MPG vs "Level" Speed
 

Today I calibrated my ScanGauge II (+4%) and since it was a calm day wind wise I headed over to what passes for a level stretch around here (you can see the difference depending on direction). I used the current trip feature to get an average for the whole stretch, resetting it once up to speed. I wanted to go all the way from 35 mph to 75 mph (all in top gear), but the short test area made getting up to speed quickly enough difficult, once I got up to 70. Someone more comfortable in high speed turns no doubt could do better, entering the straight area with more momentum. Unfortunately, it started raining soon after I began the second set of runs, but here's what I have so far. Have to go back again sometime, but since I have modifications in the works I guess it'll most likely mean starting over. Surprisingly high economy at 35 mph. Makes me suspicious.

https://www.fuelly.com/attachments/fo...2cb4851dc2.jpg

Based on these results I've been making an effort to drive (even) slower when on a level and it does seem to help fuel economy over the 50-55 I was doing before. I've changed a couple of the routes I take to lower speed limit ones to make this easier, although other cars still limit it a lot.

I just went to the local hardware store and I realized I was doing the same sorts of speeds my moped did (30-35)! Seems to work well. Didn't have podcasts on my moped, though. Big difference.

Just looking at your plots, I suspect the dip you see around 40 mph is the result of timing retard. Most likely it is mapped to provide maximum resistance to detonation at that rpm against that kind of loading, whereas just the drop to 35 mph reduces the drag loading substantially, allowing less timing retard to essentially just put along.

In other words, you might actually get better fuel economy in your next lowest gear at 40 mph. Quirky, but worth investigating IMO.

I'll give it a shot, thanks for the tip.

I don't suppose anyone else would be willing to generate a plot like the one I did to provide some comparison? The mpg vs speed graphs in my books don't show 35 mph being optimal, that's for sure. Maybe it's related to the mediocre coasting performance I've noticed. I'm waiting on my mirror mods to rerun it myself.

Today I went on a trip to the nearest Lowes (about 66 miles away) and I used the limited speed option on my TomTom to get a route on which to try out the lower speed techniques. Took 2 hrs at an average speed of 35 mph, but I managed 50 mpg (on the long parts of the trip, bumming around shopping brought it to 49.5 for the day). I brought plenty of podcasts to pass the time. Kind of dull, but it does seem to work.

I'll be doing some experimenting next week, and will try to generate charts like yours. I started slowing down when I discovered my mileage was linearly inversely related to my speed. I got 26 mpg at 72 mph, 32 mpg at 62 mph, and 38 mpg at 55 mph. I know my mileage rises to the mid 40s at 40 mph, but forgot the specifics. Pulsing and gliding raises my highway mpg to something like 43 mpg with a 50 mpg average. This is all in my Sentra.

It'll be awesome to see another car's chart for comparison. :thumbup: I was totally surprised that the lower speeds continued to be better. What car computer are you using? I've never measured pulse and glide, since I'm too lazy to do it in practice.

I've gone through some earlier notes, and my older data look pretty linear, peaking at 35 mph, just like you. The data are a mixture of tank fill-ups and Scangauge readings, but they look something like this:

72 mph = 26 mpg
62 mph = 32 mpg
55 mph = 38 mpg
40 mph = 45.6 mpg
35 mph = 50 mpg
high speed pulse and glide (avg 55 mph) = 43.5 mpg
low speed p&g (avg 28 mph) = 62 mpg

I'm using the Scangauge II in my Sentra SE-R. I get similar results with my Scion xBbut about 5-8 mpg higher overall.

Here's the mileage to speed chart for my SE-R. All values obtained in 6th gear.
https://www.fuelly.com/attachments/fo...ce5c4cee69.jpg

And here's the corresponding chart for my xB. There's an error in the data. The first bar should be 70 mph, not 72 mph. Note the 30 mph value drops off the straight line. That's because it's a 4th gear value and all the others were obtained in 5th gear.
https://www.fuelly.com/attachments/fo...f7ddcb4b82.jpg

I hope you don't mind, but I find xy graphs easier to interpret. Your graph is a lot more linear than mine. Interesting. The books I've read always show a local maximum at 35 mph, and then a global maximum in the 45-50 mph area. So much for that. Thanks for the info!

https://www.fuelly.com/attachments/fo...bb09c29223.jpg

https://www.fuelly.com/attachments/fo...f3c955d09c.jpg

Yes, I like x-y graphs, too, but I hadn't figured out how to create them in Excel by my previous post. I looked up how to make them on the Internet, and can do it now.

Pulse and glide creates another line about 5 mpg higher for each speed between 30 and 55 mph. However, there's so much variability, the P&G line can be 10 or 15 mpg better at the lower speeds, and the actual speeds to achieve P&G averages are 10-20 mph deltas.

Once you get away from anything basic the number of possible combinations sure explodes fast. Where are you using for your testing? Finding a straight and level spot here was tough.

The data points are optimal ones. If I hit traffic, or drive the cars on different roads, the mileage data almost always go down. I have a country road test loop that's a frontage road to a freeway. It rarely has traffic on it, and is reasonably level, since it's at sea level. I run it both directions, and get a ten mile drive that's pretty consistent. I used to drive the loop twice for a 20 mile drive with slightly higher mileages, but 10 miles is enough to show the differences of different P&G techniques, and different cruise control speeds. When I do the high speed tests (55 mph and over), I do them on the freeway that parallels the frontage road.

I coast as much as possible at the end of my frontage road drives, and my mpg figures look higher as a result. When I do the high speed drives, I get the car up to 55 mph, reset the Scangauge's current trip data, and record the mileage 1/4 mile before I take the exit to turn around. I get better highway mileage driving southbound, so I average high speed mileage going both directions.

I gathered some more data today with my Sentra, using cruise control in different gears, and a low speed P&G average. Here's the chart.
https://www.fuelly.com/attachments/fo...e40c0885f2.jpg

The speed you show for P&G is the average travel speed, right? Just checking. That really helps put things into perspective. We need a data logging system really badly. :)

Yes, that P&G speed is the average between 25 mph and 40 mph, which is usually in the 28-30 mph range, depending on the downhills, length of coasting, etc.

There's a high speed P&G blip at about 46 mph on my Scion, and about 55 mph on the Sentra, where the mpg for each is about 5 mpg higher than the cruising mpg for the same average speed. On a good day, I've been able to coax 59 mpg out of the Scion over 165 miles, and I averaged 51.5 mpg over 1213 miles from Gallup NM back to the Bay Area in the Scion. That was done with a lot of coasting down long downhills that brought my mileage up. On those long downhills, I can pulse for 5-8 seconds, and coast for minutes.

All of which is why I use average MPG calculations. That, and I don't have a ScanGauge.

Max... Am I the only person at gassavers who doesn't know what vehicle you are talking about?

Good point. I only have the one in the garage (on the left), so I guess I have been taking the context for granted. It's my 2008 Hyundai Accent, Gold Digger Repellent.

another possibility is that he shifts at about that point.