Automotive News / Industry / Future Vehicle Discussion Automotive news and discussion about upcoming vehicles

Honda: Our odometers roll too quick

Thread Tools
 
Search this Thread
 
Old Feb 20, 2007 | 07:10 AM
  #16  
2000GTP's Avatar
Registered User
 
Joined: Mar 2005
Posts: 12,312
From: Aurora, IL
Originally Posted by unvc92camarors
That'd be nice. You just gotta learn that a 1/4 tank really means "holy sh**, get to the gas station!"
Yeah, pretty much.

I can't believe the situation involving the odometers and how that could have sneaked through the cracks.
Old Feb 20, 2007 | 09:01 AM
  #17  
Evilfrog's Avatar
Registered User
 
Joined: Jun 2001
Posts: 750
From: Alton IL
Originally Posted by unvc92camarors
That'd be nice. You just gotta learn that a 1/4 tank really means "holy sh**, get to the gas station!"
I think my cobalt's is pretty linear. However; I still myself rushing for a gas station at the 1/4 marker. It's like breeded into me that that is time to get gas.
Old Feb 20, 2007 | 09:52 AM
  #18  
96_Camaro_B4C's Avatar
Registered User
 
Joined: Aug 2002
Posts: 3,650
From: Indianapolis, IN
Originally Posted by Evilfrog
I think my cobalt's is pretty linear. However; I still myself rushing for a gas station at the 1/4 marker. It's like breeded into me that that is time to get gas.
My Ion is pretty trustworthy as well. Same for my fiancee's L200. Not sure about my parents' Malibu or Equinox.

I'm fairly sure they've gone to a more linear approach, though, generally speaking. My Camaro is easily the worst offender I've seen, though my old '94 Caprice LT1 seemed to have a similar behavior.
Old Feb 20, 2007 | 10:49 AM
  #19  
Buttercup's Avatar
Registered User
 
Joined: Jun 1999
Posts: 939
From: Lowcountry
Originally Posted by Threxx
though if deliberate would **** me off
There's no question that it's deliberate. They are consistently off across all models and years. That's not random error, that's bias error. These are all digital controls with minimal random error and this bias error is therefore a controlled one.

The only question is who dictated this and why. We can only guess why they chose it. Many of the probable reasons for this "policy" do not show the best of intentions.
Old Feb 20, 2007 | 10:56 AM
  #20  
Threxx's Avatar
Registered User
 
Joined: Sep 1998
Posts: 4,320
From: Memphis
Originally Posted by Buttercup
There's no question that it's deliberate. They are consistently off across all models and years. That's not random error, that's bias error. These are all digital controls with minimal random error and this bias error is therefore a controlled one.
The digital controls still rely on a multitude of analog inputs to read correctly and if those analog inputs were designed or measured incorrectly then that could explain why.

Though I will say if anything makes a case against Honda it's that this went on for so LONG - they HAD to have eventually noticed this range of error, and consciously decided to do nothing about it.
Old Feb 20, 2007 | 11:06 AM
  #21  
graham's Avatar
Registered User
 
Joined: Feb 1999
Posts: 2,887
From: northeast Miss.
Its a pulsing sensor on the tailshaft half of the trans. And like a CD its an either/or readout. IIRC the pulses are called clicks and the pcm uses these clicks for a preset formula to twist the speedo needle. Which is why GM vehicles have been dead on since the early 90's. Save for tire pressure and wear of course.


(this post may be +/- 4% correct)
Old Feb 20, 2007 | 11:17 AM
  #22  
Buttercup's Avatar
Registered User
 
Joined: Jun 1999
Posts: 939
From: Lowcountry
Originally Posted by Threxx
The digital controls still rely on a multitude of analog inputs to read correctly and if those analog inputs were designed or measured incorrectly then that could explain why.
This is the internet so you can take it for what it's worth but I am familiar with controls. There is always some noise and even a half-*** engineer does his best to eliminate it. This does not appear to be random error from noise but bias, any other engineer will tell you this. You don't even need an engineer, any statistician will also tell you this. This is a text book example of bias error, at least based on the reports that we are seeing. There is still some random error and always will be but this certainly appears to be predominantly bias. Bias is controlled, especially when dealing with such a highly engineered product as cars.

You aren't going to tell me that a company such as Honda which is pretty damn good at robust engineering just happened to accidentally screw up something as simple as odometer calibration and do it by the same amount across all models for many years?
Old Feb 20, 2007 | 12:17 PM
  #23  
poSSum's Avatar
Disciple
 
Joined: Nov 2000
Posts: 1,479
Originally Posted by rlchv70
Whether this was done to reduce waranty repairs is up for debate. Mathematically, it would reduce warranty costs by up to 4%.
It would also increase calculated MPG by the same percentage.
Old Feb 20, 2007 | 12:20 PM
  #24  
96_Camaro_B4C's Avatar
Registered User
 
Joined: Aug 2002
Posts: 3,650
From: Indianapolis, IN
Originally Posted by poSSum
It would also increase calculated MPG by the same percentage.
Which is not trivial. If your Civic is really getting 30 mpg, you'd be calculating it as 31.2...
Old Feb 20, 2007 | 01:02 PM
  #25  
OctaneZ28's Avatar
Disciple
 
Joined: Mar 2000
Posts: 873
From: Chicagoland, IL www.5thGen.org
Originally Posted by unvc92camarors
That'd be nice. You just gotta learn that a 1/4 tank really means "holy sh**, get to the gas station!"
Hmm, my cars are the opposite way....
Needle on E means I can go 40 miles safely yet!
Old Feb 20, 2007 | 01:08 PM
  #26  
Z28x's Avatar
Registered User
 
Joined: Jan 2000
Posts: 10,285
From: Albany, NY
LOL.... Honda Mile = Metric system
Old Feb 23, 2007 | 08:37 PM
  #27  
3SuperSports's Avatar
Registered User
 
Joined: Jan 2001
Posts: 164
The only problem with reasoning that they'd do it on purpose in order to end the warranty quicker, is that Honda has extended the warranties on parts they found to be faulty in the past. The EGR valve (which was replaced on my wife's 2001 Accord around 75,000 mi.) and the transmission, which we've had no problems with.
Old Feb 23, 2007 | 08:45 PM
  #28  
landstuhltaylor's Avatar
Registered User
 
Joined: Jun 2006
Posts: 199
From: Methuen/UMass Dartmouth
it would be interesting to know how much the average motrcycle is off
Old Feb 23, 2007 | 08:55 PM
  #29  
bombebomb's Avatar
Registered User
 
Joined: May 2006
Posts: 1,855
From: Ohio
My gas gauge is all jacked up in the camaro, when its full it reads full and a quarter, then once it starts moving it just screams down to empty, then it can be below empty and still be running on what i asume were fumes.
Old Feb 23, 2007 | 09:46 PM
  #30  
Slappy3243's Avatar
Registered User
 
Joined: Jan 2001
Posts: 1,398
From: Fairfax Station, VA. Formally Long Island :(
I kind of wonder about my GTO sometimes. I mean, I am fairly certain GM recalibrated the sensors and computer to properly measure in miles but it is a car that natively uses the metric system in Australia. I am probably just being silly but you never know.

I also have the article from Car and Driver (I think so anyway) where they did an accuracy test between brands. Japanese imports were always the most far off. Honda, Nissan, and Toyota were the worst I believe followed by the German cars. GM was the extremely accurate with Ford close behind it.



All times are GMT -5. The time now is 10:02 PM.