How are you making so many MPGs!?!

I smoke the pack if I can, take off hard, get in front, then right as I hit the speed limit I throttle back and glide with no arrows. If I see a hill I accelerate to 10km/h over the limit then hold the throttle at about 3.5L/100km to try and hit the crest of the hill at the limit then glide down the other side and back down to 5 under the limit, then quick pulse again. If I keep this up I can average 4.0L/100km or better in traffic. Unfortunately it is hard to concentrate on doing this all the time so often I just relax and drive.

 

I found an interesting thread about the 100+ MPG marathon of a few years back. PC Forum member krausdb was a participant (and organizer, I think) and had a very interesting summary that I think is applicable here:

This was, I think, a Gen II Prius with tires inflated to 60 and they averaged about 35 MPH on a 12 mile real-world road loop with gradual rises and falls and about a dozen traffic lights. Their target P&G range was 30 - 40 MPH, not the 50-60 or 50-70 we've been theoretically talking about here.

All 14 pages of the story, a nice read

 

Wfolta, thank you for all that you have contributed to this thread.

12 seconds to go from 70 to 60. Wow. That's longer than I would have thought. Nice. (I will post my numbers soon).

I definitely understand what you are saying about the engine accelerating at it's most efficient rate in a certain gear ratio. I understand that. However, I still seem to feel that having a light foot (the opposite of a heavy foot) consumes less fuel even though it is not at the most efficient point of the gear ratio at every instantaneous moment. (example: my car "feels" most efficient when I accelerate with the pedal at about 50% of full throttle; however I experience better mpg if I accelerate at 20% of full throttle).

I understand what you mean about pushing on the pedal to stop regen or engine braking. That makes perfect sense, thank you. And I do believe the P&G experts go a step further and use the battery and electric motor to sustain a constant speed (under 45mpg).

I definitely know that wind resistance increases as a square of velocity. You are right that it has a MUCH greater effect for every mph that you go faster, and that is why there is much more wind resistance at 70mph than at 45mph.

I understand what you are trying to say in your quick math with the 8sec 20mpg and 8sec 100mpg, but I don't understand where you got those numbers (from the Prius in-dash display?). 20+100=120, and 120/2 =60. I follow you there. Interesting. You need to try this and post actual speedometer beginning and ending points also though because obviously the acceleration and deceleration (that you are talking about) don't both equal the same 8sec time interval.
I'm interested to see the results though. If anyone else has already done this, please share your experience. Thank you. I've always thought that driving at a constant speed was more efficient (we're talking freeway speeds here), but your "quick math" is making me think twice. I would love to hear from others or see real data. Thanks.

Update: I just noticed your quote and link in Post #42. Thanks!
Reading all about these magic in-dash screens is making me want to go out and buy a Prius right now. It's just a matter of time until I give in and trade in my 115K mile Corolla (that is paid off and could probably go up to 300K+ miles ... the reason I have been keeping it: paid off and next to zero maintenance).

 

Yes, I was using the instantaneous MPG gauge beside the speedometer. It pegs out at 100 MPG, so when you're gliding, you may actually be getting WAY over that (though not infinite, because you're still running electronics, stereo, AC, etc, etc).

Unfortunately, the overall MPG is probably not as simple as averaging the low and high number together, even if they last for the same time, because ultimately, it's gasoline used over distance that counts.

Check out the posting right before your latest. (I probably posted it as you were working on your reply.)

I quote what some real P&G experts did to set a record. As far as I can tell, they use the instantaneous MPG gauge and pick a range (I think each pip is 5 MPG) and accelerate to keep the MPG in that range until they hit their target top speed. So they're not always accelerating (increasing velocity) at the same rate, but they are shooting for a certain consistency in terms of the engine's workload.

 

Oy vey! Such an ordeal just to save some fossil juice. Don't get me wrong, I appreciate the thought and time you put into this. But can we get a Staples "Easy Button" for this machine?

It seems that if mere mortals can figure out how to exceed the rated mileage by more than 20%, an onboard computer should be able to pull it off.

