Trike Front End Design and Adjustment

How to build a Tremmel
http://129.79.22.9/terratrike/index.html
See adjusting track on the ICE web site.
Ackerman Steering Compensation by Ian Sims of Greenspeed. An article in the OzHPV newsletter HUFF.
Spreadsheet for Ackermann steering design and one specific to direct handlebars on the kingpins.
From: Giles Puckett giles@research.canon.com.au

I've never had toe-in do fully things to handling, but it does slow you down. I find it best to have between zero and 1mm on a trike with Greenspeed-type crossed-arms linkage; maybe a little more might be in order with a straight tie rod between the wheels.
A couple of tips I've learned by trial and terror: make sure the measurements are done with somebody sitting in the trike. That's especially important if front suspension is present! Also, have a second person try to pull the front edges of the wheels apart. This takes up any play in the steering linkages, and is what will happen on the road, unless the trike's kingpins are perfectly centre-point (which they shouldn't be, for just this reason)
If the wheels camber at all (which most do) then the vertical placement of the measuring device is important. I prefer not to use one, for this reason, and also because it is easy to bump things. Sighting to the ground in the plane of the wheel seems to work OK.
I use a bit of lino with a checkerboard pattern on it, which helps me see things clearly when looking down on the wheels from above. One wheel is aligned to the grid, and the second wheel sighted down to a straight edge on the ground. Then measure to the next grid line which is only a few mm away.

From: EGrov@aol.com
Sent: Saturday, July 17, 1999 5:55 PM
Subject: [trikes] Re: Three wheel independent braking

Last week the subject of independent braking, especially brake steering, was being discussed and I realize that I may have a piece of the puzzle. It seems that people were getting different results from ostensibly the same setups.
My experience with my Dragonflyer might apply: My very first right-out-of-the-box descent (25-30 mph estimate) produced a severe shimmy (and a change of underwear). A consultation with Earth Cycles pointed towards a possible toe-out problem. I corrected what appeared to be a 1.5 mm toe-out situation (!) by introducing a (approximately) 1.5 mm toe-in (one half turn on both track rods) and the shimmy stopped.
I began to learn to ride the trike. 300 miles later (that's almost 9 months of intermittent use for me) I realized that I had scrubbed a wide flat into my front Comp Pools - a lot of rubber had gone! The rear was still a pristine dome. I had also been puzzled by the mention of 'brake steering' on this site, since my tadpole trike would only brake in a straight line, with either front wheel brake squeezed alone or together. Alarmed by the loss of tread, I re-measured the toe-in and found it to be (approximately) 3.7 mm! This is almost twice the maximum recommended by Earth Cycles and much more than I thought I was riding with. I reduced the toe-in by about 1.5 mm (one half a revolution of the ball joint on one steering rod [track rod]) and found a radical difference in handling. I could now brake steer - violently, if desired, but could no longer brake in a straight line with only one lever pulled, no matter how gently squeezed. The tires no longer made what I realized must have been a scrubbing sound when I turned the trike and I seemed (?) to be a little faster (and just as smooth) on the major hill in Prospect Park. The handling was so different that, on the next ride, I believe it contributed to a flip in a tight situation with an unobservant car (both the Dragonflyer and I came thru in good shape - only road rash for us both).
So, my supposition is that (a) the amount of toe-in is SUPER critical to the steering performance of a tadpole trike; (b) With such wildly varying results from very small adjustments in toe-in, this could explain the wildly varying experiences of most of the responses I read - everything from rapidly wearing Comp Pools to "no problems" (no experience)? with brake steering.
Casually out-of-true (horizontally) wheels can introduce almost the (Dragonflyer, at least) maximum (2 mm) toe-in by themselves. My careful, repeated and averaged (+/- . 15 mm) caliper measurements (with a jury-rigged extension beam) no longer seem sufficient to the task. Are you all much better and more accurate mechanics than I? How many of you play with your toe-in? Might there be something to this?

