Struggling to get some understeer out

[dsmithwc04]

Had a good time at the Majors race at Gingerman over the weekend but I'm getting a little frustrated trying to get at least some of the understeer out of the car in particular on turns 8 and 9.

We eventually got to full front winglet angle, full soft ARB, 2 rounds preload front.
On the rear we were one hole away from minimum wing wing angle, 3 rounds preload, and had 20mm rake which was the highest rake setting we had tried with the new tires up to that point. Messing with the rebound didn't have any profound effects but they do seam to be working from our last dyno.

This was our second weekend at gingerman with car on the new AR tires and we have tried adjusting the setup on every session we've been out. We've got some things where we like them but continue to struggle with high speed understeer. We were losing serious time to everybody in turns 8 & 9 where I could visually see others pull significant distance from us only to slowly reel some of that lost time back in other areas of the track. About the only reason we got on the podium Sunday was by running consistent laps and not having cooked front tires by the end of the race. Had a good restart and got past Keith, Hannah went off and was able to get a good run on Beck on turn 10 on the final lap and got the inside on turn 11. Race monitor had me ahead by only 2 tenths at the flag.

Anyways, I've tried different lines, braking techniques, throttle application, car setup changes... but continue to have horrible understeer in the higher speed bends. At this point I'm willing to try any direction to get us closer to competitive times. More than willing to share our data if it can help.

[reidhazelton]

Cracked frame or broken component somewhere? Seems like some pretty substantial changes to see little affect.

[dsmithwc04]

A broken frame crossed our minds. We plan on completely disassembling the car this offseason and inspecting everything but I'd sure like to do one more event this year before starting that process. Perhaps it's time to call it a season and start taking it apart now...

I noticed that we lost two bolts on the bellhousing to the engine though.

[S Lathrop]

I think I know the person who designed the FE. And if the car is even close to the car he designed for USF2000, the push is built into the car. Having said that, The basic geometry of the USF2000 is very good at the front and the push could be fixed easily if rules allowed. If the issues were fixed at the front, the rear would need some changes or else the car would be very loose. But Within the USF2000 rules we got the car balanced.

What are you tire temperatures? And how do the tire pressure change from when you start a session to when you pit?

Why did you stop at 20mm rake?

Why are you running preload on your springs? For my cars, I never start with preloaded springs. And I have almost never setup any car I have engineered with preloaded fronted springs. I have used preloaded springs in the rear of several cars but only like 100 to 300 lbs. of spring preload. And I have found that the preload number is very critical. I tune that number in 10 to 30 lbs. increments depending on the type of car. I have found that there is some critical load at which things start to change and the adjustment range is very narrow.

Did the car get better when you softened the front sway bar or did it not make any difference and you just left the bar in the softest position?

I also assume that you can not use bump stops or packers and that you have only one set of springs that you can use?

How is the car in the slow corners? And where is the push starting? Is it at corner entry, mid corner or corner exit?

[dsmithwc04]

I understand where your coming from Steve but the chassis is the way it is and all the other FE drivers have to deal with the understeer built into the chassis. I'm just trying to get what little bit can be tuned figured out.

What are you tire temperatures? We were measuring a 8-10 degree max difference across the face of the front and rear tires. By the time we measured them we were in the middle 140's but that was after a long straight and then another long drive to pit wall. And how do the tire pressure change from when you start a session to when you pit? We started with 12psi front / 13psi rear cold temps and they would build around 2 psi on all four corners. We tried higher pressures from advice of another driver but got serious graining and lowered the pressures back down and the graining went almost completely away.
Why did you stop at 20mm rake? We were running more rake with the hoosiers but according to the basic setup sheet from Primus (and some others) less rake was being used with the new AR's. We tried running 5mm rake, 12mm rake and 20 mm rake. We ran out of test sessions due to an issue with a master cylinder that had a fresh rebuild on it. I will only ever bleed brakes with a pressure bleeder moving forward since pressing the pedal got the cylinder stuck in the MC.
Why are you running preload on your springs? The springs (1000lbs rear and 600lbs front) are spec and cannot be changed. We tried zero preload on the front and 1.5 rounds on the back. Thought maybe too little preload on the front was causing too much of a difference in roll ratios front/rear so added some preload in. This change caused the front tires to feel like they were skipping across the track on mid-corner high speed turns. Tried putting more preload into the rear to try to make throttle application oversteer more aggressive but it never got us where we wanted to be.

