3,500 foot decent = crazy HOT brakes and Burning Smells

Ok. Yes, I have specialized training above the level of basic technician hybrid training. I am reserving further comment until I have driven with the techstream logging current of mg2, mg1, and
Vehicle speed and deceleration rate included. . It may take a couple of weeks until I can get the data I need. I have to do coast down tests anyhow. Usually when I find someone won't let go of an idea (including myself) I find that there are truths to both sides. I want to make sure I know exactly where power is coming from and going to and at what times. I know there is a definite spike in current when in B position, but I want to know at what velocities it adds to what probably totals to amp-hours or watt-hours or watt minutes. I am willing to accept that what I was taught isn't correct, or is only correct for a certain set of circumstances. With a Prius, as complicated of a creature as it is, I am sure there are no absolutes. I'll let you know what I find.

XT1585 ?

 

Cheers Tony, I look forward to what you can uncover.

 

Take a look at this simulator Tony. Toyota Prius - Power Split Device

As I said previously, whether mg1 is motoring or generating really comes down to just one thing. And that is whether or not an external torque is being applied to it in the same direction as it is turning (hence generating) or if it is being applied in the opposite direction (hence motoring). This is absolutely fundamental and is always true.

As we are considering engine braking here, I will only consider the case where the engine is rotating. Look carefully at the simulator that I linked above. There are two basic "ways" that the engine can be caused to rotate.
1. Fuel injection, where the engine provides the motive power and "drags" mg2 around with it.
2. No fuel injection, where the car's inertia (or gravity) provides the motive power to mg2, and mg2 "drags" the engine around with it.

Whether mg1 is motoring or generating depends critically upon whether the pinion gears on the planet carrier (the orange gear wheels in the simulation) have a clockwise torque or an anticlockwise torque applied to them via this interaction between mg2 and the engine (the ring gear and the planet carrier).

To keep it simple consider only forward motion for now. There are just two situations regarding the direction of torque on the planet carrier pinions.
1. Under engine power (fuel injection). The torque on the planet carrier pinions is always anticlockwise.
2. Under engine drag (no fuel injection). The torque on those pinions is always clockwise.

So what this all boils down to is that we can completely judge whether mg1 is motoring or generating by just looking a which direction it (the blue "sun" gear at the center) is turning - with the proviso that the situation is always the exact opposite when there is no fuel injection compared to when the engine is powered. I'll summarize these cases below.

Engine powered case: If mg1 is turning clockwise then it's generating and if it's anticlockwise it's motoring.

Engine drag case: If mg1 is turning clockwise then it's motoring and if it's anticlockwise it's generating.

Play around with the simulator and you'll see that under most cases of engine drag that mg1 is motoring, especially when the engine rpm is high (as is normally the case when the system is trying to dump energy from the HV battery). It can however generate (mg1 be anticlockwise) when the road speed is high and the engine speed is low.

 

I know. It isn't that simple though. There are cases quite often where MG1 has torque applied as a motor and as a generator when rpms are + and -. During acceleration, it crosses the 0 threshold quite often. It's speed and direction are decided by what speed the engine should turn during acceleration.

XT1585 ?

 

Read everything I wrote carefully. All of what I said is true.

BTW. I'm not saying that mg1 doesn't switch between motoring and generating under normal driving conditions (engine powered). It will do so precisely as I said above, directly in response to the direction of it's rotation.

What we are primarily discussing here (in this thread) is what happens under engine braking. Please read everything that I posted before trying to argue about it.

 

While descending US 2 westbound from Stevens Pass (Washington State's Cascade Mountain range) today, I watched RPMs and speeds in various states.

The first 4 miles of descent, from summit ski hill to the switchback, is a fairly steady 5% grade. With feet off all pedals, D mode always produced gradual downhill acceleration from gravity, while B mode always caused enough braking drag to decelerate, in spite of gravity. Both of these effects happened, both when regenerating energy into the HV battery, and when a filled battery ceased regeneration so the only braking drag was from the spinning engine.

Initially, at ~60 mph with SOC at 6 then 7 bars, D mode produced 1150-1200 RPM, while B mode produced 2700-2800 RPM. Well after the battery was filled up to 8 bars of SOC (and really filled, not just rounded up from an internal 7.5), D mode produced 2200 RPM, while B mode initially produced 4600 RPM, then dropping a bit as the car gradually slowed.

I performed two full cycles of gliding up to 64 MPH in D mode, then braking back to 58 MPG in B mode. Then on the final cycle approaching the switchback, I let B mode slow that car all the way down to 47 mph, again without touching the brake pedal.

In short, at highway speed, B mode does provide very real and very significant braking even when a full traction battery shuts down regeneration, thanks to engine compression braking.

I can agree with Tony2ltr that on much steeper & slower roads, e.g. Mount Washington's slow 12% grade, B mode braking can be insufficient to control the car without friction brakes. I have experienced the same on the Pikes Peak Toll Road. But at Interstate Highway slopes and speeds, the situation is different, and B works well whether or not the traction battery is filled.

I plan to re-run this test and record new RPMs on a 7% slope later this week. From experience, that hill is steep enough that B doesn't completely displace the friction brakes, but does provide the vast majority of the required braking drag.

 

Tony's statement that started all this off was that "B" mode provided more current to the HV battery thereby charging the battery faster.

This flies in the face of the current knowledge.

Tony said he will conduct experiments to prove his case.

I am awaiting this dataset as it is of high interest to me, and I suspect others.

