Simraceway stage 2 and Evolution driving school

[kcbrown]

Since my experience with Simraceway phase 2 wasn't a whole lot different from stage 1, I thought I'd combine it with my commentary on the Evolution driving school.


Simraceway driving school, stage 2

This stage involved a few laps around their "autocross" course, which is really more like a mini track than an actual autocross setup, followed by a couple of lead-follow sessions on the track, followed by a couple of normal sessions on the track. I was hoping, actually, for more "autocross" time than we got, since it's a much safer environment to explore the capabilities of the car.

There's not a whole lot different about the track sessions this time than last, except that the speeds were up and I was pushing the car a little harder. The car continued to handle as before, in a very controllable manner. Wear on the tires was a little more pronounced at the outside edges, but not markedly so. I think it's interesting that it took this long to get to the point where the wear on the outside edges was starting to be plainly more than the inside edges. More commentary on this below.



Evolution driving school

This was unbelievable amounts of fun. They set up what amounts to a real autocross course. Lots of slaloms, lots of tight turns, and even a "skidpad" circle (which amounted to a clockwise 270 degree turn).

There was a lot that I learned from this. In a number of turns, especially going around the "skidpad", I found that I was able to get the car to come back into line by either lifting the throttle and forcing the front to tuck in and, occasionally, getting the rear to step out a little, or by giving it more throttle and causing the car to oversteer slightly and, thus, turn in that way.

I also found that being slow, smooth, and precise on the autocross is still slow. Being truly fast meant driving harder and much more abruptly than I like, and that actually reduced the fun. It may be that it just takes more time to get fast when one is doing so by concentrating on smoothness and precision. We'll see. They're doing another one at the same location in November, and I've signed up for it.


And then there's the tires. While the wear on the outside was becoming apparent after the Simraceway school, this one took it over the top. And it made something apparent that wasn't apparent before. How I drive the track and how I drive an autocross course is very different. On the track, I'm not pushing the car quote enough to reach the limits of adhesion, but I'm still getting a lot of lateral grip per the G meter. On the autocross course, I push the car until it actually does reach the limits of adhesion and, sometimes, even more than that. The latter is much harder on the tires, and it showed after this event. The wear is now enough that I've replaced the front tires to cover the rest of the year, mainly as a precautionary measure. There's plenty of tread on the first set of front tires, at least except for the very outer edges of the tires, so they'll become very useful again once I've got camber plates.

But I most definitely need camber plates. Those, along with coilovers, are going on the car early next year.

Oh, and Dave, since we talked about lateral versus braking grip and "calibrating" the G meter based on that, during the Evolution school I was getting 1.1G of braking, and about 1.05G of lateral acceleration. That I was getting less lateral acceleration than braking suggests that perhaps the way I drive on the track sometimes manages to get more out of the car than how I drive on the autocross. It'll be interesting to see what the G meter says at Laguna Seca.


In general, the Evolution school confirmed my previous thoughts on the suspension of the car. It's very, very good. There may be a little room for improvement in terms of the responsiveness, but I think the current suspension is well past the point of truly major gains, at least until one starts sacrificing major amounts of ride quality to get those gains. I remain pleasantly surprised with how well this suspension works, even in fast transitions such as what you get in slaloms. Of course, it ain't no S2000 or BRZ. Those cars will murder this thing on the autocross simply because they're smaller, and thus can take tightly-bounded corners faster because their line through those corners is shallower and thus faster (slaloms especially will highlight the difference -- you simply don't need as wide an S curve with those cars as you do with the Mustang in order to clear the cones, which means shallower steering angles and thus significantly greater potential speeds).

There do seem to be some steering angle versus speed combinations that result in a lot of understeer, and that's something I'd like to dial back a bit. The problem is, I don't know how to do that without eliminating understeer altogether, and that's something I don't want to do, because most of what I've experienced from this suspension so far is easily controlled balance. My thinking so far (backed by "experience" on the virtual track) is that it's valuable to retain some of that understeer, simply because it allows for greater control over the car. For instance, if you eliminate understeer entirely, then your trail braking is going to have to be very light, else you'll spin the car.


