Coasting in neutral

Yes, I should have been more careful in my wording and description. Should I have said that the armature windings (the stator in this particular configuration) are disconnected, no current is allowed to flow in them?

I don't believe this "D" throttle position for gliding is a safety equivalent to Neutral, because I am not understanding it as turning off the motor's armature windings. The car is still controlling the MG1-MG2 power flows and RPMs, not setting them to mechanically freewheel.

You are just finding to point in "D" where it is neither motoring or generating, power flow is zero, but that doesn't mean the car isn't controlling the MGs. This is useful for hypermiling or gliding, but it isn't useful as an emergency safety measure or recovery path from a hypothetical engine runaway, such as the alleged Toyota Sudden Unintended Accelerations back about 2010.

 

If it's similar to how a dynamo works on a bike, unless you put a load on it (ie, a light), there is no "rolling resistance", beside the bearing/bushing and the rotor inertia (which only affects when changing speed).

 

Zac᠎kly.

 

I.e. the ECUs ensure that the MGs are electrically disconnected from -- no current is allowed to flow -- to or from any (generator) load or (motor) power source.

 

Indeed - "N" has lots of real practical and safety uses. The forced (with exceptions) disconnect is the reason for that, and is exactly why it isn't (normally) able to start the engine, charge the battery, etc. Lots of crippled functionality in "N", but it's for the safety etc reason.

I was simply agreeing with part of Paul Gregory's point that there's no reason "N" should be more efficient for hypermiling than "D"-with-neutral-throttle-position, as they should be both as "neutral" as each other for power/force/torque/whatever at that instant.

And your point about D controlling and N not controlling is backed up by silvaing's screenshot - apparently the car doesn't even know the motor rpms in neutral. (Which suggests the MG1-protecting engine start must be activated by monitoring wheel speed alone).

 

I would be hugely surprised if that's the case (though I can't, at the moment, say what is going on with sylvaing's screenshot).

The information on the rotation of each MG comes from a sine-cosine resolver mounted right on each MG's shaft. It doesn't just give each rotor's RPM but even its position within about 0.09°—that information is needed to do the math for what exact phase currents to use when the car does want to assume control again.

The resolvers are harmless low-voltage devices and I can't see a reason the car would stop watching those inputs just because it's in neutral and not controlling the MGs. It has legitimate need to know their RPMs, and it needs to know their positions so it can apply proper currents when it wants to control them again.

Relying on the wheel speed sensors would be way impractical for that. Never mind that they can't resolve the position finely enough, they're not even connected to the right ECU. The HV control ECU would be trying to do real-time motor control using numbers it would have to poll the skid ECU for over the CAN bus.

Now I'll have to go out and see what MG RPMs are shown in my gen 3 when in neutral.

One thing that might have changed over the generations: in gen 1, the HV control ECU inside the cabin did all the work: the MG resolvers were wired back to it, it computed all the needed pulse timings and sent the drive pulses to the transistors inside the inverter (which was not much more than a big IPM). By gen 3, there's definitely a separate MG ECU inside the inverter case; the resolvers are connected to it, it does all the math and pulse generation in-house, and the HV control ECU in the cabin just gives orders to it. The MG ECU doesn't show up directly as anything you can talk to with Techstream (though you see its firmware version shown along with the HV control ECU's).

I wonder if, in neutral, the MG ECU still knows perfectly well what the MG rotors are doing, but the HV control ECU just isn't bothering to ask.

 

Agree that it's somewhat surprising, but I don't think it's as implausible as you do.

Thanks, I wasn't quite sure what the mechanism was.

Yes, not sure of the reason to stop watching either, but the need only arises when it wants to control them again. It can start watching at any moment, so I don't see how stopping watching is a problem.

Not the individual wheel speed sensors, where ever the main speedometer readout is coming from. The coarse old-school pre-VSC main transmission indication, rather than the fine MG2 position.

And why do you think they need to resolve position finely? We just need "we're over X mph with the engine stopped, so MG1 must be reaching Y rpm, so it's time to reactivate the MG controls so we can start the engine".

I wasn't suggesting we'd start the engine by spinning MG1 without activating the full MG control system. I just meant that implied that the speed trigger for the activation had to come from outside it.

Or maybe the entire MG ECU is cut-off and shut down. Still not seeing any reason it couldn't be, and restarted when you go back to D, or when you hit the speed limit. Killing the ECU is a very firm way of preventing a software glitch from accidentally energising something.

Or cutting off a glitch that's already happened, for a safety measure. Could be a very conscious part of the design - reset a chunk of the ECUs when you go into N.

