Ford torque tube trivia

[BillyShope]

Not really a "Helpful Tip," but there's no "Interesting Trivia" choice.

The simplicity of the old Ford torque tube suspension (used in the first half of the last century) is to be admired. It actually reduces to a triangle with 3 links: The axle assembly, the Panhard, and the rest of the car. Now, you expect to find triangles in bridges and such, but mechanisms designers generally avoid them. Ford's design (yes, I know it was used by others) manages to incorporate the necessary suspension restraints and articulation within that simple triangle.

But, as I recently admired it anew, I suddenly realized that the simplicity came at a cost which few would accept today. While many reference books show the torque tube roll axis on the car's centerline in plan view, this could not be the case. It is simply not possible to rotate one side of a triangle about the midpoint of an adjacent side. Essentially all of the body roll would be about the one axis of the triangle. In other words, it would be about a line passing through the front ball and the pivot at the chassis end of the Panhard.

This would mean a roll axis at a considerable angle from the car's centerline and, consequently, it would also mean different roll steer characteristics depending on turn direction.

Perhaps we shouldn't worry so much about roll axis migration.

http://www.racetec.cc/shope

 

[NormPeterson]

"it would be about a line passing through the front ball and the pivot at the chassis end of the Panhard"

I think that defines the roll motion of the sprung mass relative to the PHB, but not necessarily between the sprung mass and the axle since the PHB rotates about the X-axis relative to both. I don't think the effects of spring and sta-bar stiffnesses can be separated from the geometry here (consider a 2000 lb/in spring on one side and a 20 lb/in spring on the other, then swap the springs side to side).


Norm

 

[BillyShope]

Yes, I considered this, Norm, which is why I said "essentially." But, it would seem that the chassis motion from the Panhard rotation would be more of a translation than a rotation. This would be the case, for instance, with Panhard rotation without the presence of a lateral acceleration, regardless of the spring combination.

Certainly, there would be no reason to imagine the roll axis being central to the car. The triangular configuration simply doesn't allow it.

http://www.racetec.cc/shope

 

[BrianPetersen]

All variations of suspensions of this type (including torque tube and "truck arm" and ladder bar suspensions) that use a panhard rod are subject to some axle-steering effects and some effects due to the axle being pulled sideways during bump travel.

The effect can be minimized by making the panhard rod close to horizontal at nominal ride height and as long as possible, and also choosing the height of the front pivot of the torque tube (or ladder bar or truck arm). Or you can use other types of lateral location (e.g. Watts link) that are less subject to this, but then there are more parts and more complication and it defeats some of the inherent simplicity.

Keep in mind that all multi-link solid axle suspensions have some rear steering effects. If one side goes into bump or droop and the other side doesn't, the link on that side will pull that end of the axle forward, thus slightly steering the whole axle.

I've been through all this. A few years ago, I reworked the suspension on my trailer to use coils and shocks because of the horrible roads in this area (fed up with broken leaf springs). I used what amounts to torque arm and panhard rod geometry, with the panhard close to horizontal and the nominal height of the front pivot at the same height as the axle. I figured out that in the worst possible case of axle steer - which happens when the trailer is completely unloaded - the trailer wheels will be tracking about 1" out of kilter with the car. About 1/4" of this is because of the axle having been pulled sideways and the rest because the axle is being steered slightly. I figured this would be acceptable, and it has been. I haven't seen the trailer out the side windows of the car yet, so it must be good enough.

 

[BillyShope]

I would like to apologize. I'm afraid I became so fixated on the triangle I described that I began to talk about the major roll axis and the rear suspension roll axis interchangeably and without distinction.

The roll axis I described...a line passing through the ball and the chassis end of the Panhard...would be the roll axis for the rear suspension, of course. The major roll axis would be a line from the front roll center to a point midway between the ball and the chassis end of the Panhard (which would still be at an angle to the centerline).

Brian's mentioning of the truck arm and ladder bar suspensions provides a means of describing an interesting variation. First, it should be noted that the truck arm is essentially the torque tube arrangement without perfect convergence at the front. Now, if you were to take the truck arm structure and rotate it 180 in plan, placing the convergent point on the axle housing and to the left of centerline and then add a single ladder to the right of centerline, the major roll axis could again be placed on centerline. Yes, by the time you provide sufficient rigidity to the ladder, you've added some unsprung weight, but the arrangement can also provide cancellation of the driveshaft torque. The problem of changes in roll steer with direction changes persists.

