Archives

Hi all--

A few weeks ago I started thinking about the assist/charge bars on the HCHII instrumentation, and how it seemed like it takes a whole lot more time in the green charge bars to affect a change in the SoC than the white assist bars do.

For example, near my office there's a long moderate hill to some of the places I would grab a lunch. My HCHII likes to use 2-3 bars of assist for most of the half-mile to the top, and usually affects a 1-2 bar change in the SOC by the time I get to the top. On the return trip, if my SoC is near 4 bars I'll let the car ride on regen to the bottom of the hill instead of coasting, this time using 2-3 green charge bars most of the way, and lighting up all the green for a few seconds at the stoplight at the bottom. Problem is I only maybe get one bar of SoC back, often none at all. (The time to the top and back are similar.)

Has anyone measured or have any clue as to the efficiency of this setup? Say, if I use one bar of assist for 60 seconds, will it take one charge bar at 120 seconds under the same conditions to restore the used pack charge? Any ideas? --RN

(PS: Not an engineer, so my apologies if my question is a softball.)

This is a question that I have considered trying to answer in my own car, as the Prius exhibits the same behavior. Yesterday I tried to EV out of my work parking lot so that I could roll down-hill, thinking "No problem, it's only a quarter mile and I'll be regenerating all the way down once I get there!" Uhh...no. I made it maybe 2/3 of the way there and dropped from 6 bars of charge to 3, at which point I called it quits and engaged the ICE to protect my battery. Regenerating down that hill recovered maybe 1 bar, and it took the rest of the trip (lousy FE the whole way) to get her back up to the nominal 6 bars of charge. Needless to say I will not try that again!

Calculating battery efficiency could be done fairly easily if one could log current and voltage at the battery. You can find power in or out with P=V*I, and you can get energy by integrating it over time. The crude experiment that I envision involves accelerating under assist from a known battery state of charge, calculating the power taken out of the battery as it descends to a set "floor," and then calculate the power to get it back up to the original state of charge. It would only be a rough estimate using the discrete battery meter but it might be informative enough.

The more interesting but tougher experiment would be to estimate round-trip efficiency (wheels-battery-wheels) for regenerated power. I'm pretty sure that you could get most of the way there by tapping into current at the motor-generator rather than the battery and performing the same test as above, but the real question in my mind is how efficient the generator is at converting kinetic to electrical. Logging generated power while logging speed (and hence kinetic energy) would be a good start as long as one combines it with some kind of regen-free coast-down test to estimate so that we can tell how much energy is lost to various forms of drag and how much is lost to heat through the regen process. The results of such an experiment could be informative for Toyota and Honda folk alike.

I don't know how the HCHII or Prius regen works but on the Insight MT the regen stops at 30 mph. The regen on the Insight CVT continues all the way to the stop. If you use assist all the way up the hill but only regen part of the way down the hill, that might splane some of the difference....Louis

Louis--

I can definitely see that making a difference on the Insight, but the HCHII will regen almost to a dead stop. Essentially on this long downhill I mentioned I'm off the gas pedal entirely until the bottom, and there's 2-4 green regen bars all the way until I hit the brakes at the bottom, which lights all the bars. Only below like 3-5 mph will the regen cease, both because there's little wheel/IMA rotation at that speed, and because (I think) autostop is taking over. --RN

Well guys, I have no idea what the efficiency other hybrids have with regen, but with my FEH, I pretty much can control it. I don't have hills except for large bridges, so most of my Regen is on flat roads. With regard to bridges, I climb in EV as much as possible or at a low RPM, traffic permitting. On the way down, I try to accelerate with the ICE to 40mph rather quickly, with as low an RPM as possible, and then go EV as soon as possible. If I see I need some SoC or need to stay under 40mph for EV, I'll shift to "L" for heavy regen for ~3 seconds at a time. I'm either in neutral gaining speed or "L" grabing some regen.

The most efficient mode for the FEH is EV with a low battery. The reason is, a low battery can take a faster charge, and the size of the traction motor/generator in the FEH can provide all the battery can take in just a few seconds of regen.

