Welcome to the North American Subaru Impreza Owners Club Thursday April 24, 2014
Home Forums WikiNASIOC Products Store Modifications Upgrade Garage
NASIOC
Here you can view your subscribed threads, work with private messages and edit your profile and preferences Home Registration is free! Visit the NASIOC Store NASIOC Rules Search Find other members Frequently Asked Questions Calendar Archive NASIOC Upgrade Garage Logout
Go Back   NASIOC > NASIOC Technical > Transmission (AT/MT) & Driveline

Welcome to NASIOC - The world's largest online community for Subaru enthusiasts!
Welcome to the NASIOC.com Subaru forum.

You are currently viewing our forum as a guest, which gives you limited access to view most discussions and access our other features. By joining our community, free of charge, you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is free, fast and simple, so please join our community today!

If you have any problems with the registration process or your account login, please contact us.
* Registered users of the site do not see these ads.
Reply
 
Thread Tools Display Modes
Old 06-05-2010, 06:08 PM   #1
Concillian
Scooby Guru
 
Member#: 4414
Join Date: Feb 2001
Location: Dublin, CA
Vehicle:
2002 WRX Sedan
Midnight Black

Default Gearing vs. fuel consumption analysis (cruising)

I want to preface this post by saying that I know that people on NASIOC are largely not driving the cars they have because of their fuel efficiency. That being said, I think we all have times where we want to drive efficiently. This post is for those times.

Qualitatively, "everyone knows" that shorter gearing will use more fuel and taller gearing will use less when cruising on the highway. While this is a good guideline, I set out to make a more quantitative analysis to give a general idea of "how much worse?" shorter gearing is than taller gearing.

The first section of this post is covering setup of the experiment. Feel free to skip it, if you just want to see the conclusions. I know people will question the method, so I'm including it for those people. I tried to set it up to be as close a comparison as it could be.

------
I used my commute path, I-680 between Pleasanton and Fremont and logged this path on different days driving in 5th and 4th gears cruising at the same speed.
IAT on these days was very close with average IAT less than 2 degrees different. These were similar days in terms of humidity as well and they were at the same time of day on consecutive days.

This section of freeway has relatively flat sections with some mild inclines, followed by a steep uphill grade that isn't too long, but long enough. I used the cruise control to keep the speed relatively constant. Average difference between the speeds was less than 1% (technically 4th gear was 0.8% faster than 5th gear, or about 0.5 MPH faster). Cruise control has some modulation to it, but both gears had similar sigmas on RPM (around 20)

I started and stopped the logs while already at speed with cruise control engaged. Traffic was nice enough to allow me to cruise this section without having to change speed. I used road signs as my reference for starting and stopping the logs. Errors due to human error in starting and stopping the logs or slight differences in length due to lane changes and such were minimal, only 0.07% difference in log time could not be accounted for by the speed difference calculated from average RPM.

Initial experiments showed that downhill sections left significant room for error, so I chose to stop the logging just before I started down the major downhill grade.

4th gear in my tranny is 0.972
5th gear is 0.738
FD, tires, etc... are all the same, obviously.

I tried looking at various logged parameters and mathematical combinations of those parameters with RPM, load, etc... to find something better than plain IDC% for estimating fuel usage. I found nothing better than logged IDC%, so that's what I used. Calculating IDC% from logged injector pulse width resulted in the exact same number (not surprising).

To account for the small difference in speed, I removed 1 line of data every 137.5 lines from the 5th gear log. This made the logs exactly the same length and everything seems to line up nicely in the graph.

-----

Here is the resulting graph.
Blue line is IDC% (4th gear) over IDC% (5th gear). Whenever 4th is using more fuel, this ratio is more than 1, whenever 5th is using more fuel, this ratio is less than 1.
Red is 4th gear logged manifold relative pressure (psi)
Green is 5th gear logged manifold relative pressure (psi)



You can see that there are places where 4th gear is more efficient than 5th gear. The majority of these are slight inclines uphill. There is a more significant uphill section about halfway and then I hit the Sunol grade around data row 2200. The large spike down towards the end is where the 5th gear log caused enough engine load to trigger the ECU to switch to open loop fueling, which used considerably more fuel.

Cruising on relatively flat sections, 4th is using roughly 18% more fuel. Overall average difference for the whole log, 4th uses 8.2% more fuel. One of the main purposes of this was to estimate how much extra fuel would be used with the JDM 6-speed. It has a 6th gear ratio of 0.842, and if the increase in consumption is linear, this would result in a 3-8% increase in fuel consumption while cruising, or a loss of around 1-1.5MPG while cruising. However it would potentially save fuel under other conditions, as the additional ratios and closer spacing should slightly increase efficiency at slower speeds.

