Audi 80 B4: Steering and Handling

Hi,
I would like to present my topic, along with current measurement data, here in the forum.
Problem: The car noticeably pulls to the right and usually requires corrective steering. Additionally, when driving on a sloped road, it tends to drift in the direction of the slope.
The problem is an old one. After a complete overhaul of the rear axle (new bearings, alignment, new shock absorbers and coil springs), a measurement was taken.
The height was averaged according to RLF standards, and the shelf supports were fixed at the front.
The initial survey report is shown in the attachment "Eingangsprotokoll.jpg".
The only noticeable issue is that the car slightly pulls to the left, meaning it drifts slightly to the right.
In my opinion, the right-hand side of the housing assembly logically had to be slightly retracted.
It required some discussion with the friendly person ("the HA must continue to steer further to the left"). After the initial measure failed, however, the example "how does a trailer steer when reversing" was understood.
Unfortunately, repeated adjustments of the rear axle, including retracting the right side, were also unsuccessful. The data are shown as initial values in the second measurement log, "Final measurement after rear axle adjustment.jpg," along with the result of the subsequent "fine-tuning," after which I proceeded with the operation.
The car continued to drift to the right.
At home, I then re-centered the rear axle (2mm off-center, as the rear axle bearings were quite tight), and adjusted the left rear axle bearing bracket to its front stop, while moving the right rear axle bearing bracket slightly (almost 1mm) back from its stop, based on the original starting values and the final values.
1mm corresponds to a distance of 3 feet between the bearing supports.
The straight-line stability is slightly better, but not really good.
Where can I continue my search here?
I already mentioned that the problem is old – it has multiple sets of tires, defective wheel bearings in the front, shocks and strut mounts in the front, bushings in the front, etc., and has survived all of that.
With new tires, it's almost gone, but it tends to come back as the tires wear down. The front tires wear out more on the outside, but I attribute that to cornering.
The car is a first hand (owner) vehicle and has no accident history.
Sure, here's the translation:

"Best regards," or "Sincerely,".
Herbert.

Except for the toe-out and a slight amount of toe-in, the values are unremarkable.
You can ignore the 5-minute difference at the rear axle if there's nothing else you can do.

The axle housing would need to be adjusted to the right (to compensate for the spread) and to the left towards the rear (to reduce the toe-in). Unfortunately, another measurement will be required afterward.
Beim A4 z.B. schreibt Audi sogar (Bei Kundenbeanstandung) vor, den Nachlauf RE zu vergrössern, wenn das Fahrzeug n. RE is increasing.


First, it's necessary to check to what extent the fall can still be corrected.

Im Idealfall ist der eingeschlossene Winkel (Sturz + Spreizung) Li u. RE, same here.

I would mark all the points with a permanent marker (to be able to find the current "as-is" state) and then start making adjustments until the car runs as it should.
Subsequently, measure it.

Hello.
Due to the geometry and adjustment options of the vehicle, camber and toe cannot be adjusted independently. Changing the camber automatically changes the toe. In other words, while I can improve the difference in toe, I simultaneously change the difference in camber angles by the same amount.
The pivot points of the strut mounts and the suspension joints form a trapezoid, and I can only change the length of the base. Therefore, shifting the axle makes no sense, and even if it did, the play in the mounting wouldn't allow it. A 0.5-degree change would correspond to approximately 5 mm.
I could reduce the left wheel camber to approximately 0.20', which would increase the left wheel toe to about 3°. However, this would bring the camber difference to the limit. I cannot assess the impact of this change.
I don't understand Audi's argument for increasing the rear toe (RE) when the vehicle pulls to the right. In my opinion, the wheel with the larger rear toe will tend to straighten itself out. With positive camber, as is the case here, I would therefore expect the pull to the right to be amplified.
What do you mean?
Sure, here's the translation:

"Best regards," or "Sincerely,".
Herbert.

Quote:

Due to the geometry and adjustment options of the vehicle, camber and toe are not independently adjustable.

Can the shock absorber be adjusted relative to the wheel bearing housing, or is the shock absorber not a two-piece design?

With a split fork leg, the rebound damping is adjustable at the fork leg, so it should be possible to adjust it.
It depends on how much has already been displayed.