Did the car get better when you softened the front sway bar or did it not make any difference and you just left the bar in the softest position? Yes, it did get marginally better with softer sway bar.

How is the car in the slow corners? Slow corners I can throttle steer the car on corner exit. Corner entry there is still some push. We kept adding rear brake pressure bias but never got to a point where the rear would want to come around during braking. Most of our test sessions were spent either bedding the new rotors/pads, scrubbing the sticker tires, or fixing the master cylinder issue so we didn't have enough sessions to really try to many bias adjustments. I'm not confident with on track adjustments because the bias adjuster would sometimes bind not making any adjustment then suddenly brake loose and move the bias fulcrum point so we adjusted it in between sessions and ended up in the Sunday race at 53% Front brake bias according to our brake pressure sensors.
And where is the push starting? At Gingerman, we got a lot of push on corner entry at turn 2 which is a slowish corner, corner entry turn 3 was a really bad push , corner entry turn 5 we get corner entry push. I tried several techniques at turn 5. The best times that we got was a late threshold braking making turn 5 almost a straight, then turning towards the outside curbing for corner 6 making a slower entry but getting on the throttle earlier. Otherwise we had major understeer and slower times. After corner exit on turn 7 we had a big push mid corner through turn 8 which got worse towards corner entry turn 9. To get any kind of good speed exiting turn 9 I have be cautious on how much throttle I can give through 7 and 8 until I'm able to commit to full throttle very late while exiting corner 9 to make a run into turn 10.
Is it at corner entry, mid corner or corner exit? Almost all corner entry points are understeering, slow mid corners are not bad since I can apply some throttle steer to point the car where I want it but once we get to faster sweeping type corners it's horrible understeer.

I will say that I'm not a very experienced driver yet. This was only my 4th weekend ever in an open wheel car and I had no racing experience before other than a few track days in a miata and one weekend in a FM several years ago. While I openly admit I have plenty of learning to do, which will come with seat time, I feel confident I can put down much faster times with a better setup on the car. Hopefully, I am not being seen as the type to blame the car for mistakes that are more driver error. I would happily listen to any critiquing of my driving if the advice was making me faster.

[S Lathrop]

Why are you running preload on your springs? The springs (1000lbs rear and 600lbs front) are spec and cannot be changed. We tried zero preload on the front and 1.5 rounds on the back. Thought maybe too little preload on the front was causing too much of a difference in roll ratios front/rear so added some preload in. This change caused the front tires to feel like they were skipping across the track on mid-corner high speed turns. Tried putting more preload into the rear to try to make throttle application oversteer more aggressive but it never got us where we wanted to be.

How is the car in the slow corners? Slow corners I can throttle steer the car on corner exit. Corner entry there is still some push. We kept adding rear brake pressure bias but never got to a point where the rear would want to come around during braking. Most of our test sessions were spent either bedding the new rotors/pads, scrubbing the sticker tires, or fixing the master cylinder issue so we didn't have enough sessions to really try to many bias adjustments. I'm not confident with on track adjustments because the bias adjuster would sometimes bind not making any adjustment then suddenly brake loose and move the bias fulcrum point so we adjusted it in between sessions and ended up in the Sunday race at 53% Front brake bias according to our brake pressure sensors.
in a miata and one weekend in a FM several years ago. While I openly admit I have plenty of learning to do, which will come with seat time, I feel confident I can put down much faster times with a better setup on the car. Hopefully, I am not being seen as the type to blame the car for mistakes that are more driver error. I would happily listen to any critiquing of my driving if the advice was making me faster.

Do you know what the motion ratios are for the bell cranks? It would be interesting to see what the wheel rates of the springs are relative to the weight on the front and rear corners. The USF2000 has a rising rate motion ratio ration at the front.

If I understand you explanation about front spring preload, the front worked worse after you increased the preload to 1.5 turns (something close to 100 lbs. force) or was that when you decreased the load from 2 turns? Have you tried no preload? Or have you tried no preload and more front sway bar? I would suggest that you try no preload and full front sway bar. If the sway bar is a "U" type where the adjustment is a slide or change holes along the bar, adjustments are not very sensitive. If it is a blade type bar the adjustments are very non-linear and change very dramatically from 45 degrees to full stiff.

Also, raising the front end a bit might make the front a bit more responsive. But that would depend on where the roll center is located.

