On my way out to vacation I got a chance to do ~90km on the highway in relatively flat terrain (we usually go to the mountains, so most of my long distance driving is rolling and hilly terrain). It was not a pool table, but pretty close, it was much flatter than anything I've driven on in the last few years. I decided to use this chance to check the efficiency difference between steady, cruise control driving and PNG (P&G in Neutral, ie with engine on).
Since this came up at the spur of the moment, the test is not exactly ideal. The only thing I could come up with was to watch the engine coolant's temperature as a measure of efficiency. The outside temperature was ~21°C (68°F) with a strong side wind. I had my radiator's lower grille blocked (upper grille and intercooler were unblocked). Each try lasted at least 5-6 km (3-4 mi) to allow everything to stablize.
- Driving on CC at a steady 95 km/h (59 mph) leveled the coolant temp at 92°C (198°F),
- Driving on CC at a steady 90 km/h (56 mph) gave 91°C (196°F),
- Pulsing to 100-105 km/h, then engine-on coasting to 80-90 km/h quickly brought the temp down to 86°C (187°F), and after a while to 84°C (183°F)
During vacation I planned ahead so that on the way back I could do a better test. This time I made sure the distances were the same on each try and I used the car's OBC to measure fuel consumption and average speed. I tried to make sure that average speed during PNG was the same as the set CC speed in the previous run. Unfortunately, I wasn't able to do as many tests as I had planned - during the fourth run our infant son started crying and I had to make an emergency stop for an hour
Outdoor temperature was 32°C (90°F) with a headwind, lower grille block open. Tests #1-3 were 12 km (7.5 mi), test #4 was 8 km (5 mi).
- Cruise control at 94 km/h (59 mph), engine load 40-60%, consumption 3.5 l/100km (67 mpg), coolant temp 81°C (178°F),
- Pulsing to 105 km/h (66 mph), engine-on coasting to 85 km/h (53 mph), average speed 94 km/h (59 mph), engine load ~80% while pulsing and ~25% while idling, consumption 3.1 l/100km (76 mpg), coolant temp 79°C (174°F),
- CC at 90 km/h (56 mph), consumption 3.4 l/100km (69 mpg), coolant temp 82°C (180°F),
- PNG between 80-100 km/h (50-63 mph), average speed 90 km/h (56 mph), consumption 3.4 l/100km (69 mpg), coolant temp 81°C (178°F).
NOTE As I found out during the tests, #3 was slightly downhill towards a large river and #4 was climbing back out of that river basin. This explains why PNG had the same average fuel consumption as CC at the same avg speed.
Summing up, of all of those tests only #1 and #2 on the return trip are anywhere close to ideal tests. They imply that driving at a steady 94 km/h consumes 13% more fuel than PNG between 80-100 km/h with the same average speed. Again, this was in flat terrain during hot weather.
This means that PNG is still efficient on flat roads at highway speeds up to 100 km/h. Since the speed limit on this highway was 140 km/h I could have tried to find the break even point, but that wouldn't have been fun, since I don't like going that fast. My guess is that CC starts to take over when the engine is at constant 80% load, assuming that gearing allows rpms to stay in the ideal range for best BSFC. In rolling and hilly terrain I'm sure that PNG would allow better performance, since it would be possible for the driver to syncronise pulses and glides with hills.
As a side note, please praise me for going at only 2/3 of the PSL
