KE20/25 Rack and Pinion Conversion Guide. Driving well in my car!

[Rob KE25]

A number of people have asked me about my R and P conversion, so a brief thread on it.

A few things up front:

1. This is a guide of what I did, you take responsibility for if you choose to copy. (My disclaimer...)

2. I am not an engineer, not even a mechanic. But I did do a lot of thinking and research into steering geometry before embarking on this challenge/ adventure. Including chatting alot to a kind gentleman who built his race car from sheet metal, and has the fastest time in his class for his car around Baskerville raceway (56 seconds around Baskerville, with just 200hp.) Also borrowed a few ideas from Carroll Smiths great books (Drive to Win, Tune to Win and Engineer to Win). I was advised a number of years ago to buy these books ($110 for the 3). They have saved me much more than that in mistakes already!

3. The pics I took this morning are the best I have.....didn't take any while doing it.

4. I've busted my leg, so have time to finally do this thread. Hope I can work out how to add the pics again, abit computer illiterate. I'll try get it done this arvo.

5. Open to discussion. Thanks for the feedback and discussion others have given me in my KE25 build.

List of things needed:

KE 70 steering parts from the steering wheel to the rack ends.
AE 86 LCA's
2 x Split Lock Collars (Part number SC 220)- 32mm or 1 1/4 inch I think they were.
2 or 3 sheets of 3mm and 6mm metal, say 150mm square.
KE25 (or KE20) steering knuckles.

Expert stuff:
Welder (and a very good welder guy).
1 hour on a lathe.
1 hour on a Milling Machine.

[Rob KE25]

ATTACH][/ATTACH]Where I started:

Focussed on the Rack positioning. Three positions need to be considered:
1. Distance behind the crossmember
2. Height in relation to the LCA inner pivot points
3. Length of the bar

Edit: just realised something after writting this post up. I'm working a bit backwards here. The positioning is relative to the LCA inner pivot points. I'll fill you in on the LCA inner pivotpoints in the next post. So you need to do the next post first, ie before this one.


After heaps of 1/2 scale drawings, I came up with this position for the Rack:

1. Distance:
The Rack is 85mm behind the Centre of the LCA bushes. Another way of saying this is 85mm behind the centre of the crossmember, looking laterally. So I'm guessing roughly a 20mm gap between crossmember and the rack. Hope that makes sense.

NOTE: I had a AE86 crossmember in my garage, and these had the rack about 115mm behind, not 85mm. I chose 85mm for a few reasons, one being it suited what I wanted for ackerman better, and secondly the KE20 steering knuckles (108mm modified) are considerably shorter than AE86 (132.5mm) and AE86 p/s ones (115, 120mmish- can't remember exactly. Also, this way it doesn't put too much strain on the KE20 knuckles and rack ends when turning with lots of lock.

2. Height:
The Rack is 6mm ABOVE the LCA inner pivot point. Saying that another way is the centre of the rack ends are 6mm higher from the ground than the centre of the LCA inner pivot points.

NOTE: AE86 racks are 7mm higher. So very similar. In my 1/2 scale drawings, this measurement seemed to make the biggest difference to any bump steer, so really try get this to within 1mm either side of 6mm.

3. Length:
Did 2 attempts. first one I shortened the rack by how much closer the LCA inner pivot points are in my design, but it didn't work. A lot of bump steer, possibly more than a standard AE86. But it was toe in on compression, not toe out. Yuck.

Fitted a spare rack bar I had (standard length) and brilliant! No bumpsteer at all in compression and the in the first 25mm of droop. Then slight toe out with more droop. (this is with 40mm RCA's).




NOTE: Without the 40mm RCA's it would still have no bump steer in compression at all, nor at least for the first 65mm in droop. But this isn't much good for roll centres, and understeer would result in hard cornering (due to tire scrub, not bump steer obviously). Besides, a bit of extra ackerman on the inside wheel in turns never hurt... check out the competitive IP cars like the front line escorts. The theory behind that is too much for this thread.

