Bump Steer remediation

Rupunzell

Bernice Loui
How does this "modification" alter bump steer? What else might this modification alter?

This is VERY high risk weld up modification due to SO many things that can and does go wrong with welding... which has been stated to at nauseam the most difficult method producing a reliable joint. As for regulations, practicing proper mechanical design, engineering and fabrication methods should always be applied regardless of "regulations"..


Bernice
 

fiatfactory

Steve Cecchele
Are you aware of a source for the FIA homologated upright?

As far as I know, that item is actually the production part for the Fiat 130TC abarth. I know they use a different upright / hub / cv/ bearing and retainer to the regular 105/125 ritmo (which is actually a 138 chassis) as I know for certain the 105/125 upright is the same as the x19 five speed front upright.

I know the 130TC uses a bearing the same as x19 five speed rear / lancia up front (so the bearing is larger again) and also the same lockring / bearing retainer as the lancia. (which you can see in the ritmo 75 homologation docs)

Because the ritmo (you call it a strada in the states) is actually a 138 chassis, loads of suspension parts interchange and swap. The semi big brake option for x19 is 105 ritmo / regata 100s (regata is a four door ritmo 138 chassis - ritmo is the hatch version) that uses a 257 solid disc... ritno 130 TC uses a 257 vented disc stock so all those bits would be a direct bolt on (save for thread pitch differences in the component parts).

To get rid of bump steer, well the drop spindle... I honestly haven't ever thought to plot a camber / bump steer curve for a change as you've made, but if you say it improves it I'll believe you as I'm not going to do the math etc myself.

I've seen it done other ways, which to me seem "more clever"... Bob Boig in his (usa championship winning) X19 used a Lancia beta steernig rack (made by ZF) where the central pivot / TRE is in the centre of the car.... the effective length is so long the bump curve must equate to minimal change over regular suspension travel.

SteveC
 

Dr.Jeff

True Classic
How does this "modification" alter bump steer? What else might this modification alter?
If I'm seeing it right - I believe you are correct, it won't do anything to improve bump steer. I'm assuming the idea is to also move the strut's mounting location on the other end of the upright (relocate the mounting tabs on the strut itself), so the spindle's centerline is moved up relative to the rest of the suspension. That would allow lowering the ride height without messing up the geometry of the suspension's main arms or effecting travel length. But the steering arm (on the upright) hasn't be relocated accordingly, so the tie rods will be displaced - causing the bump steer issue. Maybe a flipped tie rod mounting on the steering arm will bring it back to it's normal level?
 

Chris Taunton

Daily Driver
yeah, welll steering com[ponents and modifications to them here in Australia is a bit of a no-no... I 'spose it's related to insurance laws for public liability.

I've hit an upright hard enough to bend it in the area you've extended, and the cast iron bent, it didn't crack... the lower control arm (on a 128 coupe so the same as an X19) bent and deformed but didnt crack.

I'm not even sure you will need the drop spindles to tell you the truth... how low do you want your 128 to be?

If your already going X19 rear struts on the front, your already dropping the car a pretty huge amount...and loosing quite a lot of overall suspension travel by having a shorter strut shaft, you will also need to use the X19 bump stop assembly from a series 1 rear to get the required bump stop height, otherise with the sedan bump stop / top mounts you'll be riding on the bump stp rubbers.... the struts you'll be using also have quite a lot of camber built in compared to a 128, and a load of adjustment in the elongated upper hole.

Just be warned a 128 sedan front chassis is no where near as stiff in the chassis on the front of an X19... there is a lot of flex...more in an older body and they do tend to crack.

SteveC

Hey Steve,

Thanks for your concern. I understand why OZ takes this stuff so seriously. Every country is different and has different regulatory priorities. Ours has focused almost entirely on exhaust emissions. I do not know the best approach to politics or government, so I refuse to comment on the wisdom of other countries' regulations.

My friend. tested these under a 5 ton (10,000 lb) load. The only way I will bend or break one is by crashing. If I break one, I'll have bigger problems.

These spindles do not lower the ride height at all because the spring acts on the top of the spindle (the strut). The same modification on an 105/115 chassis Alfa would lower the ride height because the spring acts on the lower control arm. The hub centerline would be higher than stock in relation to the LCA.