This is the first mainline car I am aware of that exceeding the EPA ratings was even a possibilty. My 2001 Prizm was rated at 43 highway, and while I was very happy with the 35 it was getting, the 43 number was as optimistic as hoping that Toyota will give us all free cars for being such fans.

I found that if I use the Pulse to drive up a ramp of a flatbed truck and then I can Glide for miles and miles without a drop of gasoline. Sure, I miss my family, but you just never know where these trucks will end up.

 

If the onboard computer could do it with no human forethought or tradeoffs, then the rated mileage would need to be increased.

While the conversation may get into minute details, it really boils down to four things:

1. Maintenance: Tire pressure (at least Toyota spec, no more than tire limits, suit to comfort level), oil changes, etc, and don't haul around a lot of junk in the car that you don't really need in there.

2. Moderation/Tradeoffs: keep A/C higher, perhaps, be conscious of what speed you're choosing and don't just drive as fast as you can, etc.

3. Look ahead and Anticipate: don't accelerate hard if you're just going to turn around in a half mile and brake, try to maintain your momentum, brake gradually (subject to safety), and try to glide (the one Prius-specific issue) as much as possible.

4. Don't let incorrect perceptions lower your mileage or make it a pain: don't be a slave to the instantaneous MPG, don't be a slave to the ECO Bar, don't think you're saving gas by accelerating slowly. The displays are instantaneous guides, but per #3, you can see ahead and anticipate. Don't get over-enamored with driving on battery only, since the battery will eventually (sooner, rather than later) have to recharge, so if you go heavy on it you may lower your overall mileage.

And that's the bottom line: it takes a human in the loop because the car cannot tell how fast you feel comfortable accelerating in various circumstances. The car can't know that traffic is stopped a mile down the road, and can't tell that you're about to crest a hill with smoothly flowing traffic.

The car essentially gives you a 40+ MPG baseline (summer), and if you drive like a maniac that's all you'll get. Drive reasonably (and you don't have to be the Slow Guy do do it) and you get 50-55. Get into it as a sport/hobby and you can get 60+. Your call.

 

Wfolta, I performed my test today and to my complete surprise, my numbers were much more favorable than I had predicted. Using the stopwatch on my iphone, just after entering the freeway and having no cars around me, I accelerated from 50mph to 70mph as I normally would (with my brain in fuel-efficient mode), and it took me 30 seconds (+- 1sec), then after checking that there was no one in my rear-view mirror, I took my foot off of the gas entirely and to my complete surprise it took 30 sec (+- 1sec) to decelerate back with engine braking. I was astonished that the two were equal and it lasted a full 30sec. Then I accelerated back up to 70mph, shifted into neutral (something I have much experience with - do not try if you haven't shifted into neutral and back before at freeway speeds) and timed the decel to 50mph. 45 seconds. At that point, I was at my [long] freeway offramp, so I just let it continue ... at 50mph and under, the rate of decel was MUCH less than when I was at 70mph (which should be expected). Interesting. I noticed that 60 to 50 lasted 1.5 times as long as the 70 to 60 decel interval. Conclusion: Wind resistance is significant at 70mph, and P&G is significantly more effective at speeds under 50mph. Future research: I'll test a 50 to 40 and a 40 to 30 deceleration tomorrow. Heck, I'll do a 30 to 20, 20 to 10, and 10 to 0 and graph the results (although I doubt I'll find a flat stretch of road long enough to do the 20 to 10 or 10 to 0, I'll try my best). [on a road with a low speed limit, of course - safety first]

Okay, now I am going to click on the link you gave me and go read that other thread. Thanks.

 

Would you consider doing a differential test of pulse and glide versus the equivalent steady-state, cruise control speed?

There is an SAE paper (2009-01-1322) on pulse and glide but I can not replicate their numbers. Using cruise control to handle acceleration and "N" for glide, I'm finding an 11% improvement for 25-43 versus the equivalent steady state speed:



This is the relative mileage results I got with my 1.6L, NHW11:


BTW, if you have a Garmin nuvi GPS with a USB interface, I have a Perl program that can convert the track data into excel loadable data. This provides a way to record the speed for post drive analysis.