I'm awfully tempted to not change a thing, with the exception of camber plates. But at least with coilovers, I'll have the ability to experiment with various settings to find the best compromise for my tastes. So I'm going to proceed with my plan as it is. I'll continue to work with the stock suspension through the end of the year. By then, I may have discovered some additional things I want to improve. Certainly, stiffening the springs and using better dampers without substantially changing anything else (especially ride height) should yield some improvements here, but what's already there is so good that I don't think those improvements will feel like they're "major".


Another interesting thing of note: this experience has taught me that the autocross venue is a much better place to test changes to the suspension, because you can put the car through most of its handling regime without having to worry about damaging anything.

And one final note, something I found encouraging: the instructors seem to think that I actually have some talent at this.


Laguna Seca is coming up tomorrow, and I can't wait to drive it. I've got many hundreds of laps around it in the virtual world. It'll be very interesting to see how different it is to drive it in the real world. As usual, I'm going to approach it conservatively, and drive the car a little under its limits as I have been at Sonoma. I tend to have more fun that way, because I enjoy being smooth and in control.

 

[kcbrown]

Also, you guys wanted me to comment on the Ford Racing TrackCal tune that I have in the car.

It's excellent. The throttle is very linear (too linear, actually -- I'll describe why in a moment). Engine braking is full and immediate, the way it should be. The car's throttle feels much like it did in my 1992 GT.

The end result is that the car was extremely easy to control with the throttle through the corners of the autocross course.


There is one operating regime where the mapping could use a little nonlinearity: at the very bottom. It seems that when you have the throttle barely opened, the mapping from that to output power is such that you can get some "flapping" from it. This seems to affect the cruise control as well, and the end result is that if you're descending a hill in 6th gear at highway speed, you might feel a slight "surging" sensation.

The fix for that would be to extend the low-end region of the throttle a little, to allow a bit more precision of control at the very bottom. That would, of course, have to come at the expense of the rest of the range, but I don't think it would need much there.

That's my only real complaint about the tune. For high performance driving, it's excellent.

 

[Norm Peterson]

I also found that being slow, smooth, and precise on the autocross is still slow. Being truly fast meant driving harder and much more abruptly than I like, and that actually reduced the fun. It may be that it just takes more time to get fast when one is doing so by concentrating on smoothness and precision. We'll see. They're doing another one at the same location in November, and I've signed up for it.

I've been told that a really fast autocross run will feel like you aren't ever quite in control, and there certainly won't be enough time to be deliberate with your inputs (you still have to be smooth, just faster with your inputs).

Which makes sense when you realize that to the maximum extent possible you're operating out toward peak-grip slip angles in the corners and near peak slip% under braking and acceleration, and course features are little more than a second or two apart. You absolutely will feel the sliding around that this much slip implies, and the rather frantic pace of autocross maneuvers exaggerates this. It may feel . . . imprecise or even a bit sloppy (for lack of better terms). At least when you're first experiencing it and the car isn't a complete understeer-pig.

There is one operating regime where the mapping could use a little nonlinearity: at the very bottom. It seems that when you have the throttle barely opened, the mapping from that to output power is such that you can get some "flapping" from it. This seems to affect the cruise control as well, and the end result is that if you're descending a hill in 6th gear at highway speed, you might feel a slight "surging" sensation.

I'm afraid I don't understand what you're experiencing here, or even whether you want more response or less down at small throttle openings.



Norm

 

[kcbrown]

I'm afraid I don't understand what you're experiencing here, or even whether you want more response or less down at small throttle openings.

Well, it's a digital system, right? The throttle is actually a potentiometer with an analog to digital converter on it. Because it has an AtoD converter on it, the output of the thing is a discrete value which depends on its position.

A linear response curve maps the output of the throttle to the input of the throttle body in a linear way. If the output range is 0-100, for instance, and we call that X, and the power range of the throttle body is also 0-100 and we call that Y, a linear mapping would be something like Y = X.

But here's the problem: the system is discrete. Between 1 and 2, for instance, there is no value. You have to choose one or the other. And going from 1 to 2 means doubling the power output, while going from, say, 10 to 11, means increasing the power by 10%.

If you're at very low throttle settings, minute changes to those settings will be more detectable. And if you're attempting to maintain a speed by varying the throttle slightly while at those very low throttle settings, the end result is that you'll either slightly undershoot your desired power setting, or slightly overshoot it, and at those lower power settings, you'll undershoot or overshoot by more than you would at somewhat higher power settings. You'll get "flapping".

