Page 4 of 7

Re: Regen vs Neutral and speed for max range

Posted: Thu May 04, 2017 8:39 am
by JoeS
Aerowhatt, thank you for the excellent and very logical suggestions, especially the one driving in Eco which reduces the pedal sensitivity. I'm pretty tied up for the next couple of days but will attempt to do this every time I go to town.

Re: Regen vs Neutral and speed for max range

Posted: Fri May 05, 2017 10:49 am
by JoeS
Brief update: using SoC over this short stretch doesn't work, as both CaniOn and EVBatMon only show a 0.5% resolution, which doesn't change over the course of this downhill, using either technique.

Will keep trying to see if Wh/km is doable, as controlling the amount of energy going into the car during acceleration is difficult/unmeasurable?, as is keeping a steady foot on the accelerator.

Science Saved Me A Ticket. Good thing I was going slow doing this testing, as yesterday there was a barely-visible Sheriff standing in the shadows with a handheld laser gun point up that final downstretch...

Re: Regen vs Neutral and speed for max range

Posted: Fri May 05, 2017 1:59 pm
by Aerowhatt
JoeS wrote:Brief update: using SoC over this short stretch doesn't work, as both CaniOn and EVBatMon only show a 0.5% resolution, which doesn't change over the course of this downhill, using either technique.

Will keep trying to see if Wh/km is doable, as controlling the amount of energy going into the car during acceleration is difficult/unmeasurable?, as is keeping a steady foot on the accelerator.

Science Saved Me A Ticket. Good thing I was going slow doing this testing, as yesterday there was a barely-visible Sheriff standing in the shadows with a handheld laser gun point up that final downstretch...


I was afraid of that. It's a pretty short road at only 1.3 miles. Wh/km should be able to show up a difference even on a short hop. That's what's nice about my "mostly" descent, it's 27 to 34 miles long depending on the chosen starting point. Sections that I look hard at because of their consistency of conditions are 5 miles or more each. Problem is I can't test it at 35 mph. The speed limits for the most part are 45 to 65 mph. Drivers around here tend to add at least 5 mph to that. Just not a safe option to go slower!

I did a little refresher on aerodynamic drag while waiting in stopped traffic today. Comparing 7,500 ft elevation to sea level. Air density
is a nice round 80% of sea level at 7,500ft (the averaged altitude of my journey). Since air density is a single multiplier in the drag equation it gets simple. Basically, given the same velocity one would be fighting 80% of the resistance that they would be at sea level.

Aerowhatt

Re: Regen vs Neutral and speed for max range

Posted: Thu May 18, 2017 9:25 pm
by JoeS
Aerowhatt, thanks for doing the drag comparison. Hey, the only reason for this whole discussion is the aerodynamic drag effect!

Well, I've been totally unsuccessful in trying to get any meaningful data by trying to keep the accelerator in a constant position and a constant speed. The combination of curvy and undulating road, short distance, and slow speed makes it not doable, IMO. On this stretch, it is soooo much easier to just kick the car into Neutral and not worry about it. For maximizing range, this low speed test doesn't address the comparison we're trying for, anyway. Neutral gives me a steady-state current draw of 1.3A (correction, I had written 1.8A) at around 50%SoC.

During my recent Tesla trip I played with Neutral as well as holding the speed down on steep slopes and regenning like crazy. The problem with Neutral and the Tesla is that its Cd is so low that the thing just keeps going faster and faster...

Without a controlled test on the same stretch of road on which one has the ability to go both at 70 mph and 35 mph to do comparative measurements, we're still at an impasse. An early Sunday morning on the Interstate near me might work at 35mph for a few minutes...

Re: Regen vs Neutral and speed for max range

Posted: Wed Aug 15, 2018 11:50 am
by JoeS
Forum readers will recall that Aerowhatt and I have been embroiled in a discussion as to whether regeneration and slower downhill descents will increase an i-MiEV's range as compared to putting the i-MiEV into Neutral and simply coasting down the hill. Aerowhatt is a proponent of coasting and, from a practical standpoint, I also do that all the time; however, the theoretical challenge on the table is still there.

The discussion boils down to whether there is a higher or lower amount of stored energy which results from a lowered aerodynamic drag (from going slower) and regen power plus the power consumed by the engaged motor/inverter combination when compared to simply coasting whereby the car is still consuming a steady (1.3A?) irrespective of speed. [That's an awkward sentence that needs to be rewritten]

The metric to prove/disprove this was the change in State of Charge between the top of the hill and the bottom of the hill. Unfortunately, the resolution of this reading (via CaniOn) is only 0.5%, which is inadequate for short-stretch controlled testing. In my case, I don't have a long-enough downhill where the car will coast at a fixed speed in Neutral.

Never one to give up I've gone back to looking at a short stretch (0.4 mile?) of freeway with a nice steady downhill where the i-MiEV holds a constant 65mph in Neutral. I'm willing to go back on a Sunday morning with negligible traffic and kick the i-MiEV into regen and thus go down this hill slowly in an attempt to prove (or disprove) the hypothesis.