With the USF2000, 120 to 140 lbs. preload at the rear was all we needed to balance that car At the rear you may have close to 300 lbs. preload. That may be what it takes to dirt track the car around a corner but it is not great for traction. With the USF2000 we had has several spring options, something you don't. The preload front and rear may be killing grip to the point where the car just slides and it slides enough at the front to keep the car balanced.

In general, preload does nothing for mechanical grip. When you use preload you need to know whether you are just changing the rate at which loads are transferred around the chassis in cornering or you are droop limiting the suspension and you are using that condition to brings about a change in the handling.

Also what toe settings are you using front and rear? Toe settings can be used to change balance, especially when you apply power. The responsiveness of the front end is also sensitive to toe settings.

The more you understand how the suspension of you car works and why it does the things it does, the better you can uses the very limited tools you have to work around the inherent problems.

One thing I do for every car I work with is run a torsion (twist) test of the chassis. I do this with a fully assembled car, on wheel stands and the shocks replaced with struts. Once you have a base line, it is very easy to run a quick check to see if you have a frame crack or maybe a loose belly pan. It takes less than an hour to run the test. The equipment took me maybe 5 hours to make.

[dsmithwc04]

1mm toe out front / 2mm toe in rear

Any chance you can expand on the twist test equipment?

[S Lathrop]

1mm toe out front / 2mm toe in rear

Any chance you can expand on the twist test equipment?

"
I built 4 stands like this drawing. The stand is mostly 1.5" by 1/8" angle iron. The flat on the top of the stand is where I clamped a pair of 14 ft. 2x6 beams. I had adjustable "feet for the rear stands but you could just as easily use a couple blocks of wood to raise the rear up.

To measure the deflection, I used a bunch of dial indicators. Harbor Freight has them and the stands for really good prices. At a minimum I would use 4 indicators. 2 at the front and 2 at the rear. You know the distance between the indicators and the angle is calculated from the displacement of the dial indicators. Subtract the angle of the deflection at the rear from the front and you have how much the chassis twisted.

I used a 50 pound weight placed on the beam. I took measurements at several locations for the weight. Static (zero for the dial indicators) was with the weight close to the stand and then moved the weight to a couple positions between the wheel and the end of the beam. 50 pounds 6 ft from the center of the chassis is 300 ft. pounds of torque. At 8 ft. you get 400 ft. lbs. Much more and you will bend something.

This system was used in the days of aluminum monocoque chassis to monitor the condition of the tubs. Over time the rivet bonds of the tub would loosen and eventually the tub would be stripped and rebuilt. This system was how we determined when to rebuild the tubs.

[dsmithwc04]

One thing should probably consider, we weighed in at 1306 after sundays race but had almost 2 gallons of extra fuel onboard. Could the extra weight be that big of a difference compared to the lighter drivers?

[S Lathrop]

One thing should probably consider, we weighed in at 1306 after sundays race but had almost 2 gallons of extra fuel onboard. Could the extra weight be that big of a difference compared to the lighter drivers?

12 pounds of fuel in a 1300 pound car. I don't think that will compute. Maybe a full tank to an empty tank might.

[R. Pare]

From the sounds of things, the craked frame mentioned is a distinct possibility.

However, I would also check to make sure that something is not binding in the front end - binding or with a very high stiction/friction value.

[S Lathrop]

From the sounds of things, the craked frame mentioned is a distinct possibility.

However, I would also check to make sure that something is not binding in the front end - binding or with a very high stiction/friction value.

One quick way to check for what Richard is talking about is to measure the ride height by first lifting the car to see how high one can get the car to stay and then push the car down and see where the car then stays. Knowing the wheel rates at the 2 ends of the car, you can the calculate how much weight it would take to move the car that difference in ride height.

For a really good car, it would take something like 20 pounds or less to move the ride height the difference between up and down ride heights.. 40 is not uncommon and over 50 is in the troublesome range.

I would unwind any preload to run this test.

[jcorsico]

In general, preload does nothing for mechanical grip. When you use preload you need to know whether you are just changing the rate at which loads are transferred around the chassis in cornering or you are droop limiting the suspension and you are using that condition to brings about a change in the handling.

I don't really understand this sentence. I'm probably confused.

