-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Yes, if you note the times when you are doing various activities so you can correlate the trip record with what you are doing. That would allow you to integrate the total usage over the trip, or at least do some averaging.
I think it would be better to do the same hill under the same conditions for an accurate look at what regen puts back in.
As far as I can see, elevation data is only recorded every 10 minutes. It probably won't have enough resolution to be helpful unless it's a very long climb, except as a rough indication.
I suppose you could park at the bottom for over 10 minutes, drive up the hill, park for over 10 minutes, drive down, and park for over 10 minutes. That way it would be obvious where the trips started and ended, and what elevation they were at.
William13
Active Member
1. The 56 kWh is the "nominal" capacity of roadster battery pack. I.e. mAh * V * number of batteries. The car will not allow all of this to be used as it would damage the batteries. The 53 kWh is the amount available in "Range" mode. This will also damage the batteries though not as much.
2. The regeneration will not work immediately if batteries are too full or too cold.
3. Regeneration is not as efficient as coasting. Coasting would imply a slower speed. Air resistance goes up very quickly with relatively small changes in speed. Someone who wants to get from point A to point B slowly can be more efficient on a bike.
4. The Leaf argument that it can carry 5 people is hollow unless the mileage is tested with the weight of five people in the car. How far would the range drop? I suspect quite a large percent. Cars have only one or two passengers a lot. I expect 90% of the time.
5. If you want a high efficiency electric hope that the Wave 2 gets built by Lion Motors. Aptera is unlike to survive and the Wave 2 is equally unlikely to sell well until energy is very expensive.
2. The regeneration will not work immediately if batteries are too full or too cold.
3. Regeneration is not as efficient as coasting. Coasting would imply a slower speed. Air resistance goes up very quickly with relatively small changes in speed. Someone who wants to get from point A to point B slowly can be more efficient on a bike.
4. The Leaf argument that it can carry 5 people is hollow unless the mileage is tested with the weight of five people in the car. How far would the range drop? I suspect quite a large percent. Cars have only one or two passengers a lot. I expect 90% of the time.
5. If you want a high efficiency electric hope that the Wave 2 gets built by Lion Motors. Aptera is unlike to survive and the Wave 2 is equally unlikely to sell well until energy is very expensive.
The log's aren't detailed enough. But! the data on the CAN bus is...
If someone wants to come up with a test case that the group agrees is valid. And someone in SoCal will drive the route. I will bring my CAN tools, log the data, and post the raw data. I'll log the entire run (so I can extract stuff out later, if needed) Someone else will need sign up to parse the data for the answer.
Here's the CAN data I can get:
- GPS: time, lat, long, elevation, heading (every 2sec)
- TIRES: pressure, temp, vehicle speed (possibly rpm per wheel) (100ms)
- ESS: Voltage & current (100ms)
- Motor: torque (and possibly current) (100ms)
- PEM: (I'll need to find regen on/off) (100ms)
- TEMPS: motor, pem, maxBrick, coolant, outside (2sec)
I can get faster gps data updates, but I don't know if the GPS is providing location fixes any faster than 1 fix/2 seconds
Last edited:
EVNow
Well-Known Member
According to this blog regen is 64% efficient, max.
The Magic of Tesla Roadster Regenerative Braking | Blog | Tesla Motors
I doubt Leaf's regen is any more or less efficient - may be a few % points here and there - with little difference in practice.
Last edited:
smorgasbord
Active Member
I live at an elevation that gives me everyday opportunity to test this out in the real world.
One part of my drive is 3.3 miles long, with a net elevation difference of 1565'. Going up I used 11 miles of Ideal Range. Coming back down I regained 3 miles, for a net difference of 8 miles used along a 6.6 mile route. This is top up, no lights, radio on, traveling about 25-35 mph on twisty roads, not touching the brake pedal either way (no stop signs on the route!). A loss of 1.4 miles along 6.6 miles is only 21%, which means a 79% efficiency. Of course, Ideal Range is probably not exactly SOC and the numbers are rounded off, etc.
I didn't use the Torque/Power/Accel screens at all. Maybe I'll get a passenger and have them note things and record the screens.
smorgasbord,
I'm guessing that the miles driven up the hill shouldn't be used in the regen efficiency calculation.
One way to think of ideal miles is as energy units. So, at the top of the hill you have 7.7 energy units in the bank (11 – 3.3) from gaining altitude. Going down the hill you use up 3.3 of them overcoming wind resistance and rolling friction, etc. At the bottom you have only three left in the bank instead of the expected 4.4 that would result from 100% regen efficiency. Dividing 3 by 4.4 should yield an efficiency estimate.
I'm guessing that the miles driven up the hill shouldn't be used in the regen efficiency calculation.
One way to think of ideal miles is as energy units. So, at the top of the hill you have 7.7 energy units in the bank (11 – 3.3) from gaining altitude. Going down the hill you use up 3.3 of them overcoming wind resistance and rolling friction, etc. At the bottom you have only three left in the bank instead of the expected 4.4 that would result from 100% regen efficiency. Dividing 3 by 4.4 should yield an efficiency estimate.
tomsax
Member
The 64% number quoted above is energy recovered from dropping elevation divided by energy spent gaining elevation.
The calculation by smorgasbord is something else entirely. It's a little hard to know how valuable it is because it will certainly change with different hill sizes, whereas the efficiency of regen ideally would be independent of the height of the hill.
Edit: I agree with Bud's post below (June 12, 3:00 am) in that his original regen efficiency calculation is more useful than what I do in the following paragraph, which is some sort of combined regen/drivetrain efficiency and less interesting.
The calculation by Bud is maybe closer to the mark. By Bud's reasoning, an extra 7.7 ideal miles were used to climb, but since smorgasbord reports gaining 3 ideal miles on the way down, instead of spending 3.3 to drive the same distance on level ground, that's 6.3 energy units recovered from the drop. 6.3/7.7 = 81%.
However, both calculation assume one ideal mile per actual mile, which depends on speed and driving conditions. On flat, level highway in moderate temperature at 57 mph, I get about 1 for 1, but at lower speeds, you'll use fewer ideal miles. The Australian record of over 313 miles on a single charge was done at about 35 mph.
If you assume 313/244 = 1.28 miles per ideal at 35 mph, then the energy loss up the hill is 11 - 3.3/1.28 = 8.43 ideal, and the gain going down the hill is 3 + 3.3/1.28 = 5.57 ideal, for a net efficiency of 5.57/8.43 = 66%. Doing the same calculation but using the later record set in California of 347.2 miles at 25 mph yields 61%. These nicely bracket the 64% estimate in Tesla's blog.
But even these numbers have significant error bars for two reasons. First, we don't really know what efficiency to expect driving the same 6.6 miles on level ground, given the variable speed and "twisty roads". Second, the 11 and 3 ideal miles could be 10.51 and 3.49 or 11.49 and 2.51. Redoing the calcs with the limits on the rounded ideal mile numbers yields 53% to 57% in the former case and 71% to 76% in the latter.
The calculation by smorgasbord is something else entirely. It's a little hard to know how valuable it is because it will certainly change with different hill sizes, whereas the efficiency of regen ideally would be independent of the height of the hill.
Edit: I agree with Bud's post below (June 12, 3:00 am) in that his original regen efficiency calculation is more useful than what I do in the following paragraph, which is some sort of combined regen/drivetrain efficiency and less interesting.
The calculation by Bud is maybe closer to the mark. By Bud's reasoning, an extra 7.7 ideal miles were used to climb, but since smorgasbord reports gaining 3 ideal miles on the way down, instead of spending 3.3 to drive the same distance on level ground, that's 6.3 energy units recovered from the drop. 6.3/7.7 = 81%.
However, both calculation assume one ideal mile per actual mile, which depends on speed and driving conditions. On flat, level highway in moderate temperature at 57 mph, I get about 1 for 1, but at lower speeds, you'll use fewer ideal miles. The Australian record of over 313 miles on a single charge was done at about 35 mph.
If you assume 313/244 = 1.28 miles per ideal at 35 mph, then the energy loss up the hill is 11 - 3.3/1.28 = 8.43 ideal, and the gain going down the hill is 3 + 3.3/1.28 = 5.57 ideal, for a net efficiency of 5.57/8.43 = 66%. Doing the same calculation but using the later record set in California of 347.2 miles at 25 mph yields 61%. These nicely bracket the 64% estimate in Tesla's blog.
But even these numbers have significant error bars for two reasons. First, we don't really know what efficiency to expect driving the same 6.6 miles on level ground, given the variable speed and "twisty roads". Second, the 11 and 3 ideal miles could be 10.51 and 3.49 or 11.49 and 2.51. Redoing the calcs with the limits on the rounded ideal mile numbers yields 53% to 57% in the former case and 71% to 76% in the latter.
Last edited:
That's 2007 information - very preliminary, I would think. Is the battery-to-motor efficiency really 80%? I thought it was higher than that.
80% is battery to wheel, so battery to inverter to motor to drive line to wheels. Seems about right, and remember efficiency changes depending on how the motor is loaded.
EVNow
Well-Known Member
We've a hill we can drive @ 40mph both ways. About 500 ft elevation. I'll drive that and get the kwh spent (using m/kwh from Leaf dash). Then compare that with a round trip on flat road @ 40 mph, same distance. I can then compare and get a good estimate for regen efficiency.
Tom can do the same on Roadster (and Rav4 EV) - though I don't know whether you can get required stats for calculation. BTW, Tom, I'm talking about 43rd st to Eastlake.
Tom can do the same on Roadster (and Rav4 EV) - though I don't know whether you can get required stats for calculation. BTW, Tom, I'm talking about 43rd st to Eastlake.
Tom Saxton is still one of my heroes, even if he did have a few moments of brain fade.
The bright guy in this one is Smorgasbord, because he figured out a clever way to approximate regen efficiency, using approximate symmetries to reduce inaccuracy. I was just trying to help with the calculations, because his result didn’t match expectations. This analysis turned out to be quite tricky.
The potential energy from altitude, the 7.7 energy units, becomes zero at the bottom of the hill. Where did it go? 3.3 units were used up by pushing the car (using gravity) downhill against wind and rolling resistance. The remaining 4.4 energy units were used up by powering the regen, which dumped only 3.0 units into the battery. If you divide those numbers, which I provided, the result is 68%.
Calculating the efficiency of converting the potential energy was not the question.
The potential energy was probably not as great as calculated, because of inefficiencies. If it were 7.6 energy units, the regen efficiency would come out as 4.3 / 3.0 or 69.8%.
I’m sure that Tesla’s dyno could provide accurate regen efficiency over a wide range of power and temperature. But being able to get an approximate data point independently is interesting and often instructive.
In the engineering trade, getting an approximate independent result is often extremely important, esp. in a world where guys accept computer spitouts without questioning them.
Physicists are often even “worse” about estimation accuracy, because sometimes getting within an order of magnitude is close enough to get the picture.
The bright guy in this one is Smorgasbord, because he figured out a clever way to approximate regen efficiency, using approximate symmetries to reduce inaccuracy. I was just trying to help with the calculations, because his result didn’t match expectations. This analysis turned out to be quite tricky.
The potential energy from altitude, the 7.7 energy units, becomes zero at the bottom of the hill. Where did it go? 3.3 units were used up by pushing the car (using gravity) downhill against wind and rolling resistance. The remaining 4.4 energy units were used up by powering the regen, which dumped only 3.0 units into the battery. If you divide those numbers, which I provided, the result is 68%.
Calculating the efficiency of converting the potential energy was not the question.
The potential energy was probably not as great as calculated, because of inefficiencies. If it were 7.6 energy units, the regen efficiency would come out as 4.3 / 3.0 or 69.8%.
I’m sure that Tesla’s dyno could provide accurate regen efficiency over a wide range of power and temperature. But being able to get an approximate data point independently is interesting and often instructive.
In the engineering trade, getting an approximate independent result is often extremely important, esp. in a world where guys accept computer spitouts without questioning them.
Physicists are often even “worse” about estimation accuracy, because sometimes getting within an order of magnitude is close enough to get the picture.
smorgasbord
Active Member
OK, so I downloaded Doug_G's nifty log parser and plotted what I believe is the time going down the hill:
The State of Charge goes from 49% to 50%. If 100% is 245 miles, then 1% could be reported as 3 miles, which is what I saw. If SOC is stored only in 1% increments, that explains why the Ideal Range "jumps" by 3 miles when I most of the way down.
If anyone is really interested, PM me and I'll email you the csv file. Actually, now that I have this tool, it'll be easier for me to note the date and time of the up/down travel starts and zero right in.
Doug, thanks for a cool tool. If I may ask for one thing, I'd like to see the ability to enter a date & time to start the graph at. Right now, it's a lot of clicking on Next Drive and fiddling with Display Span and Forward/Back to get to a particular starting point in a drive.
And Mt. Hamilton (elev. 4200') is fairly close by. Maybe one weekend I'll take a drive up/down.
The State of Charge goes from 49% to 50%. If 100% is 245 miles, then 1% could be reported as 3 miles, which is what I saw. If SOC is stored only in 1% increments, that explains why the Ideal Range "jumps" by 3 miles when I most of the way down.
If anyone is really interested, PM me and I'll email you the csv file. Actually, now that I have this tool, it'll be easier for me to note the date and time of the up/down travel starts and zero right in.
Doug, thanks for a cool tool. If I may ask for one thing, I'd like to see the ability to enter a date & time to start the graph at. Right now, it's a lot of clicking on Next Drive and fiddling with Display Span and Forward/Back to get to a particular starting point in a drive.
And Mt. Hamilton (elev. 4200') is fairly close by. Maybe one weekend I'll take a drive up/down.
Last edited:
