Mahler9th

More on installation/set up...

I have read the AIM documentation on GPS-based gear calculation/set up a few times and hope to give it a try today. But I want to do it during street driving, not track driving.

I presume that if I am getting satellites as reflected in GPS speed indication, the procedure will work. Is that correct?

Any suggestions for different procedures?

ProCoach

Sorry I didn't answer your original question.

Depending on the efficacy and efficiency of the driver as measured, tenths to over a second, depending on the length of the course...

There is ONE major reason why a driver "that drives a lot and is very consistent" IS consistent. They are driving to their idea of what they and the car are capable of...

This is because they have formed a belief system, through knowledge and repitition, that they can brake at this mark, slow to this speed, negotiate the corner at this speed and apply throttle at this point.

And they do that over and over and over again, yielding the same (or within a very narrow window of) time.

Armed with the knowledge that you will gain from these measures, both you and Steve will find areas of opportunity you may not previously have considered. When you do, and when you can execute the "new" plan, you will go faster times the number of those opportunities present on each track.

I have done and seen this for DECADES, and know it to be true. The data just makes it a hell of a lot easier to be objective and knowledgeable about evaluating your performance, formulating those "new" plans and then it's up to you!

linzman

Depends. I used to think I was really consistent and pretty quick, til I started racing.
If you could translate this into carrying 2 MPH more speed through turn 1 at Watkins Glen, that carries all the way to the bus stop and you're talking seconds, not tenths. I don't know your driving, but I think you'll be pleasantly surprised once you start analyzing and using the data.

ProCoach

NOW, you're talkin'!

Mahler9th

I spent the majority of my "climbing the curve" track time in a somewhat unique environment compared to what we have today. I guess you could call it a "competitive" DE environment.

Cars were classed pretty well, and at the end of the two day weekend there were timed solo runs. Event trophies and season championships. This environment had a high percentage of participants that were focused on the craft and getting faster... both car and driver. Especially the latter.

There was a pretty consistent focus on using as much of the car as you could, and a pretty prevalent sense of what a car's limits might be... so the goal was always to work toward those minimum lap times.

Like many of my friends from those days, I never settled for consistency at some level well underneath the car's ultimate capability, just comfortably underneath.

This is the culture I have adopted for better or worse. I get a strong sense of what my car will do in its current configuration, and set my desired level of risk and drive to it. So when "on" I don't think braking data could help me get seconds, but rather parts of seconds. When fully on, perhaps just tenths.

Tenths aren't as important to me nowadays.

Anyway, I look forward to finally experimenting a bit now that I have a logging system. My initial plan will be to focus on video with quantitative parameters rather than squiggly lines, but I am open minded, and look forward to going to my first seminar to hear what experts have to say.

Meanwhile, I want to finish getting my new system configured and installed.

Matt Romanowski

I agree with all of these. Until you have data from your driving, you don't have a real record of what you did. It's amazing to find how much difference there can be from perception to reality, especially when the driver is stressed and focused on driving. Memory can be very interesting.

To add what Peter said about braking (which I completely agree with), using G Sum is a great way to see how much traction you're leaving on the table on corner entry. I've yet to find a person who didn't learn something from their data to make them go faster.

Mahler9th

Happy New Year. It's a sunny day and I think I will take my race car out for a street drive to further test/configure my new system.

Can anybody comment on this or point me to an online resource? I will give the AIM procedure a try, but if there is a better or different approach, can anyone point me in a better direction?

"More on installation/set up...

I have read the AIM documentation on GPS-based gear calculation/set up a few times and hope to give it a try today. But I want to do it during street driving, not track driving.

I presume that if I am getting satellites as reflected in GPS speed indication, the procedure will work. Is that correct?

Any suggestions for different procedures?"

ProCoach

I have not successfully done that, on the street.

Each gear MUST be engaged for a minimum of five seconds (and I prefer ten)

You MUST not allow revs to drop much below driven speed between shifts, otherwise it gets confused.

After operating in each gear for at least five seconds (and try for more), you can drive the car normally through the gears, but the correlation between speed and RPM must be kept close after the initial run-in through the gears.

When you gear down to stop, stall the engine at the point where the speed stops (vehicle stops). DO NOT lift the power from the EVO4, allow it to make the calculations after no RPM signal exists, no GPS speed is registered but the power is still on.