The most interesting conclusion to be drawn here is that while RPM hurts cruising mileage, boost clearly hurts mileage worse, even when it's not enough to trigger open loop fueling. In this chart, almost all times where 4th is more efficient, 5th is over 0psi. If you're cruising and want to drive efficiently, this data suggests that you should downshift pretty much anytime you're in boost.

The exception to this would obviously be if downshifting puts you into open loop fueling (my ECU switches to open loop around 4000-4500 RPM). You can see how much extra fuel consumption happens when switching open loop by that spike downwards, it's very significant.

I also made this comparison because I thought I could use the data to refine my current driving. I've suspected 4th gear was better for fuel economy than 5th when driving up that grade. Now there's very clear data to support it.
* Registered users of the site do not see these ads.

Last edited by Concillian; 06-05-2010 at 06:28 PM.
Concillian is offline   Reply With Quote
Old 06-05-2010, 07:28 PM   #2
Unabomber
Big Ron
Moderator
 
Member#: 18062
Join Date: Apr 2002
Chapter/Region: MAIC
Location: I can save you a ton of cash
Vehicle:
on car parts so PM
me b4 j00 buy

Default

I like your data, but my main problem with your findings is the cruise control function. When it's used, it keeps the RPMs low and the boost high to climb hills. This nets the best MPG where load is increased...that's how the DOT gets it's MPG figures, so it's pretty much a fact. Road and Track backed this up 3 or so years ago with a question about a Saabaru going up hills and they said the same thing. Highest gear + boost = better MPG than lower gear with no boost + higher RPM.

Not hatin, just saying what I know and it seems not to jive with what you are saying if in fact I did read it correctly.
Unabomber is offline   Reply With Quote
Old 06-05-2010, 08:35 PM   #3
Concillian
Scooby Guru
 
Member#: 4414
Join Date: Feb 2001
Location: Dublin, CA
Vehicle:
2002 WRX Sedan
Midnight Black

Default

Quote:
Originally Posted by Unabomber View Post
I like your data, but my main problem with your findings is the cruise control function. When it's used, it keeps the RPMs low and the boost high to climb hills. This nets the best MPG where load is increased...that's how the DOT gets it's MPG figures, so it's pretty much a fact.
Err... I do not understand your cruise control issue. It keeps RPMs (approximately) constant. The turbo only boosts when loads are high enough to require boost. Hills = load, so yes, it goes into boost.

I don't get how this ties into the DOT, how much of the DOT testing is a 5+% grade uphill? DOT highway testing is surely at significantly lower loads than a sustained uphill climb like this. They probably only blip into positive pressure for short periods of time, and the data I collected shows in this case, the advantage to 4th is very brief. It isn't until the steeper sustained climb that the difference is enough to justify shifting. I don't doubt at all that highway tests by the DOT show 5th as superior on average. That does not mean that it's the superior gear for fuel economy in every situation.

Quote:
Road and Track backed this up 3 or so years ago with a question about a Saabaru going up hills and they said the same thing. Highest gear + boost = better MPG than lower gear with no boost + higher RPM.
And some journalist has datalogged this? Provided evidence? Journalists pull stuff out of their ass all the time. There is clear evidence here. I am an engineer who's expertise is computing statistical relevance in experiments. I was a part of the Society of Automotive Engineers high mileage vehicle team when I went to college ~15 years ago doing tests just like this on our vehicle... I think I have my own qualifications on how to design and analyze experiments evaluating fuel consumption that outweigh a journalist's... excuse me... a car journalist.

If you see a mistake I made, or a suggestion to change my methodology to improve signal to noise, I'm all ears... "Some journalist said otherwise" however, is a pretty weak piece of data to hold onto from the opposite viewpoint.

Last edited by Concillian; 08-24-2012 at 09:38 PM.
Concillian is offline   Reply With Quote
Old 04-08-2013, 02:12 PM   #4
car_freak85
Scooby Specialist
 
Member#: 13530
Join Date: Dec 2001
Chapter/Region: NWIC
Location: Cascadia
Vehicle:
EA63 , EA81T, EG33,
EJ205

Default

I'm very interested in the data collected here! It isn't everyday that a NASIOC member posts actual, imperical, primary source data. I applaude your efforts good Sir!

On a seperate note, you make some pretty foul generalizations about journalists... I agree that there are poor ones out there, but ask most of them why they got into their profession and they won't say it was to trick readers or distort data (but it does happen).

As a enginnering major with a focus on vehicle deaign and a minor in journalism, i'm on both sides of this argument... That said, there are a TON of engineers (where I work and in general) who think their shizz doesn't stink and will poke holes all day in other people's work.

All that aside, using CC does remove the human error factor, which is good for simplifying your exparament, but the human brain is better at driving for fuel economy.

Driven in 5th up to 0 psi then down shifting to 4th would net a better average mpg than holding just 4th or just 5th gear.

