Drive Accord Honda Forums banner

1 - 20 of 21 Posts

·
Registered
2021 Accord Touring Hybrid
Joined
·
49 Posts
Discussion Starter · #1 ·
I know there is already a thread for mpg in general, this is about something else. I realize that with only about 2500 miles on my car over 4 months, it is barely broken in. (The mileage includes 1 road trip of about 900 miles highway, the rest is local). However, I am finding a large discrepancy between mileage as calculated by the car's trip computer, and manual calculation. It is consistently about 2 mpg higher on the trip computer. My last car (2016 Accord LX, non-hybrid) always had a discrepancy also, but only about 0.5 mpg. At first I thought it might be because I changed gas stations, thinking the pumps might be metering wrong, but I have since switched back, and the results were the same. I am consistent in the way I add fuel as well, always removing the nozzle after it clicks, without "topping off." Wondering if anyone else is seeing a discrepancy of this magnitude.
 

·
Registered
2019 Accord Sport 2.0 6MT
Joined
·
124 Posts
I’m seeing about the same discrepancy and I’ve seen others report the same. I have a 2019 2.0 6MT.
 

·
Registered
Joined
·
2,700 Posts
I've a 9th gen and sometimes I see that.
I stick to the calculated old-fashioned way, as it's nearly foolproof.
 

·
Registered
Joined
·
425 Posts
... I am finding a large discrepancy between mileage as calculated by the car's trip computer, and manual calculation. It is consistently about 2 mpg higher on the trip computer.
Unfortunately, it is very difficult to get great accuracy in such a measurement. But we can settle for consistency. Mine seems to be quite consistently 3% too low (about 1.4 to 1.5 mpg).
I've a 9th gen and sometimes I see that.
I stick to the calculated old-fashioned way, as it's nearly foolproof.
Except you need to average over several tanks, since exactly when a pump cuts off is less consistent than the trip computer. You'll see this if one fill-up is abnormally high when compared to the trip computer, and then the next is abnormally low. Or vice versa.

I actually track 3 values: what the pump says, what the trip computer says, and an adjusted value where I "shift" some, usually a half gallon, between two fill-ups where this happens. Here the blue line shows the truest trends, and the green one shows the truest values. The red is what I'd get if I believed the pump each time.
531317
 

·
Registered
Joined
·
574 Posts
Interesting. I see about a 0.5 MPG difference between hand calculation and the car's calculation since the last fillup, but the car's calculation is LOWER than the actual fuel economy. I kind of expected the car to show the higher figure, but it doesn't. My Wife's new CRV is similar - it shows a lower mpg value than what is actually happening.

- Jack
 

·
Registered
Joined
·
895 Posts
Every car that I've had that has had a mpg display has over calculated the actual mpg that the car was getting... by about the same rate that you are seeing- 2mpg-ish.

It's always better and more accurate to do it manually.
 

·
Registered
Joined
·
2,700 Posts
Except you need to average over several tanks, since exactly when a pump cuts off is less consistent than the trip computer. You'll see this if one fill-up is abnormally high when compared to the trip computer, and then the next is abnormally low. Or vice versa.
Nope...not "mathematically"you don't....
All I care about is when I put(approx.) 16 gallons in, what was my MPG over the miles driven"...
If I'm strictly using actual miles driven divided by actual gallons pumped..that IS the mileage...regardless of what a computer ever "says" it is.
Any theory about where that pump or another one clicks off, is unreliable data..and should never be part of the equation
 

·
Registered
Joined
·
400 Posts
So, my 2018 Accord Hybrid Touring also has a significant discrepancy, but in my case the trip computer consistently shows worse mpg than the reality (as calculated from gas station receipts). My actual mpg is always 2-3 mpg BETTER than what the trip computer says.

I suppose it is better for the reality to be better than the computer, but I always feel a bit shortchanged when I watch the trip computer's display. I'd FEEL better if the trip computer was more accurate, and even a bit over-optimistic.....
 