That part's unavoidable, because the power control input is a discrete value. What is changeable is, firstly, the sensitivity of the throttle at those lower values -- right now, a tiny change in the input throttle results in a large percentage increase in the power -- and secondly, the response speed of the system to changes at those settings. If that speed is reduced somewhat, it'll make the transition from, say, 1% power, to 2% power, less abrupt, and the "surging" will become less apparent.


What I want is for the above changes to be made for the very lowest part of the power curve, so as to allow for easier and smoother control over it. I guess that translates to "less throttle sensitivity" at those parts of the power curve, but that term is perhaps oversimplifying it, hence my explanation above.

 

[claudermilk]

Evolution driving school

This was unbelievable amounts of fun. They set up what amounts to a real autocross course. Lots of slaloms, lots of tight turns, and even a "skidpad" circle (which amounted to a clockwise 270 degree turn).

There was a lot that I learned from this. In a number of turns, especially going around the "skidpad", I found that I was able to get the car to come back into line by either lifting the throttle and forcing the front to tuck in and, occasionally, getting the rear to step out a little, or by giving it more throttle and causing the car to oversteer slightly and, thus, turn in that way.

I also found that being slow, smooth, and precise on the autocross is still slow. Being truly fast meant driving harder and much more abruptly than I like, and that actually reduced the fun. It may be that it just takes more time to get fast when one is doing so by concentrating on smoothness and precision. We'll see. They're doing another one at the same location in November, and I've signed up for it.

Shall I say "told you so"? It sounds like you found the school to be as valuable as I did; it's certainly worth the cost.

Yes, slow is slow. But smooth and precise ends up being faster; they are trying to get a good foundation laid. Autocross is a violent, frantic environment compared to open track. However, you still need to be smooth (i.e., don't jerk the wheel or stab the pedals).

And then there's the tires. While the wear on the outside was becoming apparent after the Simraceway school, this one took it over the top. And it made something apparent that wasn't apparent before. How I drive the track and how I drive an autocross course is very different. On the track, I'm not pushing the car quote enough to reach the limits of adhesion, but I'm still getting a lot of lateral grip per the G meter. On the autocross course, I push the car until it actually does reach the limits of adhesion and, sometimes, even more than that. The latter is much harder on the tires, and it showed after this event. The wear is now enough that I've replaced the front tires to cover the rest of the year, mainly as a precautionary measure. There's plenty of tread on the first set of front tires, at least except for the very outer edges of the tires, so they'll become very useful again once I've got camber plates.

But I most definitely need camber plates. Those, along with coilovers, are going on the car early next year.

<snip>


In general, the Evolution school confirmed my previous thoughts on the suspension of the car. It's very, very good. There may be a little room for improvement in terms of the responsiveness, but I think the current suspension is well past the point of truly major gains, at least until one starts sacrificing major amounts of ride quality to get those gains. I remain pleasantly surprised with how well this suspension works, even in fast transitions such as what you get in slaloms. Of course, it ain't no S2000 or BRZ. Those cars will murder this thing on the autocross simply because they're smaller, and thus can take tightly-bounded corners faster because their line through those corners is shallower and thus faster (slaloms especially will highlight the difference -- you simply don't need as wide an S curve with those cars as you do with the Mustang in order to clear the cones, which means shallower steering angles and thus significantly greater potential speeds).

Another "said so"? I honestly think a major part of the improvement I am noticing is the camber plates; and much of that I think is from going to a large solid bearing in place of a soft rubber bushing--that moreso than the -2.75 deg camber I set at the track.

While I agree that the basic suspension on these cars is a very good start, the compromise for ride quality hinders it. Switching to better dampers and stiffer springs really helps a ton. You will see once you make the change yourself.

I'm awfully tempted to not change a thing, with the exception of camber plates. But at least with coilovers, I'll have the ability to experiment with various settings to find the best compromise for my tastes. So I'm going to proceed with my plan as it is. I'll continue to work with the stock suspension through the end of the year. By then, I may have discovered some additional things I want to improve. Certainly, stiffening the springs and using better dampers without substantially changing anything else (especially ride height) should yield some improvements here, but what's already there is so good that I don't think those improvements will feel like they're "major".

A good plan, and I think you will be surprised once you do make some parts changes.