I always envisioned that preload did nothing until it caused the suspension to be droop limited. In my mind, there are really three different ways you can set up a suspension in relation to preload:

1) No preload whatsoever. You set the spring length so that, when the car is at full droop, there is no pressure on the spring. When you lower the car off the jack stands, the spring compresses until it carries the full weight of that corner. For example, assume the corner weight is 200 lbs, and the spring compresses 1" to build 200 lbs of force.

2) Moderate preload. You set the spring length so that, when the car is at full droop, there is some pressure on the spring. When you lower the car off the jack stands, the spring compresses until it carries the full weight of that corner. Using the same example, assume you added 100 lbs of preload. This means the spring will compress only 1/2" to build the required 200 lbs of force.

3) Lots of preload. You set the spring length so that, when the car is at full droop, there is a lot of pressure on the spring. When you lower the car off the jack stands, the spring doesn't compress at all, since there is so much built in force. Using the same example, assume you added 300 lbs of preload. When you lowered the car off the jackstand, and the tire pushed up on the spring, it would only push up with 200 lbs - meaning the spring wouldn't compress further. You could take this to the extreme, by adding so much preload that the wheel would never compress at all under any on-track scenario.

In scenarios (1) and (2), because the spring rate is the same, the rate at which loads are transferred throughout the chassis in a corner are identical, until the car becomes droop limited. For example, if you enter a turn and the car begins to roll, weight will be transferred to the outside tire, compressing the outside tire. Since the spring rate is the same in both scenarios, the tire will compress the same amount in both scenarios (and the inside tire will droop the same amount in both scenarios). There will then be a significant non-linearity when the inside tire becomes droop limited. The inside tire will then get lifted off the ground, and all weight transferred to the outside wheel.

By adding preload, all you have done is change the point at which this non-linearity occurs. With no preload, it probably never occurs (since you probably never run out of droop on the inside tire). With preload, it occurs earlier in the corner. The more preload you add, the earlier it occurs. But, in any event, all preload does is change the point in time when the car becomes droop limited. Before that occurs, the car behaves the same whether or not there is any preload. Right?

Scenario (3) is a little harder to envision, but I think it's basically a situation where the droop limiting comes into play almost immediately, since there is no droop travel in the suspension at all. In other words, you would instantly transfer all weight to the outside tire and have virtually no weight on the inside tire. Depending on the amount of preload you have, you may compress the outside tire, or maybe not (as mentioned above, you could dial in so much preload that you'd never even compress the outside tire).

Viewed in this fashion, preload is nothing more than a super-stiff swaybar that is either on or off. Before the suspension is droop limited, the preload does nothing. After the suspension becomes droop limited, it's like having a super-stiff swaybar because all load has been transferred to the outside tire. By adjusting the amount of preload, all you are adjusting is when (if ever) this super stiff sway bar is applied in a turn.

Please rip this to shreds if I'm wrong!

[dsmithwc04]

Ran the shocks by a local racing shock specialist to be put on the dyno. The results..... all four are dead. All four gas pressures were under 100 psi, had a lot of cavitation, and were 8% off side to side in the rear. I sent them off to get rebuilt today and should get them back before the Road America event. Will be interesting to see how this affects our lap times.

[S Lathrop]

jcorsico,

There is a fourth spring load option, and that is where there is a gap between the spring and the spring perch at full droop.

Option # 3 is called zero drooping the suspension.

Over the years I have tried all the options you listed and the 2 others I mentioned.

I have taken the time to start with no preload and wind measured amounts in, usually say 50 pounds increments at first. At the rear, you will start to feel a difference in handling when you get to a certain load. With Lola T97, some where around 250 lbs. preload things started to change. On the team one driver liked 315 lbs. and the other was not happy with more than 275. Given the motion ratio and weight of the cars, the amount of preload was no where sufficient to support the cars. I have found that for a FC type car, things change some where around 120 lbs.

I have also solved a loose problem with "negative " preload. That is winding the spring perch off by an amount that would produce a given preload if done the opposite direction.

What this tells me is that the amount of preload from negative to the point where there is no suspension movement changes how the car handles. And handling changes long before you top a shock out. Maybe the way to think about this is the amount of preload influences how the corner load on the inside wheel is transferred to the outside wheels.

As soon as the shock tops out, a bunch of things change immediately. Not the least of which is the instantaneous roll center and swing arm length. For the car to roll any further, the entire end of the car has to go down. Or the inside wheel just comes off the ground.