I have a spreadsheet for such an arrangement on Page 43.

http://www.racetec.cc/shope

 

[JayMaechtlen]

Well, my '56 Buick had a torque tube same as Ford, except (istr) it used coils instead of transverse leaf springs.
It drove ok, no noticeable difference between right and left turns. But, I wasn't a racer.
Your no-link suspension is interesting, but would probably benefit from a single link on the driver side to keep the axle square with the car. That would seem lighter and easier than triangulating to the single mounting pivot of the not-a-link. If you're obsessed with panhard link location, why not use a Watts linkage?


Jay Maechtlen

http://www.laserpubs.com/techcomm

 

[BillyShope]

"Well, my '56 Buick had a torque tube same as Ford, except (istr) it used coils instead of transverse leaf springs.
It drove ok, no noticeable difference between right and left turns. But, I wasn't a racer."
I had a '53 Buick, which was probably the same, but, again, I was no racer. I remember, though, that it was much more difficult to keep on the narrow two lane roads than the '52 Pontiac I had owned before."
"Your no-link suspension is interesting, but would probably benefit from a single link on the driver side to keep the axle square with the car. That would seem lighter and easier than triangulating to the single mounting pivot of the not-a-link.
It's not really a "no-link" suspension. I was careless in my description. The truck arms, in their normal configuration, are solidly attached to the axle housing. When I flipped them around, I added pivots at the front. This would provide the necessary articulation and also 'keep the axle square.'[/b]
"If you're obsessed with panhard link location, why not use a Watts linkage?"
Yes, this is the easy out and would have solved the "problem" for Ford. Obviously, that complication...and expense...was not required.

Note that the scheme I described in my previous post has no Panhard.

http://www.racetec.cc/shope

 

[BrianPetersen]

I think the Ford system used the leaf springs for lateral location but (I THINK) they were free to float fore-and-aft and rotationally relative to the axle (unlike leaf springs on later vehicles that are securely attached to the axle and take up all axle-locating forces in all directions). The torque tube took up the axle wind-up forces, the fore/aft location forces, and the keep-the-axle-square forces.

I seem to recall that some really old cars used a transverse leaf spring with a shackle at each end, and the axle basically just hung from the shackles with a link on each side to locate the axle fore-and-aft on each side. There wasn't anything to securely hold the axle rigidly located side-to-side. Could be wrong. We are talking Ford Model T suspension here - LONG before my time.

A coil spring system cannot use the springs to do any locating or guiding, so then something is needed to locate the axle side-to-side. The '63 Chevy pickup truck-arm setup used a short panhard rod. No idea what the 50's Buicks used.

There's another torque-tube suspension system that used coil springs which is from much, much later (1970's) and that one is from my era. It's hardly a performance application, though. The rear drive GM T-cars (Opel Kadett, then Chevrolet Chevette, then many others) used a torque tube with a soft mounting at the chassis end, a panhard rod, and a short trailing arm on each side to locate the axle fore-and-aft. The torque tube mounting had to be soft because the "steering" effects of the trailing arm on each side didn't necessarily coincide with where the torque tube pivot ended up.

 

[BillyShope]

Brian, I'm going to forward your comments to a friend (also in his mid-70s) who was a line mechanic in a Ford garage while in his teens and who must have come into contact with all the early Ford suspensions. I saw them, but was not interested enough, at the time, to really "see" them.

I know his memory is becoming selective. I tried to remind him of the air pressure generating system they had at the garage where he worked. I was fascinated by it (perhaps an early sign I was going to be an engineer). It was based on a Ford V8, with one bank the engine and the other the compressor. He couldn't remember it.

The Fords to which I had reference had a transverse leaf. (Chevy used parallel leafs to control lateral location). I suppose you could use a shackle at the end of one half of a transverse leaf and consider the other half to be a Panhard, of sorts. I'll be surprised if my friend tells me that was the case.

http://www.racetec.cc/shope

 

[evelrod]

I'm not sure about Buick, I've owned a 55, a 57 and, a 37 all t-tube with coils except the 37 which had parallel leafs.

