There you go young man. That’s an estimate for the original E-type front suspension, fairly crude as it’s based on points measured off a screenshot with a Mickey Mouse ruler - should be considered more directional than precise data. I can only do analysis in the cross-car plane because I don’t have fore-aft locations, nor do I have any bush rate data. With those caveats, I’ve plotted variation for three important parameters as the attachment points for upper and lower A-arms (UCA and LCA) are moved up and down 10mm at the chassis end:
roll centre height (green lines - affects load transfer in a corner. Units are mm on right axis). Lower arm looks to have a stronger effect (higher sensitivity) than upper, but in opposite directions, so you could balance effects by working one against the other.
ride camber (blue lines - change of camber with vertical suspension travel. Strictly speaking for cornering study it should be roll camber, I used ride camber because it’s easier to check on a physical car if anyone wants to validate. Roll and ride values will be very similar. Units are deg per metre travel to make whole numbers, left axis). Sensitivity very similar for upper and lower.
Ride steer (red / orange lines - change of left wheel steer angle with vertical suspension travel, also known as bump steer. Comments as for ride camber). Interesting that the sensitivity to upper arm is higher than to lower arm, possibly because the steering rack is relatively high.
I hope there’s some interest there, good luck untangling it!
C
Here’s the same process but control arm attachments moved in and out 10mm. Obviously less sensitive than vertical changes, which could be good or bad depending what you want to do.
Sorry, the charts are a bit congested. Zero on the horizontal axis (“0 lateral”) is the nominal position of the control arm pivot as I measured it on the picture, vertical or horizontal depending which chart. Then -10 and +10 on that axis are the movements away from nominal - that’s a vertical change on the chart “sensitivity in z direction”, and a horizontal change on the chart “sensitivity in y direction”. The dimensions are just relative to each other, not directly related to control arm angle.
The vertical axes are the parameters, so for example on z sensitivity chart - the red dotted line shows ride steer is 11.8 deg/m at nominal condition*. If the upper arm (UCA) pickup point is raised 10mm it goes up to 32deg/m. With UCA pickup lowered 10mm from nominal it goes negative, -7.8 deg/m.
As a side comment, the positive values for ride steer denote toe-in as the wheel rises in bump. That’s unusual because it tends to make the car feel twitchy, perhaps to an uncomfortable level at high speed. A small amount of negative ride steer (toe out in bump) is more common because it gives a reassuring feel as you lean on it through corner entry and get a consistent message from the front wheels. It’s likely to be faster because of the improved precision and driver confidence.
No, that conversation won’t be necessary. We’ll be tuning the dampers on the road, they’ll be consistent in any likely use. Car dampers aren’t usually affected by ambient temperatures except after standing in very cold conditions when they can feel a bit stiff for a mile or two, until the oil warms up and viscosity drops to a normal level. (Ready to be corrected on that by a Minnesota resident).
I don’t expect to encounter internal temperature problems as I won’t be valving the dampers very aggressively. I’ll see it first on the dyno and adjust the setting accordingly. I didn’t actually plan to use exotic mono tube dampers, I was quite happy to use conventional twin tube like the originals. Jaguar, Lotus and others get excellent results with careful valve tuning. I was pushed down the mono tube route because I couldn’t find a twin tube that I could take apart to adjust the valve disc stacks, or change the piston, That’s where the real power and magic of the damper lies, more than the adjuster screw which typically acts on the low speed bleed system.
You’re very polite, Terry. I showed the charts to my wife, who has a very good scientific brain, and she gave me that look that wives do. No words, just The Look. Apparently there are levels of chart-geek,and I’m pretty far along. I’ll try to do it like a normal person.