I have had some very good handling cars without preload when others with the same car were winding many turns of preload into the springs. The one advantage I felt that I had with my setups was that I had better tire life and towards the end of races I had a better car that my competition. This is why I asked the question in the first place. Does any one know exactly how much preload is necessary and when do you have too much.

In looking at some of the setups for some cars, I can see why people are using a lot of preload and it may not be possible to get a good balanced car with out a lot of preload. But I have not found that to be the case. Then again I have never engineered a FE or any other car with such limited options for setups. FVs are way more complex to setup than a FE and they can be made to handle very well.

One thing for certain is that you have to have shocks that are built to be topped out as will happen with significant amounts of preload. And some shocks are not built to take any topping out.

dsmithwc04: Good that you found an explanation for your problems. Funny how important proper preparation and maintenance can be.

[dsmithwc04]

Ya, I take maintenance and prep work very seriously. We put atleast 25+ hours into the car in between events. Most of that was constantly making changes to the setup, measuring how each flat affects tge alignment for quick changes at the track, developing understanding of our data system with a plan for how to actually use it at the track, etc.... I never approached the shocks because I had a different shop dyno them and they never mentioned low gas pressure or severe cavitation. Im pretty new to open wheel racecars and am learning as I go. Just another notch on the "lesson learned" category. Im glad i went to a different shop, saved us another wild goose chase. Cant wait to feel what this car feels like when it works the way it is supposed to.

[S Lathrop]

Ya, I take maintenance and prep work very seriously. We put atleast 25+ hours into the car in between events. Most of that was constantly making changes to the setup, measuring how each flat affects tge alignment for quick changes at the track, developing understanding of our data system with a plan for how to actually use it at the track, etc.... I never approached the shocks because I had a different shop dyno them and they never mentioned low gas pressure or severe cavitation. Im pretty new to open wheel racecars and am learning as I go. Just another notch on the "lesson learned" category. Im glad i went to a different shop, saved us another wild goose chase. Cant wait to feel what this car feels like when it works the way it is supposed to.

Shock gas pressure should be an almost daily check. Running them with low pressure and cavitation will damage the internal parts and necessitation replacement.

[dsmithwc04]

Shock gas pressure should be an almost daily check. Running them with low pressure and cavitation will damage the internal parts and necessitation replacement.

Cant, shocks are sealed.

[marshall9]

A dyno can check a sealed shock, as you did.

I am light years away from being a Lathrop, but was wondering if a little toe out would have helped?

Steve?




Now, of course, you have to start over, because if the shocks were that bad, everything you did was a bandaid for that issue.

[dsmithwc04]

A dyno can check a sealed shock, as you did.

I am light years away from being a Lathrop, but was wondering if a little toe out would have helped?

Steve?




Now, of course, you have to start over, because if the shocks were that bad, everything you did was a bandaid for that issue.

Not a very easy way to check them at the track no?

We had 1mm toe out. Then went to 2 mm toe out and didn't make it any better.

And yes, I will have to start from scratch in some ways but the seat time was necessary anyways. Plus, I will know the difference between garbage shocks and brand spankin new ones, thus giving me a reference point to work forward with and perhaps increase the learning curve. Call me Mr. Optimistic! I'm just excited to have a car that I can have an accurate bench mark on where I stand against the competition.

[marshall9]

Not a very easy way to check them at the track no?

We had 1mm toe out. Then went to 2 mm toe out and didn't make it any better.

And yes, I will have to start from scratch in some ways but the seat time was necessary anyways. Plus, I will know the difference between garbage shocks and brand spankin new ones, thus giving me a reference point to work forward with and perhaps increase the learning curve. Call me Mr. Optimistic! I'm just excited to have a car that I can have an accurate bench mark on where I stand against the competition.

Sounds great, enjoy ! Being a set up minimalist, I haven't had anyone like Steve to ever really help me dial a car in. Those that do, do so often differently for every track. It requires a great deal of knowledge from the engineer and driver to interpret the changes. I asked the toe question, because given a solid car with known shocks and springs, I start with basics to get close. If I ever was to be searching for 10ths instead of seconds, I guess I would then hire someone to get it dead on. I have no illusions that the money wouldn't be wasted at my level of driving talent.

Keep us informed, I learn a lot from posts like this, especially when the brainy engineers chime in. Thanks.

[S Lathrop]

A dyno can check a sealed shock, as you did.

I am light years away from being a Lathrop, but was wondering if a little toe out would have helped?

