Death Wobble - speedway tie rod may be suspect

Well you are correct when you say that the axle should be square if you have measured diagonals but the issue here is accuracy with a tape. To do this correctly you would have to support the vehicle up on jack stands and drop plumb bobs from corresponding points and take a measurement. If the rear is lightly out of square you probably wont notice it, the rear has a fixed lateral rotational axis and the differential takes care of the tire scrub. But if the front is slightly out of square this gets one tire ahead of the other. Since one tire contacts the surface before the other and they are on a vertical rotation axis individually, but laterally connected the goofy action starts. The position of the front axle does not have to be off very much to initiate steering or shimmy issues because of the spinning effect of tires that are mounted at an angle.

One thing people fail to take into consideration is the relationship of the axles to the vehicle centerline. Because of varied tread widths you cannot just square a car by using the wheels or tire sidewall as a starting point. A bent wheel or wider or narrower tracks will throw off your accuracy.

My suggestion to square a car on the garage floor is to jack it up on jack stands and find the center line of the vehicle.

First remove the wheels to make it easier to work around. Now pick two similar points on the front frame rails by the cross member and two points in corresponding location at the rear frame rails. Plumb bob down to the ground and mark. You can now check for a diamond in the frame by Xing these points. Unless your chassis was in an accident or built by eyeball the numbers should be equal.

Now if you run a chalk line front to rear on each of these points you can build a grid of your chassis which is easy to visualize. By taking a tape measure and measuring the spread between the chalk lines you can half the distance and determine the centerline of the vehicle. By marking the center points on each end you then can use a chalk line and determine the vehicle centerline.

From the vehicle center line you can measure side to side by using a carpenter square faced off the hub flange to check the off set of the front or rear hubs, telling you if the axles are centered to the frame or skewed l-r. With these lines you now can reference the front an rear axle position.

On the center line about half way between the front and rear axle, take a large carpenters square and mark a line perpendicular to and across the vehicle center line. This will give a fixed position to measure front to rear that is square to the vehicle center line.

Now by taking you plumb bob you can choose a similar location point on the R and L forward face of the front and the rear axle and mark this point on the ground. Again with your chalk line now mark a line perpendicular to the center line on both the front axle location marks and the rear axle location marks.

This will indicate the position of the axles to the center line of the vehicle. Now by measuring between the fixed position line that you marked right angle to the center line and the axle lines you can determine if you have the axles positioned parallel to each other and square to the vehicle center line or they are out of alignment.

By using colored chalk lines you can be kind of creative and colors will give you a good picture of the foot print of your car.

Now adjust everything up if necessary. The answer your question about squaring the spindles to the axle I cheat a slight amount. I have a right angle piece of sheet metal cut out to clear the spindle and square the front face of the axle rib to the backing plate or the brake rotor to determine that the spindles are pointed straight ahead and a close right angle to the axle.

Now This is a pretty good way to check you car out with no fancy tools and be pretty accurate. The more time you take the more accurate your findings will be.

I have been going over this tie rod bending bit and the only thing that I can come up with is, as you said the tie rod is not bending, but it has a slight bowing. That is from the force applied to the rod not being on the same plane as the reactionary force and the bowing the tie rod tries to find a neutral point. This has more to do with the steering arm angle side to side than the tube itself. The steering arms should have the tie rod bolt heads perpendicular to the ground not a an angle.

 

Yes, it might be all related...too much caster makes that contact patch extend too far forward and "maybe" that's why some guys have reduced their wobble with real low air pressure which sort of changes that patch? But those flat tires may help because of the damping effect? I think someone said that about soft tires. I am no engineer, so I don't know how a tire behaves if the contact patch is WAY out too far.

One important thing I have learned on this years thread is that I am going to go with the lesser caster posted above. I have never heard anybody give those preferences before. I have always heard 7 to 8 for the last 30 years.....and by the way, in those 30+ years it was always called caster shake or caster wobble, not death wobble, so I just assumed it is all about caster.


I hope someone with bad wobble could just try that 4 degree setting FIRST and see what happens, without fixing anything else.

 

is it fixed yet?

 

Yes it is working well.

Thanks for the time to input all that alignment stuff Dick. Very good information.

 

Excellent explanation of how to measure for square and center..... should copy that into a tech thread of its own.