·
Registered
Joined
·
574 Posts
Nope...not "mathematically"you don't....
All I care about is when I put(approx.) 16 gallons in, what was my MPG over the miles driven"...
If I'm strictly using actual miles driven divided by actual gallons pumped..that IS the mileage...regardless of what a computer ever "says" it is.
Any theory about where that pump or another one clicks off, is unreliable data..and should never be part of the equation
However, even if you don't accept the idea that a pump cutoff might cause a different rate of gallons used vs miles, other things, such as driving in headwinds or tailwinds, lots of hills, high speed driving, stop and go driving, outside temperatures; that might all be different between different fillups, will cause significant changes in the mpg figures you get on each tankfull of gas. For that reason, I average my fuel economy both over the last tankful of gas AND, the total miles/total fuel used since I bought the car. The overall mpg is what I use to judge fuel economy, but the last tankful mpg tells me if something has suddenly gone horribly wrong with my vehicle. If there is a significant difference between the two, I ask myself, "What were the driving conditions on the last tankful?"

- Jack
 

·
Registered
Joined
·
64 Posts
Nope...not "mathematically"you don't....
All I care about is when I put(approx.) 16 gallons in, what was my MPG over the miles driven"...
If I'm strictly using actual miles driven divided by actual gallons pumped..that IS the mileage...regardless of what a computer ever "says" it is.
Any theory about where that pump or another one clicks off, is unreliable data..and should never be part of the equation
JeffJo is right. There is an error introduced by the automatic shutoff of the pump. You don't know if you filled up to the exact same volume of gas as when you started your trip. This is why you do from full tank to fill tank, unfortunately you have no accurate way of knowing if the tank is rally full or not...just approximate. If the error is 1/2 gallon, that could be 4% of the tank. But if you do, let's say 10 tanks of gas, your error of 1/2 gallon stays the same since it's only the first and last time you fill up.... Which now becomes, say .04% error.

My '19 HAH mpg is consistently better at the pump than the trip mater. It's on average about 2.5 mpg better than the trip meter, and that's measured over 31k miles. I've seen it as high as 4.1 mpg over.

Also, I pump at the same 2 pumps all the time and I NEVER top off.
 

·
Registered
Joined
·
2,700 Posts
JeffJo is right. There is an error introduced by the automatic shutoff of the pump. You don't know if you filled up to the exact same volume of gas as when you started your trip. This is why you do from full tank to fill tank, unfortunately you have no accurate way of knowing if the tank is rally full or not...just approximate. If the error is 1/2 gallon, that could be 4% of the tank. But if you do, let's say 10 tanks of gas, your error of 1/2 gallon stays the same since it's only the first and last time you fill up.... Which now becomes, say .04% error.

My '19 HAH mpg is consistently better at the pump than the trip mater. It's on average about 2.5 mpg better than the trip meter, and that's measured over 31k miles. I've seen it as high as 4.1 mpg over.

Also, I pump at the same 2 pumps all the time and I NEVER top off.
It matters not if the tank is full..period.
That's if all you care about is actual MPG...not "miles per tank"...
Which is what the original question was:
"However, I am finding a large discrepancy between mileage as calculated by the car's trip computer, and manual calculation."
The thread title is, and was:
Miles per gallon discrepancy

In this instance, his manual calculation should be based on actual gallons pumped, not the size of the tank.
Remember:
He's not looking for miles per tank he's looking for MPG...at the end of the day, that's all he really wants.
The tank is the tank...who cares what the tank yields? ...as that's not the question.
This not a EV....where your concern should be capacity of the "tank"(meaning the battery capacity)
The MPG calculation is going to be more "accurate" base on a larger measurement, sure, but not exponentially.
For example, The OP is not going to lose 2 mpg just because he did one average of a full tank and another average on a tank with two gallons left in it...that won't skew the data enough.
 

·
Registered
Joined
·
425 Posts
Except you need to average over several tanks, since exactly when a pump cuts off is less consistent than the trip computer.
Nope...not "mathematically"you don't....
Well, I'm a mathematician and an engineer, which is why I say that mathematically, you do. :) Mathematically, "click-off-error" affects your calculation, but not the actual mileage you say you are looking for.

