While the rest of your analysis is correct, it does not require input current to slow down a motor. To speed up the ICE, you need to slow down MG1 (remember that the car is preventing MG1 from over-revving!). MG1 actually generates current when braking itself in order to keep the ICE spinning, and that current typically goes to MG2.

It's when the ICE kicks in power at speed (high gear equivalent, but delivering power; not fuel-cut nor heretical mode) that MG1 has to work with the ICE (forcing that energy to the MG2/wheels) where MG1 is typically sucking current from MG2 or the battery.

In low gear mode MG1 is taking part of the ICE energy to generate additional current for MG2 (or the battery) which lets the ICE (and MG1) turn fast while the wheels (and MG2) turn slow.

-- Alan

 

FastMover — I suspect that your TCH was switching in and out of fuel-cut mode as you drove. It does this very easily and almost seamlessly on slightly undulating roads with the cruise control on. I doubt that you were in "heretical" mode at all.

alan_in_tempe — I don't agree! I believe that in both these cases electrical power is flowing from MG2 to MG1. In fuel-cut mode, it takes actual work to spin the ICE; i.e., power must flow to the ICE. One way that this can be achieved is by using MG2 as a generator to drive MG1 so as to torque the ICE to spin. The "engine braking" felt really is engine braking, applied indirectly via MG2. If MG1 were sending electrical power to MG2, the car would speed up. As evidence I attach two Toyota US patents:

Patent #5 914 575 is about fuel-cut engine braking. See the Abstract; as well as column 3, lines 54 - 61, where it's explicitly stated that power flows from MG2 to MG1.

Patent #6 131 680 includes claims about "heretical" mode operation. See the Abstract; as well as the top of column 3; column 49, claim 4; column 50, claim 8; and column 53, claim 19, where it's explicitly stated that power flows from MG2 to MG1.

By the way, the above does not preclude some of the electrical power from MG2 from being diverted to charge the NiMH battery at the same time.

Stan

 

It does not sound quite right that you have a fuel flow in Fuel-Cut mode, nor that you realize an increase in power with the slight change in the accelerator pressure -- even downhill when a small acceleration increase is desired. I am not confusing manifold pressure and fuel flow. My TCH has a fuel flow transducer installed -- one the the perks of working in aerospace.

The (My) conclusion of Heretical Mode is based on: 1) Fuel Flow .GT. 1 GPH; 2) ICE online as indicated by speedo center display; and 3) Power transfer indication "Battery" to "Wheel" concurrent with 2) above. There is a chance that this is an MG1 condition in the forward direction, but the nomogram indicates that this would be very low drive ratio to the axle, and would not explain the MPG incication of 55-40 MPG. I do not have a tach installed yet, but the "ear" says the ICE is turning very, very slow.

 

It was me, not jbollt! However, you are somewhat correct, but not entirely. It depends on combined speeds of MG1, the ICE and the wheels, and the power needed at the wheels.

When the wheels are going, say 42 MPH, and the ICE is 0 RPM, then MG1 is spinning near its max RPM. If MG1 is "open circuit", then it is just spinning, not generating nor consuming power, being spun by power from either the wheels or MG2 or both. To keep this simple, assume we hold a steady 42 MPH on a slight decline to provide all the needed power so that MG2 is also free spinning (virtual open circuit like MG1). With out changing anything else, if you add some resistance to MG1 (no longer an open circuit), then current will flow and MG1 will try to slow down - it is acting as a generator (I am also ignoring the phasing which is irrelevant to this discussion - it is handled by the inverter, just as the inverter applies this "resistance" across the MG1 coils). If MG1 was shorted out completely, it would stop turning almost immediately, but we are just adding resistance to an open circuit (infinite resistance), not a short (zero resistance). The less resistance, the more MG1 tries to slow down, and the more current it generates. In this scenario, where the wheels are staying at 42 MPH, and MG1 is slowing down, then the ICE has to start turning from the force of MG1 trying to slow down. There is no need to drive MG1 to turn the ICE! The current generated by MG1 can be fed to MG2 to reduce the drag to the wheels caused by MG1 against the ICE.

On the other hand, if you were on an incline and the wheels were decelerating (or MG2 was being driving by battery current), then braking MG1 (generating current with MG1) to start the ICE would effectively be braking the car. In this scenario, current is fed to MG1 to prevent the car from slowing down as the ICE is started up.

I believe the typical scenario with the car near 42 MPH steady state, in EV mode switching to fuel cut ICE turning mode, current will be fed from MG1 and the battery, to MG2, while the ICE is spun up. Only when more than minimal power is needed to be delivered to the wheels while the car is above 40 MPH would current be needed into MG1 (and MG2!) to spin the ICE. Otherwise, MG1 is generating and MG2 is sinking current.

To directly address your comment regarding the work needed to spin up the ICE while the car is in forward motion, the power comes from the wheels and the battery providing power to MG2 in order to maintain wheel speed, but MG1 is usually generating power when being slowed down in order to spin up the ICE (reducing the current needed from the battery for MG2).