My reason for going to the trouble to have these modified is to bring the camber gain curve closer to stock and raise the roll center at the lower ride height. In the PBS X1/9 race prep book, they accomplished this by raising the inner suspension pickup points in the body by the amount they shortened the struts. The SCCA rules did not allow the spindle to be modified, and did not allow larger than 13" wheels in G or F Production classes. Other classes allowed alternate spindles.

On a side note, the GTA knuckle risers were used because those cars ran 13" wheels because low profile race tires were not yet available. They needed 13" wheels to fit the wider tires. There was no room inside the smaller than original wheels for an extended lower ball joint mount.

So, this change is small. It allows the standard geometry with ride height 7/8" lower than stock height. The X1/9 rear springs and struts lower the 128 2". My geometry will be somewhat better because of this change. The ride height on my car is adjustable. I don't expect to set it lower than 2". It will depend on how the geometry works out.

I am aware of the weakness of the 128 unibody. I am looking at different ways that racers have braced it.

Thanks again for your help and expert advise.

Chris
 

Chris Taunton

Daily Driver
As far as I know, that item is actually the production part for the Fiat 130TC abarth. I know they use a different upright / hub / cv/ bearing and retainer to the regular 105/125 ritmo (which is actually a 138 chassis) as I know for certain the 105/125 upright is the same as the x19 five speed front upright.

I know the 130TC uses a bearing the same as x19 five speed rear / lancia up front (so the bearing is larger again) and also the same lockring / bearing retainer as the lancia. (which you can see in the ritmo 75 homologation docs)

Because the ritmo (you call it a strada in the states) is actually a 138 chassis, loads of suspension parts interchange and swap. The semi big brake option for x19 is 105 ritmo / regata 100s (regata is a four door ritmo 138 chassis - ritmo is the hatch version) that uses a 257 solid disc... ritno 130 TC uses a 257 vented disc stock so all those bits would be a direct bolt on (save for thread pitch differences in the component parts).

To get rid of bump steer, well the drop spindle... I honestly haven't ever thought to plot a camber / bump steer curve for a change as you've made, but if you say it improves it I'll believe you as I'm not going to do the math etc myself.

I've seen it done other ways, which to me seem "more clever"... Bob Boig in his (usa championship winning) X19 used a Lancia beta steernig rack (made by ZF) where the central pivot / TRE is in the centre of the car.... the effective length is so long the bump curve must equate to minimal change over regular suspension travel.

SteveC

I'll see if I can find a source for the 130TC uprights.

Bob's X1/9 would toe the outer wheel out (good) unless the LCA inner mount was lower than the outside. Most of the advise I've seen on the Xweb has centered on very stiff springs rates (500-800 lbs/in) to limit suspension movement and thereby limit bump steer. This is a valid approach for autocross. Make it stiff and dial in static camber and toe settings.

I am not trying to win an autocross national championship with Gus. I am not a national champion level driver! My springs are 220 lbs/in.

So, I am trying to fix the geometry in a way that makes sense for a dual purpose car. I will autocross it for fun in a local club and do some track days, but mostly I will drive it on mountain roads for fun and commute to work. I have a racecar for SCCA competition.

Chris
 

Chris Taunton

Daily Driver
Here is a couple of pictures of the completed uprights.
imagejpeg_0(2).jpg
imagejpeg_0(2).jpg
imagejpeg_1(1).jpg


Chris
 

Rupunzell

Bernice Loui
Single force load testing is NOT a good way to test/certify the weld up on these uprights are going to stay together over many hundreds of thousands cycles of loading and loading in many ways a suspension upright can and will be loaded. If there are voids and cracks inside the weld, a one time load of 10,000 pounds might not reveal this and the load test is not valid unless one knows precisely what the part was designed to be loaded with to produce the calculated/modeled deflection under that load in very specific ways.

What are the test results from applying said test force on this modified upright, where was the force applied and how did this upright change shape during loading?

Bump steer in the common racer industry definition is change in toe over suspension travel. It is commonly adjusted by altering the location where the steering attachment meets the upright. It is often measured in~ thousands to hundreds of an inch~ and the bump steer curve can be tuned to achieve a specific chassis dynamic to meet the needs of driver, track and etc..