Steve?




Now, of course, you have to start over, because if the shocks were that bad, everything you did was a bandaid for that issue.

Rear toe out works and works well.

I dynoed a set of FE shocks years ago. I found a very good match with 2 of the 4 shocks and those were a good fit at the front. The other 2 did not match anything and would not have been very good used as a pair. But if the one of the odd shocks could have found a good mate, then using that pair at the rear would have been an unfair advantage. Having all 4 shocks go south does not speak very well when you have a spec can and spec shock.

It sounds like the FE shocks are going to need a visit to the Enterprise shock service center yearly if not more often.

It is possible to bounce on a car and carefully watch how the car springs back (rebound damping). When you loose the gas pressure, the rebound will be very sudden and the car will over shoot (rebound to a point higher than where is started statically). At a minimum, this should be a weekly test of the shocks.

[jcorsico]

What this tells me is that the amount of preload from negative to the point where there is no suspension movement changes how the car handles. And handling changes long before you top a shock out. Maybe the way to think about this is the amount of preload influences how the corner load on the inside wheel is transferred to the outside wheels.

Okay; so if the handling changes before the suspension becomes droop limited, how does it change? I'm not trying to be difficult; just trying to understand what happens.

Said differently, what does adding preload do from a handling perspective? You would use it to cure what problem?

Thanks again Steve.

Jon

[S Lathrop]

Okay; so if the handling changes before the suspension becomes droop limited, how does it change? I'm not trying to be difficult; just trying to understand what happens.

Said differently, what does adding preload do from a handling perspective? You would use it to cure what problem?

Thanks again Steve.

Jon

The easy way to think about this is in the case of negative preload. Say the car has 300 lbs. on the corner static. Un-sprung weight is say 50 lbs. as the car rolls, at some point the shock will top out. But instead of lifting the wheel off the ground, 50 pounds is still on the ground. If I had the spring perch turned up so it touched the spring, I would have zero weight on the ground. If all things are equal in this example except the spring preload, the same force will give the same amount of roll. The 50 pounds does not vanish but shows up on the other 3 wheels, more likely the 2 out side wheels. But the weight on the other 3 wheels is different for the 2 cases.

As you start increasing the preload, the weight transfer from that wheel to the other wheels will change for a constant cornering force. It is strictly speculation on my part because I have not and can not model the suspension dynamics, but I believe that between no cornering force to any level of cornering force, the weight transfer is proportional to the distance the suspension rises up to the point the shock is fully extended and the weight on that corner goes to zero. Thus more preload, the shorter the suspension travel. Less suspension travel, the faster that wheel unloads. Also we have not changed the roll resistance with spring preload (springs and sway bars are unchanged). The roll will therefore be proportional to the cornering forces.

Now once we top the shock out, the roll rate changes and significantly. The sway bar rate actually drops to half its original value because the bar is fixed at the inner end and only movement on the outside wheel affects the sway bar rate. But for the car to roll at all. the out side wheel must compress as the inside wheel remains fixed down. This effectively doubles the roll rate from the springs. Also the roll center has migrated to the inside wheel, contact patch.

Once we top out the inside wheel, we get no more help from that wheel in the corner. It is not pressing down against the ground.

When you look at a trace of shock movement, you can see the effect of preload when it gets to the point of topping out the shock. With the T97 Lola, at 315 pounds the tops of the shock traces were flat for some distance but at 275 you could hardly find any traces of topping out the shock. Here we are influencing the handling by pulling the tire off the ground for some increment of time and that is altering the handling in the corner. This is at the rear of the car. This was used to eliminate push in slow corners. It made the car loose for an instant to help through a slow corner.

It preload does that at the rear, why is it not doing the same at the front? That was why I asked the question as to why so much preload in the front end.

[dsmithwc04]

The easy way to think about this is in the case of negative preload. Say the car has 300 lbs. on the corner static. Un-sprung weight is say 50 lbs. as the car rolls, at some point the shock will top out. But instead of lifting the wheel off the ground, 50 pounds is still on the ground. If I had the spring perch turned up so it touched the spring, I would have zero weight on the ground. If all things are equal in this example except the spring preload, the same force will give the same amount of roll. The 50 pounds does not vanish but shows up on the other 3 wheels, more likely the 2 out side wheels. But the weight on the other 3 wheels is different for the 2 cases.