 

Speaking of which, I'm not sure if it was mentioned and I don't want to read the whole post over again, but what is your caster at ride height Tudor.

Also, I'd like to add a bit as far as ideas I have on this subject. I'd guess there are three different types of "wobble".

One, which seems the most common where the tires are cycling in and out towards each other.

Another where the tires both go side to side together, which I'd assume is more prevalent in cross steer cars, most likely caused by play or lack of a panhard bar.

Third is one that is more of an up and down bounce alternating from side to side, which I assume is something you'd see more in a car with a rigid tube axle and hairpins with no shocks.

The first type, from my experience, seems to be more common in rear steer cars (probably do to the tie rod being in compression, with the tires acting against it as the car goes down the road, which is compounded by hitting a bump) and in each car there seems to be either aftermarket or severely dropped steering arms and/or a tie rod/center link that is bowed, bent for clearance or otherwise weakened. I'm betting that in most cases the amount of flex or deflection in these components cause or at least contribute to the problem.

That's my .02

 

Tudor I think this pic will help make your point and Dick this is what we are talking about.
What we are looking at in the pic is a 1-inch DOM tie rod assembly lying on the ground, similar to what most hot rod shops sell, and a 1 1/4 -inch .250 wall moly tie rod on my traditional pre-war. There is a ball point pen indicating where the paint was rubbed off from the DOM tie rod flexing and hitting the bottom of the wishbone when the car was going perfectly straight. FYI there is about 1-inch of clearence between the bottom of the wishbone and the top of the tie rods drag link.
Now picture this I could hang my arm and head over the door at low speed; holding the wheel perfectly still and I could watch the tie rod flexing up and down and the wheels violently toeing in and out so much so that the outside diameter of one of the tires would rub the headlight bucket; the tire and the bucket are more then the width of my fist apart, shown in an accompaying photo is the rub mark and the distances.
All this business occured after the box was done and the spindle bushings were honed to near zero, actually I honed them to about what we do piston wrist pins too. The wheel bearings were checked continually and the alignment was adjust all over the place trying to figure if that would help. I could get it to mask the condition but the 1 1/4-inch tie rod totally eliminated the wobble for my car.

I also agree with the setting up the spring properly and to that end I recommend a book that used to be available from Mopar's Direct Connection. It taught me all sorts of stuff about leaf springs mainly that they are a type of bearing and smooth, good working suspension is critical in setting up a leaf spring rearend in a drag racing super stocker.
It is so easy to take a leaf spring for granted; don't do it you'll chase your problem around in circles. There is a whole sciene to spring packs and I highly recommend the reading material if you can hunt down a copy.

Now for some clarity. In my last post I must have been misunderstood by some. I meant if my steering box wears with time and the sector picks up more then 2.5-degrees of freeplay I will run the chance of getting a wobble back, but that wobble feels like a loose steering box and it stops when the wheel is turned off center. A quick adjustment of the box takes care of that. So no I am not running it at high speed with both hands holding on for dear life. My roadster is absolutely rock solid up to 122 regardless of the terrain and it steers through turns with a slight bit of understeer, as I stated, due to the bias plys sidewalls flexing and the drag links very slight and predictable deflection. I wouldn't have written in to offer an opinion if I had an ill handling car because obviously I would have still had a huge problem to sort out myself.

To everyone and especially you Dick. I mean you no disrespect and I admire your opinions and that you take the time to post very detailed ideas. Sometimes when we write in text we aren't face to face and we cannot gauge the candor of the person we are seaking to, so I wanted to be clear that if we were all standing around looking at these cars it would be easier to see and appreciate eachothers points of view. I have found that I write in very pointed and matter of fact terms and sometimes it seems, to others, as though I may be going mad over here at my keyboard when in fact it is just the persona I write in. So I justed wanted to get across that I admire anyone what writes intelligently, as you do, and has the ability to share a lifetime of knowledge with others.

As for alignment; of course I agree alignment is critical and mine is spot on with zero toe and in my case it was a pre-war hot rod and the caster happens to be 12-degrees. I don't recommend that because of the heavy steering effort but that is the way Hanks and Belond built it and in keeping it as close to the way they ran it as possible, the 12-degrees are staying.

