GTM Owners Club Forum
http://www.gtmdrivers.com/forum/

Libra Steering Geometry
http://www.gtmdrivers.com/forum/libra-steering-geometry-t4052.html
Page 1 of 3

Author:  Ferg (861) [ Fri Jul 06, 2018 3:18 pm ]
Post subject:  Libra Steering Geometry

Odd question perhaps.
I'm giving my Libra a long awaited refresh. (My wife has used it for commuting for 4 years.) The only thing about the car I have ever found annoying is the marginal self-centring and low speed tyre scrub on full lock.
I seem to recall there being some dispute from Bryn as to the steering arms and maybe fitting them to the opposite sides of the car to the Build manual.
Any opinions on the low speed front castor behaviour?

Author:  sanzomat [ Fri Jul 06, 2018 5:17 pm ]
Post subject:  Re: Libra Steering Geometry

Can't comment on the Libra but I believe the Spyder is the same in this respect. I know exactly what you mean and mine does that too, but, and make what you will of this, it varies considerably depending on what wheels you use. I have three sets of wheels each with very slightly different offsets and also different tyres fitted so not sure whether its is the wheels or the tyres. I've also changed the tyres on some of the sets of rims and that made a difference too.

My (very unscientific) findings:
Original GTM supplied KN Racing 7.5 x 16 rims with 205/45 Goodyear Eagle F1 - distinct lack of low speed centering/Castor feel and tyre scrub on full lock very evident/noisy
KN Racing rims with 205 45 Yoko AD08R, Similar self centring issue but no tyre scrub noise
MG TF 11 spokes (7x16) with well worn 195/45 Goodyear Eagle F1 - self centring felt much better but tyre scrub felt worse and made a dreadful noise on full lock to the point I had to check if the tyres were rubbing on the arches - they weren't
MG TF 11 spokes with Avon ZZR 195/50 - feel great - nice self centring and no tyre scrub/noise on full lock. Much nicer turn in and steering response generally at all speeds. (note these tyres have a very stiff sidewall and can feel a bit firm and "track focussed though"
MGF 6 square spoke with 215/40 Barum Bavarious (ditch finder specials). I was given a set of four of these wheels for free from an MGF a friend was breaking and there is too much tread on them to throw away so I use these for general road use and as track day wets - to be fair they are actually pretty good. These do have the self centring issue but not as bad as with the KNs

So - rather than changing geometry or steering arms, try some 11 spokes or something with a similar offset to cure the self centring and try different tyres to cure the scrub?

Just my two penneth.

Author:  Spider [ Sat Jul 07, 2018 9:08 am ]
Post subject:  Re: Libra Steering Geometry

If you swap steering arms and the toe setting changes, you have also changed the Ackerman.
If you swap steering arms and the tie rod height changes, you have also changed the bump steer curve.

Low speed tyre scrub is almost entirely due to Ackerman not equal to 100%. On a Libra/Spyder with 2311mm WB and 1383mm track, 100% Ackerman will give 16.64 degrees of steer on the outside wheel when the inside wheel is steered 20 degrees. Not saying 100% Ackerman is the best setting for Libra/Spyder but any large variation from 100% will create low speed tyre scrub.

Low speed self centring is a product of caster angle and scrub radius. If either is zero, changing the other has no effect and self centring will be near zero. Increase either to increase self centring. Steering effort will increase. Reduce wheel offset (add spacers) to increase scrub radius.

At higher speeds (when cornering force becomes significant) increased "caster angle" (to increase "trail" or "caster") will have more effect on self centring.

Author:  Ferg (861) [ Mon Jul 09, 2018 7:28 pm ]
Post subject:  Re: Libra Steering Geometry

All interesting stuff.
I'm loathe to swap wheels, I run Compomotive MOs and think they are THE best wheel on a Libra looks wise. I tend to run Toyos on it, but recently changed to some 'unknown' make that was recommended to me and they have certainly improved the scrub. I remember (all them years ago... :-) )self-centering was marginal on SVA.

Author:  roger [ Thu Jul 19, 2018 6:21 pm ]
Post subject:  Re: Libra Steering Geometry

Hi Ferg

My car had really good self centre. I had MGF hairpin wheels 16". Not sure what the offset was. Maybe 35mm?

I have one thought. Are you sure you have your front wishbones the right way round? I remember quite a few cars didn't and they lacked some self centre. One person I know drove like that for many years before Duds noticed the error.

Author:  Spider [ Tue Aug 21, 2018 6:42 am ]
Post subject:  Re: Libra Steering Geometry

I checked my Spyder on the weekend and without doing any measurements, the tie rod ends are visibly outboard of a longitudinal plane through the upper and lower ball joints. This indicates there will be at least some toe-out in turns which is what you want.

