Pulse and Glide plus Warp Stealth in the Prius II for maximum FE …

[hobbit]

You've mentioned Rick and his miniscanner experiments before.
I went looking around a little bit -- found some results at
Greenhybrid, not IC, and it helps a whole lot to spell his name
right... so Wayne, please check
http://www.greenhybrid.com/discuss/showthread.php?t=452 [also t=459]
and tell me if that's the thread you were thinking of -- and if
not, I would really appreciate if you could dig up the specific
reference. I found *nothing* at Insight Central.
.
In the meantime, I've asked Graham how he calculates %load. The
OBDII garbage I have from Alex Peper seems to be calculating
it way higher than is reasonable, and claims a 23% load during
*warp stealth* f'krissake when the engine is being DRAGGED.
So it's becoming really important to me to learn exactly where
some of those old numbers are landing.
.
I don't think you can quite apply the same "throttle setting"
methodology to an eCVT. The throttle control basically does
what it needs to, quite decoupled from what your foot is doing.
I will point out that when the Prius opens the throttle quickly
like it does, it is not to *full* -- in fact the ballpark opening
appears to be 20 to 25 percent, as claimed by OBDII and also as
measured as a fraction of the full voltage swing of the TPS pot.
So it is quite possible that the Prius is already doing your
"lower throttle opening" game, and saving the full-throated roar
for when the driver really punches it. My quest is to optimize
best use of that "gentle" region.
.
One thing that may help with understanding HSD, within certain
confines, is that pushing harder with your foot has the effect
of selecting a lower gear ratio. This is totally non-intuitive,
but given what I see with RPM in relation to wheel speeds in
a scenario where the throttle opening largely remains UNCHANGED,
that's got to be how it works. RPM goes up, power [thus hill
climbing ability] goes up, but all within that narrow range of
engine *torque*.
.
_H*

[xcel]

Hi Hobbit:

___Here you go: OBDII Scanner.

Originally Posted by Rick Reese

For my commute (No battery only operation due to too high a speed) I found the % engine load, RPM, % throttle, and amperage in/out of the main battery the most useful items to keep in the main display. Keep in mind this is my first attempt. Anyway the scanner is mounted on the steering column as Graham recommends and its a nice design (even adjusts the backlighting with a light sensor automatically). For the Prius driving for efficiency is a bit different than the Insight. Since it is an automatic with a variable gear ratio you can't shoot for a specific rpm. The % engine load however directly correlates to fuel economy. As a result I looked for a % engine load that provided a high fuel economy yet was high enough to accelerate on down hills and gradually loose speed on the up hills without the engine lowering the drive ratio. I found an engine rpm of 1300-1400 seams to be best for fuel economy. I found too good ranges. The first was a 32-36% engine load which would produce 90 mpg values on the instant bar graph at 1300-1400 rpm. In areas with hills however the engine would lower the drive ratio on hills which would drop the fuel economy as the engine rpm hit 1700-2000 depending on the hill. The better range in hilly areas was around a 45% engine load. This would give a instant value of around 75 mpg at 1300-1400 rpm. At this load however you would maintain a higher average speed and might hit 75-80 on a down slope. The kinetic energy however would allow you to crest most hills with the rpm still around 1300-1400 so you don't drop to the 25 mpg range when the engine gears down. I should mention that I took my throttle control rope out of the Insight and put it on the Prius for the trip. Interestingly enough the same % throttle did not always yield the same % engine load. (I was usually around 19-23 % throttle) but the engine load was not consistent given the same terrain.

I was thinking if we could get enough orders perhaps we could persuade Graham to build one of these scanners for the Insight. I think the amperage in/out of the battery could be very useful in determining the best fuel economy range. In other words balance the trickle charging with discharging. If I set the tension on my rope too tight the Insight will drain the battery over say 30-50 miles but if I set it just right I can cover the areas in the foothills without depleting the battery and let it charge back up on the flatter sections.

Have fun, Rick

___% Load with the SG-II is not a great way to measure ICE load either. Running a max 19 - 20 TP (10 to 89 min/max) at maybe 1,900 RPM in a nice and easy Pulse shows ~ 65% load in the Accord. In other words, pretty useless other then as an instantaneous type load gauge.

___When I speak of a throttle setting, I am not speaking of maintaining a particular throttle setting or TP set point but DWL using the iFCD as the guide. I do not really care what the ICE is doing internally if I can instead maintain focus on an iFCD to maximize FE no matter the car. As long as the iFCD is a real world number or at least closely follows the actual iFE, I should be able to maintain it at the highest level for any given speed … Unfortunately, most of my Prius highway driving experiences arose while sitting in the passenger seat watching you or Dan driving your own Prius II’s

___Good Luck

___Wayne

[hobbit]

Okay, thanks; I must have still been using the wrong name to
search at IC. The greenhybrid post is almost equivalent with fewer
specifics.
.
I'll note that 1300-1400 rpm at 22% or so TPS and upwards of
75 MPG on the iFCD [only at interstate speeds, it's always lower
if you're going slower] is at the *low* end of my "sweet spot
range". Very minimal demand. It also corresponds roughly to
your "40 mpg pulse rate"...
.
This stuff shouldn't be that different between Prius I and Prius II,
I would think... but I've never actually driven a Classic,
especially armed with a vacuum gauge...
.
_H*

[hobbit]

Is there anyone here with an SG-enabled Prius that could try
a couple of things for me? Mainly, accelerate from a stop while
keeping RPM at, say, 1600, and take note of what the SG reports
as % load during that. It will probably vary a bit but I'd like
to know the ballpark of what SG thinks your ECU thinks.
.
Ideally, I'd be able to get this data for both a Classic and a
3rd-gen and see if there's any difference.
.
_H*

