Planning – Steering – Stock Data
Planning – Steering – Calculating Final Steering Ratio

After having done the math, I wanted to confirm that I was correct in thinking that steering arm length does truly matter when looking at total steering ratio.  Furthermore, I wanted to make sure that the computed results were confirmed by actual movement.  I am not much of a math person, so I don’t trust those results completely.  Mounting parts on the car to test things was out of the question because of the previously revealed problem:

So, I had to do some modeling to get what I needed.  Perhaps I could have found a computer program.  Perhaps I could have drawn it out.  Perhaps I could have been really clever and done it all in Excel.  Any of these would work, but none would allow me to easily see what is going on.  For that, I turned to tongue depressors, foam insulation board and brads.  The height of sophistication and truly cutting edge technology, I know.

Here is what I did.  Hopefully (mostly) self explanatory, as it was intentionally simple.

I made three sticks, one for the Pitman arm, one for the steering arm and one as the center link.  I anchored the arms on the line I drew on the board, then attached the center link through a hole five inches up on the Pitman arm and 6½” up on the steering arm.  Pushing the brad through the holes makes a mark in the foam board that I could measure from.

I then moved the top of the Pitman arm a set number of degrees (leaving the bottom anchored) and attached the center link to it.  Again keeping the bottom anchored, I moved the steering arm so that the center link could be attached.  Finally, I took out all the sticks, connected the holes with lines and measured them with a protractor.  The results were exactly the same as those given by my mathematical method.  EXCELLENT!

My modeling days aren’t over yet, however.  I will be trying to make a model that has a tie rod involved, instead of the current situation where the center link acts directly on the steering arm.

Previous post:
Planning – Steering – Stock Data

As I have posted before, the rated ratio of the steering box or rack will not always translate to the final steering ratio.  In my Internet searches, I found many posts on multiple forums claiming that steering arm length doesn’t matter.  I am here to not only say that it does, but to show you how I came to that conclusion.  This will probably take more than one post, so please be patient.  I will list all the posts as I go so that anyone interested can more easily find the whole story.

The difficulty that I had on my first go around was understanding the difference between rotating motion and linear motion.  Since the steering box output shaft rotates the Pitman arm, it is rotating motion.  Through the center link it is transferred into linear motion, then the outer tied rod end pivots, or rotates, with the ball joint as the center point.

The first process I used to investigate how this all works was via simple math.  If unrestricted, the Pitman or idler arm would turn a full circle.  The arm length is the radius of that circle and with simple 2 pi R, I computed the circumference.  What the circumference gave me was the ability to measure travel of the Pitman arm end for any angle.  For example, 90° is ¼ of the whole circle, so the the Pitman arm moves ¼ of the circumference.  It is impossible for the steering arm to move a greater distance as they two are rigidly connected.

For the next step, I computed the circumference of the circle that the steering arm would create.  I then took the distance that the Pitman arm traveled in 90°, divided it by on fourth the  steering arm circumference (same ¼ travel for 90°) giving me what I labeled the reduction factor.  This is the percentage change in travel that the steering arm goes through compared to the Pitman arm.  Dividing the steering box ratio by this number gives the final steering ratio at the wheels.

All calculations here are based on the steering box system, but by removing the Pitman arm variable, you can likely transfer it to rack and pinion calculations as well.  Furthermore, the calculations ignore Ackermann and angled tie rods.  Because of this, I did not stop at this point, and will show the other methods by which I verified my findings in later posts.

As you can see in the table above, the longer the steering arm, the slower the steering.  This method indicates that turning the Pitman arm X° does not always result in the steering arm, and hence wheel, turning X°.  From the factory, the Mazda system resulted in X° at the Pitman arm becoming .909X° of steering.  The longer steering arms of the V8Mongrel spindles makes the system even worse.

I found these results interesting, but did not want to jump to conclusions.  So I decided it was time to build a model.  That will be the next post.

I have recently spent a lot of time searching magazines, books and the Internet for information on all things steering.  I have collected a lot of numbers that I will be transferring to the site in the next few posts.  There won’t be much commentary, as that will come at the end when all the numbers are in.  Hopefully by that time I will have a plan of attack.

