Cruise Control and Hypermiling

[GdB]

1. CC is safer! Everyone driving constant speed is safer, up hills and all.

2. CC always saves fuel for EV's, even up hills. Saving fuel by slowing down uphill is comparing apples and oranges. All economy comparisons must be done with equal average speed. The record i-MieV range was achieved at 30 MPH constant speed, but that is not practical. This is simple calculus math. If you integrate (calculus math) a constant speed trip energy usage versus a varying speed trip energy usage, for equal equal average speed the constant speed trip will always use less energy for any EV (motor efficiency always very high unlike ICE). I can post the math to prove it, but I have already done it my head a long time ago while driving on a long trip.

Without consideration of time, physics, and math, this discussion is meaningless.

I can closely follow trucks and probably go 200 miles, but is that safe?

Accidents are on the rise thanks to smart phones. Efforts to save fuel are often misguided and unsafe. Choosing a more efficient car is far more effective. I drive my Electric motorcycle at maximum speed on the highway, 85-95MPH, and it still gets over 120MPGe.

Unless you are retired, consider how much it costs per hour to drive slower. You are making less per hour than people in the poorest countries.

FWIW, maintaining a constant speed is not necessarily the most efficient way of driving the car when shooting for maximum range - separate discussion.

Yes, I was about to add that CC can actually decrease your mileage as it keeps your speed up to the set limit even when going uphill. It is more efficient to let your speed slow while driving into a hill and then when you crest to get back up to desired speed, and maybe even go over a bit on the downside.

Point me to some studies that show CC does NOT save on milage on EV's in most cases.

This is from an article on 16 ways to save energy in EV and Hybrid vehicles
" 16. Do not use the cruise control!
It will try to maintain 70 mph, even uphill. On overpasses, I let my speed bleed off slightly on the way up, and regain it on the way down, while keeping the instant mpg needle in the sweet spot. Savings potential: 5% - 10%"

For us, whatever we can do to keep the needle as far down in the green as possible is the way to get the greatest range - Cruise control doesn't take this basic concept into account at all. I'm not surprised to learn that this saves a minimum of 5% and perhaps as much as ten

Don

This behavior is also potentially reckless if your in traffic. http://www.smh.com.au/news/national/motorists-risking-their-lives/2008/08/23/1219262622708.html Just the URL I got form wiki about hypermiling safety.

 

[Don]

2. CC always saves fuel for EV's, even up hills. Saving fuel by slowing down uphill is comparing apples and oranges. All economy comparisons must be done with equal average speed.

I don't agree. If I can bleed off a little speed going uphill so as to not need to get deeper into the battery and then make up the lost speed on the downhill side when I can do so again without needing to get deeper into the battery and thereby have a net energy savings, I am still going "an equal average speed" as you say I must do . . . . but I've saved energy over using the cruise control. If the cruise is set for 35 mph and I go 30 up the hill and 40 down the hill, I'm still averaging 35 aren't I?. The regenerative braking the cruise would do on the downhill side isn't 100% effective . . . . I've lost some of that 'extra' I 'spent' getting up the hill at the CC set speed and I didn't get it back regenerating at the set speed while going downhill

Using the CC in ICE vehicles in hilly country is even more wasteful, as you don't get anything back on the downhill side

I don't think anyone who is really good (efficient) at hypermiling is doing it using cruise control and for good reason

Don

 

[Aerowhatt]

2. CC always saves fuel for EV's, even up hills. Saving fuel by slowing down uphill is comparing apples and oranges. All economy comparisons must be done with equal average speed.

Why? If I can bleed off a little speed going uphill so as to not need to get deeper into the battery and then make up the speed on the downhill side when I can do so again without needing to get deeper into the battery and thereby have a net energy savings, I am still going "an equal average speed" as you say I must do . . . . but I've saved energy over using the cruise control. The regenerative braking the cruise would do on the downhill side isn't 100% effective . . . . I've lost some of that 'extra' I 'spent' getting up the hill at the CC set speed and I didn't get it back regenerating while going downhill

Using the CC in ICE vehicles in hilly country is even more wasteful, as you don't get anything back on the downhill side

Don

Exactamundo . . . less power used by going constant power uphill and loosing some speed. Then coasting, D with light throttle to keep the needle at zero (for sticklers of not driving in N) making up average speed lost uphill and use regen to slow if needed on the downhill side (Tickets are far more expensive than electricity). Another thing not mentioned yet is that pulling higher amps to maintain speed uphill reduces discharge efficiency too in comparison to the lower amp rate discharge.

