Uphill vs. downhill makes the biggest difference

I am new to Prius, just completed my 4th day. I agree that there is a hugh difference between going uphill and downhill. The hybrid drvetrain seems to exaggerate these differences vs a conventional engine/transmission.

I find driving uphill to be the most difficult adjustment. It seem hard to drive at a steady speed. It feels like I have to accelerate going up hill to keep from slowing down.

I have a scangauge and have been monitoring ICE RPM. The RPM jumps up quickly when climbing a hill. I happen to have a rather steep 1/4 mile hill to get to my house. Peak RPM was 3600. I assume that a lot of that is going to spin MG1 to send power to MG2 to provide needed torque. This is where the Prius HSD with it's power splitting device is clearly challenged, can't just shift into a lower gear.

My old car, a manual transmission Scion tC, could climb the hill steadily at what ever speed I wanted.

I am enjoying learning the ins and outs of the Prius. I expect that it will take a while to master it. Any advice from the experienced?

My MPG experience is similar to others with a clear drop when going up. My overal average so far is about 50 and I'm satisfied with that.

 

I don't know what you mean by "challenged".

The Prius runs the engine at almost wide open throttle most of the time as that is the most efficient. So it is always in the right gear, if you attempted to change to a higher gear (did you mean Higher or lower) the engine would not have enough power to climb the hill.

If the engine is running at 3600 RPM that is because it needs to in order to produce enough power. Don't forget it only has the equivalent of about a 1.3L engine (it's 1.8 but the Atkinson cycle pushes out about 1/3 of the inlet charge)

kevin

 

I did mean lower gear. The street is not only steep but winds up the hill. In the Scion, I would shift down into second to provide plenty of torque for a fairly slow controlled drive.

Even a 1.3l engine could climb that hill steadily with the right (conventional) gearing.

As I said, I am new to Prius. From what I have read, the torque from the ICE that goes directly to the wheels can't be adjusted. Therefore the increased RPM goes to send power via MG1 to MG2 to boost torque. Yes, I assume that it is in the right gear, but there is some loss of efficiency when so much of the power has to be converted to electricity and then back into mechanical energy, doesn't there? It certainly feels like it is working hard. Obviously, there are many other circumstances where it is much more efficient than a conventional setup.

My main comment is that I have had a hard time finding a sweet spot where I can feel comfortable driving slowly up the hill. I've only driven the hill twice, so I assume that I will get more of a feel for it.

 

I did a trip yesterday when I thought I'd try and get the best mpg I could. 35 miles each way, cold start. Parked at destination for about an hour so another semi-cold start before return trip(though temperature was warm, about 25).
Outward average was 76mpg, overall after 70 miles was 79.2mpg so return trip must have been over 80.
About 25 miles each way was on a motorway(interstate?). I cruised at between 57 and 62, mainly watching the instantaneous mpg and just trying to feather it as much as possible whilst maintaining reasonable speed.
The interesting bit for this thread was a stretch of 3 miles which I know well and have always considered flat. I maintained a constant speed of 62mph on this stretch with the HSI just a midge over the centreline and the instantaneous mpg showing 100mpg. (Is this warp-stealth?). I held this for a full 3 minutes or so.
Anyway I thought I'd check the gradient using Garmin data I've got from previous trips and that actually shows that there is a very slight downhill gradient, a drop of 32 metres over 3 miles.
Knowing that now, I will take advantage of it every time I drive that route.
It makes me wonder (and I've raised this in another thread) if there is a benefit in knowing in advance about gradients, particularly those which are so slight as to be not apparent, but enough that you can drive the Prius to take advantage of.
I'm thinking of buying some sort of altimeter!

 

I believe everyone understands we need more fuel to drive uphill to gain the potential energy.
How much fuel do we need? That's the question!

FYI,
Please refer to my post for Gen2 Prius.
We need 106 mL of gasoline to get 100m altitude change.

When we drive downhill, we have the extra 106 mL of gasoline energy to a goal 100m below starting point.

The Gen3 Prius is more efficient, therefore I think 100mL/100m will be the number.
I'll try to get the number on my Gen3 in the near future.

