try and assist you after ignoring you for the past 15 minutes when you try and start a
chainsaw
|
|
|
v8kid
|
| posted on 5/6/13 at 02:26 AM |
|
|
Where does the wheel go?
You'd be surprised how quickly the sales people at B&Q try and assist you after ignoring you for the past 15 minutes when you try and start a
chainsaw
|
|
|
Slimy38
|
| posted on 5/6/13 at 05:26 AM |
|
|
Take a look at MEV's tR1ke pictures here;
http://www.roadtrackrace.co.uk/rtr_tr1ke_gallery.html
There are a couple of it going round quite a tight corner, with the suspension looking like it's almost bottoming out on the outside wheel.
Without actually measuring, I'd say their car uses at least 10 degrees of roll, so I'm not sure dialling out the roll will make that much
difference.
I would be more inclined to see what's happening on the rear. This might be a silly question, but when you're at maximum roll, could you
be losing rear traction? I'm guessing a moped tyre isn't going to be that big, and while it's not exactly heavy it's still
more than the original moped was carrying.
|
|
|
smart51
|
| posted on 5/6/13 at 06:13 AM |
|
|
The rear suspension is a trailing arm. The small tube and the curved arm connect to the engine and wheel assembly through rubber bushes. The rod
ends connect to the chassis. The rubber donut allows the engine to move up and down by a centimetre either way. I'm guessing to allow for
torque reactions.
The weight o n the rear wheel is the same as it was on the scooter. The extra weight is over the two front wheels. 250cc scooters are heavier than
50cc mopeds. There's no sign of losing grip from either end, even if I chuck it into the corners, though I could be wrong. Frone tyres are
135s, rear is 140 but curved.
|
|
|
Slimy38
|
| posted on 5/6/13 at 07:12 AM |
|
|
quote:
Originally posted by smart51
rear is 140 but curved.
Slightly deviating from your original question, wouldn't a standard car profile be a better tyre? Ignore me if you're already planning on
it being a future update. 
The use of a scooter rear end makes a bit more sense to me now, I thought you had butchered a tiny little 30mph sewing machine!! Is there anything you
could adopt from the larger engined ones? I'd have thought Honda's 750cc beastie would need something substantial on the rear end.
Although a quick image search suggests they revert to a regular shaft drive for the bigger engines.
Just guessing here, is it possible it could be a problem with the basic front end setup feeling like a rear end problem? I'm the same as you, I
don't know how a trike should handle but I can imagine incorrect settings on the front could make things a bit twitchy. Are you happy with toe
settings, bump steer behaviour, etc?
[Edited on 5/6/13 by Slimy38]
|
|
|
smart51
|
| posted on 5/6/13 at 07:35 AM |
|
|
I'm planning to put a car tyre on the back when it needs changing. Perhaps it will improve things a bit but the severity of the problem makes
me think it's more than that.
I haven't ruled out the front end. There's no noticable bump steer (since I moved the steering rack). I tried toe out, parallel and a
little toe in before I made the new swing arm. It had a little effect but not much. I'm running 1° of negative camber at the front. I'm
wondering if +ve camber at the front would help neutralise what is in effect +ve camber at the rear. There is no camber gain at the rear with a
straight trailing arm.
|
|
|
v8kid
|
| posted on 5/6/13 at 08:02 AM |
|
|
Ok get it now the whole engine/gearbox/wheel assembly pivots on the swinging arm.
I associate weaving and instability with rear toe out or in your case rear wheel steer but I guess you are saying that cannot happen However your
description of what is happening as you turn in to corner sounds very much like what happened with my car.
Strange!
You'd be surprised how quickly the sales people at B&Q try and assist you after ignoring you for the past 15 minutes when you try and start a
chainsaw
|
|
|
MikeRJ
|
| posted on 5/6/13 at 08:15 AM |
|
|
quote:
Originally posted by smart51
The rear suspension is a trailing arm. The small tube and the curved arm connect to the engine and wheel assembly through rubber bushes.
This is the bit that caught my attention. In the original application the swing arm won't see any significant lateral forces, so is it possible
these bushes are simply not up to the job? If this set-up is like I imagine it, any deflection in these bushes will lead directly rear wheel steering
under lateral loads, effectively causing over-steer.
How easy would it be to replace these rubber bushes with e.g. solid nylon bushes to prove (or eliminate) this? It may be that for longevity you need
to consider needle rollers as pretty much all larger bikes use.