From: "Ian Sims" ian@greenspeed.com.au
Date: Sat, 17 Jul 1999 22:19:35 +1000
Organization: GREENSPEED
I'm not sure what is going on here - it seems rather odd. I have not had these sort of problems with my trikes. The only time I have experienced any shimmy was when I'd forgotten to tighten a steering rod end, after the hasty replacement of a damaged rod, and the bolt fell out on a down hill run! Now we use self locking nuts.
I've had trikes with both toe in and toe out, and the only real indication of misalignment has been more tyre noise on cornering. The recommended amount of toe in for GREENSPEED trikes is 0 to 2mm. However this should be measured WITH THE RIDER ON THE TRIKE!
The best way of measuring it is to use a TRAMMEL. This has two pointers on a bar which sit against the back of the tyres at axle height. A mark is made on each tyre with a pen, the trike rolled forward 1/2 a wheel revolution, and the marks compared against the trammel pointers. This method eliminates any inaccuracies due to wheel run out etc..
If the adjustment will not hold firm over at least a few months of normal riding, without hitting anything, then I would be looking for the cause. However I do not have much experience with trikes other than our own, and I guess all makes are different in the way they behave, despite having similar appearances.
Fortunately my employees LOVE road testing each trike to the max, before it's packed and shipped :-)
From: Anthony Sowers
Subject: [trikes] Trike toe adjustment - Dynamic!
Awhile back, there was a good deal of discussion here about toe in adjustment. There was, I believe, general agreement that it was a critical adjustment and that the best method was the use of a trammel bar.
The toe of the front wheels on a trike IS critical. Too much toe creates drag. toe-out creats drag. Where is the proper adjustment?
I had personal experience in which the trike handled well, but was about 2 MPH slower on upgrades than my bike. Adjusting the trike toe to the recommended tiny toe-in helped a LOT, but left me wondering if there wasn't a better way? After all, the behavior of the trike on the road may not be what the manufacturer assumed when they specified a static toe setting. Further, MY trike may have idiosyncracies that the manufacturer doesn't know about (perhaps extra play in the king pin bearings or loose steering ball joints).
What is needed is a way to find the optimum toe based on the trikes behavior ON THE ROAD. There is no low cost way to measure the drag contributed by the relative angles of the front wheels. It turns out, however, that there IS a way to determine the optimum dynamic toe.
The process is very simple:
1. Loosen a front wheel spindle so that the wheel can move about 1/8" to the right or left. Retighten the spindle jam nut in this position
2. lubricate the spindle, if necessary, so that the wheel will shift back and forth quite readily.
3. Ride the trike and observe whether the wheel rides to the inside or the outside. If it rides to the inside, there is too much toe. If it rides to the outside, there is too little toe.
4. Readjust the steering rods to change the toe as indicated by the wheel position.
5. I find that riding on a crowned road is helpful. On the left side of the crown, the wheel should move to the right. On the right side of the crown, the wheel should ride to the left. On the flat top of the crown, the wheel should ride happily either right or left and not change.
6. Readjust the spindle to eliminate the play in the wheel. You are done!