Thanks,
Bob Wilson

 

Hi Bob,

Using cruise control for the acceleration will not result in a figure that is even close to real world numbers. Cruise is way too aggresive in the acceleration to give a meaningful number. You will have to find a different way to if you want to come up with a number that has any credibility. One suggestion is a constant RPM acceleration. I know that you are trying to take the "human" factor out of the equation, but using cruise will just not give a number any where close to what is achievable.

 

Thanks, however you may want to read the SAE paper. Their protocol based upon 'time to accelerate' and it matches the cruise control 'time to accelerate.' In fact, they reported a decrease in performance from lengthening the acceleration (aka., gentler acceleration.) Their protocol was 30-40 mph using both 10 and 20 second acceleration intervals (they did not report a specific RPM or power setting, just the duration.) Still, you are hitting around an area I have some interest from my BSFC studies.

There is strong data from my earlier BSFC chart that suggests peak BSFC occurs at significantly lower power levels in the 1.6L engine than the Toyota SAE papers report. This is also probably true with the 1.8L engine. I think this is a not well understood aspect of the Atkinson cycle. Mapping this is one of my 'projects.'

Bob Wilson

 

Bob, I would love to do a test, but I have no way of measuring the actual fuel consumption (I have a Corolla, not a Prius. I will have a Prius within the the next 12 months, but not yet).

Wfolta, I attempted the 30-20, 20-10, and 10-0 today, however I was unsuccessful because I could not find a road long enough. I found a road with very light traffic and a low speed limit that appeared flat, but at 30mph, 20mph, or 10mph, my car just kept rolling and rolling and rolling until I ran out of road. (car was in neutral) That was in one direction. Going in the opposite direction, I was able to get a number for the 30-20, but for the 20-10 and 10-0 I ran out of road in the opposite direction as well. The approximate length of the road was about one mile. Unfortunately, there are not many roads in Orange County that are near flat for more than a mile (and where you won't get rear-ended by an agressive driver).
Results (including yesterday): In neutral, it took 15sec to decel from 70mph to 60, 30sec to decel from 60 to 50, and then to decel from 30-20 took over one minute, and 20-10 and 10-0, it kept "going and going" until I ran out of road (which was limited to one mile).

Conclusion (coupled with results I posted above): Wind resistance at 70mph and 50mph is significant, it is much less a factor at speeds below 50mph, it is very slight at 30mph, and at 20mph and below, it is undetectable in the absence of advanced equipment.

edit: Bob, I noticed that all of the data points on your graph were performed under 45mph.
edit2: Now that I think of it, I could easily have performed the 50-40 and 40-30 on that road. I'll have to do that tomorrow. Yes, I'm sure many people have already done this and posted very precise graphs, but I'm having fun performing my own experiment.

 

Wow, lots of experimenting and graphs going on here!

Bad news is I may be without my Prius for a while -- starting: not sure when. The Prius has a much smaller turning radius than I'm used to and... I managed to turn into a concrete post in the parking garage. Sigh. Not sure when or for how long it will be in the body shop.

Good news is my mileage continues to climb as I'm trying new things and the car is braking in. On this tank, with a mixture of highway, city, and traffic jams, I'm at about 55 MPG, with a personal record of 57 MPG for a while. Seems to me that I should be able to achieve 60 MPG overall in another month or so.

 

Brothers and Sisters, can I get a "D'oh!!" from the congregation?

Sorry to hear that. Hope the damage was not excessive. Best of luck.

 

the Pri is designed to be ultra efficient which means reducing drag as much as possible, the beltless design, etc. all play a part.

so to coast from a stop even at 10 mph will take a while. i commented on this waaaay back in 2004 that my Pri seemed to coast in neutral farther and faster than any other car i had owned.

 

It is more complicated than that. In your example the total MPG is only 33. I explained this here:
http://ffh.squarespace.com/blog/2009/7/8/the-myth-of-pulse-and-glide.html

 

Add to fuel, and motor oil...