[Edited on 5/6/13 by MikeRJ]
|
|
|
smart51
|
| posted on 5/6/13 at 09:05 AM |
|
|
The bushes are in the engine casing. I imagine they could be pressed out if you had the right tools. Looking at the numbers, if each bush compressed
by 0.75mm (they're quite small) then the rear "toe angle" would change by 0.5°. Multiplying that up, with the steering straight
ahead and the bushes compressed by this amount, the car would turn on a 300m radius, which is about what I am seeing. (I turn in, the car leans into
the corner, I turn out to compensate, the car goes round a 300m radius corner with the front wheels about straight). For info, 50 MPH on a 300m
radius corner = 0.4g of cornering force.
|
|
|
froggy
|
| posted on 5/6/13 at 10:52 AM |
|
|
Based on how much movement I see when testing scooters your rear end must be flexing quite a bit under load .
[IMG]http://i144.photobucket.com/albums/r187/froggy_0[IMG]
|
|
|
blakep82
|
| posted on 5/6/13 at 11:07 AM |
|
|
quote:
Originally posted by smart51
The rear suspension is a trailing arm. The small tube and the curved arm connect to the engine and wheel assembly through rubber bushes. The rod
ends connect to the chassis. The rubber donut allows the engine to move up and down by a centimetre either way. I'm guessing to allow for
torque reactions.
The weight o n the rear wheel is the same as it was on the scooter. The extra weight is over the two front wheels. 250cc scooters are heavier than
50cc mopeds. There's no sign of losing grip from either end, even if I chuck it into the corners, though I could be wrong. Frone tyres are
135s, rear is 140 but curved.
Could be well off the mark here, but is the something in the back end that deals with cornering?
I know you've made your own rear arm, but presumably based on the original. Stick with me on this...
A car and I guess your trike, you steer right the body leans to the left slightly
A moped/scooter/bike, you turn right, you lean right
There something in the design of bike tyres or the arm that helps steer the bike the same way as it leans? Which in your case is the wrong way
Stiffen up the suspension would by my first try, and maybe some different tyres
________________________
IVA manual link http://www.businesslink.gov.uk/bdotg/action/detail?type=RESOURCES&itemId=1081997083
don't write OT on a new thread title, you're creating the topic, everything you write is very much ON topic!
|
|
|
ray.h.
|
| posted on 5/6/13 at 11:10 AM |
|
|
imho. in my experience with scooter type suspension as opposed to motorcycle suspension. scooter engine/transmissions are quite softly mounted using
rubbers rather than nylon or hard plastic or even needle rollers. in a scooter application this is perfectly adequate with most of the loads in the
vertical plain. when using the same set up in a trike most of the loads are lateral which the soft bushes and mounts aren't designed to take.
replacing them with something stiffer seems pretty easy. my only concern is the increased side loadings on the engine castings once you take out the
damping effect of the rubber bushes.
|
|
|
smart51
|
| posted on 15/6/13 at 03:39 PM |
|
|
Some improvement
I've fitted an A frame to the engine casing using the mounts for the rear stand. The ends of the A frame are mounted to the chassis using rod
ends. The stability is noticeably better but still not quite right at speed. Clearly there was, and perhaps still is, a degree of rear steering.
I'll have another look later to see if there's anything else I can do to stiffen it up.
|
|
|
Ivan
|
| posted on 15/6/13 at 04:45 PM |
|
|
Maybe I am miss-understanding your original post but it seems from your description that the problem is more when you are traveling straight than when
cornering, in which case I would guess that your problem is a serious amount of bump steer.
|
|
|
smart51
|
| posted on 15/6/13 at 05:02 PM |
|
|
The original problem was both straight and cornering. Straight seems to have been sorted now. There is no bump steer. I adjusted the rack position
until it went away while I was setting up self centering for MSVA. I can stand on the chassis and bounce up and down. The front wheels stay pointing
straight ahead. I think either there is still a bit of rear end movement or it is now actual roll oversteer.
|
|
|
MikeR
|
| posted on 15/6/13 at 05:09 PM |
|
|
read this and remembered something a friend told of about riding a bike on the motorway. you get a flat spot as the bike is vertical. when you do your
first corner it can be scary as you go onto a narrow ridge of rubber so don't have masses of grip.. could the lean of the corner force you onto
a narrow strip of rubber.
|
|
|
smart51
|
| posted on 16/6/13 at 01:48 PM |
|
|
I may have been a little pessimistic yesterday. I've been for a quick blat across mixed roads and it is quite a lot better with the A frame.