At this point, it doesn't matter what the trammel bar or any other measurement says. You have found the best setting for YOUR trike in YOUR riding conditions. You might record the trammel bar reading and use it to return to the proper setting in the future, if you have confidence that your measurement technique is truly repeatable. As for me, I'll simply use the above technique, which works in a remarkably short period of time.
Qn: Why do bikes/trikes use cone and ball bearing construction of the fork/headset assembly.
A: From: Jeff Wills
Turn the question around: why do bikes use complex and hard-to-align cup & cone bearing systems where a straight tube and bushing would probably do the job? (He says rhetorically.)
I surmise that a bicycle uses the aforementioned bearings to minimize the bearing drag on the fork. Anyone who's ridden a bicycle with a slightly tight headset knows that it does not respond to balance inputs - it becomes nearly impossible to ride hands-off. I'd guess that cup & cone bearings are a necessity on a bicycle.
On a trike (or quad, for that matter) absolutely minimal resistance isn't necessary because the steering inputs are generally larger and you steer the trike along the straight line (body english is impossible). Bushings and kingpins also spread the load over a greater area, so they should be more durable than ball bearings. Heck, kingpins were used at the front of every Volkswagen until 1967- ask a VW mechanic how many of those wore out.
On the Aerocoupe trikes, we used 1/2" diameter kingpins, a 3" long stack of Oilite bushings, and a 1/4" steel spindle mount. All of that worked great.
From: Peter Ross
To brake a tadpole trike safely using just one brake on one front wheel the steering geometry MUST produce Centre Point Steering.
OK, I hear you say, surely all trikes have this?
Not so. Until competition from Greenspeed obliged us to change in about 1994, all Trice had the steering axis (when projected to the ground) about one inch (25mm) inboard of the centreline of the tyre contact patch. Both brakes were connected to the same lever, and so long as they were properly adjusted this gave no problem. However bump steer could occur if one wheel hit a bump when the other did not.
It is easy to forget that a newcomer to trike design might not know about Centre Point Steering, (where the projected steering axis meets the centreline of the steering patch at ground level).
Peter Ross -UK Designer of the 1986-1998 Trice
From: Rick Horwitz
Twitchy or sensitive steering can possibly be caused by a few issues:
Axle/Kingpin relationship- If you have a homemade trike- Insure that the front axles are aligned at the kingpins or slightly trailing. If they lead, this will cause instability and oversteering at any speed.
Steering gain- High steering gain makes steering very sensitive.
Caster angle- Steering appears too light indicates lack of Caster angle. 12 degrees of caster is an excellent starting point. Increasing the caster angle increases the force by which the front wheels want to return straight.
From Peter Ross
Whilst in theory an accurate steering geometry is important if the front tyres are not to be worn out, there is IMHO much too much importance attached to this matter.
All these calculations assume that the vehicle is cornering at virtually zero speed without any weight transfer between the two front wheels, and consequently with equal sideways tyre deflection.
As we all know, cornering can be up to the speed at which the tyres both lose their grip, and at these speeds there is considerable weight transfer to the outer tyre (this will also depend on the height of the centre of gravity). The inner tyre has less and less weight on it, and thus the tyre is deflecting far less than the outer tyre.
This means that the slip angle of the two tyres is quite different, and to compensate for this racing car designers deliberately REDUCE the angle of the inner tyre by fiddling with the geometry, so that it does not reach a point where it loses adhesion completely due to being turned too far.
This is sometimes referred to as 'anti-Ackermann'. In the extreme you would have the outer tyre being the only one doing any effective work to keep the vehicle turning, with the inner tyre 'stalled' and unable to grip at all.
I hope you are not going to ask me how to adjust the geometry on YOUR trike because I don't know. At my level it is a case of suck it and see, but true Ackermann is not always the answer if you want to race your trike.
From Bill van Kats - Victoria, BC Canada
I did some checking into my reference books, and dug up another neat little modification to Ackermann steering that might be of interest:
In some (repeat: some!) high speed applications, the axle stubs of the front wheels are actually are designed to converge on an imaginary line drawn from the center of gravity of the vehicle to the center of the turn, parallel to the effective rear axle, so that the vehicle becomes the weight at the end of a virtual pendulum. The idea behind this is for the vehicle to form the tangent to the turn at the center of gravity, so that if the front and/or rear ends slide(s), the vehicle will follow the turn at a slightly larger radius, giving the driver a better chance to maintain control. When the rear axle is the reference point, instead of the center of gravity, and the front end slides, the vehicle follows the tangent in a straight line, and the rear end sometimes spins away from the center of the turn. If the rear end breaks free, the vehicle spins in a loop towards the center of the turn, and sometimes winds up facing the direction it came from, once the dust settles (or snow, as in my case. I experimented at great length one winter on a half-cleared icy parking lot in my younger days).
An Experiment In Dynamics For Those Who Want It: A good way to visualize this is to take a cord or string, attach one end to an irregular object with most of its weight at one end, like a hammer or something (Need I mention to please be careful, and do this somewhere safe, and away from others, possibly outside?). Find the center of gravity of the object, attach the string at that point, and turn around slowly, keeping an eye on the object. It should maintain a certain equilibrium once your speed has stabilized. Attaching the string at any other point will have the effect that the heavier end will swing to the outside, and any variation of speed will cause the object to begin oscillating unpredictably.
By the way, the above only works if the center of gravity occurs inside the wheelbase. An overloaded vehicle, or one towing a trailer will not react in the same way.
From Hiroshi Sakurai
Subject: Re: Ackerman steering compensation
Ted Bennett or Lisa Kaskan wrote: Because the left and right wheels turn on different radii in a turn, the inner wheel, to reduce scrubbing, must be turned more sharply than the right. This known as Ackerman steering compensation, and an approximation is achieved by arranging the steering arms so that they point at the centerline of the rear axle.

