Jump to content


Photo

Rough fuel usage estimate from speed, weight, elevation and MPG?


  • Please log in to reply
17 replies to this topic

#1 ivanalesi

ivanalesi
  • Member

  • 1,767 posts
  • Joined: August 04

Posted 17 May 2012 - 08:28

Hi guys,

A bit unusual question, but I ask here because there are lots of engine guys with automotive industry experience.
So, we have the speed + elevation from GPS and the normal fuel usage data + weight of the vehicle supplied by the manufacturer. No rpms or throttle position.
How do you think one can make a very rough estimate on the current fuel usage on the fly?
I guess whatever is the formula will need lots of tweaking to be done and I was thinking about using Rfactor for it to simulate driving styles and so on. It will be lots of work I guess:)

Thanks in advance!:)

Advertisement

#2 Greg Locock

Greg Locock
  • Member

  • 4,498 posts
  • Joined: March 03

Posted 17 May 2012 - 10:42

So yoo have speed/height vs time, and some numbers from the manufacturer? what numbers?


#3 munks

munks
  • Member

  • 344 posts
  • Joined: January 03

Posted 17 May 2012 - 14:25

It sounds like you're just talking about the city/highway mileage estimates plus the published weight. At best, the highway mileage might be slightly useful as an estimate of steady-state driving around which you can make some other estimates.

#4 ivanalesi

ivanalesi
  • Member

  • 1,767 posts
  • Joined: August 04

Posted 17 May 2012 - 21:33

Yes, correct:( Only city, mixed and highway MPG from the manufacturer.
When it's constant driving w/o stop and go, like on highway, that's easy. Problem is acceleration in the city and I thought about getting some good real data from the different classes of cars and then adjusting guesstimates of the other cars to the real collected data.

#5 Magoo

Magoo
  • Member

  • 2,461 posts
  • Joined: October 10

Posted 17 May 2012 - 22:09

http://www.epa.gov/o...m/420p05001.pdf

http://scholar.lib.v...tricted/ETD.pdf

http://dspace.mit.ed....pdf?sequence=2




#6 Greg Locock

Greg Locock
  • Member

  • 4,498 posts
  • Joined: March 03

Posted 18 May 2012 - 00:25

http://www.epa.gov/o...m/420p05001.pdf

http://scholar.lib.v...tricted/ETD.pdf

http://dspace.mit.ed....pdf?sequence=2


That EPA one looks pretty usable, but ignoring WOT rather throws the baby out with the bathwater.

#7 desmo

desmo
  • Tech Forum Host

  • 12,922 posts
  • Joined: January 00

Posted 18 May 2012 - 04:10

I very seldom use WOT. Almost never. I doubt I am too far off the norm in that respect.

#8 Greg Locock

Greg Locock
  • Member

  • 4,498 posts
  • Joined: March 03

Posted 18 May 2012 - 04:31

I very seldom use WOT. Almost never. I doubt I am too far off the norm in that respect.

I once looked at customer CANBUS data for 2000 hours of real people driving on real roads. Hardly any of that data was over 3500 rpm, and very little was WOT. There was no instance, ever, of anyone using the full power of the engine, ie 5500 rpm WOT.

Now admittedly you don't actually need to fully depress the throttle pedal to get engine WOT (the rest is used for the transmission schedule), but that was pretty definitive.

Now, if you looked at the data from the development engineers then we are a bunch of pedal mashers from way back.

Edited by Greg Locock, 18 May 2012 - 04:32.


#9 desmo

desmo
  • Tech Forum Host

  • 12,922 posts
  • Joined: January 00

Posted 18 May 2012 - 04:47

That sounds right, I never go much over 4000 rpm with another 2000 under redline, and then only accelerating onto a highway on a short on ramp because I like to enter the flow of traffic going a little faster than traffic. I drive an old Honda Civic with a throttle cable and a 5 speed manual so I don't see WOT unless I mash the pedal down.

#10 Magoo

Magoo
  • Member

  • 2,461 posts
  • Joined: October 10

Posted 18 May 2012 - 10:14

The links were posted merely to show the sort of things being done. One can take it from there as one desires.

...Back in the day, decades before PCs and modeling, there was a guy over at Chrysler Highland Park Engineering with a pepper mill calculator and a stack of white paper on his desk. George Wallace. They would ask him stuff like what would another 7 hp do for the Plymouth Superbird at Talladega in mph on a cloudy day, or how many tons of HVAC does the Apollo Lunar Excursion Module require in March as opposed to September. He would string together these crazy calculations that looked like Drake's equation and his answers would come out within one percent. A legend. I was at an SAE presentation a few week ago by the original Ramchargers where they were reminiscing about some of his feats. They relied upon him a great deal.