Up next I will show how I computed final steering ratio.

After posting my check list of the strengths and weaknesses of a rack and pinion compared to a steering box, I got a suggestion from Jason of 240SS fame.  It seems my list missed two important considerations: feel and ratio.  In this post I am going to attempt to address those two in general terms, and then, once my research is complete, I will post up the numerical data I have been collecting and how it applies to the V8Mongrel.

Feel
This is a tough quality to address in the planning stages because it is largely subjective.  There is no numerical expression of steering feel that I am aware of, nor is there really any consensus as to what good feel is.  A lot of feel is actually effort and what may seem like excellent feel to one person may just be heavy steering to another.

That said, there are some objective qualities that become part of the overall result that is called steering feel.  Most of those qualities can be surmised by one word – tightness.  Good steering feel depends on a system where turning the steering wheel results in an immediate and predictable response from the front wheels.

Immediate response depends upon a lack of free play in the system.  Given equal steering columns and tie rods, it is my understanding that it is in this regard that a rack and pinion system can be superior.  Note, can be, not is.  Quality still matters and a good steering box can have less free play than a cheap, sloppy rack.  Based on what I have read so far, the way in which the gears in a rack and pinion system interact allow for less free play than the gears in a steering box.  Ultimately, the rack can be better.

One other advantage for the rack and pinion is that it seems to require less structure for mounting.  The Mazda factory competition preparation manual for the SA22C chassis calls for a reinforcement plate on the frame rail where the steering box mounts.  I have never seen such a requirement for a rack and pinion, and the off-the-shelf rack mounting brackets certainly don’t appear all that substantial.  Just like a suspension system, mounting point deflection results in a lack of response, and the evidence shows this to be more of a concern for steering boxes, it appears another advantage to the rack and pinion.

Finally, the rack and pinion also has fewer connection points in the total steering system.  The inner tie rod mounts to the rack, and that is it.  The steering box system, by contrast, has the Pitman arm mount to the steering box, the Pitman arm to center link, the idler arm to center link and on the center link the same two inner tie rod mounts.  Every connections point is a place where play can happen.  With fewer connection points, the rack and pinion has less chances for response dulling slop.

Ratio
Steering ratio, like steering feel, isn’t a simple consideration.  While steering feel is complicated by a lack of objective criteria to apply, steering ratio is complicated by the number of objective variables which must be applied.  The ratio of the rack or steering box is often thought to be the same as the steering ratio, but it is not.  Very unfortunate, as this little fact is actually one of the sources of my current steering dilemma.

I will provide the supporting data in the next post since it is incomplete at this time, but the basic truth, no matter which system is being used, is that the longer the steering arm on the spindle, the slower the steering will be.  So a steering box rated at 16:1 (where 16° of steering wheel angle translates into 1° of front wheel angle) is not an absolute ratio.  It depends on the ratio of the Pitman arm to steering arm length.  For a rack and pinion, the final steering ratio depends on the length of the steering arm used versus the length of the steering arm that the rack was rated with.  Since this is a somewhat nebulous concept, racing racks are also rate in inches along with the stated ratio.  A three inch rack moves the inner tie rod end three inches per complete turn of the input pinion.

Looking at which system has an advantage in terms of ratio is difficult.  Theoretically, the answer is neither.  Given unlimited funds, either one can be built to whatever ratio one might prefer.  Furthermore, both can have their ratios fine tuned by using steering arms of different lengths.

Unlimited funds would be nice, but it isn’t what I have.  As a result, when it comes to steering ratio, the steering box system wins.  While the racing rack suppliers provide ample information to determine the total steering ratio, an OE rack doesn’t.  Looking through car magazines can get you a listed ratio and number of turns lock-to-lock, but the means little if you don’t know the actual steering arm length and what was the arm length used for rating the rack ratio.

On the other hand, steering boxes typically have a known, internal gear ratio, which is easily converted into an overall steering ratio with application of basic math to the Pitman and steering arm lengths.  Knowing the OE Pitman arm or steering arm dimensions is not required as those can be changed independent of the steering box, affecting only final steering ratio, not the rated ratio of the unit.  Since most popular steering boxes have used a number of different Pitman arm lengths over time, this also adds a measure of tuning that is not so easily available with the rack and pinion.

