Don't forget that when your hands are on the steering wheel there is a moment arm of 6 or 7 inches that reduces the force you feel. So your comparison isn't really valid.

As was stated in the above post, there is a 6 or 7 inch lever arm that you're dealing with when using the steering wheel. This will reduce the force that you feel. Imagine you're driving down the road at 80, 90, or maybe even 100 (come on, it's a sports car, 100mph isn't unheard of), and you hit a bump/pothole/expansion joint. You've got 50+ pounds of mass in each wheel, plus (what i'm assuming is a good assumption) sticky ass r-compound tires. The tires won't let go, so the force is absorbed by the sidewall, the mounts, and the gear itself.

Now on to the torsion bar. It is made to actuate the valve. The valve routes pressure to the correct side of the piston, turning your wheels. The t-bar in a power rack isn't designed to turn the car without any hydraulic pressure. When you have no pressure, the t-bar does ALL of the steering of your car (it's the only mechanical link between your steering wheel and your pinion). There are no stops to prevent the torsion bar from winding up excessively when there isn't any hydraulic pressure, and why should there be? If there were stops, you wouldn't be able to use all of the steering capability of the rack, and that wouldn't be very safe, now would it?

Finally, the pinion. Case hardening is only ~.005-.001" deep. The internal material microstructure hasn't been changed by the heat treatment process, which means it is just mild steel under that very, very thin top layer. All it will take is one good hard impact without any hydraulic pressure to absorb the energy to split through the brittle top layer. After that, the mild steel will fail very quickly. When you're running a manual rack, you must have a through hardened pinion to keep things safe, because there is no hydraulic pressure pushing the rack side to side, absorbing the steering force. The pinion is taking all of the steering force, and if you don't have a through hardened pinion, it will fail very quickly.

I will concede that a full conversion is safer than looping the lines, because the torsion bar is taken out of the equation by welding the input shaft to the pinion. The vast majority of the time, there is absolutely no warning for when the torsion bar will break.

The way that we test the t-bars is by fixing one end of the bar, and using a motor with a cam mounted to it's output shaft, adjust so the t-bar will load up to a certain angle. To test for failure, we check the torque required to reach that angle. Anything over 20% loss in torque is considered a failure. Most of the time, when the bar is heat treated, it will keep torque up until the very end, and then it will break. No warning. Nothing. Sometimes, the torque will fall off, but that is very rare.

Before I started at the company, they were doing max input torque testing on the input shaft. Just for S&G, they took a few guys from the steering group over to the test building, mounted a torque cell on a gear with a steering wheel on the other side of the torque cell, and went at it to see if any human could break it with just a steering wheel. They set the rack at one of it's stops, and twisted. All but one of them averaged between 100 and 115 newtons. One of the guys was like 6'6, 250 and a body builder. He put 155 newtons into the gear and managed to snap the input shaft. That is a non-welded, heat treated input shaft.

How they took the t-bar out of the equation is beyond me, but he snapped the input shaft, not the t-bar.

interesting

Anyone have the pictures from the first post?

Heres the pics again, sorry for the delays

more

Hi folks-

I've got a problem....I've gotten everything off according to the writeup, including the end cap on the long end after removing the wire retainer. However, when I try to remove the rack, there's a positive stop preventing me from pulling out of the carrier. Anyone know what this could be?

Thanks,
Chris

Figured it out, thanks much to Bacon.
c

Anytime,

Once you get it torn down to the point you pushing the rack out, you will have to hit it from the opposite end pretty hard to get the seal out. It may seem like you hitting it pretty hard but it needs a few good shots. Don’t worry you wont hurt it
:-)

One more thing :^). I know I saw this on a search, but I'm still looking 20 minutes later. Anyone got a pic of the weld that needs to be made to the quill?

Thanks,
Chris

i have a picture somewhere i can dig up. its pretty obvious where to make the weld once u have it all apart though

:-)

https://www.rx7club.com/attachment.p...0&d=1214424612

other side
https://www.rx7club.com/attachment.p...1&d=1214424984

bacon wins a crispy slice

Killer, that's what I was looking for. Have a great eve fellas.
c

Sorry to resurrect this, but I thought people would be interested. To remove the huge hex cover, the head of a 1/2x13 bolt (snagged a grade 8 one from lowe's) fit perfectly. You just put the head of the bolt in the hex, put two nuts on the bolt, and then crank on the nut closer to the bolt head. It took some serious muscle, but it gave before the bolt did. This way you don't have to weld a fakey SST.