OVERALL NOTE ABOUT MY RACK POSTION: i mounted my rack and LCA inner pivot points rather low (because of the tall 3sge motor). The LCA inner pivot points are 15mm below the standard position in the KE25 Crossmember. They are also a lot further out, but I'll get to that later.

Below you can see the split lock collars. I had them facing backwards, so I could get the rack out without having to lift the motor. You might notice my crossmember is not very high. Alot has been cut out of the hieght so I could get the motor low. Obviously we put a lot of metal in that crossmember to strengthen it again.

[Rob KE25]

I think the pics are working...

[Rob KE25]

This post considers the Position of the LCA inner pivot points.

There are two positions that need to be considered here.

1. The height of the LCA inner pivot points
2. The lateral postion (how far from the centre line).



1. The height of the LCA inner pivot points is not that crucial in that if it was 5mm higher or lower than standard it wouldn't make a drastic difference to the Roll Centre. But of course, if you went too high, the LCA's could fowl with the crossmember. And if that didn't happen, you have to remember that the rack does sit higher than the LCA inner pivot points. And you don't want the rack to be touching the motor.

What I did: I lowered the LCA inner pivot points to about 25mm from the bottom of the crossmember. Soley for the reason my motor is tall, and therefore hangs low. (see pics of previous post). My rack would have hit the sump if I kept it at a standard height.

2. Lateral position. I located the Lateral postion on a Milling Machine as it was hard to do accurately by hand. I also had them drill the hole on a 3 degree angle, as I knew I would be running about 3 degrees more castor. With the crossmember on the milling machine, (the back of the crossmember pointing up) we drilled the holes at 289mm from the centre line. This resulted with the holes on the front of the crossmember being about 286mm from the centre line due to the 3degrees. One reason I chose to do this was because a toyota guru friend had this done to his TA22 race car and is very successful. Just in the last few months he's won at Philip Island and other tracks around the country.

If I didn't do this, I would set it at 286 or 287mm and drill straight, as there is very little room between tires and guards.

NOTE: if you measure your rack, you will find that these measurements are alot further out than the standard pivot points. Infact, you may need to extend the crossmember with extra metal as I have on my crossmember. Check out the pics below.

[Rob KE25]

Next step:

STEERING KNUCKLES.

I modified the standard ones, didn't worry about the KE 70 ones.

Background
A bit of my car history here to give you a background why.

I have XT130 corona struts, with coilovers. Strut/Disc/Caliper setup is the Corona/Peugot/Hilux setup. This setup has worked well on the race track. Check Toymods on the net for how this is done.

Anyway, I did this setup for a few reasons. Spoke to an old racer from Melbourne as he used to race these KE25's years ago and is also a 3sg expert, having done rwd conversions over 18 years ago! (Think he still is on Speed week sometimes). He said he chose to use bigger corona struts (not KE25) because they are stronger and less likely to deform. Sure, he's talking racing with sticky rubber. Now a problem is that the KE25 knuckles bolt hole spacing is closer than the strut bolt holes. KE70 Knuckles would fit the bolt holes, but may have problems with disc clearance to the bottom of the LCA. Also, the KE70 /AE86 knuckles will give really annoying amounts of bump steer. So I had the KE25 ones modified.

What I did: (well, my mate Jonny Rochester on the lathe)

1. Welded in the rear bolt hole of each knuckle. Redrilled it (lathe) slightly further back to fit the strut and placed the metal locating collar back in.

2. Then on the lathe also redrilled the current 12mm taper into a 14mm (common toyota mitsubishi etc) size.

NOTE: I questioned this racing guy (you've probably worked out who he is) about the strength of redrilling the holes, and he said he never had one fail in 5 full years of competitive racing. I've checked mine out a few times, and all looks good. Truth is, I think the bolts and locating collars would be the weak points.

The finished product.


