User Tag List

Page 3 of 4 FirstFirst 1234 LastLast
Results 41 to 60 of 72

Thread: Speedo giving incorrect reading

  1. #41

    Default

    Quote Originally Posted by LittleredMPS View Post
    From what I have read, car manufacturers only have to be within 10% of actual speed.

    To combat this nearly all manufacturers set their cars up to the lower side to avoid customers complaining about speeding fines.

    Mazda sets theirs pretty close to 10% under on most of mazda range from what I know.

    As far as it being a set rate of difference instead of a percentage...well...I have no explanation for that one.
    To be honest I'm pretty pissed off about this whole speedo thing not reading correctly. I've had my BK MPS 3 for less than a week and noticed pretty quickly that the speedo is reading more than the actual speed (it's very obvious). I've rang Mazda about this before and they say they've done this on purpose for regulations, etc (can't remember exactly what she said).

    If you're anything like me then I usually do 5 to 10kph more than what the speedo tells me so really, I run the risk of a speeding ticket. They should've just got it fairly accurate at the start so we know exactly how fast we're going.

    Anyway I'm taking the car into Mazda for a warranty job tomorrow so I'll definately mention this to them and see what they say. Hopefully they can do something about it but I don't like my chances.

  2. #42

    Default

    I took a slightly different route.

    When I changed my rims to 19" they had 225/40 tyres on them just like my stock 18's.

    I worked out the percentage increase...which is about 4.5% and then ran it through a roadworks speed camera sign board.

    It read exactly the same as my speedo at 60kmh.

    That was good enough for me

  3. #43
    Join Date
    May 2009
    Location
    Cowra, NSW
    Age
    74
    Posts
    684

    Default

    Speedo error is pretty common on all cars, from my experiences driving a variety of company cars with a variety of different satnavs the speedo's are all are out. I think a Mitsubishi 380 was the only hire car that was accurate. Fords are the worst by far. From my work to Forbes ( which I did for a year three times a week) it was 92 kilometres in a VY Commodore and 98 kilometres in a BA Falcon. Not the same VY or BA either, different cars with still the same error. That was door to door same route exactly. Go figure!!!!

    I run 235/40/18's on my MPS and the speedo about 4km out at 100kph, at 100kph satnav the speedo reads 96kph. My son's MPS3 runs 225/40/18's and is much worse about 10kph out at 100kph. When he had my standard MPS 6 wheels and tyres on his car the speedo was almost spot on.

    Standard tyres are 235/45/18 so with aftermarket tyres and an already reading high speedo "is it no wonder there is a huge speedo error"

  4. #44

    Default

    Okay so I went into Mazda this morning and enquired about the inaccurate speedo. No luck, they said it was normal and if it's under 10% then they can't do anything about it.

    I guess I'll have to live with it. Funnily though when I mentioned this at the receiption, another customer said she gets the same problem as well.

    So is the speedo inaccurate by percentage or a fixed speed (eg. always 5-10kph difference)?

  5. #45
    Join Date
    Nov 2008
    Location
    Craigieburn
    Age
    50
    Posts
    1,642
    Blog Entries
    1

    Default

    I've found the MPS3 error at 100K/h to be about
    9% on 215/45R18 (OEM)
    7% on 225/45R18

    And that it seems to be a steady 7Km/h error across the board on 225/45 that is 60 indicated 53 actual, 110 indicated, 103 actual.

    Soon I will be able to give to give some real data as I'll be comparing the ECU speed reported vs the speedo guage vs speed calculated over a 5Km speedo test trap on cruise at 100K/h ECU reported....

  6. #46
    Join Date
    May 2009
    Location
    Cowra, NSW
    Age
    74
    Posts
    684

    Default

    Wouldn't it be simpler to use Satnav.

    Maybe one day in the future big brother will use Satnav to catch speeding drivers - can be done very easily!!!!!

  7. #47
    Join Date
    Nov 2008
    Location
    Craigieburn
    Age
    50
    Posts
    1,642
    Blog Entries
    1

    Default

    Quote Originally Posted by rd415 View Post
    Wouldn't it be simpler to use Satnav.

    Maybe one day in the future big brother will use Satnav to catch speeding drivers - can be done very easily!!!!!
    Will be cross referencing that too. If you're going to test accuracies I'd be relying on distance/time measurements - stopwatch - any day over a satnav.

  8. #48
    Join Date
    May 2009
    Location
    Cowra, NSW
    Age
    74
    Posts
    684

    Default

    Quote Originally Posted by Nexus View Post
    Will be cross referencing that too. If you're going to test accuracies I'd be relying on distance/time measurements - stopwatch - any day over a satnav.
    Disagree - I have two satnavs that I use regularly - one on my mobile phone and relatively new MIO 360, both give exactly the same readings of speeds/distances even when used on a variety of cars. I use the satnav in my 1924 Dodge as an accurate speedo.

    I rallied an Ford Escort for a while and used a RTA marked distance to set the trip computer - one specifically designed for testing taxi meters. After the first days rally I was continually about 300 metres out over a 5km distance. Knowing a satnav can pinpoint distances within 30m - good ones down to a metre, I'd rather trust a satnav than a stopwatch anyday.