Much of what is published in that PBS book regarding suspension is simply not ideal. It is very likely why the PBS x1/9 was NOT successful in ways it could have been. Lowering the car the way they did then applying soft springs (200 lb/in) with trying to gain suspension travel is not going to achieve good suspension geometry or tire behavior. Unless the suspension has been designed to specifically have both compliance with camber control and it's dynamic roll center well controlled which the VAST majority of production cars never do, trying to "modify" a production car suspension to achieve this is not gonna work. This is why using high spring rates for smooth tracks works. It limits suspension travel to where the suspension behaves reasonable to allow the tires a chance to function properly. Bigger stability bars or "anti-roll bars" alone often never achieve this and they can cause other problems as they are a trade off suspension tuning device. Dampers are another suspension tuning area that is often complex and affects the dynamics of the suspension and chassis lots.

As for that 200lb/in so often quoted from that PBS book, where did they get that number from and how? Stock springs are about 30% lower than that number.

Lowering the dynamic CG too much can put the CG below road level causing the tire to essentially dig and reducing grip. This is one of the many reasons why lower is not always better.

For a combo car, it might be FAR better to leave the stock 128 suspension geometry alone, Consider replacing the combo stability bar with A-arm configuration to what was used in the 128SL or exxe and spend the $ required to get a proper damper (like Koni 8610 or 8611 or similar) with adjustable housings for 2.5" springs and set-up, tune up the whole chassis-suspension as needed.


Bernice
 

Dr.Jeff

True Classic
These spindles do not lower the ride height at all because the spring acts on the top of the spindle (the strut).
But if they are combined with struts that have relocated lower mounting tabs (i.e. the spindles sit higher on the struts), then the ride height would be lower and the suspension geometry remain standard. However I see that you already have shorter struts ("The X1/9 rear springs and struts lower the 128 2".") so the lower tabs won't need to be modified.

Here is what I'm referring to, but they can't be moved up very much so this is limited:
rear-shock-absorber-set-of-2-pieces-fiat-x1-9-1300ccm.jpg
 

Rupunzell

Bernice Loui
Moving the upright bracket on the strut housing is noted in the PBS x1/9 prep book. This was done to lower the front while preserving suspension travel.. Except that alters more than just lower the chassis.

The PBS goal was to lower the CG overall by lowering the body around the suspension.. Except, the reality of chassis-suspension and all related dynamics is a LOT more complex than that.


Bernice
 

MikeHynes

True Classic
Bernice,
I have not reviewed the 1978-79 GCR lately, maybe PBS was limited with the mods they could have made to the front suspension? Maybe that's why they didn't optimize the front suspension? I know Boig had a different idea, and from what I've seen, a better one. His mods were much more extensive. I think his interpretation of the rules was much more "open" than PBS's? Perhaps the GCR had some significant changes? (I still can't figure out how he was allowed to use a Lancia rack though.)

Don't forget PBS did their mods - a long time ago. I can't tell you why they used such light springs, but I can tell you that they were using Carrara inserts and tires that are very much out of date today. Perhaps the light springs worked well with what was available at the time? I dono.

Things change so fast that what was used last year may be "out of date" this year. I believe the PBS book was published in 1979 - 41 years ago :eek:

Yea, probably out of date by today's standards.
 

Rupunzell

Bernice Loui
SO much has changed from the time PBS wrote that book, yet SO many continue to use it as the "bible" for modifications.

It would be great to know who were the ones at PBS that came up with those alterations, why and what they think of those alterations today.

Today, there are far more and better performance items available that could never have been back then. IMO, wise would be to take advantage of the current performance tech, fully understand what is need for a given goal and go with that. Follow the book is easy default but might not produce the expected results or meet the desired goals.

Example being just for the exxe, number of modified cars with Youtube videos today with uber performance based on an exxe... as if more and more folks are discovering just how good that initial design was then applying what was denied decades ago when the exxe first appeared. Other than the Dallara x1/9 & Abarth rally cars and one road car (not Stradale), back then, most were IMO never modified properly to what was possible. Similar could be applied to the 128, which appears to have happened with the Yugo variant in more recent times.


Bernice
 

Chris Taunton

Daily Driver
Single force load testing is NOT a good way to test/certify the weld up on these uprights are going to stay together over many hundreds of thousands cycles of loading and loading in many ways a suspension upright can and will be loaded. If there are voids and cracks inside the weld, a one time load of 10,000 pounds might not reveal this and the load test is not valid unless one knows precisely what the part was designed to be loaded with to produce the calculated/modeled deflection under that load in very specific ways.