It feels stable in a straight line except at 70 MPH and even then it just feels like it's very windy. Bear in mind that the top speed is about
75. It handles lumpy urban streets with aplomb and is fun over speed humps because the rear wheel goes between them. No matter what the road defect,
you only get one bounce.
Give a quick but smallish turn of the steering wheel and it turns in quite nicely. But it as it rolls, which it does quite slowly, the radius of turn
tightens, and at an accelerating rate the more it rolls. It's not alarming but it's not "right".
I've checked the rear tyre and there's no sign of it flattening. In fact there are are a few scuffs off the centre which suggest a degree
of lean at the rear. I wonder if there could be some twist in the trailing arm as viewed from the rear?
Edited to correct the 'autocorrect' which can't be turned off 
[Edited on 16-6-2013 by smart51]
|
|
|
smart51
|
| posted on 17/6/13 at 07:57 AM |
|
|
Here's a photo of the A frame in situ. If the whole engine / rear wheel assembly is leaning over in a corner due to twising of the mount, what
can I do to the A frame to stiffen it in torsion?
 
Rear suspension A frame
|
|
|
MikeRJ
|
| posted on 3/7/13 at 10:49 PM |
|
|
It's not clear to me which are the original suspension bushes, are they the big black "void" bushes in the middle?
If not then please ignore all the following, but if it is then:
1) It's not surprising you were seeing a lot of rear wheel steering, those bushes must have made the handling of the scooter a bit woolly, let
along a trike.
2) Your A frame location is going to be introducing all sorts of nasty stresses as suspension compresses, since the A frame pivot points aren't
in-line with the bushes. In fact if it wasn't a void bush the suspension would now be locked solid, since the bush is now effectively serving
double duty as a suspension spring. If the void bush runs out of travel before the main spring, then something is going to get broken, probably the
gearbox casting.
3) The entire assembly will be trying to pivot around the A frame fixing on the gearbox under cornering loads, which will still be causing the void
bushes to deflect and induce rear wheel steering.
If this is the case I'd get rid of the A frame ASAP and get solid bushes/bearings made to replace the void bushes.
|
|
|
smart51
|
| posted on 4/7/13 at 06:17 AM |
|
|
Hi Mike,
The enormous bush you can see is there to allow the engine to move up and down a bit under shock loads. The trailing arm to engine bushes a really
quite small and are hidden from view in the photo. The A frame pivots and the trailing arm pivots are not quite in line but the difference is very
small. All the movement can be taken up by the mechanism that allows the engine to move.
I've just fitted a stiffer anti roll bar at the front. When I next go out, I'll see if that makes any difference.
|
|
|
Slimy38
|
| posted on 4/7/13 at 07:30 AM |
|
|
quote:
Originally posted by MikeR
read this and remembered something a friend told of about riding a bike on the motorway. you get a flat spot as the bike is vertical. when you do your
first corner it can be scary as you go onto a narrow ridge of rubber so don't have masses of grip.. could the lean of the corner force you onto
a narrow strip of rubber.
That happens when a bike is predominantly used for long distances on straight roads. You're actually 'squaring off' a round profile
wheel, and while I'm not sure about the narrow ridge, there is a definite 'step' as you tip the bike over.
However, a trike will never get up to the angles necessary to induce this kind of effect. In fact, the usual approach for trikes using bike parts is
to replace the normally round profile bike tyre with a regular 'flat' car tyre.
smart51, I noticed you've said there is some wear off centre, I wonder whether it might be worth investigating that a little? Draw some chalk
lines across the tyre, and see how much is getting removed. If the wear on the chalk marks are further up than the lean of the trike suggests, then
there is definitely something odd going on with the back end.
|
|
|
smart51
|
| posted on 4/7/13 at 07:38 AM |
|
|
Chalk marks are a great idea. I'll try to find some chalk. Thanks.
|
|
|
Mr Whippy
|
| posted on 4/7/13 at 12:17 PM |
|
|
Reading through everything it seems really the problem is having tyres for a leaning bike on a vehicle that doesn't lean to turn. When riding a
motor bike at speed you don't actually turn the handlebar to turn, you lean the bike over and the bar turns with the bike. Try it, without
leaning the bike the handle bar will not turn.
Now you have a trike that both can turn like a car and a bike so when you turn the wheels and the body rolls the two just feed back on each other,
it's no wonder its a handful. Put narrow car tyres on and it will be fine.
[Edited on 4/7/13 by Mr Whippy]
|
|
|