As you start increasing the preload, the weight transfer from that wheel to the other wheels will change for a constant cornering force. It is strictly speculation on my part because I have not and can not model the suspension dynamics, but I believe that between no cornering force to any level of cornering force, the weight transfer is proportional to the distance the suspension rises up to the point the shock is fully extended and the weight on that corner goes to zero. Thus more preload, the shorter the suspension travel. Less suspension travel, the faster that wheel unloads. Also we have not changed the roll resistance with spring preload (springs and sway bars are unchanged). The roll will therefore be proportional to the cornering forces.

Now once we top the shock out, the roll rate changes and significantly. The sway bar rate actually drops to half its original value because the bar is fixed at the inner end and only movement on the outside wheel affects the sway bar rate. But for the car to roll at all. the out side wheel must compress as the inside wheel remains fixed down. This effectively doubles the roll rate from the springs. Also the roll center has migrated to the inside wheel, contact patch.

Once we top out the inside wheel, we get no more help from that wheel in the corner. It is not pressing down against the ground.

When you look at a trace of shock movement, you can see the effect of preload when it gets to the point of topping out the shock. With the T97 Lola, at 315 pounds the tops of the shock traces were flat for some distance but at 275 you could hardly find any traces of topping out the shock. Here we are influencing the handling by pulling the tire off the ground for some increment of time and that is altering the handling in the corner. This is at the rear of the car. This was used to eliminate push in slow corners. It made the car loose for an instant to help through a slow corner.

It preload does that at the rear, why is it not doing the same at the front? That was why I asked the question as to why so much preload in the front end.

After skimming through your post, a question popped in my head. Could one of the ways to tune for correct preload be to watch for flat spots in the shock data being caused by the preload preventing enough suspension movement to allow the tire to apply force to the pavement? Another thought, it looks easy enough to write a math channel to look for wheel speed sensor data and shock data to calculate wheel slip and the shock topping out to provide easy to read points of a front wheel coming off the ground. Although it is probably just as easy to look for the flat spots in the shock data anyway.

[S Lathrop]

After skimming through your post, a question popped in my head. Could one of the ways to tune for correct preload be to watch for flat spots in the shock data being caused by the preload preventing enough suspension movement to allow the tire to apply force to the pavement? Another thought, it looks easy enough to write a math channel to look for wheel speed sensor data and shock data to calculate wheel slip and the shock topping out to provide easy to read points of a front wheel coming off the ground. Although it is probably just as easy to look for the flat spots in the shock data anyway.

Topping out is easy to see on a shock trace, if your logging rate is sufficiently fast, say 100 to 200 times a second. If you don't log at a high enough rate, the topping out events will not be recorded.

Wheel speed will not be all that useful because the time intervals we are looking to influence are quite short. Think the middle of a corner that is only a few car lengths long. Also you are not likely to be applying braking or acceleration at that moment in time. It would be very unusual if the tire were actually lifted clear of the ground and stopped rotating.

Before looking at the data, take a minute or so to carefully write down what the car felt like, what was it doing that you liked and didn't. Especially note how the car handled in each phase of the corner, from initial brake application to the second shift after the corner exit. See what questions you want to find answers to before you start looking at data. Take a few critical corners and do this analysis thoroughly. Ask how the last setup changed what happened and was it what you expected. Without a frame work to analyze the data, you are likely just wasting time hopping that an answer will jump out at you.

I think that driver's tails are the best data source because it is what the driver feels that is all important and there is not a sensor inserted into the driver that is logging pucker feelings.

[dsmithwc04]

There's a sensor Im definately not adding to my AiM system.

[jcorsico]

When you look at a trace of shock movement, you can see the effect of preload when it gets to the point of topping out the shock. With the T97 Lola, at 315 pounds the tops of the shock traces were flat for some distance but at 275 you could hardly find any traces of topping out the shock. Here we are influencing the handling by pulling the tire off the ground for some increment of time and that is altering the handling in the corner. This is at the rear of the car. This was used to eliminate push in slow corners. It made the car loose for an instant to help through a slow corner.

It preload does that at the rear, why is it not doing the same at the front? That was why I asked the question as to why so much preload in the front end.

Hugely helpful. Thank you Steve. This makes perfect sense to me.

It would argue that, if the car is understeering, that you would not want to add preload. Actually, you would want to do the opposite - reduce preload (at least until the point that inside tire is no longer becoming droop limited).

