92.2156 mpg from a Prius II on my first attempt?
#12
07-09-2005, 07:37 PM
Re: 92.2156 mpg from a Prius II on my first attempt?
Edited: Recalculating with data from krausdb's post.
Assuming that the MFD mpg data is systematically 1.5% high and assuming that the random error is +/- 0.5% then I get that the Wayne's mpg is between 85 and 94 mpg. Don't put too much faith in these calculations until we know more about measurement error of the MFD. But, I'm now more confident that Wayne's mileage was really in the high 80s to low 90s, which is really excellent. I'm more tempted to buy a Prius after these results (and looking at krausdb's excellent results in hypermiling with a prius II.
--Walter
PS This remains a calculation that is more sensitive to random errors in the MFD than you might guess. I've left the old calcs below for anyone who thinks their MFD error is larger.
Hi Wayne, Your calculations are correct, but I wouldn't put too much faith in the answer because of uncertainties in your data. According to Douglas, there can be errors in the mpg in the MFD of +/- 15% with 5-7% common. Redoing the calculations with +5%error in the first MFD reading and -5% error in the 2nd MFD reading I get:
mpg for 375 miles: 62.685 gallons for 375 miles: 5.982
mpg for 501.1 miles:62.225 gallons for 501.1 miles 8.053
gallons for 126.1 miles 2.07
mpg for 126.1 miles: 60.896
Lesson: This kind of calculation can be sensitive to errors in the MFD mpg.
How likely are my numbers here? Not very!! There is a low probability that the errofs in the two MFD readings just happen to be in the direction to make your mpg as bad as possible. To be fair I calculated your mpg with the opposite errors (to make your true mpg as large as possible with +/- 5% error in MFD).
Then you get 187 mpg!!!
So what mpg do I think you got? The most likely is your calculation of 92.1127 mpg, but with a fairly large standard deviation. If I get ambitious, I could calculate the standard deviation (with lots of wild guesses on the statistics).
Additional knowledge that really should be included in the statistical calculations is that Wayne is really good at this, so I imagine the reality is well over 65mpg. [Now looking at the breakdown in segments, I would guess that +/- 5% random error is high because of the consistency in the segment readings and would guess that the truth is closer to 90 mpg than to 65].
When your friend finally fills his tank it would be interesting if you report how the MFD and actual mpg compare.
--Walter
Assuming that the MFD mpg data is systematically 1.5% high and assuming that the random error is +/- 0.5% then I get that the Wayne's mpg is between 85 and 94 mpg. Don't put too much faith in these calculations until we know more about measurement error of the MFD. But, I'm now more confident that Wayne's mileage was really in the high 80s to low 90s, which is really excellent. I'm more tempted to buy a Prius after these results (and looking at krausdb's excellent results in hypermiling with a prius II.
--Walter
PS This remains a calculation that is more sensitive to random errors in the MFD than you might guess. I've left the old calcs below for anyone who thinks their MFD error is larger.
Hi Wayne, Your calculations are correct, but I wouldn't put too much faith in the answer because of uncertainties in your data. According to Douglas, there can be errors in the mpg in the MFD of +/- 15% with 5-7% common. Redoing the calculations with +5%error in the first MFD reading and -5% error in the 2nd MFD reading I get:
mpg for 375 miles: 62.685 gallons for 375 miles: 5.982
mpg for 501.1 miles:62.225 gallons for 501.1 miles 8.053
gallons for 126.1 miles 2.07
mpg for 126.1 miles: 60.896
Lesson: This kind of calculation can be sensitive to errors in the MFD mpg.
How likely are my numbers here? Not very!! There is a low probability that the errofs in the two MFD readings just happen to be in the direction to make your mpg as bad as possible. To be fair I calculated your mpg with the opposite errors (to make your true mpg as large as possible with +/- 5% error in MFD).
Then you get 187 mpg!!!
So what mpg do I think you got? The most likely is your calculation of 92.1127 mpg, but with a fairly large standard deviation. If I get ambitious, I could calculate the standard deviation (with lots of wild guesses on the statistics).
Additional knowledge that really should be included in the statistical calculations is that Wayne is really good at this, so I imagine the reality is well over 65mpg. [Now looking at the breakdown in segments, I would guess that +/- 5% random error is high because of the consistency in the segment readings and would guess that the truth is closer to 90 mpg than to 65].
When your friend finally fills his tank it would be interesting if you report how the MFD and actual mpg compare.
--Walter
Last edited by Walter; 07-10-2005 at 07:46 AM. Reason: Recalculation with data from krausdb
#13
07-09-2005, 08:11 PM
Re: 92.2156 mpg from a Prius II on my first attempt?
