Effect of gradient on fuel consumption...

[anko]

An engine running for 10 minutes at 80% load and then for 10 minutes at 60% load may operator a little bit more or less efficient than an engine running all 20 minutes at 70% load.

To be honest, I don't see what friction and / or losses have to do with it, until the gradient becomes so steep that you could actually cost or even recover energy during the descents.

Okay. There is definitely a misunderstanding here. I'm not sure why you think a car's engine would be running at 60% while going down hill?

I was thinking of a situation of 10 minutes at 40% load and 10 minutes of regen, vs 20 minutes at 5% load. Friction and losses would come into play because we can't regen all the potential kinetic energy we stored going up the hill.

Like I am not sure why you think you can maintain a decent speed at 5% average engine load ;-)

Regardless, the 60%, 70% and 80% were just my way of expressing my thoughts on how the extra load uphill could translate in a reduced load downhill. Like I already said, when the descent is so strong that you can coast /regen downhill it becomes a little bit off a different story. But on a section of road, 25 miles long where the overall incline is 150 meters, how much steep sections will there be? Also, when I take my car with me on vacation in the Alps, I do a lot of driving in the mountains there. But yet, my fuel consumption is not higher in that period than it is in our own little flat country. This was also the case with my previous, non-regnenning car.

Finally, if the effect you describe (losses due to local height differences) exists, it most likely also exists in the second part of Maby's trip. And therefor it still does not explain the 25% difference in fuel consumption.

 

[AndyInOz]

... it's a 150m climb over a distance of almost 30 miles.

That completely messed up my head. 150 metres over 30 miles, that's as flat as a billiard table...

Then I realised that m stands for miles in this conversation.

D'Oh!

 

[Sunder]

Like I am not sure why you think you can maintain a decent speed at 5% average engine load ;-)

Call it too many hours in a misspent youth (and fortune) tuning cars, followed by a currently being misspent adulthood designing electric bikes. Check out this link: http://phors.locost7.info/phors06.htm

You can see near the bottom that at 30mph (48km/h), the power required to maintain the speed is only 1.16hp, or 870w. Coincidentally 1% of our 87kw available.

At 55mph (or 88km/h) you need 7.14hp, or 5.3kw, or about 6% of the available power of the motor.

So, yes, you can sustain a decent speed at very low engine outputs.

Yes, the page is doing the calculations for a sports car, but the difference between a Prius and a Hummer is only about 2.2x, so even if you doubled those figures, cruising at 60km/h, the most common speed zone here in Australia is still less than 5% of the engine's available power.

Regardless, the 60%, 70% and 80% were just my way of expressing my thoughts on how the extra load uphill could translate in a reduced load downhill. Like I already said, when the descent is so strong that you can coast /regen downhill it becomes a little bit off a different story. But on a section of road, 25 miles long where the overall incline is 150 meters, how much steep sections will there be? Also, when I take my car with me on vacation in the Alps, I do a lot of driving in the mountains there. But yet, my fuel consumption is not higher in that period than it is in our own little flat country. This was also the case with my previous, non-regnenning car.

Finally, if the effect you describe (losses due to local height differences) exists, it most likely also exists in the second part of Maby's trip. And therefor it still does not explain the 25% difference in fuel consumption.

I'm not sure how I'm going to argue against anecdote. 150m over 40,000m (25 miles) is 1:233, or about 0.4% grade. As Andy says, that's pretty much as flat as you can get a billiard table.

Go to this calculator and play with the figures: http://bikecalculator.com/

At 150w, completely flat ground, you can go 28.44km/h

Adjust the gradient to 0.4%, and the speed drops to 26.56km/h.

Force the speed back up to 28.44km/h, and your power rises to 176watts, or an increase of 17.3%. Add a bit of fudge factors, and 25% isn't out of the realm of possibility.

Does that surprise you? I wish there was an equivalent for cars, as mass and wind resistance is slightly different for cars - but it won't be as much as you expect.

 

[anko]

At 88 km/h our engine has about half of 87 (I thought it was 89) kW available, due to the long gear ratio. So you need to double the engine load once more to get the same kW.
Our battery can deliver 5.3 kW output during roughly an hour and a half. At 88 km/h this would translate to 132 km travelled. Okay, there will be losses, but even when these losses are 50% (which would be huge for an electric drive train) it will still be 66 km.
Set your CC to 88 km/h and see where your power meter in average points to. I am pretty sure it is way more than 1/12th of the green part of the dial (which is 60 kW). Even at a cruising speed of 60 km/h.

But again, this was all not about engine load. It was about the impact of gradients.

Your example with bikes is interesting, but at 28 km/h the effect of drag is much much smaller than at 60 km/h (or another decent speed). So, the relative impact of an incline is much bigger at low speeds / smaller at higher speeds.

 

[Sunder]

What you say is true, but the coefficient of drag on a car is also much smaller than the coefficient of drag on a cyclist. So at higher speeds, the impact on a car would be smaller than the impact on a bike.

Also, you might be surprised at how inefficient electric motors are, when they are not at their peak efficiency. I'm guessing if you gave me a perfectly flat road, with no starts or stops, no wind, no corners etc, I probably could get 66km from a full charge. Such a road doesn't exist, I'm afraid.

Anyway, I guess it's the difference between maths and anecdote.

May I instead argue anecdote with anecdote?

Where did the extra 25% fuel go for Maby then? You are saying there's no difference between driving in the Alps and driving on the flat. Maby is saying that a 0.4% gradient - which wouldn't be visible to a naked eye - is costing him 25%.

Whose anecdote is wrong?

 

[anko]

Where did the extra 25% fuel go for Maby then? You are saying there's no difference between driving in the Alps and driving on the flat. Maby is saying that a 0.4% gradient - which wouldn't be visible to a naked eye - is costing him 25%.

Whose anecdote is wrong?

Agreed. Maybe both are right (or wrong), to a point where they meet somewhere in the middle

There are many other factors that could have contributed, a cold engine being one of them. Last Sunday, I have done a trip that I have done many times. But most times, I started off with a full battery and it was warmer. This time not so. So, my engine ran from the moment I departed. Initially fuel consumption was off the charts (obviously) like 23 l / 100 km. It dropped quickly when I left city limits. Then it kept getting lower, but only very slowly. Most of the trip it spent between 10 l / 100 km and 9 l/100 km. In the last couple of km's it dropped to 8.7 l / 100 km.

it is kinda funny to see how people (including myself ) bring up these and other other factors or ignore them, depending on whether they agree or not agree with the outcome of an experiment . There have been a few times where 'results' presented by myself or others where dismissed because 'they were not obtained in a controlled environment', where the same people that dismissed these results later presented their own results obtained in a similar uncontrolled environment. We all seek the truth, but we search hardest for the truth we like most, I guess :lol:

 

[Sunder]

Where did the extra 25% fuel go for Maby then? You are saying there's no difference between driving in the Alps and driving on the flat. Maby is saying that a 0.4% gradient - which wouldn't be visible to a naked eye - is costing him 25%.

Whose anecdote is wrong?

Agreed. Maybe both are right (or wrong), to a point where they meet somewhere in the middle

I'd say so. 25% would be high for a pure 0.4% gradient, but I really suspect that the road is a series of 1m falls over 100m, followed by 1.2m rises over 100m, non visible to the naked eye, but enough to chip in a bit here and there.

There are many other factors that could have contributed

Also totally agree. And I'm trying not to get bogged down in them so as to not be a pedant, and not spend too much time on here. It's just too difficult to control for these things.

At the end of the day, it's all academic anyway. We want to win the argument, but will it help us get better efficiency or performance out of the car? Unlikely. I'm happy to leave the debate here. Thanks. .

 

[maby]

...

May I instead argue anecdote with anecdote?

Where did the extra 25% fuel go for Maby then? You are saying there's no difference between driving in the Alps and driving on the flat. Maby is saying that a 0.4% gradient - which wouldn't be visible to a naked eye - is costing him 25%.

Whose anecdote is wrong?