Aerowhatt

 

[Don]

Another thing not mentioned yet is that pulling higher amps to maintain speed uphill reduces discharge efficiency too in comparison to the lower amp rate discharge.

Very true, though it's not as large a hit with a Lithium pack as it would be with lead acid technology, but it's still another loss to contend with - Peukert's Law

Don

 

[JoeS]

OK, started a new thread by moving some posts in here out of the Cruise Control hardware topic <http://myimiev.com/forum/viewtopic.php?f=14&t=384>. That thread is dedicated to implementing Cruise Control into the i-MiEV.

GdB, you brought up two related subjects and I'll quickly dispense with the first because it's going to be fun addressing the second. First, the first:

1. CC is safer! Everyone driving constant speed is safer, up hills and all.

I agree with that statement. Having very recently put on over 12,000 miles driving around the US and up and down the West Coast in my Tesla, almost all of it under Cruise Control (without AutoPilot), I can also attest to the fact that not only trucks, but most cars, actually slow down going up hills. Just about everyone not using Cruise Control seems to slow down on upgrades - I don't know whether that's because many upgrades also have curves to them and most people slow down on curves. In any case, going up hills I usually had to adjust my CC as I caught up to traffic. In the old days it would simply have been because cars were too wimpy to maintain 70mph on 6% upgrades (unlike our i-MiEV, which zooms up hills with the best of them ). In any case, I agree that minimizing speed differential is, in general, safer for all. 'Nuff, let's focus on hypermiling...

(To be continued, as I'm presently out of time. Let me simply say that over the years I have altered my perspective and have come to the realization that there IS a difference in hypermiling techniques between BEV and ICE, but still consider CC both inferior in some cases but superior in other situations, for reasons I'll discuss. Rolling up my sleeves, but I think we will all end up agreeing on some fundamental principles - especially, GdB, if we throw in math and physics into the arguments...)

 

[PV1]

Traffic allowing, I tend to do the following:

Shave a few MPH off going uphill then regain it.
Coast when possible.
Follow other vehicles with 2-3 car-lengths in between on the highway. Pickup trucks or SUVs are easier to draft as they don't create air ripples. That is, IF I can find someone else on the road doing the same speed as I am.

Unlike cruise control, which tends to overshoot the set speed when cresting a hill, I can pre-emptively back off the power as I crest and not overshoot my preferred speed. On the one route I take to work, I've gotten so good with throttle control that through hills and turns, I can keep the speedometer right on 48 MPH, outperforming the cruise control in a BMW i3 (only EV I've driven the same route with that had CC).

Using the CC in ICE vehicles in hilly country is even more wasteful, as you don't get anything back on the downhill side

The worst I've experienced is a Ford Escape on a hilly main road. Don't even attempt it. However, my work's 2011 Ford Focus does pretty good. With the PZEV, going downhill with the cruise on will shut the fuel off and use vehicle momentum to keep the engine spinning (the resulting holdback shows just how much energy it takes for an ICE to simply spin). I drove it for a week, all highway at 65 MPH, 80 miles a day. When I started, the dash said 28 MPG (80,000 miles on the odometer but I don't know when the MPG was last reset), and when I parked it, it was at 35.5 MPG and still climbing.

 

[JoeS]

All right, my sleeves are rolled up! First, my hypermiling credentials -

Internal Combustion Engine - Gen1 Honda Insight: 77.0mpg at LIFETIME 94,520 miles for a vehicle rated at 53mpg using the present-day standard.

i-MiEV at a time when I was interested in hypermiling it (nowadays, I don't bother except for range-pushing trips)

Before continuing, there is one very specific difference between Internal Combustion Engine (ICE) vehicles and Battery Electric Vehicles (BEVs) that I'd like to first put on the table: an ICE vehicle is at its maximum efficiency when coasting with the engine turned of - after all, the car is moving but not consuming any fuel. This is not true for a BEV with Regeneration, as regen, in the right circumstances, can actually add 'fuel' to the vehicle while it is moving.

The second concept I'd like to put on the table is that any time a BEV applies regen, it is recovering less energy than it expended for the same speed increments. What I mean by this is that, for example and ignoring both rolling resistance and aerodynamic drag, if a vehicle accelerates from 20 mph to 30mph it expends a certain amount of energy. If that vehicle now decelerates at the same rate from 30mph to 20mph it recovers a percentage of that energy and not the full amount previously expended. Losses are attributable to inverter/motor/converter efficiencies, with the motor acting both as a motor as well as a generator. Perhaps someone would like to put some numbers on the losses associated with deceleration compared with acceleration?