 

In gen 3 anyway (I can't speak directly to a Prime), the main speedometer readout is polled by the combination meter from the skid ECU, based on the four wheel speed sensors.

(The skid ECU also generates a pulse-frequency "vehicle speed signal" of the kind various accessories have been built to expect, and the combination meter has a repeater circuit for that. But it doesn't rely on that for the number it displays, preferring to just ask the skid ECU for that.)

Well, in my gen 3 anyway, perfectly cromulent MG RPMs are shown while in neutral. So either that is something Toyota changed for the Prime, or something else is going on in sylvaing's screenshot.

My money would be on the latter, but I can't say for sure. Maybe a few more people with Primes, using different scan tools or apps, could also check?

It's funny, while my gen 3 sits stationary in neutral, the MG1 RPM is what you'd expect from the engine RPM, but the MG2 RPM does not show a steady zero. That number pops up randomly zero or a few RPM positive or negative. Of course, those resolvers are so sensitive that the rotor bouncing a fraction of a degree within the mechanical gear lash can be read as an instantaneous "RPM", and the scan tool only shows a few updates a second, so for a little fraction of a second it can look like the rotor's actually turning.

If I stick a foot out the door and Fred Flintstone the car back and forth in the parking space, the MG2 RPM corresponds.

 

Right, well, one doesn't get the impression that they've done anything that radical in the HSD between G3 and G5, or non-plugin and plugin, but I can have a go with Techstream on my G5 at some point.

A "few" RPM is basically not moving. But maybe there is a deliberate attempt to hide "idle noise" in that app display.

 

Sitting in the driveway with 70% EV and warmed up enough to switch into HV mode with the engine staying off. I tried to start the engine while in N using EV HV EV Auto Eco Normal Power Defrost and Charge Mode. The engine didn't start.
@sylvaing I was able to shut off defrost by pressing the defrost button which switched to Auto than I pressed Auto to switch that off. When switching on defrost while in N it sounded a lot like pre conditioning. I think the heat pump twirled up while defrost was running.

 

This has not changed.

But what I noticed is that the Prime has made it esaier to "coasr" with your right foot. All you have to do is apply enough slight pressure on the pedal to remove the bar on the CHG/ECO/PWR display. And voilà, you are coasting, using 0kWh (on my HA).

 

I ran the Prime out of EV range 2 days ago 3/10s of a mile from home. Yesterday I had 2 miles of EV range showing. After running 2 miles at 25 mph I watched the EV range drop by 1/10s of a mile and at 2.3 miles driving I started using creep home speed. About 1 volt below the voltage the engine usually starts at, I shifted to N and rolled to the side of the road. I was drawing about 700 ma (DrPrius) . Typically after startup in EV I see 4 ma being used by the car (DrPrius). I had daytime lights on, but after shutting them off I still was seeing 700 ma draw. Traction Pack voltage fluctuated up at first from voltage rebound a few 10th than started to drop slowly by 100th of a volt. I sat for about 15 minutes and traction pack lost a couple 10th of a volt, I switched on the HVAC first on defrost and at the beginning as the heat pump twirled up, I was looking at between 4 to 5 amps, bouncing around between the two, and after perhaps 2 minutes I was being shown 1.6 amp (DrPrius) which is what I normally see switching on HVAC Auto after initially putting the car in Ready Mode. I let defrost run several minutes without noticeable difference in traction pack voltage besides the 10ths of a volt loss I mentioned above. Than I switched HVAC to Auto and let that run a few minutes, again without noticing much of a difference in traction pack voltage beside the 10ths of a volt here and there loss.
None of the Modes switched started the engine.
The one thing I did notice and was watching closely was the 12 volt batteries voltage drop.
The 12 volt dropped 2 volts in the approximately 15 minutes while idling in N, and at 10.2 volts showing I ended the test.

As soon as I put the car in D the engine started. I drove about 45 miles and stopped 3 times. I got my first experience driving locally without any EV range. ( I'd done it on long trips before but never locally full trips).

Without any EV range, HV mode and especially B mode work a lot differently than they do using B mode for motor braking only in EV mode.

Hwy miles were all in the 60 + mpg ( a nice change from winter mpgs ). and temps started in the low 60s F and fell into the high 50s F.

 

2012 Toyota Prius (not a plug-in or Prime) on 1 tank of gas (11.9 gal):
TripA distance driven: 911 miles.
TripA gas mileage: 76.5 mpg.
I did not use Neutral to achieve this, nor use EV above 12 mph from a stop.