Knowledge needed of rear torsion-bar housing

I am trying to find a detail of the type 1 rear swing-axle torsion HOUSING located on the pan/frame. I'm trying to gain some knowledge as to the fixed "female" splines OR torsion-bar-HOUSING-indexing geometry. I'll call these the "female" splines or what the torsion-bar engages; the splines on the torsion-BAR being called the "male" splines.
I am trying to understand better the differences between the outer and inner torsion-bar "male" splines relative to what I perceive to be the "fixed" position of the HOUSING index.
Again torsion BARS in the REAR swing-axle suspension.
If I haven't expressed this clearly, well I'll plead ignorance. This pursuit was triggered by my understanding/misunderstanding of the different spline count on either end of the torsion bar.
I looked here at an old post (by Gordon?) indicating adjustment or
"clocking". Checked STF and Samba as well. Frankly I'm still a bit baffled.
This is one area I never had to deal with in any of my 5 V-dubs in the past.

The torsion bar has 40 splines on the inner end and 44 splines on the outer end. This allows you to rotate each end back and forth a different number of splines to get a finer adjustment than would be available if both ends had the same number of splines.

For example, on the drivers side torsion bar, rotating just the inner splines (holding the spring plate and torsion bar fixed together) counter-clockwise 1 notch moves the spring plate up 9 degrees. Rotating just the outer splines (holding the torsion bar and torsion housing fixed together) counter-clockwise 1 notch moves the spring plate up 8 degrees 10 minutes. So if you went 1 notch counter-clockwise on the inner splines and 1 notch clockwise on the outer spines, the total change would be to move the spring plate up 0 degrees 50 minutes. This 50 minutes change (+1/-1) equates to a drop in rear ride height of 0.55 cm.

Because each end of the torsion bar can rotate 360 degrees, there is no "fixed geometry". The stock ride height is set by measuring the initial installed angle of the spring plate and adjusting from there to get the recommended angle. This is with the spring plate floating (off the bottom stop).

Stock angles are: (but these cars are lighter so this doesn't really apply except as a starting point.)
13 degrees long axle Swing Axle 24mm bars
17 degrees 30 minutes +/-50 minutes for Short Swing Axel 22mm bars
20 degrees 50 minutes for the short 21mm Swing Axel diameter bars
20 degrees 50 minutes for the long IRS 22mm diameter bars

Here's a good description including a chart for how many notches are changed to get a specific change in ride height. http://www.mydune-buggy.com/home/dl_files/reartorsionsetup.pdf

Justin-your explanation provided some clarity as to the adjustment. Thank you.
Regarding your comment '...there is no "fixed geometry". ... ', what do the inner splines of the torsion-bar engage with. What does it slide into-a torsion bar must have some resistance?

There's a female 40-spline socket welded inside the torsion bar housing. So the inner splines of the torsion bar are held fixed to the chassis by this socket and the outer splines of the torsion bar are held fixed to the spring plate's socket. The spring plate is free to rotate up and down relative to the chassis but because the two are connected by the torsion bar there is a resistance to that rotation.

"Indexing" the torsion bar by pulling one end's splines out of their socket and rotating it a certain number of splines and pushing it back together changes the up or down angle of the spring plate relative to the chassis causing a corresponding change in ride height.

Basically, you want to put the torsion bar into the housing first, then slide the spring plate on. Measure the angle of the spring plate. Find out how much difference there is between your measurement and those stock measurements I gave. Let's say your initial measurement is 20 degrees down angle on the drivers side. The angle you want is 17.5 degrees. A difference of 2.5 degrees upward. From that chart in the PDF linked above, a 2.5 degree change requires the inner splines to be rotated 3 notches counter-clockwise, and the outer splines to be rotated 3 notches clockwise. (If your initial angle was 14 degrees down, that's a 2.5 degree change downward and you would need to rotate the inner splines 3 notches clockwise and the outer splines 3 notches counter-clockwise.)

The trick is you can only change one end at a time. If you pull both ends loose from their sockets you'll lose your initial setting and have to start over. You can count the notches by feel if you have a delicate touch or you can do it by measuring the angle. For example, your starting angle is 20 degrees down. Do the outer splines first. Pull the spring plate off the torsion bar (the inner splines must stay engaged in the chassis housing). You need 3 notches clockwise so that's 3*8.16=24.5 degrees. Adding to your initial angle that's 44.5 degrees down angle. Rotate the spring plate until you get that reading and push the spring plate back on. Now pull the inner splines loose inside the housing (must keep the spring plate and torsion bar together). Rotate the spring plate/torsion bar until you get the desired 17.5 degree down angle and push the torsion bar inner splines back in place.

Because of the coarse splines, your angle measurements don't have to be that exact. In the example above the outer splines will slide in place somewhere between 40.5 and 48.5 degrees. So just get it close and you should be able to wiggle them in. Once they're back in the measurement will be correct.

Justin-again, MANY thanks for the clarity. It's not often that I see such in-depth responses here. I think one of my main problems was to get "a visual" in my mind as to the female socket ("fix point") you described in the first sentence of your response. Your adjustment info is quite clear. I can tell that you are really in to the tecnical end.
Anyway, I REALLY appreciate the knowledge and input you've given here. Thank you!!
Scott S.
(In my endeavor for "self-education" on this subject I stumbled across a web-site for a modified spring-plate for lowered vehicles; designer is an ME and seems fairly knowledgeable. I'll post a bit later with that info!)
Again, many thanks Justin.

Along this subject line I know that by adjusting the splines you affect the pre load/ride height but does anyone know where to get softer torsion bars to smooth out the rear ride since our cars are a little lighter? Recommendations?
Thanks.....
Doug

The easy way is to use adjustable spring plates for fine tuning.

Get an angle finder app for an iphone or similar andriod device and use it to index the bars. Once you get both sides close and similar, the spring plate adjustment will do the rest

Leon-in my endeavor to understand all of the above better, I also wondered about the torsional limits of the torsion-bar given "x" material of the bar itself. Knowing this makes it calculable. Anyway, see if this helps:
http://swayaway.com/TechRoom_VWguides.php
http://swayaway.net/joomla/images/SAW_Catalog_Part3.pdf
SS

It's not the material the bar is made of that determines spring rate. It's the diameter and length. Thicker diameter bars or shorter length bars cause the spring rate to increase.

Spring rate for solid torsion bar or sway bar:

Spring Rate = (1129000 * D^4) / (La^2 * Lb)

D = bar diameter in inches
La = length of arms (center of bar to center of axle along the spring plate)
Lb = effective length of bar (not the entire length, the length of the constant diameter portion between the splined ends)

So a 25mm diameter bar, 20 inches long (21.75 full length), and 16 inch arms gives a spring rate of ~207 lbs/in (actually wheel rate in this example). Sway-a-way has conveniently calculated all this for you in those tables.

A hollow torsion bar can be calculated by replacing D^4 with (OD^4 - ID^4). Where OD = outside diameter and ID = inside diameter.

Justin-I thought that a material strength "value" would have to have been plugged-in to this equation along the way(?). I do understand that the diameter differences would change the springing.
Thanks, SS

OK sure, if you actually knew what specific material/alloy was being used, you could find the Modulus of Rigidity (in psi) and use that to calculate that constant in the formula above.

That constant is calculated as (Pi * G / 32) where G is the Modulus of Rigidity. The 1129000 value is for the spring steel commonly used in these bars (G = 11500000 psi).

Just curious;

Does the "Modulus of Rigidity" change depending on how (and how deeply) the torsion bar was hardened or is it assumed that the hardening is constant throughout the bar?

gn

Yard work, just checking back in now. Thanks for the reply Justin!
SS

Now, we're getting way outside my knowledge base here, but I think the elastic constants are insensitive to temper. Hardness would change the max twist the bar is capable of.

Case hardening would be bad for a torsion bar. You'd have a brittle shell at the point of highest shear. The surface would tend to fracture with less twist and cause the bar to fail.

With all this talk of hardness and rigidity, I can believe how far this thread has gotten without drifting!

I had the same thought, Troy. This could be a record.

They must be busy this weekend

obviously, that should have said.... "can't believe" in my posting.

I can't even PRETEND to know enough about this topic to START a hi-jack/drift. But I will say that I once took out a torsion bar to replace until my neighbor ran over himself.

Seems that my neighbor ignored the fact that I was in the middle of my project when he came into the garage and asked me if I would run a package to the Post Office for him. I told him "NO", so he ran over himself.

Besides, I don't think many of the SoCal SOCers even look through the 'Technical' threads.

Scott,

regarding your question about how to reduce the stiffness, or spring rate:

One of the TD guys took his rear torsion bars to a machine shop and had them shave a little off them on a lathe. .010, if I recall it right.

Same idea as yours--the car's a bit lighter than the Bug, so the springs ought to be a tad softer.