I would be interested in data without using CC and see what you can achieve on the same loop controlled by a human.

Last edited by car_freak85; 04-08-2013 at 03:31 PM.
car_freak85 is offline   Reply With Quote
Old 04-08-2013, 03:14 PM   #5
Rally2
Scooby Newbie
 
Member#: 132023
Join Date: Nov 2006
Chapter/Region: W. Canada
Location: Alberta Canada
Vehicle:
2005 STI
blue

Default

Awesome info, thanks freak for digging it up and Concillian for the initial work invested!
Rally2 is offline   Reply With Quote
Old 04-08-2013, 04:13 PM   #6
Concillian
Scooby Guru
 
Member#: 4414
Join Date: Feb 2001
Location: Dublin, CA
Vehicle:
2002 WRX Sedan
Midnight Black

Default

Quote:
Originally Posted by car_freak85 View Post

Driven in 5th up to 0 psi then down shifting to 4th would net a better average mpg than holding just 4th or just 5th gear.
Obviously, and that's what I do now.

The takeaway is that going uphill in 5th lugging boost is bad for everything, not just the engine, but fuel ecomony as well. There's no good reason to do it.

Lean burn experiments on E85 have resulted in a similar conclusion.
As you lean it out, torque drops and the throttle opens more to maintain torque. This results in higher manifold pressure.... this seems to be beneficial up until right around 0psi, then fuel consumption starts increasing.

This suggests that anytime the intake manifold is over the pre-turbo air pressure, the turbo losses overwhelm any improvements in VE or gearing. Ideal seems to be shooting for right around the same pressure pre and post turbo.

Quote:
Originally Posted by car_freak85 View Post
On a seperate note, you make some pretty foul generalizations about journalists... I agree that there are poor ones out there, but ask most of them why they got into their profession and they won't say it was to trick readers or distort data (but it does happen).
Yes, generalizations are bad. I'm sure there is quite a spectrum of automotive journalists. Being an automotive journalist alone doesn't mean we should necessarily believe what they say as gospel. There needs to be more to their qualification list or an explanation of their methodology leading them to a conclusion before I believe them. Yes, I'm cynical like that. There's just as large a spectrum of engineers. I have spent some of my time trying to help PhDs with statistical analysis of their experimental data straight after they had four weeks of training on it... I've also dealt with a select few who have been faced with data that goes against what they think should happen theoretically and instead of adapting their theories, they fudge their data. Like if their theory was that wings of a bee cannot support flight of the weight of a bee, they'd yell at a bee to land if they saw one....
So, yes, generalizations are bad and I'm cynical.

Last edited by Concillian; 04-08-2013 at 04:29 PM.
Concillian is offline   Reply With Quote
Old 04-08-2013, 07:41 PM   #7
Uncle Scotty
Scooby Guru
 
Member#: 16200
Join Date: Mar 2002
Vehicle:
OK Houston
we have an Uncle

Default

no alcohol fuel makes a far bigger mpg difference than any of this ****

gimme no alcohol gas and i get 5-6mpg more at 80mph on the highway

every time
Uncle Scotty is offline   Reply With Quote
Old 04-10-2013, 12:06 PM   #8
aboothman
Scooby Specialist
 
Member#: 127745
Join Date: Oct 2006
Chapter/Region: NWIC
Location: Humboldt County
Vehicle:
06 WRX 30r 6 speed
99L EJ255 6 Speed

Default

Unabomber and Uncle Scotty: 2 people we can rely on to consistently post quantifiable, scientific and accurate data.
aboothman is offline   Reply With Quote
Old 04-10-2013, 10:05 PM   #9
Uncle Scotty
Scooby Guru
 
Member#: 16200
Join Date: Mar 2002
Vehicle:
OK Houston
we have an Uncle

Default

Quote:
Originally Posted by aboothman View Post
Unabomber and Uncle Scotty: 2 people we can rely on to consistently post quantifiable, scientific and accurate data.
its ALL...every bit....been posted here before.

and before the moderators went to sleep and let the kiddies post so many insane and unnecessary threads....which killed the search function

it would have been easy to search the data and find it within minutes and hours

so, in the end, i think i have already posted it so that makes it up to YOU to find and not post YOUR bull**** to add to the rest already here
Uncle Scotty is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Oil Consumption Analysis TestPilot Service & Maintenance 5 08-19-2005 01:42 PM
FYI: Fuel consumption calculator Ken Newbies & FAQs 0 01-26-2001 02:40 PM
Excessive fuel consumption Mr.Sniggs Legacy Forum 6 12-14-2000 10:48 AM


All times are GMT -4. The time now is 10:05 AM.


Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2014 Axivo Inc.
Copyright ©1999 - 2014, North American Subaru Impreza Owners Club, Inc.