Honda Insight Forum banner

1 - 20 of 24 Posts

·
Registered
Joined
·
75 Posts
Discussion Starter #1
We bought our brand new (UK) I2 EX a week ago, and I know you guys like to measure the 'last tank mpg', but I've just done a mixed journey (cold start, 10 miles of open road, 3 miles in town - then all the way back home again - at 64mpg (imperial/UK gallons) which is 52.5mpg (US gallons). On the way home I tucked it into 'sport' mode and hauled past a slow-moving truck.

I have a friend who's a motoring journalist and when I told him about our new I2 he said "oh, so you've 'gone green'! But, hey - you won't get anything like the mileage that Honda tell you".

Well, there's one theory out 'the window already...

My car has only done 277 miles from new, and people on this forum seem to be saying that I'll do a fair bit better after a few thousand miles. That said, I think I'm doing ok already, bearing in mind that the car and I are both rookies!

I'll start recording my 'last tank mpg' soon (I know that's what really counts), but we've not had to fill up the tank yet. My respect for this car just grows and grows...
 

·
Registered
Joined
·
234 Posts
Nice job, keep it up :thumbup:

There was probably a good year or so that I was getting pretty frustrated with this car...seemingly never able to get over the mid-to-upper 30's for a tank average. I'm finding that it's not impossible, but my normal short commute, makes it very challenging, to say the least ;)

As far as MPG goes, my one remaining complaint with this car is the overly optimistic MID. I thought it was 1 or 2 MPG off, but my last tank showed 48.0 on the MID, but was 44.2 when calculated. Almost 10% off :rolleyes:

I was like "nearly 50!" before submitting to Fuelly, and then saw the calculated :Damn:

Just out of curiosity, have you found the MID to be as far off on the 2012?
 

·
Registered
Joined
·
75 Posts
Discussion Starter #3
Thanks: your observations on the (maybe quite seriously) inaccurate MID are interesting. I can't say very much about my car yet, as we've only just filled up for the first time and I don't really know how full the tank was when we picked the new car up from the dealer.

I'll update this a tank (or two) from now, when a clearer picture of the actual and MID consumption should be starting to emerge...
 

·
Registered
Joined
·
605 Posts
The car has been growing on me as well. Part of the getting better and better has been taking on the challenge of discovering driving techniques for best mpg.

Determining an accurate mpg figure has turned out to be a little more difficult then one might think, at first. It has bee a great help to report all fill ups / gas put into the car, to my Fuelly | Share and Compare Your MPG account. That way over a stretch of say 3 - 6 fill ups I can get a reliable picture of mpg performance. Also I've programed the "A" mpg read out to reset with each fillip. And I reset the "B" read out every morning before starting the engine, to get a daily mpg readout. Then again as needed to checkout or compare various driving techniques, routes, and etc.
 

·
Registered
Joined
·
69 Posts
27k miles on 2010 Insight. MIDI indicated 52mpg, sometimes getting to 53mpg for 15/16 of tankfull. Then when refilling gas tank divided gallons into mileage and got 48mpg. I am so disappointed!. I thought I had finally broken the 50mpg barrier. It had the econ button on the entire tankfull. This tank I will be running with econ button off. I think with it on it just makes the MIDI give higher bogus numbers. It must be a feel good button.
 

·
Registered
Joined
·
75 Posts
Discussion Starter #6
55mpg (UK) at first proper fill-up

Today we filled up for the second time, so I know that the tank was full at both fill-ups. So far we're getting 55mpg (UK/Imperial gallons) = 45.8 mpg (US gallons).

My car has only travelled 420 miles from new, so I'm pleased with this - but I want 60mpg (UK) = 50 mpg (US)! I'll work hard with this tankful, and let you know what happens...
 

·
Registered
Joined
·
605 Posts
An actual 50 mpg (US) for a tank of gas has been my goal as well. It IS achievable!!

The in-accuracy of the MID mpg read out has been a topic of extensive discussion here many months ago. I did a 6 tankfull comparison test, and found that the MID readout for a 9 gal fill up was ON AVERAGE 2.9 mpg optimistic. Meaning that you can estimate your actual mpg any time by subtracting 3 from whatever the MID is saying.
 

·
Registered
Joined
·
605 Posts
It isn't just the Honda MID that overestimates mpg. Accuracy in these factory gauges is rare indeed.
How accurate are are factory/OEM MPG fuel economy gauges ? - Fuel Economy, Hypermiling, EcoModding News and Forum - EcoModder.com
Thanks for the link.
Our other car is an 2006 I1, which my wife drives. When we first got it we were both happy and amazed to see an in-dash mpg gauge. What a great addition, and how very appropriate for a car dedicated to high mpg! After driving the car for awhile I got curious about the accuracy of the gauge. I spot checked it 3-4 times and found the gauge right on for accuracy or off one way or the other by 0.1 mpg. So all the complex technical issues for an in-dash mpg gauge have been solved.....I thought.