It is widely believed so, but I calculated and found it to be wrong. The intersection should be between the front and rear axles. Then in page 294 of John C. Dixon, "Tires, Suspension and Handling," SAE, 1996, I found the following.
It is widely believed that aligning the steering arms so that their lines intersect at the rear axle will give the true Ackermann steering (the Jeantaud diagram). However, this is far from true; the actual Ackermann factor varies in a complex way with the arm angle, rack length, rack offset forward or rearward of the arm ends, whether rack is forward or rearward of the kingpins, and with the actual mean steer angle. Moving the rack forward or backward to change the tie-rod angles can be a useful way to adjust the Ackermann factor, the most important single variable being the angle between the tie-rod and the steering arm in plan view, the Ackermann factor being proportional to the deviation of this angle from 90°. With straight tie-rods, to obtain an Ackermann factor close to 1.0 may require the projected steering arm intersection point to be at about 60% of the distance to the rear axle.
This 60% matches my analysis. If you want, I can send you the derivation in a Word document and the calculation in a Mathcad file.
I have not considered the arms pointing forward yet.
From: Glenn
Manteca, California, USA
My Dad has a TerraTrike 2.0 also.
His front tires wore out the same as you described. The rear was hardly worn.
As per the Owner's manual, the toe-in was setup with a 1/4" font to back difference in rim spacing.
WE DID SOME HOME WORK:
The TerraTrike owners manual calls for 1-2 deg. of toe-in and states that this is about the same as a 1/4 in. difference when measuring between the front and back of the rims.
I can only make these numbers work out if the 1-2 deg. angle is the included angle between the wheels. Based on our findings this is in the category of "too much toe-in".
We tested a lot of different toe adjustments, both toe-in and toe-out. No real records were kept but we did start with the toe-in as recommended in the manual. (One point here. Be sure and make the front back measurements at the same height as the axles. The camber of the wheels will cause measurement errors if this is not done.)
Adjustments were made by turning the tie rods equal amounts and going for a ride. It helps to make a mark with a felt pen on each rod. Adjust each rod equally.
Starting from the recommended position we reduced toe-in in steps and rode the trike between each adjustment. The progression of adjustments went past toe-in to neutral and on to toe-out. We then worked back to toe-in and tried different amounts until the best setting was found. In this case best is what we found to be a satisfactory. YMMV.
FINDINGS:
Toe-out is bad for handling. It makes the trike twitchy. When you steer there will be a point where the trike will suddenly switch from steering in one direction to the other. The tires scrub, wear, and cost you power.
Neutral toe will feel a bit mushy. The trike will wander back and forth by small amounts. This takes a lot of concentration and constant steering inputs to keep everything moving in a straight line. This is caused by the accumulated slack and inherent springiness found in any steering system.
A small amount of toe-in yields good handling. You need just enough toe-in to counter what happens when toe is neutral.
Too much toe-in doesn't make handling any better, or really any worse. But the tires begin to scrub, tire wear accelerates, and its all paid for by having to pedal harder just to over come the scrubbing.
THE SOLUTION:
After some trial and error, where we sampled all of the above conditions, we finally settled for a toe-in setup where the difference in spacing between the front and back of the rims was 1/8 inch [3.17mm].
The benefits were immediate. Smooth handling, tire wear is imperceptible, and the effort to maintain speed is lower.
P.S. The Premo-Comets were replaced with 100psi Maxxis Hookworms. These are huge tires that at pressure roll easily on the road and deal nicely with the occasional trip into the sand. If you want to try a set of large tires in the 406 size I would advise anyone to give these serious considreation. If the colors put you off, I have seen somewhere that blackwalls are now available.
Oh! On the original question. The off center wear is due to the combination of camber and improper toe adjustment. Adjust the toe-in. Reverse the tires if you want, or not. Its your choice.
From: Rick Horwitz
To: trikes@ihpva.org
Date: Mon, 15 May 2000
Subject: [trikes] Re: Ackerman Vs. Anti-Ackerman
In the pursuit of ultimate trike handling at speed vs. ultimate efficiency- I find that a compromise is needed for Ackerman compensation. Many years ago, some of my customer were noticing that my Zephyrs became sensitive to steering at speeds greater then 35 MPH. Although the root cause was improper steering gain ( I had 3 positions for steering gain), I researched several documents and discussions pertaining to Anti-Ackerman. Most of this material manifested from automobile racing, but the logic appeared to make sense for HPVs as well. I rigged up a MkIII Zephyr with an Ackerman that intersected 8" aft of the rear axle. I rode this unit to the top of steep grade that included some aggressive corners. During my descent I was able to reach 48 MPH on the straight and was able to maintain ~40 MPH into and out of the corners. With a normal Ackerman, I was only able to handle the same corners at around ~30-35 MPH.