Another interesting thing of note: this experience has taught me that the autocross venue is a much better place to test changes to the suspension, because you can put the car through most of its handling regime without having to worry about damaging anything.

And one final note, something I found encouraging: the instructors seem to think that I actually have some talent at this.

Agreed. When your major risk is killing a traffic cone & getting a scuff that will buff out, you can really push the car. I'm not surprised that they think so.

 

[Norm Peterson]

Well, it's a digital system, right? The throttle is actually a potentiometer with an analog to digital converter on it. Because it has an AtoD converter on it, the output of the thing is a discrete value which depends on its position.

A linear response curve maps the output of the throttle to the input of the throttle body in a linear way. If the output range is 0-100, for instance, and we call that X, and the power range of the throttle body is also 0-100 and we call that Y, a linear mapping would be something like Y = X.

But here's the problem: the system is discrete. Between 1 and 2, for instance, there is no value. You have to choose one or the other. And going from 1 to 2 means doubling the power output, while going from, say, 10 to 11, means increasing the power by 10%.

You don't think that the PCM has an interpolation routine for injector pulse width and timing? The resistances in the pedal sensor are certainly continuous in nature. (I believe that there are three separate ones - with one of those increasing resistance in the opposite direction to the other two - for redundancy and error code purposes.)


Even the aftermarket speed-density system I fitted to a previous car and tuned myself (via laptop) performed spark and fuel map interpolations based on continuously variable MAP and RPM inputs, and that was over 15 years ago.

Which kind of reminds me that there are a couple of acceleration enrichment schemes that could be involved.


Norm

 

[kcbrown]

You don't think that the PCM has an interpolation routine for injector pulse width and timing? The resistances in the pedal sensor are certainly continuous in nature. (I believe that there are three separate ones - with one of those increasing resistance in the opposite direction to the other two - for redundancy and error code purposes.)

The resistances in the pedal sensor are certainly continuous in nature, but those resistances are being converted to digital values via an A converter, and those values are discrete. Now, it may be that the precision of the A converter and the range of its output is such that, for practical purposes, it's the same. If, for instance, it's converting to a 32-bit integer value and has sufficient precision in its conversion to be able to make good use of the entire space provided by that 32-bit value, then sure, one shouldn't see a problem with throttle position precision.

But the same has to be true of, e.g., the throttle body input, which is at some point going to be using a digital to analog converter to take an input value and convert it to a target throttle plate position. If that conversion process has a sufficiently small precision (e.g., if it takes, say, an 8-bit integer value as its input) that one would notice the power output difference from two adjacent values, then you could easily see "flapping" in the way I am.


I'm sure there are some smoothing functions involved in the current design, but whatever they are, they're not sufficient at very low throttle settings. The stock tune has enough smoothing that it masks/eliminates this issue, but the TrackCal tune doesn't.


Regardless, it's probably the only notable downside I've discovered thus far (there is one other minor issue, and that is that sometimes, after high performance events, the computer will throw "emissions control efficiency below threshold" codes on both banks, but that doesn't surprise me given that the car has been operating at wide open throttle much of the time. These codes don't reappear during normal driving, so it's clearly related to high performance driving, and thus I'm not bothered by it).

For high performance driving, the TrackCal tune is superb.

 

[kcbrown]

Shall I say "told you so"?

Heh. Yes. You should.

It sounds like you found the school to be as valuable as I did; it's certainly worth the cost.

It was so fun that I've signed up for the one in November. To be honest, I may wind up moving up my target date for coilovers to sometime in October, so that I can use the November Evo school for "test and tune".

Yes, slow is slow. But smooth and precise ends up being faster; they are trying to get a good foundation laid. Autocross is a violent, frantic environment compared to open track. However, you still need to be smooth (i.e., don't jerk the wheel or stab the pedals).

Right. And I was that (well, relatively speaking), even at my most frantic. It's just that I didn't enjoy it as much. I was much happier being a little slower and a little smoother with the controls. Thing is, though, that it didn't make an immense amount of difference, and it's probably one of those things that just takes getting used to. I'll learn more about that at the next Evo school. I can't wait!

Another "said so"? I honestly think a major part of the improvement I am noticing is the camber plates; and much of that I think is from going to a large solid bearing in place of a soft rubber bushing--that moreso than the -2.75 deg camber I set at the track.

That could be. Given the phenomenal grip levels I'm already getting from the car, I think it'll be interesting indeed to see how much improvement there is once I've got camber plates on the car.