Ken@Japan

 

I'm not allowed to view that post. Is the "prius-connection-detriot" forum restricted to those who were invited to that long-ago event? If so, can you copy it over to here?

Based on a different old post, I've been using a rule-of-thumb approximation of 1 extra gallon per 10,000 feet of climb, which corresponds to 124 ml/100m. After a number of round trips with over 1000 meters of climb and descent, this rule seems to be within roughly 10% of my actual results, but I haven't performed any strict tests.

How much cargo mass is assumed in to 106ml/100m rule?

 

At what speeds?



What I found using Shell 89, E10:

• 0.1286 gal/1000 ft @ 55 mph
• 0.1371 gal/1000 ft @ 65 mph
• 0.1486 gal/1000 ft @ 77 mph

These were derived from the indicated MPG numbers so they are probably ~5% low at these speed. I'll adjust them later using the calibration curve and convert them to standard units. I needed to do that anyway. FYI, the starting altitude is ~600 ft. above sea level on this course.

So what units do folks want?

• ml / 100 m
• gallons / 10,000 ft.

I have no preference about the units.

Bob Wilson

 

My rule-of-thumb is for extra fuel, compared to the same speed and distance on level road.

For your graph -- what slope? or horizontal distance?

 

I was told it is a 6% grade with a trip meter distance of 1.2 miles for a GPS confirmed 525 ft change in altitude. However if you use the altitude / tripmeter distance, it comes out as 8.3% but the trip meter is the hypotenuse. Humm, that means the grade may be a little steeper than 8.3%. A friend whose Dad works in the highway department told me 6% but lazy, I never really went back and calculated the grade. I've only been interested in altitude changes since that along with the weight maps into potential energy.

I have a little flat run out at the top because it is easier to use the intersecting road as the stopping point. The starting point is a sign outside of a highway patrol 'station.'

Bob Wilson

 

The hypotenuse vs. horizontal difference is less than 0.03% at this slope, which is invisible at the number of significant digits displayed, and at the accuracy of the measurements available.
- - - - - - - - - -
My routes nearly always have rolling hills and variable slopes. I'd prefer a fuel rule having two components, one portion for level road and a separate portion for net elevation change. If such a rule can be made reasonably accurate, absent friction and compression braking, it would be much more useful to me than a fuel-vs-slope-&-speed chart.

Ken's guideline of 106ml/100m fits the form I'm looking for. I'd like to see his linked post, blocked from my view, reposted here..

 

My bad. I replaced the link with public viewable one.

We don't need the cargo mass value there.

91km/h reading speed or 84km/h real.

Ken@Japan

 

I'll try this figure, keeping better track of segments results and trying to estimate 'wasted' energy in friction and compression braking coming down steep grades.

For those who don't want to perform the conversions, Ken's 106ml/100m corresponds to 1 gallon per 11,700 feet, or 0.085 gallon per 1000 feet.

 

As a follow up, I have gotten better about climbing the hill. I just have to keep my eye on the road and not on the HSI.

I measured the hill with a gps. About 8.3% slope over 1000 feet. Steeper in the middle.

Milage drop is less if I go slow and steady.

 

Ken, how was the 106 ml / 100 ml figure determined? Is this a theoretical calculation, or an empirical best fit to real world observations?

I've tried simple versions of both approaches with the following results.

Assuming gasoline is 720 g/l, Prius engine efficiency of 220g/kwh (from your BSFC charts?), and US-market Prius mass of 1380 kg empty / 1754 loaded, I get:
Empty: 115 ml / 100 m (10800 feet elevation gain per gallon)
Full Load: 146 ml / 100 m (8500 feet / gal)

On a round trip to the mountains, climbing 5900 feet, my MID displayed 47.2 mpg for 105.2. The return trip achieved 84.6 mpg over the same distance. The descent involved purely friction and compression braking for one 1800 feet vertical section, and for about 300 feet in other sections where I had been unable to sufficiently drain the battery beforehand.

Converting to SI units, this was a climb 1800 m, 20.07 l/km up for 169.3 km, 35.97 l/km down, and 640 meters of non-regeneration descent. This route used 8.44 liters up, 4.71 liters down, for a difference of 3.73 liters due to elevation change. That computes to 104 ml / 100 m without adjusting for non-regeneration, or 126 ml / 100m with adjustment. In English units, this is 11900 feet/gal unadjusted, or 9860 feet/gal adjusted.