Let's ignore SoC, and try CaniOn's energy consumption over distance readout.

I haven't tried it, but if we reset CaniOn at the top of the hill and then take the Wh/km (or non-metric equivalent) reading at the bottom of the hill, using both driving techniques, will that prove anything? Let's discuss it before I try it.

Aerowhatt, you had published some compelling numbers in another thread which showed that engaging the motor/inverter was terrible compared to coasting in Neutral, but I couldn't readily find them. Perhaps summarize your findings on this thread?

Incidentally, the location of my "test" hill is (using Google Maps) 37.3860966,-122.2056714, but I have to re-learn to use GPS Visualizer and come up with an exact Start and Stop point on this road for the test.

Re: Regen vs Neutral and speed for max range

Posted: Thu Aug 16, 2018 9:04 am
by Don
All in all, I think neutral is hard to beat if you have a straight downhill stretch where the car will coast at about the speed you want without the need for either slowing or accelerating on portions of it - Anytime you introduce electronics into the equation, there are losses, caused by getting DC from the power source, converting it to AC to power the motor, or regenning AC and converting it back to DC to recharge the batteries. If you were on an Eco Challenge mileage run, you would be using the coast mode as much as possible and you would be going further on less juice, no question

It's a shame Mitsu didn't give us an easy way to make use of the coast mode - A paddle on the steering wheel where you don't need to fiddle with the darned shifter every few seconds would have been really nice. But as we all know, it was a new thing and their concentration was to make it drive as close to a conventional ICE as they could. Thinking like that left lots of good, low cost ideas left on the curb - A MAX regen paddle on the left side of the wheel and a coast paddle on the right side could have probably been done for less than $100 per car. But then . . . . some elderly driver would get confused which was which and drive through a store front and everyone would be saying "Them darned EV's are just UNSAFE!"

For everyday driving in traffic where other drivers affect how and when you can coast or slow, shifting the shifter all the time is just way too much trouble, especially when all you're really doing is making and braking switch contacts - What a ridiculous way to accomplish that . . . . and then we read about people who have shifter problems after a time, where it needs to be readjusted to get it back in sync

I've never had any range problems and the wife even less so - We got used early on to just pulling the shifter all the way back into 'B' and just leaving it there. One pedal driving. You get used to where to ease up to make the curves and corners without the need to touch the brakes. I like it even better in the Volt, as I can ease out of the throttle later and then when I need even more regen to come to a complete stop, pull the paddle on the steering wheel. No question we could be adding a few percentage points to the energy savings by constantly fiddling with things and if there was a need, maybe we could be bothered - But after 6+ years driving in B mode, it's just so simple and you don't even have to think about it . . . . it's just the simple, normal way to operate, at least for us

Don

Re: Regen vs Neutral and speed for max range

Posted: Fri Aug 17, 2018 1:11 pm
by Gorfllub
Don,

I just made a "Coast/Neutral" button! Works great!

Going to start a new thread about it in modifications section.

Re: Regen vs Neutral and speed for max range

Posted: Mon Oct 08, 2018 1:05 pm
by JoeS
Found this very good article discussing regeneration and its limits, further strengthening Aerowhatt's position in our regen vs. neutral debate.

https://chargedevs.com/features/regenerative-braking-a-closer-look-at-the-methods-and-limits-of-regen/

Dang, gotta do some controlled tests to see if I can bolster my argument. For those of you wondering what this is all about, see this post which defines the hypothesis -

http://myimiev.com/forum/viewtopic.php?f=28&t=1685&start=20#p32826

Re: Regen vs Neutral and speed for max range

Posted: Mon Oct 29, 2018 9:59 pm
by JoeS
In a previous posting on this thread, I defined the hypothesis that Aerowhatt and I are debating. Summarizing:

On any given downhill, placing the i-MiEV into Neutral and coasting is more efficient than leaving the car in "gear" and regenerating, defining efficiency as based on the difference in State of Charge (SoC) at the bottom of the hill vs. SoC at the top of the hill.

Unfortunately, I then bounded it by limiting the upper speed limit to 70mph which means that for steep downgrades one must kick the car out of Neutral and do some regenning as opposed to simply coasting in Neutral and let the car find its own maximum speed for any particular grade. I also limited the lower speed limit to 35mph. Sadly, these two limits complicate matters to such an extent that, to date, I have found it impossible to perform a controlled test.

Nevertheless, not being one to give up easily, last week, after climbing up to Lake Tahoe and then from Truckee up to Donner Summit (7,000ft elevation) on Interstate 80 in my i-MiEV, I was really looking forward to the >50-mile mostly downhill run into Sacramento with grades of 4%-6% in order to take some data and see if I could make any sense out of it to either bolster or refute some of the arguments Aerowhatt and I were having.