What are the test results from applying said test force on this modified upright, where was the force applied and how did this upright change shape during loading?

Bump steer in the common racer industry definition is change in toe over suspension travel. It is commonly adjusted by altering the location where the steering attachment meets the upright. It is often measured in~ thousands to hundreds of an inch~ and the bump steer curve can be tuned to achieve a specific chassis dynamic to meet the needs of driver, track and etc..

Much of what is published in that PBS book regarding suspension is simply not ideal. It is very likely why the PBS x1/9 was NOT successful in ways it could have been. Lowering the car the way they did then applying soft springs (200 lb/in) with trying to gain suspension travel is not going to achieve good suspension geometry or tire behavior. Unless the suspension has been designed to specifically have both compliance with camber control and it's dynamic roll center well controlled which the VAST majority of production cars never do, trying to "modify" a production car suspension to achieve this is not gonna work. This is why using high spring rates for smooth tracks works. It limits suspension travel to where the suspension behaves reasonable to allow the tires a chance to function properly. Bigger stability bars or "anti-roll bars" alone often never achieve this and they can cause other problems as they are a trade off suspension tuning device. Dampers are another suspension tuning area that is often complex and affects the dynamics of the suspension and chassis lots.

As for that 200lb/in so often quoted from that PBS book, where did they get that number from and how? Stock springs are about 30% lower than that number.

Lowering the dynamic CG too much can put the CG below road level causing the tire to essentially dig and reducing grip. This is one of the many reasons why lower is not always better.

For a combo car, it might be FAR better to leave the stock 128 suspension geometry alone, Consider replacing the combo stability bar with A-arm configuration to what was used in the 128SL or exxe and spend the $ required to get a proper damper (like Koni 8610 or 8611 or similar) with adjustable housings for 2.5" springs and set-up, tune up the whole chassis-suspension as needed.


Bernice

Hey Bernice,

Thanks for weighing in. What type of test would you recommend for a modified upright like this? How would you test it? My understanding is that the fabricator put a dial gauge on the part to measure deflection and applied the force with a hydraulic press with no measured deflection.

This modified part is not what this thread was initially about, but it certainly changes the bump steer. Steve C said that bump steer on a 128 with stock geometry is not a problem once the car is lowered 2". Is that your experience also?

The coil-over struts I got from Matt at MWB use Koni 8611 inserts. The spring rate is 220 lb/in.

Nothing I am planning to do with Gus will be irreversible. I do not want to modify the car to use the 128 SL/3P or X1/9 radius rods.. I will have to make do with the 128 wagon anti-roll bar. This is not a race car, but a street car that will see some track use. I want the driving experience to be rewarding, but I am not chasing a national championship.

My original question was about who had corrected the bump steer on their lowered 128 and how did they accomplish it.

Thanks in advance for any insight you can give me.

Chris
 

Chris Taunton

Daily Driver
But if they are combined with struts that have relocated lower mounting tabs (i.e. the spindles sit higher on the struts), then the ride height would be lower and the suspension geometry remain standard. However I see that you already have shorter struts ("The X1/9 rear springs and struts lower the 128 2".") so the lower tabs won't need to be modified.

Here is what I'm referring to, but they can't be moved up very much so this is limited:
View attachment 35718

Hey Dr. Jeff,

The distance from the spring seat to the spindle matters for ride height. This is what you adjust with coil-overs. The distance from the spindle to the LCA mount does not effect ride height.

Chris
 

Chris Taunton

Daily Driver
Bernice,
I have not reviewed the 1978-79 GCR lately, maybe PBS was limited with the mods they could have made to the front suspension? Maybe that's why they didn't optimize the front suspension? I know Boig had a different idea, and from what I've seen, a better one. His mods were much more extensive. I think his interpretation of the rules was much more "open" than PBS's? Perhaps the GCR had some significant changes? (I still can't figure out how he was allowed to use a Lancia rack though.)

Don't forget PBS did their mods - a long time ago. I can't tell you why they used such light springs, but I can tell you that they were using Carrara inserts and tires that are very much out of date today. Perhaps the light springs worked well with what was available at the time? I dono.

Things change so fast that what was used last year may be "out of date" this year. I believe the PBS book was published in 1979 - 41 years ago :eek:

Yea, probably out of date by today's standards.