The problem is that with our (you have a hybrid, right?) 12.8 gallon tank, you may:
  • Fill it to 12.2 gallons, where it clicks off.
  • Drive 500 miles, using exactly 10 gallons (so the tank is at 2.2).
  • Fill it up again, but this time it clicks off at 12.5 gallons.
You actually got 50 miles per gallon, but your calculation will say 500/10.3=48.5 mpg.

But if you use 3 tanks, driving 500 miles each time, using 10 gallons each time, and again starting at 12.2 and ending at 12.5? Your calculation, as an average, will be 1500/30.3=49.5 mpg.

Any theory about where that pump or another one clicks off, is unreliable data..and should never be part of the equation
Exactly. You are assuming it clicks off at the same point every time. The point of averaging is that you don't know that, and can't estimate it. But if you use, say, three tanks? The potential error in gallons will be the same, but over three times as many miles.
+++++

[Edit]
It matters not if the tank is full..period.
What matters is the fact that it isn't filled to the same point each time.
his manual calculation should be based on actual gallons pumped, not the size of the tank.
No, it should be based on number of gallon used, not number of gallons pumped or size of the tank (which I never suggested).

Number of gallons pumped is only an estimate of number of gallons used. It is off by the difference between where you stopped pumping in the intervening fillups.
 

·
Registered
Joined
·
2,700 Posts
No, it should be based on number of gallon used, not number of gallons pumped ...

Actual gallons used is what's "pumped"! That's how your getting your volume.

From both the last time and what you pump the next time...you have no way of really knowing that unless you put it all in a certified container.
You have to trust an/or assume that that gas pump meter you use...is metering correctly.
That's really the only wild card, all things being equal.
 

·
Registered
Joined
·
425 Posts
Actual gallons used is what's "pumped"! That's how your getting your volume.
Only if you fill it to the exact same level each time. If there is a difference between fill levels - which is what happens when different pumps "click off" at different places - then there is an error.

You have to trust an/or assume that that gas pump meter you use...is metering correctly.
No one has doubted that. The difference is that different pumps, or even the same pump on different days, have a different definition of what "full" means.
 

·
Registered
Joined
·
226 Posts
Yes, I think there is a discrepancy. All of you guys have shared some very valuable information so far in this thread. I am learning a lot of new things from some of you all the time.

Well, I just completed a hypermiling tank of gas where my goal was to take on a challenge that OG1 suggested to me when he posted he got 801 to a tank of gas. So, my goal was to see if I can get past that almost unbelievable number. Well, I did it.

OG1 posted his odometer showing 801.9 miles to a tank of gas with consumption of 64.3mpg. But his manual calculation showed he got 66.81mpg after a fillup at the pump of exactly 12 gallons.

Well, here is mine. I just completed the hypermiling challenge with my odometer reading says 810.1 miles done on a tank of gas with consumption of 66.1mpg. I filled up with 12.6 gallons. My manual calculation is 64.3mpg, which is lower than what the car says; 66.1mpg.

Here are pictures of my just completed hypermiling run...



 

·
Registered
Joined
·
2,700 Posts
You guys are not getting this....
MPGs has been discussed ad nauseum since Ive been on this forum...everything from what the computers says to what the actual mileage is(read: the manual way...or old fashioned way...which has been done since the Model T)
And the general consensus is...,<drum roll, please>the computer has never been and will never be as accurate as the old fashioned way when computing MPG.

So why do we have it?
Because humans, in general, want to be as lazy, fat and stupid as possible...without using their actual brains!
Oh dear! it took me about a minute! Good god...
That's going to take time away from me ordering my triple decaffeinated half-caf latte with a mocha twist from Fourbucks!!
Oh the humanity!!!

You're arguing about tank size when that has zero to do with what the original question was....It's way too much work to do what I've outlined right??

TOO much time! Id rather push a button! I'd rather be brainless!...and then we wonder why it's inaccurate.
It's a reference guide. Period.
It's not accurate.
IT wasn't meant to be.