My Fords, a 30, a 40, a 46, a 47 all had t-tube with transverse leaf with shackles at both ends of the leaf. NO Panhard bar on any of them. I've pulled a many rear end from those cars to change blown up trannies and I had my 30 out last year to put in a 39 gearbox. NO Panhard. I've never seen a Panhard, Watts, whatever on a stock early Ford rear suspension.

Yeah, Billy. I'll be 70 in two weeks and my wife says I can't remember doodly!

Rod

 

[BillyShope]

Here's the response from my Ford mechanic friend:

"The early Ford rear suspension was, for the most part, unchanged between the Model T days of the early 1900's, and the last of the type, in 1948. The spring was a transverse leaf type "buggy spring" secured to the axle by a set of shackles at either end, and at the center of the spring by a center bolt and clamps to the rear frame crossmember. The fore and aft position was maintained by the torque tube and a pair of braces from the brake backing plates to the center of the torque tube. Any fore and aft movement was then through the motor mounts. The body was free to swing sideways, limited only by the length of the shackle bars. As the weight and speed of the Ford cars increased, a Panhard (sway) bar was added in the late 40's to both rear and front axles to reduce the swing. This system was abandoned in late 1948 with the arrival of the "new Ford", which used parallel springs at the rear and independent coils at the front, and a Hotchkiss-type rear axle. One unfortunate side effect of the old style suspension was the tendency to produce clutch "chatter" induced by the thrust of the torque tube to the engine and transmission. For all its faults, the transverse leaf suspension was very tough, and was designed for the primitive roads of its day. Henry Ford hated change, witnessed by the fact that he forbid hydraulic brakes until 1939 as being unsafe."

http://www.racetec.cc/shope

 

[evelrod]

I read you post, but I strongly disagree with all of what you friend says. Indeed, I never owned a 1948 Ford product, but I sure as heck owned a 47 and 46, both of which I reworked the suspension for drag racing...Sorry, NO PANHARD BAR. Perhaps it was added to Lincoln or Mercury or perhaps to the 1948 model year, but I just went out to the garage with a flashlight to check just in case my mind had gone all sideways in the last few years.

The 47 got a 265 SBC (as did my 40 DeLux coupe) with a Wilcap bellhousing adapter to the Ford box so I left the rear alone.
The 46 got a 52 Olds/49 Hydromatic and Olds 4.10: 1 final mounted on coil springs.

Perhaps I can manage to get a photo posted for y'all. For me, THAT is a challenge.

Rod

 

[JayMaechtlen]

Hi, Billy
You have so much good stuff on your site, I must have been looking at the wrong one. ? I'd swear I saw it said that (roughly) 'a link has pivots on both ends, so if there's only one arm and it only pivots at one end, you have a no-link suspension'.
I see p43, but I remembered seeing a drawing like that with the ladder but without the link. Did I catch it before you were done? What you show now looks plausible, but seems like it would be real sensitive to any play in any of the pivots- everything would have to be Heim joints or ball joints to maintain alignment well enough?


Jay Maechtlen

http://www.laserpubs.com/techcomm

 

[BillyShope]

Here's his response and I think we'd best leave it at this:

"Hate to disagree, but there IS a Panhard bar on the late 40's Fords. I happen to own a '47, and it has them, fore and aft, as did all the '42-'48 cars. In the Ford master parts book, it is called a "passenger stabilizer", and is clearly shown on page 183 of the book. The part number is 21A-5494. One of the major producers of aftermarket parts, Chassis Engineering Inc., used my car as a basis for offering the bars for the earlier cars. The[y] measured and photographed the bars on my '47 convertible. I also owned a '46 sedan coupe, which had the bars."

 

[NickJ67]

Peugeot used the torque tube/live axle setup with coil springs and panhard rod from the 50s through to the late 80s (304,404,504 & 505). My father had a 404SW and that had a very decent ride and handling (it was no sports car it is true) and needed no more than the occasional pair of dampers in over 200k including a large proportion of unsealed African roads. Being the estate version it had double coils, fore and aft of the axle, on each side. The pickup versions had leaf springs and no panhard rod.

Nick