Trike steering

Wally said:

Are any trikes that have progressive steering, where the wheels turn
proportionally more as the handlebar turns? I did some googling, but
couldn't really find any concrete reference, other than someone who built a
trike wondering about it.

Quoting from http://www.eland.org.uk/steer_intro.html, I take it this is
what you are referring to:

"Ideally, in the middle of the steering motion (when you're going in a
straight line) the steering should be relatively insensitive, for ease
of control at speed. So a degree of handlebar movement has only a small
effect on the steering.

But towards the extremes of the steering motion, which would only be
used for low-speed manoeuvring, handlebar movement may as well make a
big difference to steering direction.

In this way you can make the best use of available handlebar movement,
which will be limited by space available."

Further on in the same article, there are spreadsheets for ackerman
geometry and other stuff which would be useful in finding out more.

--
JimP
--
"We don't have a plan, so nothing can go wrong" - Spike Milligan

Jim Price said:

Quoting from http://www.eland.org.uk/steer_intro.html, I take it this
is what you are referring to:

<snip>

Yes, exactly so. I'm trying t find out if there are any built trikes using
such a system. I have an arrangement in my CAD model, but I don't think it
accounts for things like friction or increase of steering effort as the bars
turn, so some real-world feedback would be handy.

Quoted message said:

Further on in the same article, there are spreadsheets for ackerman
geometry and other stuff which would be useful in finding out more.

Yup, got that, but I think he says that they don't model the kingpins tilted
in two directions. Food for thought, though - it led me to tweak the model
and improve the ackerisation factor.

--
Wally
www.melodolic.com/IOTT
www.wally.myby.co.uk

Wally said:

Jim Price said:

Quoting from http://www.eland.org.uk/steer_intro.html, I take it this
is what you are referring to:

<snip>

Yes, exactly so. I'm trying t find out if there are any built trikes using
such a system. I have an arrangement in my CAD model, but I don't think it
accounts for things like friction or increase of steering effort as the bars
turn, so some real-world feedback would be handy.

Just a suggestion - you could put real world measurements into the
spreadsheet and draw a graph of several points of input angle (say 1 to
20 degrees) against the output angle. If that graph is linear, then
there is none of the effect you were describing. You could then modify
various parameters to see how they affect the graph. I'm off for some
turkey right now, but I am interested in the idea as I'm planning to put
USS on my recumbent, and even though its a two wheeler, I still want a
variable geometry to the steering as described previously. Anyway, that
means I'll probably get around to modifying the spreadsheet myself, and
if a sufficient number of tuits present themselves I'll post the results
here.

As for real world feedback, my plan (see sig) is to find out what the
theory says will give me a more or less exponential curve, and build an
adjustable version to try it out and find out what settings give the
best riding feel. Now I just have to wait for someone to come along and
tell me I'm re-inventing the wheel.

Quoted message said:

Quoted message said:

Further on in the same article, there are spreadsheets for ackerman
geometry and other stuff which would be useful in finding out more.

Yup, got that, but I think he says that they don't model the kingpins tilted
in two directions. Food for thought, though - it led me to tweak the model
and improve the ackerisation factor.

I'm glad it was of some use - are you designing your own trike?

--
JimP
--
"We don't have a plan, so nothing can go wrong" - Spike Milligan

Jim Price said:

Just a suggestion - you could put real world measurements into the
spreadsheet and draw a graph of several points of input angle (say 1
to 20 degrees) against the output angle. If that graph is linear, then
there is none of the effect you were describing. ...

I think the graph would be pretty linear - not perfect, but pretty good. At
full lock, the steering angle is about 5-10 degrees more than the handlebar
angle. I'll see what I can do to set up a series of incremental rotations
and do some measurements.

Quoted message said:

As for real world feedback, my plan (see sig) is to find out what the
theory says will give me a more or less exponential curve, ...

I'm not sure how acheivable that would be, at least with a simple mechanism.
I get the feeling it's one of those things where friction, angles and
interferences are such that it becomes impossible to turn the steering any
further. I have that concern with my present design, and that's linear (or
nearly so).

Quoted message said:

... and build
an adjustable version to try it out and find out what settings give
the best riding feel. Now I just have to wait for someone to come
along and tell me I'm re-inventing the wheel.

I'm finding that there is a limit to how much lock can be had, regardless of
whether there's a ratio to control - something would have to be moved, like
the point where the two track rods meet in the middle (move it much further
forward, perhaps).

Quoted message said:

I'm glad it was of some use - are you designing your own trike?

Yes, from a starting point of the dimensions given by one Rickey Horwitz
on...

http://www.ihpva.org/com/PracticalInnovations/index.html

I've used his kingpin angles, wheelbase and track, and some chassis
dimensions (same ground clearance and crossmember tilt). The steering
geometry has changed quite a bit, but I might try going back to his set up,
now that I've had some fiddling time with that. Using CAD makes it
relatively painless to experiment with steering geometry, because I can put
long extensions onto certain components, such that they extend to the
intersection point as the steering turns (eg, by extending the three wheel
axles and viewing the model in plan, I can see how well the ackerman thing
is working).

--
Wally
www.melodolic.com/IOTT
www.wally.myby.co.uk

Wally said:

Jim Price said:

Just a suggestion - you could put real world measurements into the
spreadsheet and draw a graph of several points of input angle (say 1
to 20 degrees) against the output angle. If that graph is linear, then
there is none of the effect you were describing. ...

I think the graph would be pretty linear - not perfect, but pretty good. At
full lock, the steering angle is about 5-10 degrees more than the handlebar
angle. I'll see what I can do to set up a series of incremental rotations
and do some measurements.

I'd expect the linearity to depend on the relative angles of the
steering arms. If they are parallel then the graph will be linear. If
not then they will depend on the angle between the link and the arm (eg
if the two are lined up at 0 or 180 degrees the steering will be
insensitive, at 90 degrees it will be maximally sensitive.

...d

Wally

Why not have joystick steering as on the Windcheetah. This works excellently
and is not constrained by the packaging of the handlebars.

best regards

Rod King

"Wally" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Are any trikes that have progressive steering, where the wheels turn
proportionally more as the handlebar turns? I did some googling, but
couldn't really find any concrete reference, other than someone who built

a

Quoted message said:

trike wondering about it.

--
Wally
www.melodolic.com/IOTT
www.wally.myby.co.uk