I am speaking loosely in terms of power/work/energy, but trying to be consistent and clear in terms of current. I can rework this entire argument in terms of voltages if that might help understanding (I am afraid it might add confusion). The inverter controls the voltages. When the voltage is raised higher than the motor-speed voltage, the motor consumes power, and when the voltage is less in magnitude, the motor generates power. Similarly, when the voltage is raised above the open circuit voltage of the battery, the battery gets charging current, and when below the open circuit voltage, the battery discharges current. The inverter manages the voltages on MG1, MG2 and the battery to balance the current from all three. It effectively controls the voltage applied to the motors when they are generating current by effectively lowering the coil resistance. The bottom line is, to slow down a spinning motor, forcing it to slow down, generates power from that motor, and slowing down MG1 is necessary to spin up the ICE in this 42 MPH scenario with the ICE starting off not turning.

-- Alan

 

Bob - I copied this from Stan's post in the FEH fuel injector shut-off thread. I've also seen the same correct and incorrect readings on my ScanGauge when in the fuel-cut mode. It always indicates fuel being used when in fact it should read just the opposite (9999mpg). Could this be what your seeing?

From Stan's post-
(1) When coasting in 'D' between 65 and 80 km/h when fuel-cut occurs:
(a) TCH's FE gauge says that the FE is 0 L/100 km
(b) SG says (correctly) that the ICE immediately goes OPEN LOOP
(c) SG says (correctly) that the ICE stays at ~1000 RPM
(d) SG says (incorrectly) that fuel consumption is still occurring at ~1 L/h
(e) SG says (incorrectly) that the ICE is at ~30-35 LOAD

I've been using the term "heretical mode" based on my driving experience and from what I can piece together from the prius web sites. I agree with the earlier post stating that without more sophisticated devices, you just don't know what MG1&2 are doing. But I just can't find another explanation for what the car is doing between and only between the 45-52mph speed range. The heretical mode just seems to fit. Here's what I find.

After the engine is warm I accelerate up to about 52mph and then slowly back off of the pedal just enough to get the engine lugging. After maybe 1/2 mile the SG indicates the engine load between 45-50, 1280rpm, closed-loop and 70-110mpg. If my speed drops below about 47mph before the car goes silent, I'll start all over again. Getting the engine to lug seems to be the key. It also works when the car is accelerating thru 45mph as long as the engine is lugging. Again, I can only do this at 45-52mph. As long as the engine load doesn't increase too much, I can usually stay in this mode for several miles while getting fantastic gas mileage. A decrease in engine load and better fuel economy are the only significant parameters I see changing on the SG. Also, engine lugging seems to bring the battery to a higher SoC than it would otherwise be in. Perhaps this is also a factor. If anyone can find a better name, or better yet, a quicker way to get into this mode let me know.

-Ed

 

alan_in_tempe — Sorry about the mixup of the two names! Precisely your arguments about power flow in fuel-cut mode had occurred to me, and yet I had come to the conclusion that I stated. Let me think it over a bit more, and refer to the Toyota patents again too. It seems that they are wrong in what they say in US Patent #5 914 575, column 3, lines 54 - 61. In particular, I want to look at column 13, lines 25 on. You're certainly quite right that loading MG1 would cause it to slow down, and that in this condition it is acting as a generator and not as a motor. I want to reconcile this with Toyota's claims. I'll get back to you!

FastMover — Okay, you have, I think, convinced me that the behavior you're seeing likely is heretical mode operation. Given, that is, that fuel is actually being consumed (see LOL TCH's remarks about ScanGauge's fuel-consumption readout errors during fuel-cut — but I don't think you're using ScanGauge anyway). If the MFD shows the ICE icon, then it is using fuel, as you say, and could be in heretical mode. But, a power flow arrow from the battery to the wheel doesn't sound right for heretical mode. Your high FE readings do indicate a very efficient mode of operation; however, the heretical mode of operation isn't very different in efficiency from normal operation. That's why I'm still hedging my bets here!

Stan

 

Remember that good patents claim and describe the minimum necessary to exclude others from the invention, and any patent that gives unnecessary (rather than just sufficient) details provides less (narrower) protection to the patent holder. Don't trust patents for details of implementation of a patented device. Patent law require full disclosure of what is protected, but not disclosure of all non-covered improvements on the claims. Detailed descriptions in patents do not have to cover every possible or every implemented embodiment of the patent in detail!

-- Alan

NB -- I am not a patent lawyer or agent, but US Patent #5774833 is one of mine, and may suggest how much I understand patents.

 

alan_in_tempe — I have now spent some more time reading most of Toyota's US Patent #5 914 575 to see what it says about how they spin the ICE using MG1 and/or MG2 when operating in fuel-cut mode. I'm considering the case where the car is coasting, say on a continuous downhill slope in 'D' at a speed above ~64 km/h (~40 mph), with no pressure on either the brake pedal or the accelerator pedal, so that this speed can be maintained for a prolonged period under gravity. Then pure-EV-mode operation is excluded, and the ICE must be spun (without fuel or spark) in order to protect MG1 from over-revving. This provides engine braking. The battery is not used in this circumstance — it neither charges nor discharges, since neither could be maintained for an extended period of time given the finite battery capacity. The question is whether electrical power flows from MG2 to MG1 or from MG1 to MG2 under these conditions.