A quick example shows how the ratios between rack and pinion and steering box differ from overall steering ratio.  A mid-90s Jeep Grand Cherokee has a steering box ratio of 12.7:1.  A Mitsubishi Evolution IX has a rack ratio of 13:1.  One is renowned for having the quickest steering of any production car; and it isn’t the one with the faster rated steering ratio.

I am not sure that there are any conclusions to be drawn just yet, but I’d like to thank Jason for bringing up these two point.  I think that most of my data collection is now complete, so the next post should at least be moving towards some conclusion and hopefully a plan of action.  There are still more variables to go, including bump steer and Ackermann, which, thanks to this ongoing lesson from the school of hard knocks, I am not going to ignore or try to consider without understanding their place in the entire system.

I am still working on the design of the steering system and haven’t yet come to any conclusions.  Here is a table I have made that I am using to help me narrow down the potential solutions to one of the two major steering system types.

Unfortunately, there is no conclusive winner from this list, although the steering box system likely has a small lead.

While the website and the garage have been quiet, it is not because I have forgotten about either.  I have been working on my understanding of the complexities of steering system modification and design.  To be honest, I put suspension design (and lots of other things) ahead of steering design, figuring that I could just make things fit.  I was wrong.  Steering design is not easy and a system that is just thrown together won’t work well.

Temporarily, progress on the project no longer requires turning wrenches, it requires turning pages.

I enjoy reading in general and especially about cars.  I have lots more reading to do, but once I feel like I have found some good direction, I will start posting what I find.  I won’t post information I am not sure of because I don’t want it to mislead anyone.

In the meantime, I will see if I can find some parts of the project that haven’t been documented and make entries for them if I have the time.

Previous steering posts:
Assessment – Steering – Center Link
Planning – Steering – Center Link
Planning – Steering – Still Considering the Options
Planning – Steering – Starting on the Other Options

While searching through a big box of parts in the garage, I discovered a welcome surprise.  Mixed amongst the many components taken off the car during the course of the project, I found some spares that had come along with the original purchase.  Included was a complete steering assembly!  Tie rods, adjusters, and most importantly, an extra center link!  Now, rather than rely upon my meager math skills, I can find a solution via trial and error (or cut and weld, if you prefer) while having a perfect template for the dimensions that must remain unchanged.

Sometimes a little luck is all you need.

Previous posts in the saga of the steering system that no longer fits:
Finding the problem and my list of potential solutions
Finding out how much really needs to be done

After having realized that the entire center link would need to be ground down to nothing in order for it to work without hitting the oil pan, I moved onto the next item on the list – fabricate from scratch a center link with new reamed holes for the existing idler and Pitman arms.

I have been unable to find online any resource to tell me the taper of the holes in the center link.  Furthermore, tapered reamers are $80 for the most common ones, so there is significant expense involved in going this route.  Add in the fact that the tapered holes are different sizes (smaller for the idler arm than the Pitman arm) and I might have to buy two reamers.  I would imagine that the taper is the same, but I am not sure that one tool will do both sizes.  Lastly, I believe that reaming should be done on a mill, not a cheap drill press like I have.  This idea quickly started looking like a dead end.

I was getting frustrated and at the suggestion of my wonderful wife, decided to do some other work on the car.  After I did that, I decided that I wouldn’t worry about the order of the list of potential solutions, I’d just pick one that fancied me.  I skipped almost to the end of the list and started looking into – convert to a rack and pinion, also needs a new steering column and likely modifications to the engine subframe, thus requires removal thereof.

Luckily, in my garage, I found this:

I have no idea about the mounting, how things would actually fit, but in a time when things weren’t going particularly well with the project, I also found some good news.

Look pretty close to the same size, don’t they?  The power steering line that is closest to the center link is obviously not configured for minimum size.  I would substitute a banjo bolt for that fitting to save space.

The rack is from a Focus SVT.  ST170 in Ford speak.  It has a slightly quicker ratio than the one in the standard ZX3 I own and was supposed to be an upgrade for that car.  At this point, the rack is just a fantasy, and I have already found one major problem.