There are some other complications to this, such as transmission losses missing from the theoretical calculation, and significant temperature change between ascent and descent, and dissimilar warmups, on the road test. Those will be subjects for future investigations.

Ignore this paragraph. On hindsight, unmatched entrance/exit ramps can account for the difference. [On a different subject, the uphill leg should have been about 0.5 mile shorter than the downhill leg, due to a late reset of the trip meter. But that didn't happen. Do the tires distort on steep uphill and downhill grades?]

 

edit: please note that my data is about Gen2 Prius.

I did latter one. I used Excel spreadsheet changing the parameter to see a reasonable flat result for the adjusted FE number.

I did a different way as follows;
My car inspection sheet says 1290kg empty. I'm a skinny person, so I set 1350kg test vehicle weight.
The 100m potential energy: mgh=1350X9.8X100/1000000=1.32MJ
The 106mL gasoline energy: E=0.106X34.6=3.67MJ
The vehicle efficiency: 1.32*100/3.67=36.1% or 230g/kWh

My test highway was gentle up/down hills, so I did not use brakes and the battery level was always six.

I'm not sure about that, but I have heard something like "static load tire diameter" and "dynamic load tire diameter". The "static load tire diameter" is approx 92% of no load tire diameter, and the "dynamic load tire diameter" depends on a lot of factors, such as load, speed and so on.
I think the front tire load becomes low on the uphill leg, then the tire diameter goes large, so we see short trip meter reading.

Ken@Japan

 

'might' not be broke in? ... there's no 'might' about it ... you car's not even nearly broken in yet. Give it another 9 or 10 thousand ... THEN it'll be broke in. I didn't start hitting the 60mpg's until we had over 10k miles on it.

 

Using a gross trip approach:

105.2 mi / 47.2 MPG = 2.23 gal. climb fuel burn
105.2 mi / 84.6 MPG = 1.24 gal. descent fuel burn
0.99 gal. difference
(0.99 / 5,900) * (1.69 / 1.69) = 1.67 gal / 10,000 ft
NOTE: I would prefer to take out the flat land fuel burn since there may be things going on a descent that save more than the 1.24 gal.
​

Another result:

Using this we get:

10000/11900 ~= 0.840 gal / 10,000 ft.
10000/9860 ~= 1.014 gal / 10,000 ft.
​

Taking my hill climb data at 55 mph and MPG on flat land:

1.2 mi / 59 MPG (@55 mph) = 0.0203 gal. on flatland
0.0646(*) - 0.0203 = 0.0443 gal. climb fuel burn
0.0443 gal / 525 ft * (19.04 / 19.04) = 0.843 gal / 10,000 ft
​

Conversion of Ken@Japan's numbers:

106 ml/100m = 0.855 gal / 10,000 ft.
​

I suspect Ken@Japan is using straight gas whereas fuzzy1 and I both use E10.

So if we make a table to look for consensus:

	Column 1	Column 2	Column 3	Column 4
0	gal/10k ft.	method	by
1	1.67	MPG differential	bwilson4web
2	1.014	adjusted	fuzzy1
3	0.855	efficiency analysis	Ken@Japan
4	0.843	differential climb vs flatland fuel	55 mph	bwilson4web
5	0.840	unadjusted	fuzzy1

.
Averaging the center three or lower four, we get:

0.904 gal / 10,000 ft (middle three)
0.888 gal / 10,000 ft (lower four)
0.846 gal / 10,000 ft (lower three)
​

This is what we'd call an informed guess, back of the envelope, or best estimate of hill climb fuel consumed as a function of altitude. Without a series of benchmarks at different speeds along with or using HSI settings, we don't have enough information to call out an exact hill climb curve. Regardless, a rule of thumb like this can help with trip planning ... YMMV (your mileage may vary.)

Bob Wilson

* - adjusted for 5% calibration error in MFD

 

My question here should be ignored. With hindsight, it is now obvious that the difference was in unmatched entrance and exit ramps to divided highways.