brake

«1

Comments

  • WOW! That's great video! Is that chasecam software used for overlay? What camera equipment did it have?
  • And what software used for the speedometers, G-meter, etc? That's not racechrono2avi it is nicer!!
  • edited September 2009
    Don't forget RaceChrono2AVI is free! This one is probably a commercial one (they tend to be expensive). I think it's ChaseCam software. But of course there is something to learn and copy here never the less :)
  • I'm trying the 30 days trial version of ChaseCam DashWare and yes, it is make with that software. The problem is I can't open .csv files, it say

    Data Logger File Invalid
    Error reading data logger file 'C:\Document and Settings\Administrator\Desktop\20090905-1247-all-laps-export.csv'

    Some time ago I have tried to open a .CSV with another program (Bob's track builder) and can't to open, so I think something it is not good in these .CSV

    PS:Sorry offtopic
  • edited September 2009
    I too am playing with the demo version of DashWare. You have to delete the header lines off of the CSV export. Start with the 1st line above the data and include all of the data and it will work.

    OP - How did you approximate Brake Pressure? I am thinking about adding a column to the CSV that will map max negative long G to 100% brakes and map negative values on that scale.
  • edited September 2009
    I was playing with DashWare I love it, only had a problem with G-meter, the lateral G it fixes on 45 and don't move, longitudinal G works fine.

    The trial can't to create a video, I thougt to buy is too expensive 149$ is exaggerated :(

    rickerbr the brake/throttle is in reality the G-Longitudinal Acel. so this g-meter bar take data from G-force.
  • Click on the icon that looks like a power plug in the gauge pane. I suspect the latititudinal Gs are mapped to the wrong column. I looked at the OPs video again and it looks like they mapped any negative longitudinal Gs to 100% brake, as it is all or nothing.
  • Nothing to do with lateral G... It is selected correctly but for a reason don't read fine the lateral Gs

    For brake/throttle I make 2 rectangles and maped:

    Brake, min -0.2 max -1
    Throttle, min 0 max 0.2

    So the results is how I remember that touch pedals, also this max/min depends on car and you have to adjust.
  • edited September 2009
    Throttle Pos % I can get from OBD. I ended up just using long (G (if -)/max long G) * 100. That way the heaviest braking in a given session/log is 100% and everything else is a fraction of that. I may add the if < -0.2 (or so) instead of 0 so you don't get some of the noise of false brake indication during a coasting period.

    Software looks good (custom gauges etc). $149 is cheaper than traqmate, but still plenty spendy. Some of the folks on the forum indicate that traqmate's traqstudio is buggy and they charge for upgrades etc.
  • edited September 2009
    I do that with Dashware, I capture the video with another program (Camtasia)

    http://www.youtube.com/watch?v=5m-JGgYLCAw

    It is really intuitive and easy to overlay, and for sync video with data
  • edited October 2009
    nice video..it seems you picked all this up very quick. Your acceleration and braking bars seem excellent!!!...I doubt you have obd2 like I use for throttle so you are deriving this from the gps???...nice job! Can you also shed some light on how you did it???
  • edited September 2009
    I did it in excel and then exported the sheet as CSV (which converts the formlas to absolute values).

    The formula in excel looks like this

    =IF(N1 < 0, ABS(N1/MIN(N1:N65535) * 100), 0)

    Where N is the column that has longitudinal acceleration. Basically what this does is determine if the longitudinal acceleration is negative (deceleration), if so, get it as a percentage of the maximum deceleration that occurs in the exported data, if not, return 0. I just created a new column with a header of brake % and pasted that formula into the column at every row, and mapped the new column to the brake indicator.

    Obviously this is not going to give you absolute brake pedal pressure (can't do that without a pressure transducer, a modern ABS controller probably has this, but it is not exposed via OBD2), but it will indicate a proportional amount of braking effort to the maximum encountered during the export. For track use this is probably good enough, if you are logging street data, the low braking effort will cause a bunch of noise in the brake indicator (coast downs vs actual braking etc).

    I also added a column via excel to map OBD speed vs rpm into selected gear. You need to know the gear ratios and final drive ratio as well as tire radius (which is determinable from the size). I can post that formula later.

    I am working on using the mass airflow rate (MAF), rpm, intake air temp and displacement to get approximate manifold absolute pressure (MAP). My car is turbocharged (VW 2.0T) and has a MAP sensor but doesn't expose the MAP value via OBD2. I would like to add a boost gauge to the display with this calculated data.
  • edited October 2009
    .
  • I will post a spreadsheet on Monday (left my notes at work for the formulas) that will take in all the values and generate a lookup table to be used in excel to populate the gear. The way I calculate the gear doesn't require you to know speed@rpm in each gear, only the ratios and the drive wheel tire size. Once you have the lookup table for a car, you don't need to re-generate it unless you change tire size or gearing. This might be a good candidate for doing the gear calculation in the RaceChrono program for the export.
  • This reallt looks great.

