Steering

Raptor steering setup

The steering is THE MOST important feature of a recumbent trike. 

On the original Raptor the steering looked like this. Actually the handling with this very simple design is pretty good. I made my first trike just like it. But the drawback is huge brake-steer. When you use the left brake the trike immediately goes left, and right with the right brake. It is difficult to apply the same force to both brakes. Especially in an emergency situation and fast downhill. 


You can use this Raptor design. If you link the front brakes you'll be fine. It is very easy to do.and the simplicity makes it great fun. On the flats the brake steer is not that big a problem once you know how to push against the bars to keep a straight line.

To get rid of brake steer we need to learn more about steering geometry. 

• caster/trail
• steering axis inclination/kingpin angle
• camber
• track
• wheelbase
• Ackermann compensation
• steering system
• seat and bottom bracket height
• riders position
• tires
• and probably a bunch of other things    

Please read more about the subject here. http://www.eland.org.uk/steer_intro.html

 Of the whole list I would say that caster/trail is the most significant factor how good a trike will roll on the straights. Like on a regular bike the steering axis has to fall between 4 and 10 cm in front of the contact point of tire/road. Basically this means the 'kingpins' not only have to be at an angle sideways, but also to the front.

The problem is that in order to achieve the correct steering geometry, the steering knuckles have to incorporate  the kingpin angle and the 90 degree angle of the wheel. Almost all designs need machines to construct steering knuckles with precisely cut shapes. Not so with the Tribolt design.

The Tribolt centerpoint steering  can be done without machines. It eliminates brake steer as much as possible. This is what I came up with. 

First lets discuss downside of the Tribolt steering design.

The rear wheel still leans a bit in corners. With outward pointing rod-end bearings the inner wheel has a downward movement while the outer wheel goes up. Together with the right amount of caster the rear wheel stays ( almost) vertical.
On the Tribolt with the forward pointing bearings the inner and outer wheel stay more on the same level when turning. This results in a rear wheel leaning somewhat in corners.
Notice it is not tilting. A tilting trike trike leans into to corner. The Tribolt leans with the corner.


Now this effect is not noticeable while riding. Many factory trikes in fact do the same. And it does not have to be bad thing. The downside is your body weight adds to the risk of rolling your trike. The plus is the rear wheel also steers with the corner to a certain amount with less scrub.

Here is what the leaning rear wheel looks like. Notice this was with the straight Raptor steering knuckles. The effect is far less in the Tribolt design!

. 

   

Centerpoint steering.

What we want to achieve is that the steering axis meets the contact patch of the tire on the road. In this picture you can see that it is slightly inside centerpoint. Due to the width of these particular hubs the wheels come further from the frame. The steering axis inclination could have been 2 degrees or so more but you want to stay between 12 and say 18 degrees. A 20 degree angle will result in a very sluggish steering trike.

With normal hubs and caliper brakes the wheel comes closer to the frame and a 15 degree angle is pretty much like it should be with 20" front wheels. Most of the brake-steer will be gone.

You can do a mock-up and see how much angle you need. My advice is to stick close to 15 degrees and take a bit of brake-steer for granted if it turns out to have some.

Greenspeed have their GT and X models with 16" wheels slightly outside centerpoint like 5mm.It results in better rolling on the straights. The steering self-centers a bit more then for example an ICE Sprint model. But the ICE geometry is about as good as it gets in the recumbent trike world.

Construction of the steering knuckles.    

The material I used was 70x70x7mm aluminum L-profile. A meter costs about 30 to 35 euro. The size is 2x150mm but we can use the L-profile for other parts like the rear dropouts. So 50cm or half a meter is the least what we should order. Or maybe a meter and we don not have to worry we are short of the stuff.

There is is no reason why we can not use steel L-profile. Of course it does not have to be 7mm thick.

The smallest L-profile that will work is 60x60mm or there about. Less and the steering angle will not reach the acquired 15 degrees. The solution would be to bolt on extra material which complicates things and make it heavier. 

First step:

Drill the bottom hole (1) close to the side but not too close for the nut to fit. If you did then you can cut off the L-profile next to it. But leave enough on for extra strength. The 7mm aluminum is strong but just to be sure. Notice in the picture the bottom hole can be more to the left. ( discard the third hole at the top please) 

Next 

Drill the top hole (2). You can do the maths with a front wheel mock-up to determine centerpoint steering. Not easy. We can shoot from the hip on this one. 15 Degrees will be OK in normal situations.

Important: Be sure the distance between top and bottom hole is at least 85mm for a 40mm cross member tube. And 90mm for a 50mm cross-T. There is a workaround if you do not have enough L-profile. But the bottom bearings need to be adjustable in height.

A too wide steering knuckle will not work very good. Chances are a 100mm wide L-profile causes problems and the knuckles hit the cross_T when cornering resulting in a huge turning radius. Try to limit the size of the knuckle to max 80x80mm.

Also, do not make the distance between the top and bottom hole too big. The higher the top bearing is, the less angle it will have obviously.

Once we drilled one knuckle, mirror it to the second piece of L-profile by clamping them together in the vice.

Next we can drill the hole for the wheel axle. I have drawn them slightly higher then the top bearing. This will make it easier to put through a too long axle bolt while constructing. But in theory the axle can be much lower and even below the bottom bearing (using the correct angle for centerpoint steering!). You will get a much higher trike of course. However there is a risk that the axle will not clear the cross-T. The best alternative height for the axle would be halfway the cross-T were there is the most clearance. So right between bottom and top bearing.

I think it is best to use the dimension in the picture to be sure you will not encounter unforeseen problems and keep experiments for later.

There should be enough material left under the wheel axle hole for the tie-rod arm. Unless you want a higher seat. In that case the tie-rod can go over the main tube.

Last we can cut off all excess on the front. Leave sufficient on round the holes please. Other then that you can save weight and reduce the turning radius by removing as much material as you want without weakening  the steering knuckles. My advice is to do the rough shape and finish the knuckles after the first test rides!! Than you will understand how the steering works and how much more you can cut off. Do not overdo it while building! Maybe try the steering when attached to the cross-T on the workbench before you cut off anything.

Please continue to the chapter Cross-T.             

Cross-T

We can now attach the rod-end bearings to the cross tube.The top bearing is directly on the tube or maybe use  washers for extra strength. The bottom bearing is adjustable in height. It needs to be or the bearings will lock up.
Temporarily you can use use normal nuts. But to get the maximum out of the tuning radius  self locking nuts should point with the narrow rounded side to the bearings.

 

To improve the turning radius even further the steering knuckle can be filed in. But normally it is already tight enough. 

Check to see of the steering knuckles are perfectly 90 degrees vertical. Now it's time to make any corrections. Don't waste time on cosmetics yet! 

Tie-rod arms.  

If we did our work right we should have material left under the axle holes for the tie-rod brackets. The female bearings need to be 5cm or so behind the cross member tube or they will hit the tube before we get to the maximum turning capability at the front of the knuckles.

The shape doesn't matter a whole lot. It depends on what material you have. 

What does matter is that you can experiment with the amount of Ackermann compensation. Here is an example of the brackets being adjustable with small bolts. 

And here the finished product.