Also notice in the last pic that the thread on my bias plys is uneven, and it was that way when I bought the car, but I wanted to see if I could cure the wobble even with tires that had a poor pattern, lever shocks and a drag link intersection angle that is way off the map for anything the so-called professionals recommended. I wanted to prove to myself that these things were wives tales and I feel I have. Furthermore anyone who has been on a run with me like the Reliability Run can a test to the fact that this old roadster flat handles mountain roads and even grooved freeways with ease.

Lastly my reference to 4X4 trucks did not mean think big heavy lumbering trucks that barely do the legal speed limit. Think 60's or 70's Blazer, Bronco, Scout for that matter a Jeep. Short wheelbases and light weight but with over built, old straight axle type front suspensions. What I was trying to point out is that the front suspension of a conventional 4X4 is basically a straight axle and therefore it is over built to the point of still being roadworthy even when the front end is badly worn. That is to say, of course the tires will wear like hell but it won't have death wobbles like so many experience with hot rods which in comparision have flimsy front end components.

Hope I didn't piss anyone off, that's not my intent and I apologize if I did.

 

Great information 777, Thanks pictures and sharing your relative experience. I also appreciated your input in post 169.

Just think, the tie rod was deflecting that much to hit the frame that close to the end, imgagine what it was doing in the center. IT CAN"T BE!!! ha ha ha
It's also amazing your tires and head light came together. The wobble is insame once it gets going.


Actually, one of the people who I have discussed this with off line is a former extreme 4x4 enthusiast. He recomended a stout tire rod and said it was very important from the get go. He had experience with death wobble with heavy wheels and jacked up suspension and attributed it to anything that allows free movement like worn out rod ends, bushings, weak tie rod, steering boxes: anything that allows free movement. He said they routinely sleeved tie rods or made heavy duty tie rods to keep everything in check.

Where did you get the reducers on the end of your 1.25 tie rod?? Looks good btw.

 

Good morning guys, I see that 777 has take some time to state his position and while this has gone in deeper depth than probably intended lets just keep this going a little longer. Tudor's post about the tie rod bending creates some skepticism but I've seen enough stuff that anything is possible. I thought this tie rod bowing over quite a bit and the mechanics that would cause this. Now with 777 stepping up and posting pictures, it would be nice to get a picture of Tudors application and compare the two cars and see why this is happening. Lets understand that I realize that the rod is bowing not bending. What is causing this? This is my position.

Using 777's pictures so its easier to relate to you can see that the rod has a slight contact from the rub marks. Now this is a fixed axle set up so the only way that the rub marks will appear is due to lack of clearance as the spindle steering arms rise and fall during cornering. As the spindle rotates the steering arm on the opposite side of the turn rises and the turn side lowers or the tube is flexing for some reason. First is process of elimination, straight ahead there is plenty of clearance so if the wheels are rotated full lock/lock the tie rod should rise but not strike the wishbone. lets assume for conversation that it doesn't.

Ok what is causing the tie rod to bow. Must be external forces, the only force is that generated is by the weight of the spinning tire. The speed of the spinning tire creates momentum as it speeds up or slows down on a fixed spindle. Now changing to a rotary spindle axis like your car this creates what I believe to be a polar moment meaning the tire wants to align itself on a vertical axis as it spins. This force developed is on the weight and the speed of rotation. This force is the determining factor in the centering effect of caster, camber angle and an unequal offsetting force is the force that is bowing the tie rod.

Now you have to take into consideration that not only the spinning force has an effect but the caster angle, camber angle and tread offset effect the force direction. The caster angle just wants this spinning tire to go to neutral but the + camber angle wants to pull the front leading point of the tire outward and thus force the rear point of the tire inward. Well there is this arm that is connected to the spindle, the lower steering arm. The length of the steering arm creates a lever to transfer the centering or pulling forces to the tie rod where the reactionary forces begin to resist this movement. Its physics and most of us didn't pay attention that day.