The calibration procedure does not allow for stopping and starting again, "dipping" of the clutch allowing the revs to drop close to idle and in any way sitting and idling during the procedure or end of the procedure.

I think the track is better (for me, mandatory) for this and while it takes a few tries to get it right, once you do it, you're done. Good luck!

Matt Romanowski

I've always been able to do the gear calc in the paddock, but most of the time we have not been geared too high. Regardless of where, it's important that you keep a slow, steady acceleration. The system is determining the ratio of RPM to speed to determine the gear, so you have to make sure you give it good info (GIGO).

For folks that have a wheel speed (which I'm an advocate of and is easy on a 911), you can use the wheel speed to get the gear calc done even easier. If you have a drive wheel, you can do it on a lift.

Mahler9th

Thanks guys. Looks like some experience and finesse is involved. Seems like you have two overlapping set of experiences.

My car is an air-cooled 911 with typical GT gearing. If I have to row through the gears shifting at red line, then I had better wait until I get to Thunderhill. If I can get accurate gear calc short shifting but maintaining 5-10 seconds in each, I suppose I could try it on the nearby freeway. I guess I'd have to carry the laptop to initiate the sequence, starting from a stop on an on ramp. Then come to a stop/stall after rowing through the gears . I think I could do that safely.

I am curious about wheel speed Matt. How would you "easily" get wheel speed on an old 911 race car?

ProCoach

I think you could get this by short shifting on the freeway (say, 2000-3000 rate of change in each gear, you don't need to floor it for it to associate speed with RPM).

You do NOT need a laptop or anything else to do this.

On the G-Dash, go to Menu>System>Restart Gear Calc (IIRC) and roll off from a stop. Then follow your plan and stall it after coming to a stop.

On the wheel speed sensor, build a bracket and mount it on the inside of the front hub. Like this: https://hooperdave.files.wordpress.c...heel-bolts.jpg

You may be able to do this on yours without adding the Channel Expansion.

Matt Romanowski

You should be able to run the wheel speeds into the speed input on the Evo4 (top left connector). No need for a channel expansion. You can get the splitter so it will do 2 wheels.

No need to go to redline. When I said slow acceleration, I meant as long as it's anything that is increasing. It can be very slow, as long as it's increasing.

This is how I do it for the fronts on a 911. I'm not sure what Peter's picture is from, but same idea. You make a bracket to go from the strut to mount the wheel speed sensor near the rotor mounting bolt heads. You use the inductive sensor, not the hall effect. One pulse per revolution with the magnet is not enough. I have a different way on the rear that works, but I don't like it as much. I'm working on a different rear setup.

ProCoach

I guess my prejudice against wheel sensors comes from decades where that was the ONLY way to get ground speed, and for generations of Pi, Stack and Mychron units, it was the only speed measure, without even the possibility of adding GPS.

CDS came up with a nifty, if not inexpensive, device that plugged into the wheel speed sensor and beacon inputs on a Pi Clubman that would substitute GPS speed for wheel speed pulses and insert the manually entered start/finish coordinates (needed only the first session out per weekend) for the beacon pulse.

I remember a few years ago debating GPS positional and speed accuracy with really smart people like Chris Brown, Bob Knox and Tom Remedios, guys that lived and breathed wheel speed sensors, having begun and spent much of their professional careers reliant on them. There is a very good explanation of all this in Chris Brown's book; "Making Sense of Squiggly Lines."

So while there are benefits to their use, particularly when diagnosing chassis efficiency under braking and setting up (or adjusting) diffs (for traction off the corner), I think the world is an infinitely easier and more reliable place with GPS as the speed source!

Matt Romanowski

I completely agree on starting with the GPS. When we did the first AiM systems on a couple of Crossle 35F, Reynards, and 911s back around 2000, using just the wheel speeds for things was difficult. GPS is so much easier, quicker, and more accurate. I like the wheels speed as an addition to the GPS, not a replacement. Their are certain things, like gear charts, gear calc, and setup things were the actual wheel speed can't be beat and GPS speed is not as good. For things like driver comparisons, gearing, etc, GPS is the best. Plus, now we don't have to do tire size adjustments and trying to compensate for tire growth.

ProCoach

Hah! That's another reason why I have this cringeworthy issue with whspd. People wouldn't measure their circumference and comparing drivers, car set up, et cetera on the same weekend between similar cars (if I wasn't the one responsible for maintaining and calibrating the systems) was a mess.