The patent exhaustively (and exhaustingly!) covers many possibilities for achieving engine braking, including the case envisaged above. [For example, they also consider the case where only MG1 is used to control the spin of the ICE, with the necessary electrical power flowing either from or to the battery in the process.] It appears that, in general, if both MG's are involved in the operation, but not the battery, then power can flow either from MG2 to MG1, or from MG1 to MG2, depending on the amount of engine braking desired — e.g., 'D' versus 'B' gearshift mode. This occurs because of the complexity of the way that the "power-split" planetary gear set splits the torque between its three shafts, and the way this "torque-split" interacts with the different shaft speeds to produce the different possible directions and amounts of power flow.

I direct your attention to the following parts of the patent that I believe address the electrical power flow question in fuel-cut mode (I've added some annotations):
(a) Column 4, lines 24-39 — "systems with 2 MG's"
(b) Column 13, lines 29-47 — "3 ways to achieve engine braking"
(c) Column 18, lines 20-41 — "coasting downhill in 'D' or 'B'"
(d) Column 21, lines 13-36 — power flows from MG2 to MG1"
(e) Column 22, lines 26-67 — "coasting downhill in 'D' or 'B'"
(f) Column 23, lines 27-33 — "prolonged coasting downhill"
(g) Column 24, lines 5-14 — "power can flow either way"
(h) Column 34, claim 14 — "no battery intervention necessary"
Complete understanding of these statements requires understanding the governing equations, but is simplified by extensive use of nomograms in the patent. I don't claim to have worked through it all thoroughly, but I have satisfied myself that Toyota are correct in their claims. This patent clearly tries to tie up all possible engine-braking possibilities lying within the scope of two-MG hybrid systems.

Stan

P.S. I'm also not a patent attorney, but I think I understand why they're written in specific ways, and how to read and interpret them.

 

So if I understand that correctly, if going down a very long hill, it would be advantageous to do something that you would not do in a "normal" car -- ride the brake pedal a bit.

Slightly pressing the brake pedal would cause the battery to be charged while going down the slope (and of course keep you close to the speed limit), yet would not put any wear on the brakes or cause them to heat up.

Is the correct?

I could have been doing this when traveling through WV a couple of weeks ago! I thought the generator was charging the battery on these long downhills.

 

abward — Not really. I deliberately pointed just to the "continuous downhill" engine-braking example to illustrate just the steady-state (i.e., long-term) behavior. The Toyota patent includes all the scenarios. When the NiMH battery is below "full" they do bleed off some electrical braking power to top it up. That would show up on the MFD display, I think. They also discuss scenarios where only a single MG is used in conjunction with the NiMH battery — power flows either from or to the battery as the situation demands. However, I passed over these cases, because they can't apply to prolonged engine braking because of the finite battery capacity, and also because the 2-MG scenario seems to make better sense. I believe that, if your battery is fully charged, both MGs are in use during fuel-cut coasting, and the MFD won't show any power flow to or from the battery.

abward & alan_in_tempe — Perhaps I can try to explain in a bit more detail what's going on in the TCH's ECUs during 2-MG fuel-cut coasting at say 80 km/h (50 mph) in 'D'. ScanGauge tells me that the ICE is being spun without fuel (indicated by an "open-loop" SG readout — be aware that SG's fuel consumption readings are wrong during fuel-cut) at ~1000 rpm. To spin the ICE at this speed requires power to be input to the ICE. Now, the gear ratios in the planetary gear set ("power split" device) split the ICE's required torque 72% to the ring gear (i.e., MG2 and the wheels) and 28% to the sun gear (i.e., MG1). Each of these gears is spinning at a rate that can be calculated by formulas from the car's road speed (here 80 km/h) and the desired 1000 rpm ICE speed. Depending on the road speed, the MG1 gear speed could be positive, zero, or negative. MG2's speed is always positive when the car is moving forwards. The product of the MG1 speed and the required net MG1 torque gives the required net MG1 power, either positive or negative. If it's positive, then MG1 is set up to act as a motor; if negative, MG1 will be used as a generator. Similarly, the product of MG2's speed and MG2's required net torque gives MG2's required power, either positive or negative, and MG2 is set up to act as either a motor or a generator as well. Assuming that the battery isn't being used, then (ignoring efficiency losses) the power output from the one MG (the "generator") is equal to the power absorbed by the other MG (the "motor"). Depending on the situation, either MG can act as the motor while the other acts as the generator to result in the desired engine braking being transferred from the ICE to the wheels! This seems counter-intuitive at first (or even second) sight. But I think that's what is going on in essence.

Stan