EDIT: The pivot joint on the control arm appears to have elongated holes.
From that point on, it shouldn't be a problem to communicate the spread. It might even be possible to squeeze out a little more profit.

If you want to improve your straight-line driving, you either need to make some changes or continue driving in circles. However, achieving this requires having the right values and parameters. That's simply undeniable.

Hi,
The wheel bearing housing is connected to the shock absorber with two screws. There is only a small amount of play due to normal tolerances. By the way, the distance between the strut mount and the axle joint hardly changes, even when the trapezoidal arm is bent. If anything, the angle between the axle half and the trapezoidal arm would need to be adjustable with a fixed trapezoid, in order to be able to adjust camber and toe independently.
Sure, here's the translation:

"Best regards," or "Sincerely,".
Herbert.

Herbert, the overall angle is incorrect.

Adjust the camber and toe-in by correcting the alignment of the axle housing and, if possible, via the lower suspension joints (ball joints).

After that, the alignment is adjusted at the shock absorber. It's not a lot. If it doesn't work, you either need to rework something and/or find the cause, or replace any bent parts.
I don't understand the problem.

Quote:

In my opinion, the wheel with the larger offset will tend to straighten out more.

This applies to the spread, not the lag.

A larger toe-in, combined with a positive steering radius, forces the wheel inward. That's the theory. Different values cause one wheel to turn inward more than the other.

Hi,

Quote:

The ball joint on the control arm has elongated holes, if I'm looking at it correctly. From there, it shouldn't be a problem to convey the spread.

Perhaps we can even squeeze out a little more profit.
Sure, toe-in/toe-out adjustment is no problem; it's achieved through the lateral play of the control arms. I've already been exploiting that for a long time. But camber and track width adjustment are a separate issue.

Sure, here's the translation:

"Best regards," or "Sincerely,".
Herbert.

Quote:

But adjusting camber and toe is one thing.

Theoretically, the adjustment should be made at the pivot joint. However, that won't help you because the overall angle will still be wrong. A solid shock absorber would actually be bent. A split shock absorber can be bent, but it doesn't have to be if it's misaligned. For example, for the Golf 2, there were supposedly specially thinner screws available for correction.

Let's just sleep on it. A new day will bring new thoughts. Thank you very much for your time and effort this evening, and please keep an open mind (or eye) .
Sure, here's the translation:

"Best regards," or "Sincerely,".
Herbert.

Even if the car has no accident history, it doesn't automatically mean that the front axle is still straight.

Here are some simple things you can do to get more involved:

- Measure the distance from the center of the tire at the top to the shock absorber (approximately from the red to the green line) and compare it to the left (LI) and right (RE) sides.

- Compare the distance between the tire and the rocker panel when the vehicle is driving straight, and when turning right.

While it's not always definitive, it's often an indicator...

As the drawing shows, the steering radius is also determined by the camber and toe angles (total angle or included angle).

If there are significant differences, the steering radius changes, and you'll experience different leverage forces acting on the steering system.
Depending on the design and the wheel sizes used, it is theoretically possible to have a positive steering knuckle radius on one side and a negative steering knuckle radius on the other.

Hi,
I should have noticed it sooner - for example, the spread was 20 degrees to the left (2 degrees 24 minutes), and 1 degree 41 minutes to the right.
Here are some approximate measurements: Distance between the dome supports: 1050 mm, Track width: 1450 mm, Height of the dome supports above the ground contact point: 900 mm. This roughly results in a total angle of 12° per side.
I'm still unclear about what the spreading angles in the survey report refer to.
hg
Herbert.

Quote:

Some approximate measurement data: Distance between the suspension mounting points: 1050 mm, Track width: 1450 mm, Height of the suspension mounting points above the ground contact point: 900 mm. This roughly results in a total angle of 12° per side. I'm still unclear about what the spreading angles in the survey report refer to.

That would mean that each leg is positioned 20 cm further inward at the top pivot points compared to the bottom.
Unfortunately, that's incorrect. The angle being measured is the angle of the imaginary steering or pivot axis (shown in green in the image) relative to a vertical line perpendicular to the road surface. The pivot axis is determined by the upper and lower pivot points, which are the pivot point of the strut mount to the pivot point of the lower ball joint (tie rod end).
You would therefore need to measure the distance between the lower pivot joints instead of the track gauge, and take into account the distance between the pivot points in order to calculate the sum of both spreads.
But what's the point? It only gives you the sum x, and nothing else.

