Self alignment!
 

Amazing what fishing line can do for aligning a car. Anyone every try it? Mine was off and bugging me. Took me two sessions to dial it in just right but now its spot on.

For those that never did it. Its simple. Block the car up on all four but on the suspension. You can do it on the ground though, just a little harder. Anyway, the basic idea is to wrap all four wheels with a string then get at least one front tire touching the line evenly. For example, if the driver front tire is turned to go left the string would only be touching the outer front of the tire and not make contact with the outer back. Dialing it in can be a pain if you do not find a way to lock down the steering wheel. If you do, then the method will take only a few minutes. The only thing that may be harder to adjust will be the wheel angle adjusted by the LCA. Thats a little more work.

Only odd thing was the old tie rod set was super easy to adjust. Once the lock clamps were open it turned very smooth and easy. The new one (newer, about a year old) was a complete pain. Go figure.

Wow. Thank you for bringing this to my attention. I have never heard of this method, but sounds great.

I replaced an inner tie rod end on my MR2 last weekend and thought I kept the alignment as it was set, but it seems too twitchy, so I think I don't have enough toe-in. I will use your method to check it out.

Here's something you DIY's might want to read concerning alignment.

http://mercedes-diy-alignment.blogspot.com/

That blog is the other method I read ALL over the place. The string method seems much less common and quicker.

I will probably still get a professional alignment but after I replace the uppers and put in new lower ball joints. But the pull I had is gone and gas milage is up. So its better than before.

At the very least, its a good way to get things close so one does not destroy tires on the way to getting an alignment after some major DIY work.

A quarter turn of the tire rod can almost be a full degree. So the slightest adjustment can make a big diff.

if you can no lock the steering wheel, then keep in mind any change (distance to string) is basically doubled. So if you check and you find that one wheel is not touching the string on the back outer side, dont adjust until it does touch. Adjust half way then play with the steering wheel to level it again then check. Otherwise you will end up with a steering wheel in a bad tilt.

For simple toe adjustments yes this works. But camber is a whole nother story.

very nice indeed!
 

Is the toe-in set keeping the wheel spreader factor in mind? I assume one has to set more toe- in to compensate for a lack of spreader assist.

When I replaced a tie rod on our (now retired not long after) '83 SD.....I used a tape measure method I invented myself. :D Seems to have worked.....since it tracked perfectly straight and the tires were not wearing incorrectly. :D

I've used a similar version of the DIY reference for years and had good results.

I'm not sure about the string...

Unless the track width of both front and rear axles are identical, you introduce error. Specifically, most cars have a wider front track than rear track dimension. This induces a toe out error. Rear wheel drive cars require a toe in, not toe out.

Secondly, if I'm reading the method correctly, you cannot measure toe-in with the string method, because the line will continue to grip the front of the tire as you transition from toe out to toe in.

I'm not even going to go into the spreader bar..... You just have to account for tolerance stack up.

Jim

And caster too!

I was surprised how much the front wheels move when you use a (DIY) spreader bar with the wheels on slip plates.

I've found that a Vice-Grip quick-grip turned backward and wood with a hole drilled in it will work for a substitute spreader bar.

What would be the effect of doing this with wheels on the ground?

Even if the car gets lifted up a bit to help ease the tie rod adjustments, no big deal.

Then one can do both sides a slight turn inward to induce toe in at the end... That should be measurable to verify...