I wanted to know where is the biggest and smallest bang for the buck in moving pivots control arm pivots, and how they interact. I pulled out bump steer, bump camber and roll centre height as major items of interest. Looking just at bump camber (called Ride Camber by serious engineers), I ran a few specs. The following relate to the blue lines in the charts I posted yesterday. Baseline run (#7037), standard E-type according to my measurements from a screenshot. This shows change of front wheel camber angle as the suspension moves up and down. The slope at zero travel (nominal suspension position) is recorded as -12.9 degrees/metre - that single number obviously doesn’t tell the whole story because the line is curved
Next I lowered the upper arm pivots, which seems to be a popular change. I moved it 10mm, #9483 shows the camber change is stronger, the slope at centre is -21.4 deg/m. The camber at 50mm bump travel is about -1.4 degrees, compared with -1.1 degrees for the standard car (7037) - suggests more cornering power for the outside wheel in a turn. The inside wheel (going to rebound) has more positive camber with 9483, also suggesting more cornering power (maybe). Finally the lower pivot in 9483 is closer to a straight line, which is likely to improve driver confidence as the wheel will be flapping around less (yes, that is a valid technical term) on an undulating road.
All the above tends to confirm why that is a popular mod - noting I only modelled 10mm change, the optimum is probably a different term.
This is what happens if you raise the upper control arm pivot 10mm (9482). The steep change of curvature doesn’t look appealing, pretty much the opposite of 9483.
Interesting that the single figure for the slope doesn’t tell the whole story, you really need to see the curve to get the whole picture
Then I did the same for the lower control arm pivots, #9484 up 10, #9485 down 10. Looks like the opposite trend to upper arm, no surprise but I wanted to understand if one had a stronger effect than the other. Unfortunately RACE software won’t currently allow me to combine plots from different files so it’s a manual operation
Thanks Clive. The upper control arm is typically dropped about 23 - 24 mm and it really is transformative to front end grip and control. The car goes from heavy understeer to neutral or slight oversteer.
One of the things the car needs is better anti dive. Have you looked at the setup on the V12 E Type? I’m not familiar with the compromises that entails as to lateral grip, but the V12’s handle better than the 6’s.
Wait there’s more, for the truly obsessed. Combining the plots confirms what you could guess from the individual plots.
I’ll run with -23/24 and include in the plot, will also add our project car
I haven’t looked at the V12, will certainly do so if I can get data for the pickup points. I only have a cross car view for the early car. We have a small amount of anti dive on our project car. I hadn’t heard that the V12 has better handling than the 6’s, thought it would be major sacrilege to utter that in some quarters!
Here’s the combination plot of the UCA and LCA moved up and down 10mm. Baseline (standard car) is the red line. Solid lines are LCA moves, dash lines UCA moves. Darker line is a down move, light line is up move. No surprises, confirms the single plots - more negative camber in bump (more cornering power) requires UCA down or LCA up. It also confirms Terry’s point, the UCA is more effective than the LCA. So a sensible person would leave it there and do something useful. Me? I wanted to see how that move interacted with other effects like roll centre heights and bump steer. That’s why I made that horribly complex chart, so I could see it all in one place.
Here’s how Crazy Chart came about - staying with bump camber, I took the value of the slope printed at the top of each of the single charts (-12.9, -21.4, -3.9 etc.) and plotted that against the control arm pivot positions (+10, 0, -10). The blue lines show that for bump camber, you can see the UCA (dotted) line slopes slightly more than the solid LCA (other lines removed for clarity). So far nothing you couldn’t deduce by looking at it, but when you add back the lines for roll centre height and bump steer (original chart) you can judge their relative strength, and how you might change two dimensions to balance one effect against another. I’m sorry, that kind of stuff puts a spring in my step. My wife has learned to live with it.
