Announcement

Collapse
No announcement yet.

Driving Data -- Was MOWOG 9 Recap

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Originally posted by the tick View Post
    Ah yes, the "weight savings" vs "Structural integrity" compromise...
    Lolz.
    Jeff Rye #64
    1999 Mazda Miata Sport -- ES
    2001 Toyota Corolla S -- HS
    Proud recipient of the 2016 Paul Weidner Spirit of the Sport award.

    Comment


    • Jeff, A quick note... Place YouTube to run the video's at 0.25x speed then use cracks in the pavement as reference landmarks. Also do this on a PC so you can start/stop the video with a click on the spacebar.

      Now, this isn't going to happen between runs unless you have software to do so.

      Got to run.

      Comment


      • Sorry Guys I've been a WEE BIT testy this week. It's been a HELLish week with needing to put out fires Tuesday through Friday. And I have a packed Monday starring at me.

        Comment


        • Can also move frame by frame with the "<" and ">" keys when paused on youtube. I just think its also a neat thing I didn't know about until this year...and am sad to admit.
          Eric Janssen

          #320 1999 Miata & 2013 Honda Civic Si

          Comment


          • I hope this isn't too soon. I showed this to Barb before we bought the MX-5...

            Comment


            • Originally posted by ryejeff View Post
              I ask because I saw a *ton* of line variation in my own runs. Interestingly, the differences in my lines generally had very small impact on my time through the segment, but may have impacted prior or subsequent element.

              For instance, look at the attached ghetto screenshot of my line through the turnaround after the slalom. When I sync these at the last slalom cone and play them forward, the green line gets by 0.05 s entering the turn, but then goes wide. By mid-to-late corner, red is ahead by 0.109s. But red travels a little extra distance toward the next left hander and green makes up most of the time, for a total delta (in red's favor) of 0.015s at the apex. (BTW: at the apex of the left hander, the two runs are traveling within 0.1mph of each other and I accelerated out at WOT.)

              Click image for larger version Name:	IMG_1085.JPG Views:	0 Size:	93.8 KB ID:	477378
              Ok Guys & Gals I'm going to try to walk you through Jeff's SoloStorm data example of Turn 5 (180 left turn after the slalom of MOWOG 10). "Reading" the lateral G data trace in regards to driven line through a corner (early vs neutral vs late apex) is hard to do and the way it's displayed is counterintuitive on a data trace. It took me about 20 tries when I was learning this stuff before really understanding and gaining the skill to be able to glance at a lateral G trace and know if I early, neutral, or late apex a corner. So I understand if the following is super confusing.

              TLDR; Either the Lateral G trace is correct, or the driven line is correct. One of them is lying to you. (I'm going to step you through why).

              OK I'm going to take Serger's examples from his book and use them as a reference. Please refer back to them (I did over & over when learning this stuff).

              Click image for larger version

Name:	Screenshot (111).png
Views:	85
Size:	157.0 KB
ID:	477412

              Click image for larger version

Name:	Screenshot (110).png
Views:	83
Size:	223.8 KB
ID:	477413

              Click image for larger version

Name:	Screenshot (112).png
Views:	81
Size:	163.6 KB
ID:	477414

              Comment


              • We are going to first look at the orange/red line.

                Ok we are going to start at the point marked "3" on the driven line; this is also the point that the Lat G trace crosses to the zero G horizontal line. Following the Lat G trace, the trace comes to a plateau that is slightly increasing in Lat G. This looks like a neutral/mid corner apex (compare with the example) with a slightly increasing steering input because the plateau is also sloping to increasing Lat G. This also could be called a mixture of mid corner & early apex. Yes I know it's counterintuitive.

                The slope of the Lat G line going back toward zero G is about equal to corner entry so the driven line appears to be correct, a nice arc with a tiny bit of being pinched.

                Comment


                • Ok now we are going to look at the green line.

                  We are going to step through this in tiny chunks.

                  Again we are starting at the zero Lat G horizontal line. The green slope is slightly steeper than the orange/red line. This would be a tighter/earlier corner entry; which the driven line also shows.

                  See where the green Lat G trace and the orange/red Lat G trace diverge? There is a vertical line also at this point. See the "knee" on the data trace. See the placement of this "knee" in relation to the whole corner? Now look at the examples. The green data trace looks like a late apex corner entry. Also the green driven line looks like you turned in earlier which coincides with the Lat G trace.

                  OK This Is Extremely Important!!!!

                  Look at the plateau on the data trace; at mid corner the Green & Orange/Red traces converge. The radii of the driven line MUST be very similar (physics). Now look at the green driven line, see how the radius is pinched/smaller mid corner. If the green driven line is correct there has to be an increase in Lat G on the data trace. Not just a little bit; a lot.

                  So which is correct?

                  Also look at the Long G trace. The green trace has less negative (braking) acceleration, but the speed decreased more in a small place and overall equaled the Orange/Red trace. This makes no sense.

                  Comment


                  • OK 99.99% of you are thinking that I'm just geeking out on the data.

                    Again I give ZERO S.... about the data. I care 100% about driver inputs and how these inputs affect the car and the path it takes through the course.

                    If I am going to compare one driver input against another I need to know that the results I am recording are accurate. I'm sorry but I look at GPS based data and I see questionable data at the speeds & radii that we work with. I don't trust the driven lines.

                    All of this is a part the Scientific Method. Being skeptical; especially if the data doesn't support each other.

                    Comment


                    • Use the application you know (SoloStorm) and share data with your friends. Just get a better data source than GPS.

                      For $300 get a RaceCapture/Track, you have real accelerometers, a better GPS, and I can help you record off the car's CAN Bus for driver inputs (2008+), and you still use the SoloStorm software.

                      If you want the driver inputs on a older car, get a RaceCapture/Pro for $600 and you can add the sensors.

                      Comment


                      • Very interesting breakdown, Todd. Thanks!

                        It took me a lot of staring and rereading, but I think you might have the data backwards. I should have been clear that this was a run from Sunday (M10) -- we were going anti-clockwise. So the traces coincide with the end of the slalom, the tight right hander, and the entry to the subsequent left. Note that the vertical line denotes "now" -- where the "x"s are positioned on the GPS trace. And that "now" is after the point marked with the 3.

                        Here's a version of the image where I have marked the GPS and g traces with corresponding points.

                        Click image for larger version  Name:	IMG_1085-mid-corner.JPG Views:	0 Size:	95.8 KB ID:	477423


                        At these marked points, I reached the same peak g. The green trace had a smaller radius. This is reflected by a lower speed. Per physics!

                        After the apex marked by the blue regions, I was able to get on the gas a little earlier in the green run. We see this in the fall off of the lat g trace and the change in slope of the speed trace. This happened just early enough to make the speed trace match up with the red run, where I held a (slightly) higher speed in the corner but delayed my acceleration.

                        FYI: while I *could* have SoloStorm compute acceleration from the GPS traces, I actually have it take acceleration data from my tablet's accelerometers. These sensors are not as good as Race Capture's sensors, but they are a lot better than computed values. The tablet is fixed to the body of the car, so lateral (and longitudinal) g are affected by pitch and roll, surface camber, and bumps.
                        Jeff Rye #64
                        1999 Mazda Miata Sport -- ES
                        2001 Toyota Corolla S -- HS
                        Proud recipient of the 2016 Paul Weidner Spirit of the Sport award.

                        Comment


                        • Hi Jeff I think we are the only people that really care about "this" LOL.

                          A couple of things...

                          -Positive lateral acceleration is a right turn. So your circles on the data trace is near where the crosshairs are. Not the left turn where you circled on the driven line map.

                          -The vertical line on the data trace is in the left turn that you circled on the driven line map.

                          -I don't know what is going on with the driven line map but the cross hairs do not align with the vertical line on the data trace. The crosshairs are aligning with the blue circles that you drawn. I tried to copy what you are showing in SoloStorm with MoTeC and you can see where the red dot is aligning on the data trace (yellow vertical line).

                          Click image for larger version

Name:	Screenshot (113).png
Views:	60
Size:	1,010.7 KB
ID:	477425

                          Comment


                          • Originally posted by Todd Giencke View Post
                            Hi Jeff I think we are the only people that really care about "this" LOL.
                            Lolz indeed!

                            If you'll allow me to clarify. I wasn't offering a hypothesis about the SoloStorm interface. I was clarifying what it is actually showing, since it looks different than your version.

                            Originally posted by Todd Giencke View Post
                            A couple of things...

                            -Positive lateral acceleration is a right turn. So your circles on the data trace is near where the crosshairs are. Not the left turn where you circled on the driven line map.
                            1. The sign for the lateral acceleration is consistent through a run, but can vary from run to run depending on whether I put the tablet into position with the sleep button toward the front or the toward the back of the car.

                            2. I circled a *right* turn. My map is North up. In the actual run we entered the visible portion of the map on the top left (coming through the slalom), slowed and made a right turn, and accelerated Southwest to set up for a left hand sweeper.

                            Originally posted by Todd Giencke View Post
                            -The vertical line on the data trace is in the left turn that you circled on the driven line map.

                            -I don't know what is going on with the driven line map but the cross hairs do not align with the vertical line on the data trace. The crosshairs are aligning with the blue circles that you drawn. I tried to copy what you are showing in SoloStorm with MoTeC and you can see where the red dot is aligning on the data trace (yellow vertical line).
                            It's not a hypothesis. The vertical line is actually part of the SoloStorm interface and identifies the point in the line graphs that corresponds to the "x"s on the map. That's just part of the SoloStorm interface!


                            At any rate, the main purpose of my posts was to point out that my GPS traces show a noticeable difference in my line choice. And, as an interesting aside, in this case the line difference ended up being very nearly a wash (only 0.05s difference).
                            Jeff Rye #64
                            1999 Mazda Miata Sport -- ES
                            2001 Toyota Corolla S -- HS
                            Proud recipient of the 2016 Paul Weidner Spirit of the Sport award.

                            Comment


                            • I'm finding the SS screenshots & explanations helpful. Great to see & hear how people are using the tool.

                              Comment


                              • Originally posted by Faust View Post
                                I'm finding the SS screenshots & explanations helpful. Great to see & hear how people are using the tool.


                                FWIW: I do not always include longitudinal and combined g. Yeah, those could give me some information about where I am on the traction circle. But more often than not, the traces that I rely on for insight are the lateral g, speed, and time delta.

                                And, BTW, you can easily configure SoloStorm to include/exclude various traces. There are many choices beyond what I use.
                                Jeff Rye #64
                                1999 Mazda Miata Sport -- ES
                                2001 Toyota Corolla S -- HS
                                Proud recipient of the 2016 Paul Weidner Spirit of the Sport award.

                                Comment

                                Working...
                                X