[jcorsico]

After skimming through your post, a question popped in my head. Could one of the ways to tune for correct preload be to watch for flat spots in the shock data being caused by the preload preventing enough suspension movement to allow the tire to apply force to the pavement? Another thought, it looks easy enough to write a math channel to look for wheel speed sensor data and shock data to calculate wheel slip and the shock topping out to provide easy to read points of a front wheel coming off the ground. Although it is probably just as easy to look for the flat spots in the shock data anyway.

The only way to really measure this would be to use a strain gauge on the pushrod. What you are trying to measure is the force the wheel is exerting on the ground.

If you have suspension position sensors, those will measure it somewhat - since you can see suspension travel and, thus, when the shock becomes fully extended. But if you are running a lot of preload, there may be times when the shock is fully extended and there is still meaningful weight on the tire. For example, if you have 200 lbs of preload, you can have 100 lbs on the wheel, and the shock will still be fully extended. Only a strain gauge would be able to measure that situation with accuracy.

[R. Pare]

Preloading to limit droop is done to decrease jacking effect at the point where push starts.

It does so in part by forcing the car - as more lateral load is generated - to roll about the inside tire contact patch, forcing the car to lower itself. That lowering decreases the jacking effect.

To get the weight transfer so that the is zero load on the inside tire, the car would have to be rather soft torsionally, over weight in the rear, and have way too much preload, too big a front bar, and/or be using zero droop. Look at old pictures of Porsche 911's for a great example of that sort of issue.

[jcorsico]

It does so in part by forcing the car - as more lateral load is generated - to roll about the inside tire contact patch, forcing the car to lower itself. That lowering decreases the jacking effect.

Richard - can you explain this again? I'm having a hard time envisioning how this happens.

Thanks again!
Jon

[S Lathrop]

The only way to really measure this would be to use a strain gauge on the pushrod. What you are trying to measure is the force the wheel is exerting on the ground.

If you have suspension position sensors, those will measure it somewhat - since you can see suspension travel and, thus, when the shock becomes fully extended. But if you are running a lot of preload, there may be times when the shock is fully extended and there is still meaningful weight on the tire. For example, if you have 200 lbs of preload, you can have 100 lbs on the wheel, and the shock will still be fully extended. Only a strain gauge would be able to measure that situation with accuracy.

To measure a lot of these effects, you will need high logging rates and most systems at the club level will not log that fast.

I have done testing with the data logger set to only log a lap or tow when I was trying to look suspension movement. Not enough memory and that was with a fairly advanced Pi system.

[R. Pare]

Richard - can you explain this again? I'm having a hard time envisioning how this happens.

Thanks again!
Jon

If the car wants to roll more - which it will if the lateral load is increased - the center of rotation - the actual physical Roll Center at that time - is at just about the center of the inside tire contact patch. As it rolls about that point, the chassis lowers, lowering the center of gravity at that end of the car, and the jacking effect, ( a self-stiffening in reaction to the lateral load) decreases.

[David Ferguson]

.

[jcorsico]

If the car wants to roll more - which it will if the lateral load is increased - the center of rotation - the actual physical Roll Center at that time - is at just about the center of the inside tire contact patch. As it rolls about that point, the chassis lowers, lowering the center of gravity at that end of the car, and the jacking effect, ( a self-stiffening in reaction to the lateral load) decreases.

With the effect that the lowered center of gravity and reduced jacking effect has some positive impact on front grip (even though overall front grip must be down because you've got little weight on the inside tire)?

[R. Pare]

If overall front grip went down, you would have even more push.

[jcorsico]

If overall front grip went down, you would have even more push.

Fair enough!

Why does anyone use droop limiting as a tuning tool? It's seems like a sledgehammer type of approach. You introduce this massive non-linearity into the suspension, and hope that the positive benefits of the non-linearity outweigh the negative.

I can see why it's a very sensitive adjustment. You're going to get no effect at all until it kicks in, and then when it kicks in, it should be dramatic. By adjusting the amount of droop limiting, you would be adjusting where in the turn the effect kicks in. It probably varies turn by turn and track by track. What a mess!

[R. Pare]

It's a crutch adjustment for an overall suspension design where the front and rear jacking effects are not in harmony with each other. Add in the front anti-dive that too many designs at this level incorporate, and you have a push problem.

[R. Pare]

One thing not mentioned as a possibility with these cars are rising-rate front front bellcranks without the rears matching.