In '09 I bought an I2. Its gauge has been inaccurate as extensively discussed. Being a daily watcher of mpg, knowing my indash gauge is inaccurate and knowing that Honda can do better I have made my complaint known to Honda and encourage others to do so as well.

At consistently 3 mpg optimistic at 50 mpg for the tankfull thats (3 / 50 = .06) 6%. Which is generally consistent with the "industry standard" alluded to in yur link.

Blue Bird (Honda Insight) | Fuelly
 

·
Registered
Joined
·
38 Posts
I started to put in what the MID says on my Fuelly, My MID seems to be about 2.4+mpg optimistic. It was 3mpg optimistic this latest fillup. With only 3 occurrences it seems that the higher the actual mpg, the more optimistic the MID is.
 

·
Registered
Joined
·
943 Posts
I started to put in what the MID says on my Fuelly, My MID seems to be about 2.4+mpg optimistic. It was 3mpg optimistic this latest fillup. With only 3 occurrences it seems that the higher the actual mpg, the more optimistic the MID is.
That seems likely, as mpg does not have a linear relationship with fuel usage. Switch it round to gallons per mile (or gallons per 100 miles) and have a think about the difference between 1 and 0.

I suspect the difference between calculated and displayed l/100km will be more consistent.
 

·
Registered
Joined
·
605 Posts
I started to put in what the MID says on my Fuelly, My MID seems to be about 2.4+mpg optimistic. It was 3mpg optimistic this latest fillup. With only 3 occurrences it seems that the higher the actual mpg, the more optimistic the MID is.
I used to get the kind of numbers reported above. But as my actual mpg got better with changes in driving style the raw MID optimism also got higher. I've also noticed that when my actual mpg is lower due to cold and water on the road, the MID optimism gets smaller.

All this suggests to me that there is a given percentage error in the MID's calculation process. If so a programming update could fix the problem and produce more accurate mpg readouts.

Perhaps enough of an improvement to where the mpg readout of the I2 would equal the accuracy of the mpg readout of the I1!! This was the content of my complaint / request to Honda USA and my local dealer, and my suggestion to anyone who shares this pet peeve.
 

·
Registered
Joined
·
75 Posts
Discussion Starter #13
Personally, I've never really trusted in-dash mpg displays.

My 2005 MINI Cooper S had one that consistently under-reported fuel use (i.e. it claimed better fuel consumption than I was actually getting) to the tune of about 1mpg or roughly 3%

That said, these gauges are useful, in that they are probably quite consistent even when inaccurate. So, as a way of seeing whether you're doing better or worse than last time, they are helpful but if you want a definitive mpg figure (e.g. for fuelly.com) there's still no substitute for measuring the volume of fuel you're putting in the tank!

If you look on these two methods as serving two distinctly different purposes, it kind of ceases to be a problem...
 

·
Registered
Joined
·
605 Posts
......That said, these gauges are useful, in that they are probably quite consistent even when inaccurate. So, as a way of seeing whether you're doing better or worse than last time, they are helpful but if you want a definitive mpg figure (e.g. for fuelly.com) there's still no substitute for measuring the volume of fuel you're putting in the tank!......
EXCELLENT !! I completely agree.
 

·
Registered
Joined
·
943 Posts
I think you all missed my point. As mpg gets higher the mpg discrepancy between MID and actual obviously gets higher. However - the MID is not getting more inaccurate at higher mpg.

What is the car actually estimating? Distance travelled and fuel consumed. As fuel consumption decreases each mpg is worth less.

Here's a quick table as to the increase in fuel used over 100 miles in 5mpg increments.

MPG, Gal/100m, Difference (Gallons)
35mpg, 2.86gal, N/A
40mpg, 2.50gal, 0.36gal
45mpg, 2.22gal, 0.28gal
50mpg, 2.00gal, 0.22gal
55mpg, 1.82gal, 0.18gal
60mpg, 1.67gal, 0.15gal

Ergo - a 3 mpg discrepancy at 40mpg is roughly the same margin of error as a 6 mpg discrepancy at 55 mpg. The trip computer isn't getting more inaccurate at all. It's just that MPG is not directly a measure of fuel consumed. It's a rate of fuel consumption.

Edit: that's not to say MID accuracy can't be improved, just that any discrepancy will always be magnified at low consumption when you're using distance / volume to gauge fuel consumption.
 