Some observations I made: With normal Ackerman I found that I was exiting out of the corners with greater acceleration, but stability was difficult to maintain. With Anti-Ackerman I had better control (almost an understeer) during high speed handling, but the added tire scrub slowed accelerations through the corners. I found that the steering was very forgiving regardless of speed. I might add that this geometry gave the steering a much heavier feel.
So, my conclusion: If you desire ultimate downhill racing and stability at speed, try the Anti-Ackerman.
If you desire efficiency and have steady hands on the bars, use the standard Ackerman.
BTW-I am using a T-bolt with a steep Anti-Ackerman (can't quantitative it). I am capable of reaching speeds in excess of 55 MPH without any loss of control. However, during any tight corner, I lose much of my speed (as predicted). I plan to build a new set of knuckles that will reduce the amount of Anti-Ackerman, but not eliminate it. As I have said before, almost everything is a compromise, even Ackerman steering compensation.
Date: Thu, 01 Jun 2000
From: Giles Puckett
Subject: Re: [trikes] toe-in on a Greenspeed?
List-Id: Recumbent Human-Powered Trikes <trikes.ihpva.org>
My method, for what it's worth:
I have a piece of lino on the garage floor with a large black and white checkerboard pattern on it. I sit in the trike (important, even if there is no suspension!) and put a bungy between the front wheel rims at the rear edges, to pull tight any play in the system. I line up one wheel on the pattern in the lino, and have my lovely assistant sight down the other one against the pattern. Adjust steering balljoints and repeat. With the play pulled out of the system I reckon toe-in should be zero for best tyre wear, but err on the side of toe-in (0.5-1mm max) rather than toe-out, as one way you just get tyre wear, but the other way you get tyre wear _and_ squirrelly handling :-)
If you can't find a bit of lino like that you can draw up lines on a piece of plywood or something like that. This would have the advantage that they can be drawn in the right place for the front wheel track.
Date: Mon, 05 Jun 2000
From: F.Lenk
Subject: [trikes] Thoughts on toe-in adjustments on delta configs.
List-Id: Recumbent Human-Powered Trikes <trikes.ihpva.org>
There has been a good deal of discussion on this list about toe-in adjustments on tadpole (2 front, 1 rear wheels) trikes. It can be difficult to get it dialed in. Misadjusted toe-in results in lower speeds and tire wear. Even if a brand new trike is delivered with toe-in correctly adjusted (for a 150# rider), it might not be adjusted for your weight. Further, as the components settle in, the toe-in can require readjustment, and as components wear, again readjustment. But at least, with the typical tadpole design, you can make these periodic adjustments. And if you slightly tweak the crossbeam by jumping a curb, off-roading, or running into something, you can possibly decrease tire wear by compensating for the damage with the toe-in adjustment.
A delta (2 rear, 1 front wheels) trike I've seen on the web has got me wondering about toe-in adjustment for the rear wheels. On all the deltas I've seen (and straight front axle, rear steer tadpoles), the toe-in is determined at axle weld-up time. The rear axle housing (paired wheels) on some of these trikes consists of 2 straight tubes with a U shape curve/brace in the middle which is where the rear gears/brake are located. This design seems pretty susceptible to getting bent if the paired wheel axle suffers some blow . Not only could the toe-in be affected, but also other geometries. Also, the alignment of the axle and gears within the axle housing might now be binding, contributing to quicker bearing wear (besides tire wear and lower speeds). It would seem the only recourse would be to replace the entire rear paired wheels axle. Some delta manufacturers make the rear axle a straight tube from side to side, use an intermediate drive gear and axle which then drives the (typically roadside) rear single drive wheel. This would _seem_ to be a more robust solution, less prone to getting bent and misaligned, though at the penalty of some additional weight. But for some hpv applications, a little more weight to increase durability, is an acceptable tradeoff.
In their favor, the distance between the rear paired wheels of many delta config recumbent trikes is narrower than the c. 35" typical width of the front tadpole paired wheels. This possibly makes it harder to damage the rear axle on the delta config. (Sure makes them easier to get through doors!)
I'd be interested in hearing personal experiences on the subject of paired, non-steering, driving wheel(s) alignment on trikes.
From: "Ian Sims" ian@greenspeed.com.au
Date: Tue, 28 Nov 2000
Organization: Greenspeed
Subject: [trikes] Re: Centre-point steering question - GLR problem??
List-Id: Recumbent Human-Powered Trikes <trikes.ihpva.org>