Here is where it gets trickey to explain. The forces acting on the spinning tire that are trying to center it or spread it apart also are placed upon the steering arm, thus the angle of the steering arm swing in relationship to the force applied determine the direction of the force. So this means that if your steering arms are positioned so they are pointing in an upward swing they are applying an upward force to the tie rod. Now the tie rod swings a straight line fairly horizontal between the opposite point, the force of the tire is not operating in a straight line but following the direction that the steering arm swings however this is not the direction that the tie rod wants to go and one force wants to overcome the opposing force going in another direction and the tie rod bows. Enough force and it could bend. This effect is exhibited even statically as stated in the Tudors original post.

My take on the reason that the tie rods are bowing in both these applications is the the angle that the steering arms take during operation and that they are creating a position where the forces transmitted thru the tie rod are unequally opposing each other creating the bow and the addition of more mass to the tie rod only over comes the force and serves to mask the issue.

777 thanks for the pictures to help this along but from your comments, your issues are from an excessive amount of caster which rolls the steering arm back so much that the force is being directed upward and thus bowing the tie rod. You have just installed a tie rod that is so thick that the bending forces are overcome. You need to take about half the caster out of you car and you wont need that boat anchor tie rod. I realize that you are trying to retain the period correctness of your car but having issues means that just because this was done in 1953 like this doesn't mean its correct. I'm assuming that the steering arms on your spindles were bent for clearance and you need to re look the arcs that they follow during rotation also.

Tudor I think you guys have a similar problem but I think that you mentioned that you have straight bolt on steering arms and these will do the same as they swing.

I have always taken the position that the tie rod end ball should swing horizontal to the axle plain or as close as possible so the steering forces are directed as straight as possible thru the tie rod.

Hope this has got you thinking, and thanks for taking the time to respond, and yes it sure is easier looking at a car that trying to second guess on the computer.

 

I'll try and post pictures tonight - Thanks Dick. I would have earlier, but I was out of town and tied up last night.

 

I would like to know also. Also are they threaded? if so what pitch and diameter.

I agree with StealthCruiser, I've always thought Tudor was dreamy.

 

Damn it Josh - you went for a ride Monday night and didn't call ? Are you going to Hodges shop on Friday night? Oh and glad this is fixed

 

Dick, Geez buddy your killing me over here.
Give me more credit for making sure I would give the drag link ample clearance with the wheels turning through their full travel and turning radiuses. The tie rod rub marks never came from rubbing on my car.

I did want to touch on steering arm geometry though but I thought most people would get lost. But I'll do a short history of what I did just so we can compare ideas.

I build tube chassis cars too so I have a full sized jig table. I set the roadster up without any wheels or tires. I used tire simulators instead to eliminate any differances between the wheel and tire assemblies. The fronts were set up with turning plates also. I then created a jig to allow me to make very precise adjustments to the steeering arms, even though we are just taking about getting something heated up and bending it. It can still be done precisely.
FYI: In my experience a difference of as little as 1/32-inch in height at the tie rod mounting face will dramatically increase bump steer on a true race car. As a point of reference most race car chassis builders sell 1/32-inch misalignment shims just for this purpose.
I then used a laser sighting system and my spindle arm jig to alter the steering arm angles so that the center point of all of the circles tracked by all the wheels lied within a common point. This turned out to work pretty well. But as a constant tinkerer I wanted to apply a bit of reverse Ackermann to try and compensate for the differences in slip angles between the inner and outer front tires when the roadster was cornering at high speed. What we found was the use of such geometry helped to reduce the tire temperatures in high-speed cornering but we found that it compromised performance in low speed cornering. So we went back to a pure Ackerman.

I don't really have an opinion on caster because I can only offer my thoughts based on this car and as we agree 12 is too much but in 1937 when this rod was first built it must have worked because Sandy Belond ran about 113 with it and Sam Hanks won just about everything there was to win in a midget. They must have been pretty sharp for their time.

 

I bought the moly rod from an alloy wharehouse here in So Cal. and made the reducers from 1 1/2-inch moly bar stock.
If you find a source for rod end adapters please let me know. It sure would make life easier then have to tap right and left hand bar stock.
I checked the normal sources, such as Jerry Bickel, Chris Alston and R.J. Racecars for these parts but came up empty.