Well then, Terry…here’s the camber change for your upper arm lowered 23mm. Red line is standard car (7037), light blue line is upper arm down 23mm (9514). As you say, it’s a pretty significant change in the camber curve - lots more negative camber in the bump condition which directionally has to be good for cornering power at the outer wheel. Also more positive camber on the inner, should be likewise in many situations. My concern about this change is that it adds a lot of bump steer (toe out in bump = roll understeer). It also raises the roll centre from 145 to 177mm, more on that later. A small amount of roll understeer is generally considered desirable but this is more than a small amount, as you can see in the steer chart. It could be that people like it by contrast with the original, which has a lot of the opposite - I believe that would feel fidgety and get quite tiring. I looked at reducing the bump steer, dropping the steering rack 15mm (9515), green line. Minimal effect on camber, good reduction in bump steer. But…it pushes the roll centre a little higher, now to 181. I don’t know if that’s a problem but it would need to be checked relative to the rear.
Next step was to pull the roll centre down a little by lowering the lower arm pivots 5mm. That’s the orange line (9516), it pulls the roll centre back down to 167 with virtually no effect on camber or steer. I can’t say how useful or necessary that would be. The typical result of raising the front roll centre would be lower roll angle and more load transfer at that end in a corner - the nett result in terms of under or over steer depends on how the curves interact with each other and the rear suspension, difficult to predict anything more specific without knowing a lot more about the car. That’s the point where you need to get in and drive it!
A note I should add about the steer and camber data - they all take 0 degrees as nominal condition so you need to add your static setting to get the actual angles in a turn.
The dotted black lines are our project car. The camber is less aggressive than the “down 23mm” car on the bump side, similar on the rebound. We’ll see how that works. On the bump steer we have a very small amount of toe-out on bump (roll understeer). The line is straighter than the original car because we spec’d a shorter steering rack with the inner ball joints optimised to the control arm positions. We’ll hope it was worth the $/mm ratio!
I use a modified steering arm, raising the track rod end. This was done by measuring actual bump steer at different positions until the bump steer was gone. I can’t comment on roll steer as I’ve never measured it. Both my race car, and my autocross car were/are very stable under all conditions. On both I used a softer front bar to induce greater roll to induce greater negative camber for increased front grip. I know it works as times are better softer.
Seems we have pretty good agreement on direction, but you got there twenty years ahead of me? Good work.
Your softer front bar makes sense, was it the same shape as the original? The layout of the original looks quite “pure”.
No reason to measure roll steer as such, bump steer is easier and they match well.
I had a look at anti-dive, tilting lower arm slightly up and upper arm slightly down at the rear as we’ve done with our project car. The result was similar - a small amount of anti-dive with no negative effect on other parameters. With that as a baseline I’ll see what we can do on the upper arm only. I’ll also look at moving the upper arm back to see if we can gain some castor, maybe there’s a package for camber gain, castor and anti-dive within the upper arm.
My data says the standard car has very low static castor (1 degree?), is that correct? That is unusual, I’d expect it to show up as loss of steering precision / stability feel under braking - is that apparent? I surmise they kept castor low to keep steering efforts low without power assist - does anyone know the real story on that? I’d like to raise it to about 5 degrees, which would also give more negative camber in a corner as a bonus.
More later.
Hi Clive. Well truth be told Jaguar used the dropped UCA in the days they were racing lightweights. They also used toe in on the rear suspension to add stability at speed, as did I. I read about these in one of the E Type books I have. Static set up recommended by Jaguar was 1/4 - 1/2 positive camber, castor 2 plus minus 1/2 degree, swivel inclination 4 degrees. Typically you cannot get static negative camber in the car without mods to the parts.
I ran a lot of castor in my race car - at least 5 degrees. I don’t know why Jaguar set it at what they did. Most, if not all the E Types I raced against had 0 castor - I recall that they all thought (on principle) that it added nothing. I’m a great follower of Carroll Smith - his book Tune To Win is the pre ground effects racer’s bible. He recommends castor. You can add additional castor by using different numbers of shims on the front and rear brackets.
As best I recall I used .92 torsion bars and a stock sway bar in the racer. In the autocross car I used a custom made hollow bar that was 80% of the stock bar, with stock springs and original torsion bars. Both bars were stock in shape.