Only if you fill it to the exact same level each time. If there is a difference between fill levels - which is what happens when different pumps "click off" at different places - then there is an error.


No one has doubted that. The difference is that different pumps, or even the same pump on different days, have a different definition of what "full" means.
You're still not understanding anything that I've said to you.....and for that reason...
 

·
Registered
2019 Accord Sport 2.0 6MT
Joined
·
124 Posts
Well, I'm a mathematician and an engineer, which is why I say that mathematically, you do. :) Mathematically, "click-off-error" affects your calculation, but not the actual mileage you say you are looking for.

The problem is that with our (you have a hybrid, right?) 12.8 gallon tank, you may:
  • Fill it to 12.2 gallons, where it clicks off.
  • Drive 500 miles, using exactly 10 gallons (so the tank is at 2.2).
  • Fill it up again, but this time it clicks off at 12.5 gallons.
You actually got 50 miles per gallon, but your calculation will say 500/10.3=48.5 mpg.

But if you use 3 tanks, driving 500 miles each time, using 10 gallons each time, and again starting at 12.2 and ending at 12.5? Your calculation, as an average, will be 1500/30.3=49.5 mpg.


Exactly. You are assuming it clicks off at the same point every time. The point of averaging is that you don't know that, and can't estimate it. But if you use, say, three tanks? The potential error in gallons will be the same, but over three times as many miles.
+++++

[Edit]

What matters is the fact that it isn't filled to the same point each time.

No, it should be based on number of gallon used, not number of gallons pumped or size of the tank (which I never suggested).

Number of gallons pumped is only an estimate of number of gallons used. It is off by the difference between where you stopped pumping in the intervening fillups.
Always to see someone with a STEM background and science writing skills make their point in here. Chemist/biomedical engineer here.

You guys are not getting this....
Lol, you are quite literally the one not getting it.

..the computer has never been and will never be as accurate as the old fashioned way when computing MPG.
I'd actually argue that the computer is more accurate. At any time the car knows exactly which injectors are firing, when they are firing, and how much fuel is passing though them each time they fire. Alternatively the car probably also knows at what speed the fuel pump is running and how much fuel is passing through it. I would imagine that measuring fuel use at the injectors or fuel pump is more accurate than the fill up method due to the reasons that JeffJo has pointed out. I'm not going to go over what he has said because I think the explanation was quite good.

You're arguing about tank size when that has zero to do with what the original question was....It's way too much work to do what I've outlined right??
Tank size is not the point and that's not what JeffJo is saying. The point is when you fill the tank you do not know if it was filled to the exact same level as last time you filled it up. When you do the manual calculation you're making that assumption and you cannot be sure that it is true.

I would take another read though his post again and try and see if you can see why you're wrong.
 

·
Registered
2021 Accord Touring Hybrid
Joined
·
49 Posts
Discussion Starter · #18 ·
Okay, I do get that different pumps, different weather, etc., can affect how "full" the tank actually is. An individual calculation could well be off due to these variances. However, since I have had the car, I have filled up 9 times. Each time, the discrepancy between manual calculation and the computer, is 2 mpg plus or minus about 10% (0.2 mpg). I think that is a fairly consistent error.
 

·
Registered
Joined
·
1,820 Posts
2019 1.5 Sport 6mt here. consistently 1.5-2mpg higher on the computer than the real world calculations over 30k
 

·
Registered
2017 Honda Accord, Coupe, Touring (V6, 6AT), 2017 Honda Accord Sedan, EXL V6 w/ Navi, Sensing
Joined
·
249 Posts
Well…the engineers do get the math (I might be biased). The click off error is damped as the number of miles and fill ups is increased in computing the average. The fact of the matter is that the lie o meters are typically optimistic relative to hand calcs. In my experience, upwards of 6% on my coupe (determined on the basis of somewhere around 20-30 tanks of fuel). The smallest degree of optimism I’ve seen in lie o meter data is in my Tundra, it’s within about 0.5%, which is pretty good considering we’re not using it to send someone to the moon.
 
1 - 20 of 21 Posts
Top