Great job. The only reason the numbers could possibly be wrong is because you don't know exactly how much gas you used. Are you sure the tank was filled to the exact point as the previous tank? If you were off by .4 gal it would drop your mileage drasticly. You need a few tanks to realy be able to determine your mileage. The more tanks you put in the more the margin or error goes down.
I'd guess by your track record you got around 70mpg. How long will you be able to borrow the car for or was it just for that one trip?
I'd guess by your track record you got around 70mpg. How long will you be able to borrow the car for or was it just for that one trip?
Last edited by tbaleno; 07-09-2005 at 08:14 PM.
#14
07-10-2005, 03:08 AM
Re: 92.2156 mpg from a Prius II on my first attempt?
After reading through these posts I just wanted to make a few comments. It has been suggested that the only way to know for sure is if the MFD rreading is correct is to ensure that the tank was filled to the same level before and after. While this is certainly true, it is very hard to do given the variables of temperature and bladder temperment while filling.
I have made calculations such as Wayne's a hundred times (that is why I developed a spreadsheet for it) and I have found after filling the tank that the calculation is quite accurate. In fact, over 47000 miles of Prius fills I can say that the MFD has never been off by more than 5% from the gallons pumped. Over that period, the average has been a 1.5% MFD overstatement. I would say that with the data I have collected, IMHO, it is more accurate to apply a 1.5% penalty to the MFD than to base a conclusion on how many gallons it took to fill the tank. That is because I have 90+ tanks of data that serve to smooth out all of the individual errors associated with filling one tank.
I have made calculations such as Wayne's a hundred times (that is why I developed a spreadsheet for it) and I have found after filling the tank that the calculation is quite accurate. In fact, over 47000 miles of Prius fills I can say that the MFD has never been off by more than 5% from the gallons pumped. Over that period, the average has been a 1.5% MFD overstatement. I would say that with the data I have collected, IMHO, it is more accurate to apply a 1.5% penalty to the MFD than to base a conclusion on how many gallons it took to fill the tank. That is because I have 90+ tanks of data that serve to smooth out all of the individual errors associated with filling one tank.
Last edited by krousdb; 07-10-2005 at 04:33 AM.
#17
07-10-2005, 08:21 AM
Re: 92.2156 mpg from a Prius II on my first attempt?
Cruising at those speeds on relatively flat ground I've noticed instant, and average FE of 2.5 l/100km ~94MPG so I'd say it's perfectly possible to get 92 MPG, given some stopping and starting. The problem around here is the hills: pretty much anywhere we go is up and down several big hills, so FE is just trashed, especially with traffic usually going 10-20 km/h over the limits and plenty of stop lights. It must be so nice living on flat land... I now envy the drivers over in Saskatchewan. 
Last edited by Schwa; 07-10-2005 at 09:50 AM.
#18
07-18-2005, 08:30 AM
Re: 92.2156 mpg from a Prius II on my first attempt?
Hmm... I see you have discovered the Prius II magic Wayne :-) Remember the 1000 mile tank we talked about last year? Perfectly doable if you run the car non-stop. But no sleep for 24 hrs of course :-)
Now about gliding, in places that are not all that flat, it would really help if you find a route that has strong climb and gentle slope down. You can then do all the climbing in one go and then glide for a very long time. The opposite would be quite bad because you need to pulse a lot to go up the gentle slope. Every pulse has a startup cost, you really don't want to do it very often. And also the big downhill will require you to brake too hard to regenerate much of the energy.
That's way route choice is very important here. I have evaluated alternative routes extensively for most of the places that I need to visit on a regular basis. My wife always get really upset when I take all these strange detour :-)
Andy
Now about gliding, in places that are not all that flat, it would really help if you find a route that has strong climb and gentle slope down. You can then do all the climbing in one go and then glide for a very long time. The opposite would be quite bad because you need to pulse a lot to go up the gentle slope. Every pulse has a startup cost, you really don't want to do it very often. And also the big downhill will require you to brake too hard to regenerate much of the energy.
That's way route choice is very important here. I have evaluated alternative routes extensively for most of the places that I need to visit on a regular basis. My wife always get really upset when I take all these strange detour :-)
Andy
#20
07-19-2005, 04:43 PM
Re: 92.2156 mpg from a Prius II on my first attempt?
Wayne,
Im not a statistician but I would buy the 92MPG number +/- 2%. And congrats to Carl, the next Prius Hypermiler in training! I'd like to meet him also next time I'm in the area.
Im not a statistician but I would buy the 92MPG number +/- 2%. And congrats to Carl, the next Prius Hypermiler in training! I'd like to meet him also next time I'm in the area.