It's measured like this, if I remember correctly.

"Toe" refers to the difference in alignment between the left and right steering angles of a wheel. The zero point (reference point) is the straight-ahead position.

When measuring the toe angle, the wheel used to have to be locked/braked so that when turning 20° left and 20° right, it moved as a fixed unit parallel to the suspension strut, thus tracing a spatial curve. The difference between the left and right sides results in the toe angle. In the past, this was measured using levels and adjustable scales. You really had to know what you were doing.

Hi,

Quote:

That would mean that each suspension leg is 20cm!! closer to the center at the top pivot points than at the bottom.

That's exactly how it is. Take a look at a vehicle designed like that.
Furthermore, older wheel alignment measurements indicated toe-out values of approximately 12°.
I just wanted to point out that the spread values in the measurement log are very strange and certainly do not reflect the actual spread. I'm holding back from using the word "nonsense" because the term "spread of 20°" is not entirely clear to me.
If you use a drawing with a negative steering radius (or at least a steering radius of zero) instead of your current drawing with a positive steering radius, and supplement it with ruler and pencil to include the distances of the suspension mounts, a mirrored wheel on the other side, and the track width, then you can estimate the toe-in/toe-out using the given dimensions and some trigonometry. Therefore, again, the toe-in/toe-out values in the report are strange.
That's why I definitely won't start adjusting anything based on some arbitrary criteria.
Your suggestion about measuring the distance between the wheel and the suspension makes sense. I'm going to improve on that by using a reference line placed above the strut mounts and using a level, which might help reveal a difference.
Otherwise, it seems to me that a second survey with better quality would be the only sensible option, but we first need to find someone who can do it.
After that, you still have to figure out which side is crooked, if any.
Sure, here's the translation:

"Best regards," or "Sincerely,".
Herbert.

Quote:

That's exactly how it is. Take a look at a corresponding vehicle. Furthermore, old wheel alignment measurements showed spreading values of approximately 12°.

I'm sorry.

Why are you misleading us by stating that the values in previous measurements deviate so drastically from the latest measurement? Unfortunately, there are no target values for the variation, and we don't always have a crystal ball or a reference vehicle available for comparison.

That means: The measurement is for the A...h, or the measuring device/only displays a single digit.. It's a shame about all the effort.

Here you can see what it looks like on an Audi B4 Cabrio.

Code:

VA: Nachlauf 20°: Links °23' | Rechts °59' | Soll -0°30' [+2°10'] +0°30' Spreizung 20°: Links 3°25' | Rechts 3°27' Spurdifferenzwinkel: Links -1°55' | Rechts °37' | Soll -0°30' [-1°00'] +0°30' Sturz: Links -0°48' | Rechts -0°48' | Soll -0°30' [-0°45'] +0°30' Sturz max.Differenz: +0°00' | Soll [0°30'] Einzelspur: Links +0°08' | Rechts +0°07' | Soll -0°05' [+0°08'] +0°05' Gesamtspur: +0°14' | Soll -0°10' [+0°15'] +0°10' Eingeschlossener winkel 20°: Links 2°37' | Rechts 2°39'

If you have any other realistic measurement records, please provide them. We will then delete all the previous irrelevant content (topic) here.

Dear Uncle BM,
We had both become fixated on seemingly precise measurements that, upon closer inspection, were not actually meaningful , and we have now realized this.
That's why we're here, and our readers will once again see that not everything should be believed just because it comes from a computer.
That's why we're not deleting the "nonsense" from before.
The old measurement data I am referring to are 300,000 kilometers old. The only thing I am certain of is that my car has not been in a press that has compressed the front of the car by several centimeters.
Sure, here's the translation:

"Best regards," or "Sincerely,".
Herbert.

Quote:

The old measurement values I am referring to are 300,000 kilometers old.

The only thing I am certain of is that my car has not been in a press that has compressed the front of the car by several hundred millimeters. }

Okay, if I've understood everything correctly, this problem has been around for a long time.

However, the last measurement report already indicates your problem, even though the spread values are about 10° too low.

Therefore, I think it might be another approach to compare the existing protocols.