^ THANK...YOU...SIR !!!

ill be giving that a go this weekend....

So I am curious as to the points that were mentioned by whitey.. It seems that research study was well executed and the data should have a good deal of validity. BUT I am curious of those who have been using the Maval rack, AND those that have done this conversion themselves....Has anyone experienced failure?? I would really like to hear from a few that track thier cars on a regular basis as this is going to provide the most stress to those components than just normal city driving.

Even if whitey is perfectly correct from an engineering standpoint, it doesn't seem to match with real world data. He suggests that failure is only a matter of time, but I've never heard of the failure of a single manually converted rack. He also said that a looped line setup is even worse and people do that all the time with no ill effects.

For me, if Damian runs this conversion on his track car and hasn't had any problems with his huge tires I'm sold.

https://www.rx7club.com/3rd-generation-specific-1993-2002-16/power-steering-removal-catching-up-writeups-part-3-a-448757/

So is there an actual Manual rack that is made without an "conversion"?

I know Mustangs and other various cars have companies that produce manual racks.

Yes but its pricey. Search on a thread from gmonsen. Atomic Rex in the UK makes it.

Yeah AR makes the $1000 billet one.

ouchhh 1000$

I completely agree. That is why I stated my question as I did. Not always do studies and valid test results always translate to 100% reality. It is near impossible to re-create a 100% accurate testing environment to translate into real-world usage.

I have never heard of one failing either. I know damian and several others on here that track there cars on a regular basis have done this and was just looking for some feedback and experience with it. This would also help document here real-world experience along with the engineering data.

I did a full manual conversion my self and so far everything is well. I have no complaints. Its obviously tough when parking but i don't regret ditching the power steering at all.

As far as it failing....I highly dough it. as long as you leave the black seals on each side of the rack to keep debris out of the quill pinion i can`t see where a problem can occur. worst case scenario the welds on the quill can brake but the shaft will still hold it in position.

WOuld it be possible to adapt a manual rack from an old 911 or a manual miata? I imagine the splines and the way the Tie rods attach are significantly different.

Bringin it back! Sorry, but im tired of looking at my looped rack and was wondering if there was anyway to get the pics back up and running?

Sorry to drag this one back but I am in the middle of doing this and found another really good write up with pics to link others too. The pics in this thread were removed so I thought this would help others. Plus it shows how to get that damn bearing off the quill.

http://forum.teamfc3s.org/showthread.php?t=54847 <--Big thanks to the FC guys!!

Back from the dead!

Doing this conversion right now, and wondering whether the black seals go back in?
I know the seal on the right gets removed, but what about the circled one?
http://www.flickr.com/photos/66462178@N04/7138053461/

No need,just will be more drag

some ideas for a clean conversion
 

I just did this conversion and wanted to share what I did for blocking off the oil lines instead of welding / crimping.

The large bolt is a M16 X 1.5mm thread pitch
The banjo bolt is a M12 X 1.5mm thread pitch

I got these from boltdepot.com. I had to cut them short but wasn't difficult.

The other 4 fittings
are M12 X 1.0mm thread pitch which was difficult to find. I ended up getting some Honda flywheel bolts. The ARP ones were pricey but you may be able to find something cheaper. Because the holes are in close proximity, I had to cut the lengths a little different so the heads wouldn't interfere.
https://cimg6.ibsrv.net/gimg/www.rx7...195521982d.jpg

https://cimg3.ibsrv.net/gimg/www.rx7...c705690899.jpg

Regarding the welding of the quill discussion, I wouldn't be at concerned about the heat input from welding weakening the quill.

For one, as earlier indicated, only the surface is hardened and that's for wear. Because the layer of hardened metal is very shallow compared the thickness of the quill, this hardening has no affect on the overall strength of the component.

Secondly, the properties of the weld metal will likely be superior to the steel used to make the quill shaft (unless the quill was drop forged or made of some exotic alloy, which doesn't make sense. Why use an expensive part when you can engineer the part to have sufficient strength with an economical alloy). If you are using SMAW (stick welding) then 60XX or 70XX electrodes should be more than sufficient.

If you truly are worried about heat input then I recommend using GMAW (mig welding) with a thin wire as this is a lower heat input process. L56 type wire has comparable strength to the electrodes mentioned above.

With either of these weld processes, you should get sufficient quench of the weld pool to achieve sufficient shear strength in the resultant weld and heat affected zone.

Granted, the part may no longer wear as well, but the affected area is under the seal bushing anyway and isn't rotating against anything.