Before continuing, I'd like to ensure we have agreement on these two concepts.

(Back to work...)

 

[Phximiev]

Ok, I'll buy into those two notions.

Back to Olympics.

 

[Aerowhatt]

Ok, I'll buy into those two notions.

Me too, except for the caveat that we should stick to reasonable "safe" driving modes. Which of course excludes coasting an ICE in neutral with the engine turned off :roll: (I consider coasting in neutral with the ICE engine at Idle both safe and acceptable however, even though most states traffic laws forbid it)

Aerowhatt

 

[Don]

Me three! - On with the dissertation Joe

Don

 

[pmaupin]

1. CC is safer! Everyone driving constant speed is safer, up hills and all.

I agree with that statement. Having very recently put on over 12,000 miles driving around the US and up and down the West Coast in my Tesla, almost all of it under Cruise Control (without AutoPilot), I can also attest to the fact that not only trucks, but most cars, actually slow down going up hills.

Most people in front of you alter speed depending on road conditions (including incline), but you maintaining a constant speed is safer. :roll:

I'll just leave this here: http://www.ncbi.nlm.nih.gov/pubmed/21376911

 

[JoeS]

Me too, except for the caveat that we should stick to reasonable "safe" driving modes. Which of course excludes coasting an ICE in neutral with the engine turned off ...

I would agree with you except for one car: the first generation Honda Insight (I know nothing about Gen2). It's normal AutoStop feature (whereby the engine turns off whenever the car stops at a traffic light) could be augmented by a mod to turn off the injectors with a momentary switch, stopping the engine but yet leaving all the car's normal system functions operational s.t., for example, if brake vacuum pressure lowered the engine would automatically restart. Recall, it has electric power steering. Used in conjunction with a clutch switch, this technique in the Insight community is called FAS (Forced AutoStop) and when hypermiling I employed it on every downgrade, often safely coasting for miles down a mountain in neutral with the engine off.

Back on topic, let's define our terms (I purposely capitalize Cruise Control and Throttle) -

Cruise Control (CC) - a car gadget which, when engaged, maintains a constant vehicle speed by continually adjusting the power applied to or subtracted from the motor

Throttle - a car gadget which, when engaged, maintains a constant accelerator pedal position

Where I'm coming from -

My experience with a Throttle comes from ten years of driving my Gen1 Honda Insight with a big klutzy thing similar to one of these:
http://www.awdirect.com/exterior-Throttle-cable-15-foot-control-cable-vcgtx15/Throttle-cables/
This Throttle control attached to the bottom of the dashboard and went to the control arm directly above the accelerator pedal. It had a nice vernier control knob to enable me to fine tune the car's speed on level ground and the big red center button was an instant disconnect. I subsequently bought a smaller more-elegant aircraft Throttle but never got around to installing it in my second Insight, nor did I ever feel the need for it on the i-MiEV. This Throttle control did not have a 'safety' disconnect (e.g., when stepping on the brake) and I used this Throttle only on long wide-open highway stretches and not in heavy traffic.

My Tesla is the first car I've ever owned that has Cruise Control. Once engaged it keeps the car's speed locked in to within one mph, and can be adjusted up or down in increments of one mph (light tap on stalk) or in five mph chunks (heavier tap on stalk), the latter being very handy when speed limits change; e.g., two heavier taps result in a ten mph speed shift. I use it all the time while driving the Tesla (even in town, primarily for self-defense so as not to get a speeding ticket). What bothers me is that these manual CC speed changes are accompanied by a significant *** or release of the accelerator pedal, IMO applying way too much power immediately or unnecessarily kicking in regen, respectively - a decrease in amplitude and thus a slightly longer delay in response would be acceptable, IMO.

From a hypermiling standpoint, how do I compare the two? Short answer - Throttle wins! Long answer - it's nuanced.

From a constant-speed standpoint the Cruise Control is superior to the Throttle, but on level ground the difference is minimal.

DISCUSSION

I don't know if you've noticed, but highways undulate: every time there's an overpass, the road goes up a bit and then goes down a bit. With the Insight under Throttle control, the car would slightly slow down going up the overpass but would then slightly speed up going down the other side. In practice and if I was really hypermiling and the overpass was significant, at the top I would disconnect the Throttle and engage FAS - gives one something to do on those long boring Interstate drives. It was quite a game to keep one interested during any trip, as the Insight's instrumentation provided a fantastic feedback platform showing instantaneous mpg via a nice big ribbon display, in addition to trip mpg and lifetime mpg. Now, was this technique superior to simply having a Cruise Control that would squirt a little more gas into the engine going up the overpass and a little less gas going down the other side? After all is said and done, considering total energy to maintain a particular average speed, for a gasoline car it's probably a wash on level ground and going up a hill, but going down a hill FAS beats everything and dramatically improves a car's trip mpg.