Mike,

You are absolutely correct. Racers develop car to a set of rules. The rules are always evolving. Things that were a beneficial compromise one year can be a bad choice the next.

What Bernice said about production car suspension geometry being far from ideal is correct. SCCA production classes allowed limited modifications. PBS went to extraordinary lengths to raise the roll centers within the rules. When I first read the book in 1979, it put me off Production class racing. I was not prepared to do that level of modification. Instead, I raced in Improved Touring with a much more restrictive rule set and a much lower required budget.

Chris
 

Chris Taunton

Daily Driver
How does this "modification" alter bump steer? What else might this modification alter?

This is VERY high risk weld up modification due to SO many things that can and does go wrong with welding... which has been stated to at nauseam the most difficult method producing a reliable joint. As for regulations, practicing proper mechanical design, engineering and fabrication methods should always be applied regardless of "regulations"..


Bernice

Bernice,

It sounds like your resources far exceed mine. What would you advise to fabricate an alternate 128 upright on a budget? Would you hire a mechanical engineer to design the parts in CAD and have them CNC machined? Have you been involved in a similar project? If so, what was the total cost?

I admit, mine is a grass roots approach. I had a friend with years of experience make them. What does it cost to do it right?

Thanks,

Chris
 

Chris Taunton

Daily Driver
Bernice,

Sorry. I forgot to answer this:

"How does this "modification" alter bump steer? What else might this modification alter?"

It alters bump steer because the relationship of the angles between the lower control arms and tie rods as the suspension moves in bump. It also changes the camber gain curve and front the roll center.

Chris
 

Chris Taunton

Daily Driver
If I'm seeing it right - I believe you are correct, it won't do anything to improve bump steer. I'm assuming the idea is to also move the strut's mounting location on the other end of the upright (relocate the mounting tabs on the strut itself), so the spindle's centerline is moved up relative to the rest of the suspension. That would allow lowering the ride height without messing up the geometry of the suspension's main arms or effecting travel length. But the steering arm (on the upright) hasn't be relocated accordingly, so the tie rods will be displaced - causing the bump steer issue. Maybe a flipped tie rod mounting on the steering arm will bring it back to it's normal level?

Absolutely correct. The ride height will be set by adjusting the spring seat on the coil overs.

Chris
 

Dr.Jeff

True Classic
The distance from the spring seat to the spindle matters for ride height.
Ya, moving the strut body relative to the spindle changes the location of the lower spring seat accordingly. This is referring mainly to a non-adjustable strut, but the concept is the same. By moving the lower mounting tabs on the strut body up, you are effectively moving the lower spring seat down compared to the location of the spindle.
 

Chris Taunton

Daily Driver
Chris, I'd like to see what you are making for drop spindles.

My old VW's get lowered significantly, throwing off the bump steer. They have a similar suspension and steering layout to the old Fiats. The easiest approach I've found is to "flip" the tie rods and mount the outer tie rod ends from below the steering knuckles instead from above. The ends have a tapered mounting bolt so when they are flipped the taper is in the wrong direction. The solution is to drill out the hole in the steering knuckle and install a tapered sleeve insert from the bottom. Then the stock tie rod end fits correctly. This is what the inserts look like:
View attachment 35520

They also offer heim joint tie rods for that vehicle. It wouldn't be too difficult to make some. You can get universal kits to help:

View attachment 35518View attachment 35519

The amount of tie rod drop needed depends on how far the car is lowered. But the rule of thumb that I've always heard is to make the tie rods parallel with the lower suspension arms.

For really low drops, they also add an extension to the lower ball joint to raise the strut relative to the suspension arms. That reduces the change in the geometry.

Dr. Jeff,

Do you know the source for the inserts you pictured in this response?

Thanks,

Chris
 

Chris Taunton

Daily Driver
thumbnail_20200830_171132.jpg
thumbnail_20200830_171301.jpg

thumbnail_20200830_171016.jpg


Here are some "before" pictures. I took these pictures before installing the coilovers from Matt at MWB. When I took these pictures, I didn't know that the previous owner had modified the front springs. It turns out the left side (pictured) was 1/2" lower that the right. The top coil on both springs had been heated with a torch. You can see in the picture that after 1" of compression, camber will start to move in a positive direction. Measured camber before removing this setup was -1 degree on both sides.

Chris
 
Top