·
Registered
Joined
·
75 Posts
Discussion Starter #16
Uriel's observations about mpg versus gallons per mile are certainly valid.

Another way to think about all of this is to consider what the MID is actually doing. It counts "fuel injector pulses x pulse width". A pulse occurs on each cylinder every two engine revolutions and the pulse width is the amount of time the computer tells the fuel injector to stay open - unless the engine is on the over-run with the throttle closed, in which case the fuel supply is shut off (i.e. injector pulse width is zero).

The more engine revolutions you get from your gallon, the more any small errors in calibration will compound - because the system has had to count the summation of more injector pulses in order to compute the mpg. It seems likely that smaller pulse widths (which our very efficient engines will be using a lot of the time) will be more prone to percentage volume error than larger injector pulses.

So maybe we should recognise that the MID is having to do a tough job on our very efficient engine - measuring lots and lots of very tiny injector pulses - and be a bit more forgiving of it!
 

·
Registered
Joined
·
622 Posts
Uriel's observations about mpg versus gallons per mile are certainly valid.

Another way to think about all of this is to consider what the MID is actually doing. It counts "fuel injector pulses x pulse width". A pulse occurs on each cylinder every two engine revolutions and the pulse width is the amount of time the computer tells the fuel injector to stay open - unless the engine is on the over-run with the throttle closed, in which case the fuel supply is shut off (i.e. injector pulse width is zero).

The more engine revolutions you get from your gallon, the more any small errors in calibration will compound - because the system has had to count the summation of more injector pulses in order to compute the mpg. It seems likely that smaller pulse widths (which our very efficient engines will be using a lot of the time) will be more prone to percentage volume error than larger injector pulses.

So maybe we should recognise that the MID is having to do a tough job on our very efficient engine - measuring lots and lots of very tiny injector pulses - and be a bit more forgiving of it!
is the MID algorithm in fact based on injector pulses?? if so, i had no idea
 

·
Registered
Joined
·
605 Posts
..Another way to think about all of this is to consider what the MID is actually doing. It counts "fuel injector pulses x pulse width". A pulse occurs on each cylinder every two engine revolutions and the pulse width is the amount of time the computer tells the fuel injector to stay open - unless the engine is on the over-run with the throttle closed, in which case the fuel supply is shut off (i.e. injector pulse width is zero)...
Thank you very much for posting this. It is the FIRST information I've seen as to how the MID is determining its mpg estimates!

It does help to explain for me two things. (1) I've noticed that MID error is greater on those tank fulls in which I did allot of "ev mode." (e.g. lots of running with "the engine on the over-run with the throttle closed, in which case the fuel supply is shut off.) (2) Since the I1 does not have an ev mode its mpg computations don't have to deal with the mathematically dreadful multiplication by zero! and thus have a better shot at good accuracy.
 

·
Registered
Joined
·
943 Posts
As Alfettaman said, the car is not directly measuring fuel usage. It's calculating it and injector pulses is the most usual way to do that (although I've not seen the specifics of the I2's system).

Worth pointing out that most modern cars have over-run fuel cutoff though. The Insight can just artificially induce / prolong it compared to most using the motor, while minimising engine friction with cylinder deactivation.
 

·
Registered
Joined
·
75 Posts
Discussion Starter #20
It seems that what we're saying is that...

(1) as mpg improves, so the precision needed to measure it accurately gets higher (see Uriel's "mpg vesus gallons/100 miles" chart) - because the differences in actual consumption are getting disproportionately smaller and smaller

(2) short injector pulse durations must surely be harder for the MID to record accurately than longer ones are, so a high mpg car is inherently more of a challenge

(3) All of this EV/lean-burn/fuel shut off/pulse-and-glide/auto stop etc is great, but can only complicate what is already a difficult measurement to make.

It's a strange conversation for i2 owners to be having, really. Here we all are, agonizing over our mpg, but get this: if you drive an ordinary car carefully in order to get 25mpg instead of your usual 20mpg, you'll save a whole gallon every 100 miles.

To save an extra gallon every 100 miles in your 50mpg i2, you're going to have to do 100mpg instead! Getting from 50mpg to 55mpg makes little difference, and will only save you a gallon every 600 miles. So at 50mpg were already saving nearly all the fuel we realistically can: it's the "20mpg brigade" who ought to be agonizing about their fuel consumption, not us.

Ironically, they can make a big difference, but we can only make a small one. So relax, enjoy your i2, feel good about having done nine-tenths of what might be possible - and spread the word!
 
1 - 20 of 24 Posts
Top