Author:  Spider [ Tue Dec 03, 2019 3:31 am ]
Post subject:  Re: Libra Steering Geometry

Getting tired of the large turning circle and tyre scrubbing at parking speeds so I measured the steering geometry a little more accurately. Shock - horror! The Ackerman is close to zero, in fact it has measurable anti-ackerman (about 5mm toe-in at full lock). This is bad on just about every count. It seems the rack position combined with the tie rod end location is producing this effect. I want to make the steering a little quicker (has the metro rack with 3.6 turns) and reduce the turning circle, so the obvious solution is to shorten the steering arms (from 125mm to 105mm). This should reduce the turning circle by almost 2m and make the steering almost 20% quicker. I modelled the system in Solidworks (needs refining) and it seems like the best I can get is about 50% Ackerman with the tie rod end as close to the brake disc as I dare (about 8mm).

I will post more later. Meanwhile, does anyone have a steering rack out of the car they can measure? I need to know how much travel it has lock to lock and length of the rack between the inner spherical joints. These numbers are about 100mm and 800mm.

Author:  sanzomat [ Tue Dec 03, 2019 8:35 am ]
Post subject:  Re: Libra Steering Geometry

I have the metro 3.6 turn one that came off when I fitted the TF 2.7 turn one. It is buried in a pile of junk in the corner of my mother in law's garage so might be a few weeks before I get around to digging it out but happy to measure it when I do!

Author:  PhilTheGeek [ Tue Dec 17, 2019 10:43 pm ]
Post subject:  Re: Libra Steering Geometry

Spider wrote:
Getting tired of the large turning circle and tyre scrubbing at parking speeds so I measured the steering geometry a little more accurately. Shock - horror! The Ackerman is close to zero, in fact it has measurable anti-ackerman (about 5mm toe-in at full lock). This is bad on just about every count. It seems the rack position combined with the tie rod end location is producing this effect. I want to make the steering a little quicker (has the metro rack with 3.6 turns) and reduce the turning circle, so the obvious solution is to shorten the steering arms (from 125mm to 105mm). This should reduce the turning circle by almost 2m and make the steering almost 20% quicker. I modelled the system in Solidworks (needs refining) and it seems like the best I can get is about 50% Ackerman with the tie rod end as close to the brake disc as I dare (about 8mm).

I will post more later. Meanwhile, does anyone have a steering rack out of the car they can measure? I need to know how much travel it has lock to lock and length of the rack between the inner spherical joints. These numbers are about 100mm and 800mm.


I'll try and remember to take a wander to my parts store at the weekend and get some measurements for you from my old 3.6 rack. That's on the basis that you'll provide a STEP/IGES file of the replacement steering arms. Hopefully in the new year I'll have a 2.5 or 3 axis CNC mill so it should be easy to make them.

Author:  Spider [ Wed Dec 18, 2019 4:59 am ]
Post subject:  Re: Libra Steering Geometry

Sounds like a deal. I haven't designed them yet. I will need to pull one off the car first. I am aiming to make my steering 15-20% quicker (and reduce turning circle by the same) Are you wanting a similar change with your 2.7 turn rack?

Author:  PhilTheGeek [ Thu Dec 19, 2019 8:07 pm ]
Post subject:  Re: Libra Steering Geometry

I'm primarily after the reduced turning circle. The steering is more sensitive with the 2.7 rack but not too much. A 15-20% improvement would still be welcome. I'd recommend buying a 2.7 rack while they're still available. Unfortunately Mike Satur doesn't seem to list them any more as they were around £130. Rimmer Brothers currently has them for £180. https://rimmerbros.com/Item--i-QAB000250

Author:  Spider [ Mon Dec 23, 2019 5:41 am ]
Post subject:  Re: Libra Steering Geometry

Replacing it looks like a lot of work and I am reasonably happy with the response of the 3.6 rack. I figure 15-20% quicker (equivalent to a 2.9-3.0 turn rack) will be enough for me.

Author:  PhilTheGeek [ Tue Dec 24, 2019 4:11 pm ]
Post subject:  Re: Libra Steering Geometry

Lock to lock is 132mm +/- 1mm. Distance between inner joints is 800mm +/- 5mm.

Author:  Spider [ Wed Dec 25, 2019 10:49 am ]
Post subject:  Re: Libra Steering Geometry

Thanks Phil, much appreciated. It will be a few weeks before I get back onto this.

Author:  Spider [ Thu Jan 16, 2020 4:40 am ]
Post subject:  Re: Libra Steering Geometry

Update. Decided to remove a steering arm to measure up so I can design a modified version. Once I got in a bit closer, I realised the steering arms were on the wrong sides - DUH!!!

Took some measurements. The tie rod end is 6 mm offset so swapping the arms side to side moves the tie rod 12 mm - outwards in my case. One of the tie rods was screwed into the rack extenders was shortened to the max and so tight (to get toe adjustment no doubt), the thread was damaged. Finally got it all back together but ran out of time to set the toe and test drive. Will report back.

EDIT. Discovered an issue. The rear bolt holding the steering arm to the upright is not in the centre of the steering arm - to the point where it overlaps the edge of the machined flat surface on top of the upright. This means the edge of the steering arm is sitting on a small shoulder and doesn't pull down properly when tightened. A couple of minutes grinding the edge of the steering arm sorts it out. This overlap only occurs when the steering arms are fitted to the correct side of the car.

Page 1 of 3 All times are UTC [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/