There are no readily available steering column universal joints that will mate with that shaft.  All the ones I see are splined.

So, it is back to more realistic options.  Still, this was a fun little what if exercise that could one day be reality.

In my previous post I wrote my plan of attack for solving the dilemma of the steering center link and oil pan wanting to occupy the same physical space.  After making that entry, I went and did some more measurements and found that the problem is greater than I originally thought.

I tied some string between the idler arm and Pitman arm, then took the steering from center to lock.  It appears that the center link moves backwards by about 3½” over the full course of travel.  Thankfully, the body of the center link is set forward from the mounting holes, as can be seen below.

The tape measure in on the side where the oil pan is located in the car.  This offset helps create more space before contact compared to the string I used between the arms during measurement.  Combinig that offset distance with the gap from the string to the pan with the steering centered, I estimate that I need about one inch of extra space to avoid all possible contact with the pan.

The problem is that the center link bar is ¾” thick.  That means that my original first option of reinforcing the side away from the pan and removing material from the contact area is unlikely to be a good one.  I would have to remove more material than currently exists!

So I looked things over and noticed that the curvature of the pan makes this more of a three dimensional puzzle than I had originally considered.  When I first looked at this, all I thought about was moving the center link forward (red arrow).

Then I realized that because the pan is curved at the point of contact, I could gain space by moving  the link down (yellow arrow).

So now the puzzle is to figure out what angle (orange arrow) will bring the greatest results with the least modification to the center link.

Math isn’t my strong suit, so I am trying to figure out how to compute what I would need to do.  If I had another center link, I’d be much more inclined to experiment, but as it stands, I only have the one pictured.  So I am either going to have to do some computations or find another link that I can sacrifice in order to know what works.

In the mean time, I have been working down the list of possibilities and will post next about what I have found.

Center Link Assessment

So with a nice, clean center link, it was time to do a test fit before applying paint.  One lesson I have learned over the course of this project is to never make something look nice before checking that it fits.  The more time spent painting before the test fit is directly proportional to the amount of grinding, hammer and clearance alterations that will need to be done in order for the part to go in properly.

So I found the factory castellated nuts, cleaned them up with the wire wheel on the grinder, and bolted up the center link to the idler and Pitman arms.

Looking good!  Of course, I have to check the range of motion.

Oh f _______________________________!  Yup, the center link and the oil pan want to be in the same place.  A fight that the center link would eventually win, and the engine would likely end up as collateral damage in the oil pan’s defeat.  This is not what I needed.  Time for an action plan.

Challenges:

1. The Pitman and idler arms mount to special, reinforced sections of the front frame rails, which would make relocating them very difficult.
2. The tapered studs are part of the Pitman and idler arms, which makes replacing them as a step to using an alternate center link more involved.  If they were on the center link, the studs would be replaced in one step.
3. The taper of the studs and holes in the center link is unknown.
4. The taper for the idler arm stud and the Pitman arm stud are different, which appears to not be standard practice.
5. The overall space for any solution is very limited in all direction, so trade offs that allow winning one inch in the right way at the price of two inches the wrong way won’t work.

Potential Solutions:

1. Modify the existing center link so as to be able to use tapered holes and ensure correct width.
2. Fabricate from scratch a center link with new reamed holes for the existing idler and Pitman arms.
3. Convert to a GM or circle track steering box and idler arm for which tapered slugs are available, making center link fabrication much easier by removing the requirement for expensive tapered reamers I don’t have.
4. Remove engine, trans and all things hooked up to it, then modify oil pan.
5. Convert to a front steer design, thus avoiding oil pan clearance issues, again using a GM box and idler, but this will require a new steering shaft that may cause new and unforeseen clearance issues.
6. Convert to a rack and pinion, also needs a new steering column and likely modifications to the engine subframe, thus requires removal thereof.
7. Use the force to steer.

Those are in order (I think) of how I will pursue them.  First I have to figure out just how much steering travel there is beyond the point of contact with the oil pan so I know just how much room I need to create.  Stay tuned.