While I agree that the basic suspension on these cars is a very good start, the compromise for ride quality hinders it. Switching to better dampers and stiffer springs really helps a ton. You will see once you make the change yourself.

Well, remember that my goal is to retain or improve the ride quality in the process. I believe this can be done, having ridden in cars with substantially more spring in them than mine and noticing that because the dampers were set up properly, they didn't really ride worse than my car over anything but the biggest bumps (and that's just a matter of physics).

My spring rate choice will involve a compromise. I'm not going for anything like 450 lb/in in the front or anything crazy like that. I might wind up merely (roughly) doubling the rates that are already on the car, and thus be in the 300 lb/in range. Frankly, that's fine with me. Cutting body roll (which thus far hasn't really bothered me much) by 20% and brake dive in half by doubling spring rates is perfectly good with me!

A good plan, and I think you will be surprised once you do make some parts changes.

I just hope I don't make things worse in the process. Above all, I want controllability. The stock suspension seems to give that to me.

 

[csamsh]

I just hope I don't make things worse in the process. Above all, I want controllability. The stock suspension seems to give that to me.

just wait....once you get some good parts on there, you will want to come back and mark that statement out with a giant red pen

 

[2013DIBGT]

Heh, Personally I would describe the stock suspension having the finesse of a water buffalo wearing roller skates.

 

[claudermilk]

Above all, I want controllability. The stock suspension seems to give that to me.

Heh, Personally I would describe the stock suspension having the finesse of a water buffalo wearing roller skates.

A lot is about perception. I started out where kc is and ended up where DIB is regarding the stock suspension. I even ended up using a similar analogy after the Goodguys event (hippos doing ballet in a phone booth). For me right now it's all about figuring out the settings. On the street it's a huge improvement, and I'm quite happy with the changes. On the track, I've got some balance tuning to do--but again, overall it's a big improvement.

 

[kcbrown]

A lot is about perception. I started out where kc is and ended up where DIB is regarding the stock suspension.

I was under the impression that your basic opinion of the stock suspension was fairly consistent from the beginning, no? I seem to remember that you had experience with other cars with coilover suspensions, which would make such an assessment from the beginning understandable...


Anyway, I'm very much looking forward to seeing what the changes I have planned will do. I just have to compute the spring rates first, and I'm wrestling with a couple of different basic models for determining those.

Maintaining the "flat ride" characteristics is likely to be the most challenging aspect of the exercise.

 

[csamsh]

Maintaining the "flat ride" characteristics is likely to be the most challenging aspect of the exercise.

Easy...just make it stiff w/good shocks. Presto

 

[claudermilk]

I was under the impression that your basic opinion of the stock suspension was fairly consistent from the beginning, no? I seem to remember that you had experience with other cars with coilover suspensions, which would make such an assessment from the beginning understandable...


Anyway, I'm very much looking forward to seeing what the changes I have planned will do. I just have to compute the spring rates first, and I'm wrestling with a couple of different basic models for determining those.

Maintaining the "flat ride" characteristics is likely to be the most challenging aspect of the exercise.

Not really. My last performance car was lowered, but on Illuminas/Eibachs with a customer rear sway & strut tower braces front & rear. It was also FWD with about 1/3 the HP. I spent a while just getting used to the characteristics of the S197. As I got more familiar and comfortable with it, I started noticing the soft setup and resulting body motion more. That came to a head at the track day where I really took note of the brake dive; after that it bugged me until I lowered the car. Now I'm playing with the settings on the Konis to find a balance I like on track--so my issues are self-induced user error as I climb that learning curve.

 

[Norm Peterson]

Hmmmm . . . with Illuminas I wonder if you'd ever notice anything at all about flat ride for all the harshness.

I think pitch and roll movements bother some people more than others, and I'm pretty firmly in the "they don't bother me much" camp as long as the corners of the car aren't trying to imitate pogo sticks.


Norm

 

[claudermilk]

LOL. Yeah, I was young and uninformed then. I wished for Konis or coilovers most of the time I ran the Illuminas, but funds were not there to fix that.

The soft stock springs didn't bother me too much until I started really pushing the car. Then, once you've noticed something you can't un-notice it. It was to the point I thought every day "jeez, did I buy a performance car or an old Cadillac?"