[hobbit]

Okay, I updated http://techno-fandom.org/~hobbit/cars/sweet/
with a little more explanation, and less of a "do exactly this"
feel in favor of a "we might be onto something here but more
research is needed" theme. With more silly pictures. Please
give it a look and let me know what should be added. What *could*
be added is a whole book about hypermiling at the bottom, but
I'll leave that to Wayne!
.
In the meantime, I got an answer from Graham about the miniscanner.
His %load figure is simply grabbed from whatever the ECU says
about it. However, if a gen-3 Prius is at a claimed "35% load"
that's completely useless for maintaining any prayer of highway
speed, so there might be something else going on here that may
be peculiar to the Classic.
.
_H*

[JimboK]

Hello, Hobbit. I've been fascinated with your findings regarding this sweet spot ever since I stumbled upon the article on your site. I don't have a mechanical or engineering background (I consider it a major accomplishment just to have changed the oil myself!), so I've read and re-read it to try to absorb it. The take home lesson seems simple enough: At speeds above those where P&G is useful, use “warp stealth” when conditions allow, and when they don’t, in the Prius with no added instrumentation, keep the instantaneous MPG between 35 & 75. I’ve been doing both whenever possible.

I now have more instrumentation available: I just bought a used ScanGauge I. I've set the SG to display RPM so I can can watch that. But not fully understanding what many of the other SG parameters are, I'm not sure what else to use to help approach Prius high-speed FE utopia. Are you familiar with what it can display? If so, which of the other parameters would be best to monitor in order to achieve and maintain this sweet spot, and what numbers should I aim for?

Originally Posted by hobbit

Is there anyone here with an SG-enabled Prius that could try
a couple of things for me? Mainly, accelerate from a stop while
keeping RPM at, say, 1600, and take note of what the SG reports
as % load during that.
_H*

Did you get what you need in response to this? If not, I'll offer to test this and anything else you might need with the SG.

BTW, I posted a similar message on PriusOnline, so if you've responded to one, you may disregard the other unless you think it worthwhile for the masses reading the thread.

Thanks!

[hobbit]

In the meantime I've gotten hold of a Vetronix Mastertech MTS3100
to borrow, i.e. the official Toyota tool, and if *that* lies to
me then something's very wrong. Under most fairly gentle
acceleration scenarios I see about 50% calculated load.
I'd still be interested in what your SG says...
.
I rarely hang on priusonline these days; mostly on priuschat,
well, until it got knocked over AGAIN and I hopped on POL for
the updates.
.
_H*

[JimboK]

Hobbit, thanks for the reply. I remember someone else suggesting 50% as a max. I think it was in CleanMPG, but I can't remember. I've searched in so many places, I can't remember what I've read where.

Anyway, for your acceleration test, how long/fast do you need me to hold that RPM?

I knew you were pretty consistent on PriusChat, but it was still down when I posted on POL. There's now a nearly identical post on PC.

Thanks again!

[Yoshi]

Hello,
I would like to hear some comments about Prius fuel injection at downhill coasting on highway.
Discussing on PriusOnline, they say ScanGauge tells them 110 - 200 MPG instead of 9999.9 above 42 mph.
http://www.priusonline.com/viewtopic...236&highlight=

Our Japanese Prius clearly shows the fuel cut, but they say some fuel is consuming for CAT.
Maybe, it is cause by the emission control system difference on the US models.

I believe the ScanGuage is a kind of macro type of instrument and it sometimes lies.
Can anyone confirm this using oscilloscope or anything?

Thank you,
Yoshi

[brick]

I wanted to throw my 2 cents in here since I just spent >100 miles on the highway in the closest thing to a HS P&G cycle that I could manage.

First let me address Yoshi's question about the ScanGauge showing fuel injection while the Prius coasts downhill. In reality I don't think the ScanGauge knows how to deal with fuel cut while the engine spins. While I was watching its iFCD in the Accord it became apparent that the device isn't looking at actual fuel injected. Instead, I think it's basing its calculation on engine displacement, RPM, and manifold absolute pressure. I have never, ever seen 9999mpg without cutting the engine completely. 5MT Insight users will attest to the fact that it can't handle lean-burn, either, which would not be a problem if it had a way to monitor injector pulse width and had some idea of the injector capacity. (Doesn't the SuperMID look at the injectors directly?) I can't confirm that through any instrumentation but this is what the empirical evidence tells me.

Finally, what I experienced last night. It turns out that the Prius II responds a whole lot better on the highway than I would have expected. Yes, a 1.3L lean burn i-dsi engine with a tall 5-speed would blow this guy out of the water. But coming down the back side of every hill in Warp Stealth gets the job done. My concerns of draining the pack were completely unfounded, as the SoC was completely stable at the top-most blue bar even after a mile+ in the 60mph range. (The only time I have seen a lower SoC so far was after an extended EV at ~30mph around town.)

However, I do think that I pay some price for gliding down those hills with the engine being spun by MG1 (I think?). It looks like the ECU has figured out my game and likes to keep the battery topped-up in anticipation of my next demand for current. The result is that there were a few times when I would have expected >60mpg at a cruise but it was hanging in the 40s instead. After a minute or two it's back to its old self, especially if I throw in a little regen for speed control and to help it along.

Barely 300 miles on her and that 105 mile segment came out just north of 56mpg in 43F temps while travelling at an average of 60mph. The Prius II is a joy to drive in this kind of aggressive cycle since it responds so well to fine accelerator input. I'm very much looking forward to seeing what she can do with 5,000+ miles on her and some spring-like daytime temps.