  9. #49
    Join Date
    Dec 2008
    Location
    Carindale
    Age
    37
    Posts
    2,043

    Default

    Quote Originally Posted by eq View Post
    tyres will effect it to - i'm running 225/40 rather than 215/45.

    Speedo say 66ks, gps says 60ks....made me look like a liberty driver (grandpa)! Lol
    dito!!!

  10. #50
    Join Date
    Nov 2008
    Location
    Craigieburn
    Age
    50
    Posts
    1,642
    Blog Entries
    1

    Default

    Quote Originally Posted by rd415 View Post
    Disagree - I have two satnavs that I use regularly - one on my mobile phone and relatively new MIO 360, both give exactly the same readings of speeds/distances even when used on a variety of cars
    I'm sorry, but you seem to have missed the point.
    time vs distance is a very simple measurement and calculation to make over which I have full control of the variables.
    satnav I have various latencies etc that will creep into the data.

    At the end of the day it is not appropriate to use anything but time vs distance as a baseline to compare from. Perhaps satnav will show very accurate results, but that's not the point.

  11. #51
    Join Date
    Nov 2008
    Location
    Brisbane
    Age
    47
    Posts
    3,387

    Default

    Quote Originally Posted by pHiLix View Post
    Okay so I went into Mazda this morning and enquired about the inaccurate speedo. No luck, they said it was normal and if it's under 10% then they can't do anything about it.

    I guess I'll have to live with it. Funnily though when I mentioned this at the receiption, another customer said she gets the same problem as well.

    So is the speedo inaccurate by percentage or a fixed speed (eg. always 5-10kph difference)?
    Grab yourself a Scangauge or Dashhawk. I use a Scangauge mainly for the speedo function. The feed straight out of the ECU is pretty spot on.

    ScanGaugeII - Trip Computers + Digital Gauges + Scan Tool

    DashHawk Vehicle Information Display

  12. #52
    Join Date
    May 2009
    Location
    Cowra, NSW
    Age
    74
    Posts
    684

    Default

    I used the time over distance calculations often, as a Auto Teacher it was the standard accurate measurement taught to the students. The big problem is that it is a "calculation" which can easily be misinterpreted.

    I taught Dyno testing to stage 3 students and used the dyno regularly to test speedo's and also the usual power outputs, even emission for a while for our local RTA. The point is that even the dyno is not that accurate. There is only one dyno in NSW that is accurate, the one in "polution control commission" that has a sealed atmospheric testing room with no possible variables.

    Stop watches have a major problem of the delay period caused by brain hand co-ordination. All of the variables of using a stop watch and a measured distance are eliminated with a good satnav.

    How do you accurately measure the distance? - off somebodies speedo or the RTA signs, which neither are accurate.

    If it works for you use it, but technology has surpassed the old calculation method.

  13. #53
    Join Date
    Nov 2008
    Location
    Craigieburn
    Age
    50
    Posts
    1,642
    Blog Entries
    1

    Default

    Quote Originally Posted by rd415 View Post
    If it works for you use it, but technology has surpassed the old calculation method.
    Are you joking, or did you not think before stating that? All of those technologies are calculation methods. All of them. The simplest one, which is the most appropriate for measuring a baseline is time vs distance. End of story.

    That's my stance from a science and analysis point of view. Use the simplest measurement method for the baseline.

    We don't triangulate the speed and location of a racecar to see who crossed the line first, we measure distance over time.

    Yes, there is a presumption that a 5K trap placed specifically for testing speedo accuracy actually is accurate to 5K. If I find it is not accurate, which would become apparent in testing, then I would be lodging a complaint with main roads.

    I'd be very surprised indeed if the 1/5 --> 1/10 second hand stopwatch latency is higher than GPS.

    I think people have too great a reliance of technology these days. You're using complex methods to measure something simple. We're not talking about measuring the time to travel 100m. We're talking about 5K here. 1/5 of a second hand operated stopwatch latency is going to be very negligible impact on the stats.

    Furthermore, please explain exactly what is to be misinterpreted about measuring distance / time???? I've travelled 100Km in one hour therefore I've averaged 100Km/h....the only misinterpretation I can see is if someone assumes that you are travelling at a constant speed when you were not. In this case we will be deliberately travelling at a constant speed +/- small margins...

    Yes, distance / time calculation works for me because that's how we measure speed. Km/h. Think about it. This is why it is the scientific baseline method.

  14. #54
    Join Date
    May 2009
    Location
    Bathurst
    Age
    42
    Posts
    1,930

    Default

    Speed over time as a mathamatical equasion is all well and good, but how do you measure distance accurately?

  15. #55
    Join Date
    May 2009
    Location
    Cowra, NSW
    Age
    74
    Posts
    684

    Default

    All good fun, but I will stick with what I have proven to be accurate and stay away from so-called accurate measured distances - I have been caught out, it is no fun using a trip computer to navigate when it is not accurate - approaching a what is a slow corner in 500 metres, to find it's another 500 metres down the road doesn't make for good section times. Purely due to an inaccurate RTA Taxi Meter calibration distance.