Now, let me explain the Cruise Control as implemented on the Tesla: the darn thing is a very tight control loop, maintaining that preset speed with incredible accuracy and instantaneous response to any terrain or surrounding environment changes. The example of the overpass is that it applies what I would term way too much pedal pressure going up the overpass (as seen by the instantaneous power gauge and graph), whereas going back down it actually engages regen, all to keep that speed spot-on! Remember, every time regen kicks in to slow the car down unnecessarily, it's robbing energy that needs to be put back in. At the accelerating end, why the need for a massive jump in current just to go up an overpass? Losses do add up in both of those situations. It is these repetitive spurts of power and regen that, in my opinion, make Tesla's Cruise Control less efficient when compared to either a human driver or Throttle control.

Now, let's address the situation where Cruise Control really works wonders as a range extender: limiting speed on downhills via regen. Remember, in a gasoline car with engine turned off and coasting down a hill, we don't much care what aerodynamic drag does other than we don't want to go much over the speed limit. With an electric car, it is far more efficient to let regen limit the car's speed and stuff energy back into the battery rather than simply putting the car into neutral and coasting. I would rather go at a constant speed and take that energy otherwise being absorbed by aerodynamic drag and stuff it into the battery. I first addressed this concept in this post:
http://myimiev.com/forum/viewtopic.php?p=20148#p20148
Edit: whoops, wrong post, it is this one:
http://myimiev.com/forum/viewtopic.php?p=19947#p19947

As an aside, I think it would be fun to play around with both a variable time-constant on Tesla's CC feedback loop as well as a decrease in response amplitude. For example, on open-highway driving give Cruise Control a few seconds to react to something like an overpass and then respond gently, yet stop the correction as soon as the situation reverses - I'd be willing to bet that there would be a measurable energy gain over the course of a few-hour trip. On the other hand, this may result in an uncontrolled oscillation and, besides, the present short time-constant would still be needed in traffic or hilly situations... (sigh), need to do the math first.

'Nuff already - flame suit on.

 

[PV1]

My parents' C-Max Hybrid has an option called ECO Cruise. What this does is sacrifice 3-5 MPH in order to stay in EV mode. It's great at staying in EV mode when hitting the bump at the end of a bridge, but for actually saving fuel on a trip, normal cruise seems to work better. The flaw with ECO Cruise is that once the engine starts, the car applies heavy acceleration to get back up to speed, and more often than not, this happens at the crest of the hill. This means that the car starts going downhill while in over-throttle, then shuts down and brings on the regen.

If it would hold the lower speed until power drops back down to allow EV mode then slowly re-gain speed (or go to zero power and coast back up to speed then regen to hold when going downhill), it would be much better at increasing MPG. Then again, for the mostly mountain driving that it is used for, a larger battery would do wonders.

However, normal cruise control does work pretty well and doesn't really over-accelerate. Over 20,000 miles, with most of those miles being normal driving on the highway, it is averaging 43.2 MPG, which I think is pretty good for a vehicle its size (larger profile but similar shape to the i-MiEV).

One long trip I've made a few times is from Pittsburgh to State College, PA. It is all highway at 55-65 MPH with a general climb over a few mountains. Heading to State College results in a fuel economy of 46-47 MPG, coming back around 49 MPG. There are certain parts of the highway where I'll shut the cruise off because pre-empting the road terrain allows for much more use of EV mode vs. having the cruise on (and some places where I need to run the battery down before a long descent, like eastbound into Altoona or westbound into Blairsville).

Another trip I did once was driving Route 31 from Somerset to Bedford. Over a distance of 41 miles, I managed 60.4 MPG at an average speed of 41 MPH with the A/C on. My technique here was no cruise control, power up the hills, let off just before the top and coast over, then regen down to hold back. I tried to spend most of the time in EV mode, trying to only let the engine run when going uphill, but once again, a larger battery would've made that much easier.

Back to hypermiling techniques, are there any mechanical/technical reasons to not put an ICE automatic transmission in neutral while moving? Every reason I've heard was because of either legality or the few second delay before one has motive power again. I've only done it while sitting in traffic on a downhill slope and creeping at 1 MPH but never at speed.