 

Just to reiterate what someone else mentioned earlier. I am not doubting the fact that you can flex the tie rod but the tie rod can't flex one way when you pull on the tires and the other way when you push them if the forces are acting along the centreline of the tie rod. You simply just can't bend something by pulling it along it's length. So either the forces aren't for some reason acting through the centreline of the tie rod (binding heim for example). Or the tie rod is already bent (it would only take a very small amount over 47") in which case it is flexing more upon compression and straightening out under tension. The "bent tie rod" theory would explain why you didn't have this problem originally. It would also of changed the toe in which you mention you have changed (maybe back to what it was before it got bent).
Having said all of that I'm no engineer and reserve the right to of just spoken a load of bs.
Either way I'm glad all seems to be good now and that you can once again enjoy driving.

 

When I took the tie rod off it was not bent. It rolled flat on a flat surface and the angle iron that was new and straight laid on it perfectly. IT WAS NOT BENT. When you apply the force in the spreading direction it is possible the reaction force occured in a different direction than the compression force causing a different bow in the tie rod.

 

Here are some shots of my front end.

 

Ok. I did say I could be talking bs. I shall go back and hide under my rock

 

I don't know man - I just know that rod flexed - not a lot but it did and it wasn't bent. Didn't mean to bark.

 

That's cool. Just make sure that when you do find the cause you let us know because it's got me scratching my head and my hair is thin enough up there as it is.

 

After reading 10 pages of VERY informative posts, one has to wonder why the cheap shot directed at Speedway in the first place.

 

What cheap shot? Why does one have to wonder? The tie rod they supplied is weak enough to be flexed or bowed when light pressure is applied to the tires pushing them together or apart with the car jacked up. Fact. I know of a car that has the same tie rod that has never been driven that the guy can do the exact same thing. Fact. Where's the cheap shot?

 

Nothing wrong with rod ends, been using them for years. Just remember to get the Teflon lined ones for your steering needs. I use the Teflon ones on everything because they are quiet and last. Also the last i checked, if running Ford spindles and kingpins i believe the correct setup is .005-.012 clearance between the spindle and the kingpin boss after adjusting with shims. My $.002 worth.

 

Granted, Speedway doesn't list the wall thickness on their 7/8" tie rods etc.

I did find a pic of a threaded tie rod tube in their catalog and the wall thickness looks to be .156.


I thought I had a chunk of 7/8" x .156 DOM tubing, but couldn't find it.
Was gonna suspend some weight from each end and hang it in the center to see how much it bent.

All I could find was some 40" long 3/4" x .156 wall DOM.

Instead of setting up the weight deal I laid the center of the tube against a rounded piece on my mill and leaned into it with my hands on each end . . . kinda teeter-totter like.
Perhaps a hundred # weight total, 50# on each end - I weigh 200#.

The 3/4" tube did bend, but not much.
Maybe 1/8" at each end.

Guessing that doing the same test with 7/8" x .156 would show virtually no bend.


I'm still amazed that you can bow the tie-rod in your car and you're on the short end of the lever.

Not saying your tie-rod doesn't bend, but it sure looks like something else is going on here.
Especially since you've added two pieces of angle iron and it still bows.

A quote from your above post:

Pulling the tires together at the front puts the tie-rod in tension and I don't understand why it would bow then.


I suspect I'm not alone in the confusion dept. and your posting a series of photos would help me and probably others understand what's going on.

 

I have only added one piece of SMALL 1/2" x 1/2" angle iron. That's why is still flexs a little. You can flex that angle by itself. It still added a lot of rigidity.

Did you see the photos I did post this morning? When you change from compression to tension it only moves slightly in the opposite direction but it does move.

What else could be going on? The wheels rotate on the spindles and the tie rod attaches the two wheels together. Push on them and tie rod flexs. Take a drink stir stick - push on the ends of it, it'll bow. It a slender piece that is 43 inches long. It is not that crazy or impossible to think it'll flex.

 

10 pages of info,gets a little confusing,so many angles here.
Here is my set up,speedway kit,laid back 12*,3/16 toe OUT.
Was told the steering arm had too much angle,needed to be on top of king pin arm. Tried it and it works like a charm,no wobble and no pulling to one side. Have not changed anything from original build so I cant tell you what helps or hinders. Radical....yes. Works.....yes. Johnny Rotten

 

Is the panhard bar attached to the frame on the driver or passenger side?

 

passenger side.

 

i'm thinking draglink from the steering box...you would think it would have some bumpsteer at some point...