    Did you know that aircraft are going to satnavs in 2010 as the current beacon system isn't working well.

  16. #56
    Join Date
    Nov 2008
    Location
    Craigieburn
    Age
    50
    Posts
    1,642
    Blog Entries
    1

    Default

    Quote Originally Posted by bd581 View Post
    Speed over time as a mathamatical equasion is all well and good, but how do you measure distance accurately?
    Good point. I find the (admittedly fairly likely) idea that a test trap made for testing isn't accurate kind of peverse.

    Don't get me wrong, I'm going to do this just to satisfy my own curiosoty as much as anything else.

    I will be using a BU-353 GPS reciever and copilot laptop for the satnav portion. Perhaps the first thing to do is see if the speed trap and satnav agree on distance then, perhaps?

    distance over time, actually.... :-)
    Last edited by Nexus; 10-11-2009 at 09:58 AM.

  17. #57
    Join Date
    Nov 2008
    Location
    Craigieburn
    Age
    50
    Posts
    1,642
    Blog Entries
    1

    Default

    First cursory tests of the GPS system yesterday

    CoPilot + BU-353 seems to report about +2.5Km/h higher than other GPS I have used before, and contradicts the expectations we have from previous observations. According to this satnav, we have a speedo error of between 3 and 4%

    Bullshit.

    When you take speed out of the equation and sit still it gets a good fix that does not wander, but is incapable of calculating speeds below 8Km/h and on the road appears to report high.

    Gatton bypass speedo test trap, again only cursory checks, but appeared to be a close match for the GPS distance reported. My first estimate is that the speedo test trap will be accurate within 1%.

    +2.5Km/h on the GPS is effectively a 2.5% margin of error on the GPS at 100Km/h.

    under 50M error over 5Km on the speed trap, with negligible timing latencies on a stop watch (0.2sec out of 180sec) will show an error of less than 1%

    Really, I would like to see the baseline measurements have an error margin of 0.5% or less.

    First indications are that the trap will be more accurate than the GPS. That's not empirical data, that's just first observations.

    ---------- Post added at 12:00 PM ---------- Previous post was at 11:29 AM ----------

    I must say it's nice having a computer to do route recalculation though - so much faster than dedicated GPS units and much more configurable :-)

    Plan out a whole road-trip holiday....

  18. #58
    Join Date
    May 2009
    Location
    Cowra, NSW
    Age
    74
    Posts
    684

    Default

    Good to hear your observations. Very interesting. I would like to see other satnav units used (using Navteq & whereis programs) as well as other measured distances - just to make it a bit more of a controlled test.

  19. #59
    Join Date
    Nov 2008
    Location
    Craigieburn
    Age
    50
    Posts
    1,642
    Blog Entries
    1

    Default

    Quote Originally Posted by rd415 View Post
    Good to hear your observations. Very interesting. I would like to see other satnav units used (using Navteq & whereis programs) as well as other measured distances - just to make it a bit more of a controlled test.
    The hard part is a dead straight 5K control distance.
    I could use the BU-353 with other software for comparison. I don't expect the antenna would be a problem. It gets good fix across a dozen satellites with it placed on the roof, and radio waves don't travel any faster or slower to different antennas - the time latencies converting that to a serial data stream and processing that stream can. Another GPS reciever would be good for comparison...

    The Gatton bypass test trap is nearly straight but undulates.

    Oakey --> Dalby would be an awesome spot for good accurate test trap; there's at least one stretch that's dead straight without undulations and loong...

    This is the thing with measured controls - 1g for instance is defined by a control weight stored in sealed environment somewhere. I forget where it's kept. Ultimately it's an arbitrary definition.

    I can measure out 5K by walking it out with one of those counter wheels, driving it according to the odometer, marking with a GPS, laser rangefiner....

    and I'll get four different results. Statistically doing it over and over with certain methods will show the results converging on an accurate location. Presumably the laser rangefinder would be the most accurate over a straight run. So much for low tech and...who was the last person to calibrate it?

    All good reasons for main roads test traps to be accurate control measurements.

  20. #60
    Join Date
    Nov 2008
    Location
    Craigieburn
    Age
    50
    Posts
    1,642
    Blog Entries
    1

    Default GPS still doing calibration?

    Um. Noticed on the return trip that the speed reported by copilot is now closer to expections. -6K on speedo reading.

    It took about 30minutes the first time I started copilot to get it's initial data ready and sync up with satellites; that's normal so I am wondering if this first real-world use 600K round trip may have been used by the software to adjust the calibration?

    It's also the first GPS solution I've had that gets 9 satellites indoors. Perhaps I've only had pov solutions before, but indoors is usually no fix?

    Anyway, back on topic soon. Designing an OBD dashboard at the moment, so next will have an ECU speed reading. For each wheel if I want to.

Page 3 of 4 FirstFirst 1234 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •