Best economy, cruise control or mild hypermile

[Borgholio]

Speaking from experience in my Prius, the best gas mileage comes from using a combination of techniques. First, I don't use Eco mode anymore. I used to when I first got it, since it forces you to accelerate slower and lowers the power usage of the A/C. But in truth, setting the A/C temp wisely (a few degrees below outside temp) will keep you cool and use less energy than setting it to frigid in Eco mode. Accelerating slowly is also (counter intuitively) worse for efficiency. It's actually more efficient to use normal (or even power) modes to accelerate briskly and get to your cruising speed quicker. You don't need to floor it, but don't feather it.

Using cruise control is a great idea when terrain is flat, but it's a good idea to disengage and accelerate slightly on the downhill, then re-engage cruise as you begin to decelerate on the uphill portion. That way you can make use of a gravity assist to save fuel, and the cruise control doesn't go crazy when you hit bottom because it couldn't predict the coming uphill.

So with the Model 3, I would expect the same to be true. Use moderate acceleration to get to your cruising speed and then use autopilot to stay there. If you encounter a hill, accelerate down and then engage autopilot as you start to lose momentum on the uphill. Set your A/C to a few degrees below ambient and just keep it to where you are comfortable, no lower than you need.

 

[hoang51]

hyper-miling in an EV seems completely crazy to me, a pointless exercise until you are on a journey where every mile counts.
I used to drive my leaf with Eco always enabled and power sipping low acceleration etc etc to build all the trees and compete on their website to get better W/mile.
Then I figured out how much I saved. Range was never an issue, so it could only come down to cost.
Driving completely normally, having fun away from lights etc dropped my monthly W/mile from 5.2 to 4.7 and a increase in cost of a small cup of coffee from Starbucks - BFD

My advice - until you are extremely range constrained, don't worry about it. Enjoy the performance and everything the car has to offer, the extra electricity cost is never going to break your bank.

I used to try and get the most mileage out of my ICE car. It's a 6 speed manual, I would shift at around 3,000 RPM or so, make it all the way up to 6th gear to save gas. But then I realized, doesn't net me much and it's a slow process, doesn't work well if there's many cars behind me. Now, every now and then, I hit VTEC (5,800+ RPM) for giggles. My gas mileage isn't breaking my bank either. So Twiglett has a point: drive and enjoy the car, not stressing to try and achieve maximum savings. Model 3 should already be saving a load of money coming from an ICE. Enjoy the car.

Once I get my Model 3, I'll be honest: I'm going to put the pedal to the metal because it's more cost effective than my current car. No matter how much energy spent in the Model 3, I'm pretty sure it's a definite savings over my current ICE. It'll cost so little to go from 0-60 MPH...

 

[Saghost]

Thanks folks for the inputs.
From the replies I'll assume the Tesla S cruise control does not "see ahead" based on GPS and autonomous driving data points.
I'd like to see an "eco" cruise control program written into the software where one enters a max and min highway speed desired, a watts per km to aim for (based on distance to destination charger/home) versus having to stop at a supercharger.

Well, you're right the S cruise control doesn't - at the moment.

Current generation Tesla cars do understand the terrain profile for routes in the navigation system, however (for the energy planner,) and can see traffic ~600 feet ahead with radar, and possibly further with the camera.

Ford introduced an "Eco-cruise" mode on the Energi twins (and maybe their hybrid siblings?) that attempts to drive more efficiently - I think it's mostly "driving with load" type optimization, letting the speed fall on climbs and rise on descents. Mercedes patented a system a while back for optimizing when to run a range extender in a PHEV based on terrain.

Given all of the above and Tesla's OTA updates, I think it's entirely plausible that at some point in the not too distant future Tesla will introduce an efficiency mode similar to those suggested here where the car plans further ahead and allows the speed to change in an effort to be more efficient (presumably as a configurable option, either linked to or replacing the current "range mode" rather than as the cruise default.)
Walter

 

[Tiberius]

You might get better/more thorough answers in the Model S forum.

 

[navguy12]

You might get better/more thorough answers in the Model S forum.

Agreed. Thanks.

 

[brianman]

Brianman, thanks for the link. Exactly what I'm aiming for. I note the last input on your link is dated 29 Dec 2013.
Has there been any further development of this concept? Thanks. Cheers. Mike.

Nope. I should probably formally forward the request to Tesla though.

 

[navguy12]

Nope. I should probably formally forward the request to Tesla though.

I think that would be a great idea.

 

[chipmunk]

I think it would be interesting to see a study on this. My belief is that you would find the math is not the same in an EV as in a hybrid. In your hill scenario, an EV doesn't need to shift to a less efficient gear to get up a hill. It's very seamless...the cruise control never 'goes crazy.' I agree with the previous comment that you are much more likely to improve efficiency by going slower than trying to optimize for hills. Don't forget, whenever you go faster, wind resistance increases non-linearly and efficiency goes down.

 

[Saghost]

I think it would be interesting to see a study on this. My belief is that you would find the math is not the same in an EV as in a hybrid. In your hill scenario, an EV doesn't need to shift to a less efficient gear to get up a hill. It's very seamless...the cruise control never 'goes crazy.' I agree with the previous comment that you are much more likely to improve efficiency by going slower than trying to optimize for hills. Don't forget, whenever you go faster, wind resistance increases non-linearly and efficiency goes down.

I don't know how generically applicable it is, but the data GM published in SAE papers on the Volt shows that there's still an efficiency curve involved, and a ~10% variation between the efficiency of the Volt motor at moderate RPMs and half it's maximum torque (best efficiency,) and either higher RPMs or maximum torque - and even more loss at really low power outputs, too.

This seems to suggest that an approach that limits the maximum power usage to the most efficient range and avoids regenerating unless decelerating is absolutely necessary could give meaningful gains at any given speed - though of course slowing down will always provide more gain, at least into the 40 mph range (there's a minimum speed where the fixed loads like HVAC and keeping the computers/inverters alive begin to dominate the equation; most EVs are at their best in the 20 mph range with no HVAC load, but that peak can rise to 40-45 in cases where you have a lot of HVAC draw.)

 

[chipmunk]

I don't know how generically applicable it is, but the data GM published in SAE papers on the Volt shows that there's still an efficiency curve involved, and a ~10% variation between the efficiency of the Volt motor at moderate RPMs and half it's maximum torque (best efficiency,) and either higher RPMs or maximum torque - and even more loss at really low power outputs, too.

This seems to suggest that an approach that limits the maximum power usage to the most efficient range and avoids regenerating unless decelerating is absolutely necessary could give meaningful gains at any given speed - though of course slowing down will always provide more gain, at least into the 40 mph range (there's a minimum speed where the fixed loads like HVAC and keeping the computers/inverters alive begin to dominate the equation; most EVs are at their best in the 20 mph range with no HVAC load, but that peak can rise to 40-45 in cases where you have a lot of HVAC draw.)

Interesting, thanks. I didn't mean to imply there was no efficiency penalty for higher power output. I still suspect there are significant differences between an EV and a hybrid when it comes to maximizing efficiency.

 

[Saghost]

Interesting, thanks. I didn't mean to imply there was no efficiency penalty for higher power output. I still suspect there are significant differences between an EV and a hybrid when it comes to maximizing efficiency.

Yup. But then again, there are differences in maximizing efficiency between different hybrids. Some techniques are much more applicable to some cars than others, depending on the details of both their system architecture and engine. (Most engines are most efficient at low-moderate RPMs and full throttle or just shy of full throttle if they have fuel enrichment. Some are not, like the lean-burn engine Honda had in first generation Insights, which was most efficient at about half throttle and ~4000 rpm.)

One of the interesting case studies for this is GM's factory programming for the first generation Volt, which uses three different strategies for various speed ranges - two of which involves cycling the engine on and off while driving at a constant speed (trading better engine BSFCs for more conversion losses going from mechanical to electrical to chemical (battery) and back through to mechanical.)

Although I'm certainly not an expert in the subject, for EVs with a fixed drive gear ratio most of the material I've seen suggests that a constant power "driving with load" strategy is likely to be the most efficient.

The idea of a cruise geared front motor and a power geared rear motor (or motors...) is very intriguing - from some early employee comments we thought for a while that's what Tesla had done with the D series cars, but only the PXXDs have different ratios, and those are only ~6% different (whereas a power/cruise combo would be more likely to be ~50% or so I'd guess from the few charts I've seen motor efficiencies.) At that point you'd have a lot more variables to play with in optimizing the car's efficiency - and using smaller motors would mean that by putting most of the load on one you could push cruising road load up close to the peak efficiency ranges (again, assuming most motor have curves like GM's with ~50% torque being optimal.)

 

[Twiglett]

This is an odd problem - being a Tesla for economy. If I wanted economy, my next car would not be a Tesla, they are performance vehicles that happen to be EVs because an EV makes an excellent performance vehicle.
Getting it for its economy is like buying a BMW 7 series or Merc S class and being worried about its economy and looking to enable stop/start and eco mode.
Give me POWERRRR

ECO mode belongs in crappy econobox gas powered hybrids

 

[kort677]

ECO mode belongs in crappy econobox gas powered hybrids

or severely limited range EVs

 

[LaCostaRacer]

I have never used cruise control in my cars for these reasons- my brain does a better job processing the driving environment than the cruise control.

Seems like Tesla could do a better job in this regard by using the standard speed set point along with a +/- mph tolerance to handle hills- mostly going up them because regen helps going down. If I wanted a mild cruise I would go 65 mph +/- 10mph. If I wanted a more hyper-active drive I would increase the set speed and lower the tolerance to 3 mph.

I noticed the Tesla web site has a cool range calculator that shows your range based on speed, temperature, tires, and A/C on/off.

 

[brianman]

In your hill scenario, .... It's very seamless...the cruise control never 'goes crazy.'

Drive I5 from Seatac to Seattle and you'll change your opinion. You can visualize the hills in the energy app clearly when you use cruise control for that route.

 

[navguy12]

Although I'm certainly not an expert in the subject, for EVs with a fixed drive gear ratio most of the material I've seen suggests that a constant power "driving with load" strategy is likely to be the most efficient.

I appreciate your inputs into the discussion. I'm used to the pulse and glide technique, but your suggesting that may be counter productive in the case of a Tesla. It will be something to enjoy learning once I have my Model 3

 

[navguy12]

This is an odd problem - being a Tesla for economy.

It's not a problem at all. EM wants his Tesla EVs to be successful across the full spectrum of potential customers. Out of the 325,000 advanced reservation holders for Model 3, one must assume a segment will be interested in the performance envelope as it pertains to maximum range under various profiles.
When I took a 2016 Volt for a test drive, the ev range (at the start) showed 82 km available. After a town and country drive of 20 kms, the range stood at 67 km available. Lack of data points and repeatability limits any conclusions, but I suspect when I learn how to drive Model 3, I'll be able to improve on the stated ranges.
I also suspect a (probably small) cadre of fellow Tesla 3 owners will attempt the same.

 

[omgwtfbyobbq]

EVs are a lot more tolerant of lead-footing IME. If I'm looking for some extra EV range in the wife's pip, that's mostly a function of keeping the average speed down and trying to time lights. The only reason I use N coasting is b/c it's hard to coast. In hybrid mode, it's doing all that and trying to keep the engine operating efficiently as I accelerate (pulse) and then coast (glide).

 

[David99]

Guys, hypermiling isn't only about saving a little energy. Knowing what techniques help to reduce energy usage and how much each does is helpful info. In any car you will get in a situation where you, due to unforeseen events, have to drive as efficient as possible to make it to the next charging station. We all have been there. So please don't automatically dismiss any questions about efficient driving.

 

[chipmunk]

Drive I5 from Seatac to Seattle and you'll change your opinion. You can visualize the hills in the energy app clearly when you use cruise control for that route.

We'll probably need to accept differing opinions. I'm familiar with that stretch of road. While I haven't driven it in an EV, I've driven up much steeper mountains. Yes, going up hills requires energy by definition (whether it comes from the battery, or from a loss in kinetic energy by scrubbing speed). In a Prius though, going up a hill with cruise control is much more dramatic as the car first loses speed, and then changes gearing and accelerator settings to catch up resulting in quite a bit of commotion. In the Tesla, speed stays constant as power is immediately applied and no gearing changes are even possible. That's what I meant. Yes, it uses extra energy to raise the elevation of the car.