As mentioned, I use the corona strut pugeot disk and hilux 4pot caliper set up. I have 14x6 12positive offset superlites and 185/60 tires (adrenalin). I use 40mm RCA (roll centre adjusters, commonly used to give lowered cars acceptable amounts of bumpsteer. Got these from David Lawson, the guy on the AE86 drivers club. Whiteline ones are only 25mm high. T3 one are 30 or 35mm). With the measurements in the thread, It all JUST fits. 195/60 tires might scrape the TE27 flares. I use Notec ajustable strut tops (beware, 2 different spacing choices here) and have them set for maximum camber, which is between 1.9 and 2.1 degrees negative. Even without the adjustable strut tops, tires might scrape flares. Be informed that Adjustable Strut Tops on KE20/25 are not alot of use with standard diametre springs, as the strut towers are quite small and the springs would hit the strut tower. Also, I found that KE 70 LCA's are 3-4mm longer than AE86. So I swapped mine over with my mates AE86. This little bit helped the tires not to scrub the flares. Maximum castor I can use is 4.5 degrees before the tire hits the front of the wheel arch with lots of wheel lock.

Thanks to Aaron Martin from AZCUSTOM for his excellent work with the bits and pieces, such as the rack and pinion, roll cage, strut brace, Adjustable panhard rod etc. Top work!

Thanks Jonny Rochester for your input and work on building this car. It wouldn't have been going without you.

[Rob KE25]

Drawing of AE86 steering assembly with measurements
A pic of the measurements of the AE86 crossmember I had at my house with rack and LCA's fitted. Note that all the measurements are with the LCA's flat on the ground and fitted on the crossmember. So they may seem slightly longer (5mm or so) than fitted in a car. This is because the LCA's are on an angle (sloping down) in cars, unless lowered dramatically.


FYI KE70 LCA's are 4-5mm longer than AE86 ones(strange, this I used to think they were the same, but slight different design too). XT130 LCA's are 4-5mm longer again, so 8-10mm longer thanAE86. I measured them all next to each other. If you had really big flares, the XT130 ones may fit, giving about 2.6- 3 degrees camber, not the 2.0 I have in my setup.

[Rob KE25]

Steering collumn placement and wiring.

This was the easier part, not so much theory and stuff like bump steer etc. And didn't need to be precise. I spent an hour working out the wires and which ones needed to be connected. I also used the KE70 indicator, but left the KE25 light switch on the dash. This part took most of 4 hours, but doing it slowly and thoughtfully.
There are only 7 or 8 wires to really connect or extend. Just need a multimetre to work it out which ones short circuit when you turn the key, or turn left or right. I've had no training with electronics, and managed. An Auto Technition could tell you what to connect in less than an hour.

I moved the bracket that holds the KE25 steering collumn (cut it off the KE25 collumn) to under the dash so that the steering collumn splined end could reach the rack.(NOTE: these splines are of differing length between early and late KE70's). Then got it welded after dummy fitting it. I only used the small flange from the KE70 collumn (see bottom pic). I took the big bell shaped one off (2x 12mm bolts), and it actually sat where I wanted it. Marked where the standard KE25 dash braket needed to be welded. Took it out of the car and welded it. All up, the inside of the car work could be done by one person in a Saturday job.


NOTE: I changed to a deep dish wheel, as the steering collumn is slighly (30mm or so) further away from the driver than a standard wheel. With the wheel like this, it is much closer to me the driver than the earlier (Autotechnica) wheel.








The silver stuff is expensive sound deadening.. hard to bond to the bitumen like standard sound deadener. You can just see the small flange lip under the sound deadener. Just bolted this to the firewall, after drilling holes through the firewall.

[Rob KE25]

The final result is a very quick steering system with no bump steer (well, it has about 5% bump steer of the original AE86 in compression- that's a 95% improvement!). The turning circle is very tight. In fact that is one thing I need to fix: the inside wheel actually scrubs the inner skin before the rack runs out of travel. I'm thinking an easy solution is to put a fabicated spacer on the ends of the rack- the opposite of machining the rack ends for more travel as some drifters do. This would be good, because it's only 1-2 hours work, and it won't change any of the steering geometry.

The steering feels pretty well between a standard car and the gokarts I've been having fun in. Heaps of feel and excellent turn in.

Keen to hear other people's ideas thoughts and different approaches.

Cheers,

Rob.

[twentyEight]

Brilliant, brilliant, brilliant!

Very well written and thought out. +Rep!

[oldcorollas]

very good write up and great work ont he conversion

is steering fast enough to not need a quickrack (quaife)?

do you find the rack lighter in steering than the original? (even with the castor and camber)

[Rob KE25]

very good write up and great work ont he conversion

is steering fast enough to not need a quickrack (quaife)?



do you find the rack lighter in steering than the original? (even with the castor and camber)

Thanks twenty Eight and oldcorollas for your words.

About the steering being quick enough or not:

Yes. Just right for me. On the tight twisty roads near my place (Tassie is full of them) I need not let go of the wheel to change hands position. That's even 2nd gear corners at 3500rpm say 50km). I worked out that with a standard KE70 rack and standard KE 70 knuckles, at full compression, the bump steer means that you have to turn the wheel about 6-7cm MORE at the top- just to get rid of the bump steer! So that's something I don't need to do. Sure, obviously we don't normally get to full bump though...

I think the intial turn in felt too quick when I first had it, as it has no bump steer. Some other drivers found it hard to drive spirited too. Takes some time to learn not to compensate. After a week of driving, I wouldn't have it any other way. About half way between a normal car and a gokart. Any quicker and it would be too twitchy at high speeds though. Now I hate driving normal cars through tight bends.


Light or heavy feel:

No. When I had the original steering it was too light for me. I like heavier steering, and after an hour and a half of spirited driving on the curvy roads 60 -110 kmh, I'll not be feeling it. (15 laps at the racetrack and I'll feel it). Drove 2 and a 1/2 hours on the highway, and it was OK, probably a bit too direct though. (That's the compromise though.. made it for corners. My petite wife (51kg) finds it a bit to "darty and tiring" on a normal A to B drive, but can reverse park it. That said, when in the mood, she'll ask to drive it when she feels like having fun.

Thinking about it... It's definately heavier than the current x police commodore I've driven, and the new FG xr6 falcon. And tighter in the steering than both of them. On the open road on the highway át 110kmh it would only be a bit heavier in feel, but a fair bit more 'direct.'

Hope that helps.

[twentyEight]

Thinking about it... It's definately heavier than the current x police commodore I've driven, and the new FG xr6 falcon.

They both have power steering though, so are going to be much lighter.

[AE25]

i like what you've done. great informative thread
but anyone noticed a potential crash waiting to happen? ie this


i believe there is a reason toyota clamps the steering housing to the crossmember.
i should point out that the rack tube you have securely clamped to the crossmember, can be knocked out of the alloy steering housing. (This was brought up in another toymods rack conversion thread, to shorten the inner end of tube rather than the outer end).
if this happens, the whole steering system will be able to slide, side to side thru the rack tube dependant on the position of the pinion on the rack.
full lock left will have no sideways travel as it'll pull the rack end hard up against the tube, full lock right will allow the wheels to turn full lock in both directions with no control by the driver, because the pinion shaft can swing off the uni's once the steering housing is free of the rack tube.
in theory, the only force that will knock the steering housing off the tube, is a blow to the front wheels that would force the wheels to turn left. or if a wheel got stuck in a rut or somehow could not turn and you forced the steering wheel to turn. or maybe just over time, the push pull nature of the rack against the housing loosens the press fit of the tube. likelyness is probably slim.. but the chance of it happening is no longer zero.

[Hurricane]

nice write up, im about to inbark on r&p on my ke26 soon.

dan.