Edited by Magoo, 18 May 2012 - 10:17.


#11 Kelpiecross

Kelpiecross
  • Member

  • 855 posts
  • Joined: October 10

Posted 18 May 2012 - 11:05

That sounds right, I never go much over 4000 rpm with another 2000 under redline, and then only accelerating onto a highway on a short on ramp because I like to enter the flow of traffic going a little faster than traffic. I drive an old Honda Civic with a throttle cable and a 5 speed manual so I don't see WOT unless I mash the pedal down.


If you drive a manual car fitted with a vacuum gauge you will find that it is surprisingly easy (especially by "lugging" the engine in a higher gear) to get full intake manifold pressure on quite small throttle plate openings - meaning that even if you did "mash the throttle" and fully open the throttle plate, the manifold pressure could not go any higher so it wouldn't make any difference to acceleration etc.

Even "blipping" the accelerator in neutral momentarily causes the manifold pressure to go to maximum.



#12 munks

munks
  • Member

  • 344 posts
  • Joined: January 03

Posted 18 May 2012 - 20:12

... you don't actually need to fully depress the throttle pedal to get engine WOT (the rest is used for the transmission schedule) ...


I didn't actually know this ... is this also true for manual trannies? Have they always done this for automatics?

#13 Greg Locock

Greg Locock
  • Member

  • 4,498 posts
  • Joined: March 03

Posted 19 May 2012 - 01:16

I didn't actually know this ... is this also true for manual trannies? Have they always done this for automatics?



It's an outcome. The airflow rate at say half the redline rpm is roughly half what it is at full power, roughly. So the butterfly only needs to be 'half' open. and as you get lower in rpm it is even worse

That's why you usually have a separate idle circuit, the main butterfly is just too coarse to be able to control the tiny airflow needed at idle.

The relationship between pedal motion and butterfly opening is also non linear, for instance mercedes used to have a very gentle take up, great for fuel economy but makes the car feel lethargic.

As kelpiecross mentioned MAP is the thing, everything else is just a way of creating it.

I'd guess manuals are similar but they might not use the same snail cam to get the pedal progression.


#14 Canuck

Canuck
  • Member

  • 1,643 posts
  • Joined: March 05

Posted 19 May 2012 - 02:21

I once looked at customer CANBUS data for 2000 hours of real people driving on real roads. Hardly any of that data was over 3500 rpm, and very little was WOT. There was no instance, ever, of anyone using the full power of the engine, ie 5500 rpm WOT

By that measure (and a few others I'm sure), I'm abnormal. My old BMW doesn't even wake up until 3500ish, but then falls on it's face at 5200. It feels quick for a day or two after I've spent a couple of weeks on the bicycle, but is soon replaced by reality and it's tired 185ish hp. :p

#15 Magoo

Magoo
  • Member

  • 2,461 posts
  • Joined: October 10

Posted 19 May 2012 - 13:07

With ETC and throttle-less SI, a good part of this is becoming ancient history.

#16 ivanalesi

ivanalesi
  • Member

  • 1,767 posts
  • Joined: August 04

Posted 19 May 2012 - 17:53

Thanks for all the suggestions, enjoying the discussion and gives me guides for potential loopholes that I may encounter. I just want to pass some trouble on Monday and then concentrate on this work:)

By that measure (and a few others I'm sure), I'm abnormal. My old BMW doesn't even wake up until 3500ish, but then falls on it's face at 5200. It feels quick for a day or two after I've spent a couple of weeks on the bicycle, but is soon replaced by reality and it's tired 185ish hp. :p


I also feel better on higher revs with optimum torque, many are the opposite and it all makes it so difficult to make a good guesstimate, so one possible solution is to differentiate aggressive, normal and cautious drivers. I also logged some different driving and the differences are quite big in rpms to shift, sometimes at more than 20% difference!

#17 munks

munks
  • Member

  • 344 posts
  • Joined: January 03

Posted 21 May 2012 - 00:14

It's an outcome. The airflow rate at say half the redline rpm is roughly half what it is at full power, roughly. So the butterfly only needs to be 'half' open. and as you get lower in rpm it is even worse


OK, thanks for the clarification and additional information, I sort of misunderstood your first statement (I was taking WOT literally even if the same airflow could be achieved with 'half' of wide open).

#18 Lee Nicolle

Lee Nicolle
  • Member

  • 5,762 posts
  • Joined: July 08

Posted 23 May 2012 - 11:25

By that measure (and a few others I'm sure), I'm abnormal. My old BMW doesn't even wake up until 3500ish, but then falls on it's face at 5200. It feels quick for a day or two after I've spent a couple of weeks on the bicycle, but is soon replaced by reality and it's tired 185ish hp. :p

Put the cam timing back where it should be and you will probably gain both bottom and top end.