    I have a good looking dash, but can not find a freeware program to merge the two videos. I guess I will have to bite the bullet and buy something.
  • while I have not tried this...it is free
    http://is.gd/3eoZf
  • I didn't realise you could use values other than 0 or 1 for brake. I should've tried it. That's something else for the wishlist:

    - Sticky gears (no neutral between gears)
    - Brake %, rather than on/off
    - Browse button for CSV files
    - Select what to modify (throttle normalisation, etc.)
    - Better documentation
    - Adjustable sensitivity (not just integers)
    - Only set brake flag if slowing for more than x rows
    - OBD columns offset to allow for lag in the OBD2 adapter

    When I get some time, I'll work on another version.
  • rickerbr...I would be interested in seeing your spreadsheet and how you formulated the gearing. I also think if we could merge all this great info from you and Andylaurence and make it available as a post process or better yet get it within rc will further enhance the product.
  • edited September 2009
    I am also working on gearing formulas given the range changes depending on acceleration vs de-acceleration.

    thresholds:
    X Increasing:
    1) 0-45
    2) 45-74
    3) 74-110
    4) 111-150

    X Decreasing
    4) 150-65
    3) 65-42
    2) 42-27
    1) 27-0

    Acceleration is determed by 3 consecutive data points
  • Gear calculation info and spreadsheet here -

    http://www.rickernet.net/racechrono/gearcalc.html
  • Good work, rickerbr. Have you tried this with real data? I don't have OBD2 in my car, so I don't have a speed/RPM input. In the data I've been given, I've had GPS speed and OBD RPM. There's a slight delay in the OBD2 data, so it's slightly out of sync, which leads to erroneous gear determination. I wonder if this is eliminated by using OBD speed, rather than GPS speed.

    Perhaps much of this could be done in RaceChrono? If we could build car profiles in the same way we can select drivers for sessions, we could set the gear ratios in the car profile and RaceChrono could determine the gear appropriately. Maybe something to add to the wishlist?
  • edited September 2009
    I've used real data from the commute to work to verify the results. Gear changes match changes in RPM with little to no change in speed. Comparing GPS Speed to OBD Speed (after converting to matching units) reveals a distinct lag in the GPS speed vs OBD Speed (even with a 5Hz GPS) but GPS speed should be good enough to get you close. The GPS speed is mostly off at very low speeds. But if you have the OBD reader for RPM, you might as well use the VSS (OBD Speed) channel as well.

    The only caveat I have with the method I posted is that I don't know what will happen with cars that the shift time is greater than the measuring period (200ms in my case). I'm 99% certain my method won't detect neutral when you put the clutch in or are between gears. So you might have a weird flicker/noise in the data during the shift itself. For me personally that isn't an issue because my worst case shift time is around 100ms for a downshift and <20ms for a upshift. I really only wanted to fill the gear value in for the F1 style gauge for Dashware.
  • I see. I had lots of issues with GPS speed and OBD RPM giving a varying gear ratio. I added a tolerance to deal with that but it's still not great. I used gear = 0 for neutral, which is the default until a gear is determined. Basically, I set the gear as neutral, then iterated through the ratios until one matched within a tolerance.

    My script also normalises the throttle data in case full throttle on your car shows as 93% instead of 100% in the OBD data and adds the brake flag (I didn't know it was a percentage) at -0.1g. I should really look at modifying that to a percentage based on maximum negative G. It's on the wishlist.
  • edited October 2009
    .
  • Try a small change in your formula.

    =IF(N1 < 0, ABS(N1/(Max(N1:N65535)) * 100), 0)
  • FOLLOW UP. I tried the original and mhoward's revised formula (thanks!) and was unsuccessful on the non-excel products mentioned above. I was finally able to do it on excel mac...so there is some difference between these spreadsheet apps.

    After I did get the brakes as % I ran it through dashware and got very erroneous results...so I went back to Andy's runtime with the on off flags. I'm sure it 's me on this too...anyone successful with getting an animated bar for brakes in dashware?

    Andy...can I come over the puddle and do yard work or pave your driveway or something to get the v2 going with the filter???...or the v2 with the %???

    ...or does anyone know a formula to apply to a column in excel that would change the erroneous random single and double brake flags of "1" to a "0" unless there are more than 5 in a row???

    tia,

    mr
  • edited September 2009
    Figured this out, there are some errors with my posted formula based on the test csv file

    Should be

    =IF(M12 < 0, ABS(M12/MIN(M$12:M$5921)) * 100, 0)

    Where M is the long accel column. This is for cell T12 which is the first line of data. The MIN function doesn't work if you set the end range past the last line in the file (which in this case is 5921).

    This will give percentages (and I tried it with both iWork '09 Numbers and Excel 2008 for Mac).
  • I've gone back to look at the VBS I wrote and modified it so it outputs braking as a number between 0 and 1. However, it doesn't import into RaceChrono2AVI. I notice that there's a column list in the RaceChrono2AVI forum and braking isn't included. Previously, I had the column heading as Brake Flag.

    Anyway v1.2 is now working, I just need to finalise what the column heading should be and whether the column should show 0-100 (%) or 0-1 (to two decimal places). The only thing I've added is the granular braking, the other wishlist items are still work in progress. The erroneous braking should be less obvious now as it'll just tickle the bottom of the gauge, rather than flashing the whole thing.
  • thanks rickerbr I will try it today...I have 17 runs to process through dashware....Andy...I could beta test that version today of it's running???

    thanks

    mr
  • edited October 2009
    Sorry for the delay. I tried to email you, but it bounced. Here's a newer version:



    http://www.bigupload.com/files/PRHMM9MV46/RCnormalise.1.2.zip.html
Sign In or Register to comment.