On flat roads driving EV, I get a restart when the level is as low as it will go. The best speed for a quick charge with regen and back to EV is at 30-40mph. As soon as I get that low battery restart of the ICE, I shift to "L" and let off the gas ("L" fake shift) for a few seconds and drop in speed of 3-5mph. Sometimes this will provide enough charge to go EV right away, so I let it go EV, but if traffic is closing in, I'll accelerate to engage the ICE and increase my speed. I always try to get one fake shift in "L" so the ICE will not be loaded down as much with the small generator/motor charging. On the second fake shift, I go EV at what ever speed I need to under 40mph. There is always enough charge in the battery for EV during the second fake shift. If I need to exceed 40mph, I do another fake shift at 44mph to pack one more charge to lighten the load on the ICE charging with the generator. As the battery fills, the load is reduced on the ICE and MPG gets much better.

If I had your Toyota's, I would start by trying this out to determine if you have the regen ability and a battery that would support it. From what I have read, you don't. Now the Toyota (THH) and Lexus 400H might have this ability. Not sure if Honda has a full hybrid yet.

GaryG

What I can say is that on the HCH-II the number of regen bars do not represent an absolute amount of regenerative current being sent to the battery's BCM. Instead as some like to describe it, is a rather nominal (relative) indication of max regen capacity that depends on its primary source (kinetic or engine load).

What this means is that 1 bar of regen comming from a braking action does not necessarily equate to 1 bar of regen derived from the engine load. It would appear that regen derived from engine load is more effective. On the 1st and 2nd Gen IMA's the ineficiencies of kinetic regen appear to also be measurably higher, especially when compared to the 4 th gen IMA on the HCH-II. So, I suspect the phenomenon is more pronouced on those vehicles as well. I was told the 3rd Gen IMA is closer to the HCH-II in kinetic regen efficiency.

It left me bit surprised at first, but since on the HCH-II the regen afinity is software controlled (and hence dynamic in its nature) it appears to be a feature inherent to its design anyway.

I agree, my Prius seems to exhibit a similar behaviour. I have yet to look in detail at the Toyota technical literature but I suspect this may also be deliberate and easy to explain as well.

Cheers;

MSantos

MSantos--

If I understand correctly, you're saying 1 bar of regen from braking recharges the pack less efficiently than 1 bar of forced charging by the engine. Do I have that right?

So which would be better for FE (assuming you need to tap into the pack - e.g., hilly southwest Ohio), taking the hit with a forced IMA-recharge, or using regen instead of full ICE-off gliding? Am I making sense?

I guess I'm asking: is it better to watch the pack like a hawk and induce regen whenever possible to stay above 4 bars SoC, or should I just drive and take the occasional hit from IMA-recharge? --RN

I think the answer to that question is an option c, though I don't know how well it would work on a HCH-II. I would say that you want to watch the pack like a hawk and stay out of it to the best of your abilities, minimizing the chances that it would ever go down near 4 bars in the first place. That is, keep all conversions between mechanical and electrical to an absolute minimum in order to reduce those losses as a result.

That's right. It would appear engine regen is more effective than braking regen.

I too, watch my SoC very carefully and I have already gone to some extremes in making sure that I do minimize the amount of assist I use. In fact, I only use an EV glide when it is clear that I can get VERY good mileage out of it, otherwise I would rather keep the charge to feed the air conditioning in AUTO (or other similar uses).

As you may have noticed we have the explicit regen that occurs when the SoC dips to 4 bars or less. Many folks have found that we can modulate this to some extent in how we drive for the purpose of reducing its impact on FE. While this type of forced regen is a penalty, it does not have to be a terrible one at all as long as we manage to keep it at one bar (at most two) of regen.

Then, there is the more sneaky regen that occurs below that of a regen bar. This one while not as severe explains why sometime later the SoC is so high even when we did not see any green bars at all. The FE penalty when this occurs is the lowest I have seen but it is undoubtledly the one I prefer since it nibbles at a tiny amount of engine load. I just wish I could summon it at will... unfortunately it is not that easy.

What I have noticed is that the HCH-2 can actually restore the SoC amazingly quick as long as the engine is doing it. The same can't be said for rechanging strictly via braking.

Cheers;

MSantos

On the Insight IMA, the green regen bars are only a rough indicator of battery current like the civics.
The MIMA system taps the battery SOC hall effect sensor, so I can see the real current.
The Insight can draw 100a in full assist, but only regens at 50A.

The amp monitoring can be done with a simple high impedance DVM, so you may want to tap in and watch the real current.

What I have noticed is that the HCH-2 can actually restore the SoC amazingly quick as long as the engine is doing it. The same can't be said for rechanging strictly via braking.

completely opposite with my car, if I have to stop and go a lot then my SoC on average will be higher as opposed to me gliding all the time, however, much of the motion captured all came from the engine in the very end.