Bob Stuart writes:-
Just so we don't cut ourselves off from a much larger body of knowledge, let's remember that car people call "Centre Point Steering" "Zero Scrub Radius" and avoid it because it reduces feedback, giving a "dead" feeling to the steering action, much like overdone power steering.

This certainly does NOT apply to our "centre point" or "zero scrub radius" steering!
MY definition of this type of steering is that the projected axis of the kingpin will go through the centre of the contact patch of the tyre in FRONT view.
This means that when the front wheel hits a bump, there is no torque reaction on the handle bars, and likewise when one wheel brakes, again there is no torque reaction in the steering system.
However, when viewed from the SIDE, the projected kingpin axis is some distance in front of the contact patch, in much the same way that the projected fork axis is some distance ahead of the centre of the tyre contact patch on most bikes.
This distance, commonly know as trail with bikes, or as caster in cars, which do not normally have an offset, is what gives the feeling of "weight" to the steering while cornering at *some speed*.
As I designed my own car, and a number of road race side car outfits before I started on trikes, I was acutely aware of the relationships between steering "feel" and steering geometry.
Thus, because the trikes are a fraction of the weight of cars, the trike trail or caster was carefully designed to give a good "feel" on the road, so one can sense, for instance, exactly HOW much grip the tyres have on the road, so you can adjust your speed accordingly.
IF you were to have ZERO caster and trail, then you would have "power steering" as Bob suggests, and there would be no feel to the steering.

From: "Ian Sims" ian@greenspeed.com.au
Date: Tue, 28 Nov 2000
Organization: Greenspeed
Subject: [trikes] Re: Centre-point steering question - GLR problem??
List-Id: Recumbent Human-Powered Trikes <trikes.ihpva.org>

Ian Humphries writes:-
Is it possible to eliminate "brake steer" completely and still not have "bump steer"?