Surely the difference is that alpine driving is more of a mixture of climbs and descents than the episode I was describing. I can well accept that over a period of days, anko saw a fuel consumption in the mountains that was similar to what he sees in his own (relatively) flat country. My bet is that if I were to measure my fuel consumption driving from the house to the boat and straight back again, I would see a significantly better figure - I would get the equivalent benefit on the down hill to the negative impact I seem to be seeing on the uphill.

My post was an observation on the fact that the impact of such a gentle gradient was greater than I would have expected - and anko has demonstrated mathematically that an impact of 10% is to be expected - even that is greater than I would have naively expected on a run where a walker would probably not recognise the gradient. It is also a warning about measuring performance naively - when I was running some tests to investigate anko's theories about the impact of SOC on fuel economics a few months ago I was doing it on this exact route but in the opposite direction. In fact, the first part of the test was on the flat and the second part started pretty much at the top of the descent - anko's calculations seem to indicate that the gradient could impact the results by at least 10%.

 

[anko]

Exactly my point. I brought up my Alpine experience, only to explain why I though that any local height differences (up 1.2 meter, down 1 meter, and such, which must exist during your 25 miles long climb to 150 meters) should not matter that much. All that matters is the fact that you eventually have increased your height by 150 meters.

 

[Sunder]

My head is now spinning with everyone arguing in every direction. So where does everyone think the missing 15% went? (10% expected compared to 25% actual).

I'm arguing the 15% is the "meaningless" dips and rises. Of any kind of energy conversion, it would most likely be the most efficient - because it's a straight kinetic to kinetic energy conversion. You might gain a kilometer an hour or two, and then lose it on the next micro incline.

Even in Alpine driving, in a perfect ideal situation, you would double your power going up, and spend no power going down. But only in a perfect situation. If you start hitting (non-regen) brakes, you'd lose some of that stored potential energy to heat.

Meh. Way too many variables to track who is arguing what.

Let's just say 25% doesn't surprise me. If it surprises you, that's okay.

 

[maby]

My head is now spinning with everyone arguing in every direction. So where does everyone think the missing 15% went? (10% expected compared to 25% actual).

I'm arguing the 15% is the "meaningless" dips and rises. Of any kind of energy conversion, it would most likely be the most efficient - because it's a straight kinetic to kinetic energy conversion. You might gain a kilometer an hour or two, and then lose it on the next micro incline.

Even in Alpine driving, in a perfect ideal situation, you would double your power going up, and spend no power going down. But only in a perfect situation. If you start hitting (non-regen) brakes, you'd lose some of that stored potential energy to heat.

Meh. Way too many variables to track who is arguing what.

Let's just say 25% doesn't surprise me. If it surprises you, that's okay.

Well, let's not get too hung up on exact percentages - I didn't set out to conduct a scientific experiment and simply noticed some figures that surprised me part way through my journey. There would be a variety of other factors to eliminate before we get to a definitive figure. Anko's calculation of 10% is a bit of a "back-of-a-fag-packet" figure - useful to put it into context but not to be considered definitive either.

 

[anko]

I'm arguing the 15% is the "meaningless" dips and rises. Of any kind of energy conversion, it would most likely be the most efficient - because it's a straight kinetic to kinetic energy conversion. You might gain a kilometer an hour or two, and then lose it on the next micro incline.

Although I doubt it, it might explain 15% difference between a trip in the Netherlands and a trip in a more hilly environment, but how can it explain 15% difference between the first half and the second half of Maby's trip? As both half of the trips most likely have these "meaningless" dips and rises.

 

[Sunder]

Anko's calculation of 10% is a bit of a "back-of-a-fag-packet" figure - useful to put it into context but not to be considered definitive either.

Yeah. I would have considered it a textbook answer, not a real world answer. The same way that in a textbook a perfectly elastic ball can bounce forever

Anyway. All fascinating, and all relevant. I have been helping a guy on an electric bike forum build a bike that can go 45km/h on the flats, but will not melt the motor on a 15% gradient hill. Not difficult if you can strap a large multi-kilowatt motor to your bike, but much more challenging when you want a stealthy micro motor.

We've been working through the impact of different winds, different stator sizes and lamination thicknesses, different voltages and different methods of commutation. Again, so many things to wrap your head around.