My best-laid plans blew up! Here's why -

I had anticipated having the ability to drive at a leisurely pace and play with both coasting and regen and was even considering going back up a particularly suitable grade to repeat this experiment (if I found one) and doing it a few times ... BUT ... what I had not counted on was the truck traffic. Interstate 80 is a major East-West transcontinental thoroughfare and what I experienced was a never-ending closely-spaced convoy of trucks significantly exceeding the state's 55mph truck speed limit on the downhills. No way was I going to drive even at 55mph hypermiling along using regen - I'd have been eaten alive!

So what I ended up doing was roaring along with the other car traffic, passing the trucks (it's only two lanes in one direction) which caught me completely unprepared as I consciously blew past a key charging stop only to find that I didn't have enough charge for a major planned side trip. No problem, as I ended up continuing on down the hill to the next CHAdeMO and having to take a different route than originally planned - I was so rattled that I forgot to take data or even CaniOn screenshots.

But, with all those downhills on the Sierra side roads, I had time to think (I know, that's dangerous...) as I watched the SoC creep up on CaniOn and the fuel gauge gain bars -

So, let's revisit this without limiting constraints -

For any serious downhill, we have two competing conditions, and are asking which is the more efficient:

Coast in Neutral vs. limit car speed using Regen

1. One argument is that the car's electronics draws significant power when the motor is engaged, negating the power gained using regen. A quick check of quiescent power draw: when the car is stopped, foot on the brake, shifting from Neutral into Drive results in no change in the amount of current drawn by the entire i-MiEV system (about 1.2A as seen by CaniOn), with the car not moving.

2. If the car is moving and regenerating, it is indeed stuffing current into the battery and increasing the State of Charge of the battery, despite any losses it may have with the drivetrain engaged. OTOH, in Neutral, the car continues drawing its 1.2A.

3. At the bottom of a steep 20-mile downhill, if the car had been in Neutral the whole time (speed unchecked), the SoC would be unchanged (or maybe even lose a bit) compared to what it was at the top of the hill.

4. OTOH, going down that steep 20-mile grade and slowing down the car using Regen will result in a significant increase in SoC by the time the car gets to the bottom of the hill. Like I tell people, it's fun watching the fuel gauge go UP when going downhill!

Thus, I still contend, on downhills it is more efficient to regen than to coast.

Flame suit on...

(Or am I so confused by now that I should throw in the towel and lock the thread on this discussion and fugetaboutit? - I'm presently on the road in the Tesla killing time in Oregon)

Re: Regen vs Neutral and speed for max range

Posted: Tue Oct 30, 2018 1:52 am
by Don
How fast does the car end up going down a long 5% grade? You are limited by aero drag and the mechanicals in the car, since there is in actuality no real 'neutral' which allows you to disconnect the drivetrain and just coast . . . . and the car's Cd is just terrible at the speeds involved in such a long run down such a steep grade. No question you would get much different results in the Tesla . . . . you might be going 140 mph at the bottom of the hill!

It's easy to see where using (our) neutral down a long grade limited by aero and friction to whatever speed that gets you is less efficient than regenning and going down the long grade even just 3 or 4 mph slower will have you with a fuller battery at the bottom, no question . . . . but, what have you proved??

That an iMiEV has a terrible Cd - We knew that already!

If this same test is done on lesser hills at 35 mph, you won't have those huge aero losses and that limiting of speed done while regenning might mean you have scrubbed off momentum recharging that you're going to have to pay (in electricity) to get back to either maintain whatever speed you want to maintain after the coast, or you pay more to make it up the next hill - I suspect the latter is true for sure. Then, you have the energy conversion losses to add into the equation too which are probably more than the cars quiescent 1.2 amps that you were losing while coasting. So, you can't make a blanket statement about coasting down hills without factoring in the aero losses based on the speeds involved - It might be more efficient to coast at 35 mph than it is to coast at 70 mph or more and regenning just enough to reduce the aero loss just enough is more efficient when aero is playing a big factor like it does at higher speeds, yet the calculations for half that speed would give you a completely different answer . . . . right? I'm pretty sure they would

I think if you're going to design an EV to win the Eco Challenge, you would want a real neutral, plus the ability to shut down all current draw in the car while coasting . . . . which just isn't safe to do in a real car driving on real roads among real traffic

Truckers love to tack on an extra 10 or 20 mph at the bottom of a hill to 'use' to make it up the next hill and they know that's more efficient than riding the brakes down the hill (where they get no energy return) and then burning extra fuel to make it up the next hill. Now you're trying to prove whether or not an EV which can return 80% or so of one of the the speed limiting forces is better than just coasting while using the cars constant quiescent current . . . . or not . . . . and that's probably very speed dependent

IMCUO (In My Completely Uneducated Opinion) I think you guys set out to try to prove something one way or the other by carefully examining the 6 or 7 factors you thought governed the situation, only to discover in the end that there are actually 9 or 10 factors to consider *and* those unexamined forces threw all your calculations out the window :lol:

But, I might be completely wrong!

Don