Basically you are looking at two different trike behaviours.
Our "centre point" steering or "zero scrub radius" is what prevents the steering from "kicking back" on bumps.
It also prevents the brakes from pulling on the steering system, i.e. physically moving the handle bars while you are holding them, which will occur when a steering system has scrub radius of say 30mm or more as I have seen one some designs, and one front wheel is braked.
The OTHER behaviour we are looking at is the behaviour of the trike when it is considered as a whole, i.e. neglecting the steering mech. completely.
Let's say the trike is sitting on ice, and you kick one front wheel from the front. Then is does not matter WHAT steering mech. you are using, the trike will still spin on the ice!
This is the situation that comes in to play at high speed. Whereas this effect will not be noticed at low speed, at higher speed it is more noticeable.
Obviously riders differ in their skill and personal preference.
Some like independent brakes, and some prefer them coupled and balanced - obviously the coupled brakes will feel safer at high speed, but not offer the safely of using the strength of both hands, or the convenience of using either hand while being able to signal with the other. Or the ability to steer with the brakes at high speed on loose surfaces (Yahoo!)
To improve brake control at high speed with our disc brake option, we are now using the Hope disc brakes instead of the Hayes, which we have found to give far better modulation - i.e. much more accurate control of the braking, yet with similar ultimate power. They can be fitted with one lever for balanced braking.
We now also offer hydraulic operation for the Sachs TOP drum brakes, and again these can be ordered with single lever operation, as we have done for many people with the use of only one hand.
So yes, Ian H., you can eliminate "brake steer" completely and still not have "bump steer".
Personally I prefer the independent brakes on my trikes, and find them fine above 80kph. I think I'd find coupled brakes too boring.
But then we are all different :-)
From: Giles Puckett giles@research.canon.com.au
Date: Mon, 27 Nov 2000
Subject: Re: [trikes] Centre-point steering question - GLR problem??
List-Id: Recumbent Human-Powered Trikes <trikes.ihpva.org>

Ian Humphries wrote:
Is it possible to eliminate "brake steer" completely and still not have "bump steer"?

Naive though I may be, I always thought the two were exactly equivalent. If the projected KP axis is slightly outboard of the contact patch, then you wouldn't have any brake or bump steer affecting the steering wheels directly. The exact location may vary with the degree of deceleration, and so be impractical to implement. (it also requires a skinny hub or rather a lot of KP inclination)
The old VW Passat used this arrangement, so if you hit a gutter at a glancing angle, the car would turn away rather than jerk the wheel out of your hands and turn into the gutter.
Note also that a KP directly in front of the contact patch does not take play out of the steering while in motion, so a little flutter may be noticed.
From: Rick Kaplan
Date: Tue, 5 Dec 2000
Subject: [trikes] Center point steering
List-Id: Recumbent Human-Powered Trikes <trikes.ihpva.org>

The recent discussion on center point steering was most valuable to me.
I was having some noticeable break steer on my homebuilt trike and took a closer look at my kingpin angles. Sure enough, the steering tubes were aimed about a half of a rim thickness inboard.
I use wheelchair hubs with 1/2" axles. The axles are mounted through a tube that has a nut welded to one end. The tube is in turn fitted (and welded) through the bottom of the steering tube (this makes stub axle replacement easier). I relocated the axle tubes inboard about 3/8".
Now the steering tubes are pointed directly at the tire contact patch. This has greatly reduced the brake steer and the trike is much less twitchy than before.
From: GEGRAT@aol.com
Date: Sat, 7 Oct 2000
Subject: Re: [trikes] bushing - 'plain bearings'
List-Id: Recumbent Human-Powered Trikes <trikes.ihpva.org>

Hi Trikers, I am a long time lover of bushing 'plain bearings' in the correct application. They are very good in periodic heavy shock loads and slow speed operation. (sounds like a spindle or head set application doesn't it) Just imagine a shock load on a ball bearing set, you have 15 to 20 round almost points of contact. If the load is a little too much or too frequent you will have 15 to 20 small divots in the bearing race. From there on it is down hill till it is replaced. A bushing has a large flat area of contact for the shock load to distribute it's force. If the speed of the rotations is slow enough to not over load the lubrication the bushing will last indefinitely. Most all engines have bushing in them on the crank and cam shaves that are feed with high pressure oil for lubricating and cooling, that last many hours. (Not that I am an engine lover, it is just an example of the use of bushings. )
From: Mark Stonich
Date: Sat, 7 Oct 2000
Subject: Re: [trikes] Plain bearings
List-Id: Recumbent Human-Powered Trikes <trikes.ihpva.org>

Plain bushings often are the first choice for IC engines, but for reasons that do nothing to make them suitable for steering mechanisms. I haven't followed F1 auto technology for a while, but back when I did, designers were equally divided between roller and plain bearings. In an auto engine there is no metal to metal contact once oil pressure and RPMs are up. Hardly the case for steering pivots.
If you like your bushings, fine. But I want something with seals, and that I can adjust.
I may be prejudiced because;
A. I have found headsets to be one of the most durable and maintainance free parts on a bicycle.
B. Though I've never used them in this exact application, every time I've built something using oil impregnated bronze bushings, they quickly developed an unacceptable amount of slop.
Date: Sat, 07 Oct 2000
From: Fred Lenk
Subject: Re: [trikes] Plain bearings - a 3rd alternative
List-Id: Recumbent Human-Powered Trikes <trikes.ihpva.org>

Another steering solution besides plain bushings and headsets: My trike's original frontend used bronze bushings and kingpin. Noticable play after 4000 miles or so, and I stopped riding it much after about 5000 miles since I then had the Pursuit. This particular steering design also used endbearings connected to the ends of the kingpin. These endbearings were still in good shape.
At about 5000 miles, front end of trike was reengineered. Now there are only endbearings supporting the steering knuckle, much like the new Catrike. These are the same endbearings that came on the original frontend! I just brush off the dust periodically and put a little Tri-Flow on them. I sometimes wonder why not more trike makers have tried the endbearing solution. Seems like a good idea, and since they are adjustable, they can compensate for small imperfections in frame squareness. They're also really easy to periodically clean thoroughly.
Website->www.lenk.net
Wed, 16 May 2001
From: Richard Ballantine Chair, British Human Power Club bikebook@ballantine.u-net.com
Subject: [trikes] Re: measuring toe-in
It's easy to measure toe-in, with the seat on and the rider aboard as well, if you measure from the outside. I made up a wooden jig in a wide U-shape, using 3 pieces of 1/2" x 1" (anything similar will do) plus a couple of corner braces to keep the sides of the U rigid. The jig is wide enough to slip over the outside of the wheels, and deep enough to clear the machine, yet reach the mid-point of the wheels. On the inside of one leg, near the end, there is a nail, which stands proud of the wood. On the opposite leg, there is a bolt, with nuts to either side, so that the length which protudes is adjustable. It is set so that when the jig is slipped over the rear portion of the wheels, and the nail and bolt make light contact, the toe-in is correct. As a cross-check, the jig can be placed over the front portion of the wheels, and a small ruler used to measure the gap between rim and bolt or nail. Alternatively, a cap of the appropriate size can be placed over the bolt or nail. It's all very ad hoc, but is quite precise and works quickly. I made up the gadget because in addition to a seat, I've got to get over a full fairing.
From: "Ian Sims" ian@greenspeed.com.au
Subject: King pin angles
Date: Sun, 9 Sep 2001
IMO it is NOT centre point steering or zero scrub radius which takes the feel out of the steering but the lack of *CASTOR*!

Generally speaking with cars, it was normal NOT to have zero scub radius, simply because it was fairly difficult to achieve with wide wheels, and not having it reduced some of the suspension loadings under cornering. Plus with hydrulic braking there was no steering reaction under braking.

However tyre blowouts did produce a fairly severe pull on the steering to one side, and after a number of deaths, car manufactures started introducing centre point steering, or at least reduced the scrub radius in their designs.

Most cars run between zero and six degrees of caster, a common passenger car figure being 2 degrees.

The more caster you run, the greater force you need to turn a corner at the steering wheel, and the more "feel" you have. E.g. if you suddenly come apon and icy patch while you are corning, the wheel will go light in your hands, indicating that the front wheels have reduced grip on the road.

This quite different from the "feel" that you might get from NOT having center point steering, which will simply give you "kick back" from the bumps on the road, which IMO is NOT helpful or useful.

Thus I use centre point steering with 11 degrees of caster, so that on such a light machine as a trike, there is a good amount of feel in the steering WITHOUT kickback from road shock, or steering bar reaction from uneven braking.

Back

Thursday, 29 January 2009