As far as tenet of what fuzzy1 posted above in relation to braking ability at different slope ratios and braking ability of "D" vs "B", I don't think this is disputed by anyone.

 

Yeah, that sounds very similar to what I've encountered too fuzzy. Descending at speed I get a much earlier response in "B" mode (starts revving the engine earlier) and it goes to much higher revs as the charge level tops out.

Some people find those high revs a bit disconcerting, but it certainly does the job. In "B" mode it can do a lot of engine braking even when the battery is topped out, particularly at high speeds

At lower speeds like say 35 MPH, "B" mode is still very effective, however it is not able to rev the engine quite as fast (otherwise it exceeds mg1's maximum speed). So if you repeat this at say 35MPH then you'll get a max of about 3200 RPM engine speed.

BTW. When it's engine braking with high RPMs like that then MG1 is definitely acting as a motor rather than a generator. So mg2 can still do some regenerative braking without there being any net regenerated power.

 

I don't watch the battery and MG1/2 currents, so have no hard data to contribute to that point.

That is not how I read this and similar passages, which triggered some of my earlier responses and today's test:

----

A bit of semantics here. By 'regenerative', I am counting only that energy flow that is stored or transferred for reuse. I am excluding the energy that is simply shuttled over to some dissipator, be it a resistor bank or a motor with a draggy load that produces no useful work.

 

I stand corrected, I actually forgot about that one.

I also half agree and half disagree with Tony's statement in that, if you have all 8 bars (i. e. you are at 80% SOC), then, yes I agree, you have no re-gen going into the battery (but you may still have re-gen as uart says above), but I disagree that you are only left with hydraulic braking -- because my understanding is that engine braking is substituted for the lost re-gen capacity. That leaves the steepness of the gradient to come in to play as to whether engine braking alone is sufficient to do the job.

I await the outcome of Tony's research.

 

And vehicle speed too. I agree with Tony on a different point, that at low speeds, the MGs don't produce much power, thus limiting the amount of regeneration (and engine braking) available.

On my Pikes Peak descent, I was mostly stuck behind other traffic going significantly slower than I wanted, but there were no safe passing opportunities. Some drivers ahead were geared down very low, in gear choices very appropriate for their vehicle powertrains, but rotten for my Prius. Before initially catching up to them, and later when pulling out onto empty road after brake cooling stops, I could reach comfortable speeds that did provide significant engine braking. It was still very short of the total needed braking drag, so friction brakes were still heavily used. But as soon as I would catch up to the prevailing traffic and have to slow down again, regen / engine drag would shrink sharply.

Having taken several brake cooling stops, the Brake Checkpoint monitor was happy with my brake rotor temperature. An F150 with Florida plates, already stinking to high heaven, was ordered into the cooling off timeout zone.

 

So I have been watching MG2, MG1, HVB current, engine power, rpms from everything. I need to work out the math and address for current from both motor generators. Current jumps up quite a bit during regen. However, it seems I may have been incorrect about the function of mg1 during regen. I have to check my current polarity for MG1 again, but it seems that while overall current jumps significantly during regen, most of MG1 power production is under acceleration, where it's power is put on the HV bus to be used to charge the HV battery or reused by MG2. Evidently, MG2 can create lots of current right down to about 4-5 mph in "B" with the brake pedal applied. At some point before MG2 current drops, the vehicle should be switched back into "D" or the idle stop command from the HV ecu doesn't occur. While I still have to go back and make sure the math is right for torque, I'll verify everything with the techstream scan tool later. There were a few anomalies I noticed and will be revisiting as well. But so far it absolutely does seem that MG1 is a power consumer during B regen while on the brake pedal,meaning it is consuming current in order to increase engine rpm. What I still have to work out is how much of the extra regen out of MG2 is recycled to mg1 VS. How much goes back to the HV battery. So here is where the issue is, so far it seems that in B current on the HV bus can almost double. Some of that current is diverted to MG1 in B for the seemingly primary purpose of spinning the engine to a higher RPM, so yes, it seems that one aspect of B position is for compression braking. The other aspect is to increase HV battery charging. I have verified that utilizing B mode increases bus current and the rate the HV battery charges. When it reaches about 80%, the regen current drops off like a rock, no matter how much heavier the brake pedal is used to float in that regen space, once the battery is full, no more current goes to it, and engine braking will eventually stop working as well. I'm not sure why yet, but since MG1 is so very small, it may be temperature derating. I have noticed, just like when I tune the EVs I build, there is an allowable discharge/charge rate in the data stream used to determine on time for the IGBTs and boost voltage in the inverter. This is another issue that makes data hard to interpret, as the current to voltage relationship changes in the boost converter. Also I know power factor (likeliness of voltage and current to be in phase) is a major issue for MG2. Just because current is high, it doesn't necessarily only change as a result of load, but is RPM dependant as well. So to my graphing arsenal boost current and voltage will be added. It will take a while, since I don't get too much uninterrupted time any more. Back to MG1 operation though...during regular regen in D position, mg1 at times is a power producer. That power drops off and current reverses if the shifter is moved to the B position, at that moment MG2 produces a lot more power, depending on the HV state of charge. Using MG1 to turn the engine at that point would be the only way to dispel some of that current. Because the P112 transaxle uses permanent magnet Motors, they produce voltage without a field set up by the mains, so there has to be a place for it to go when the battery can no longer accept a charge, the only place for that extra energy it seems, it MG1 using that bus power to turn the engine at a higher speed. However, at a certain point you would think that as the battery charges, engine rpm would increase, due to the large amount of excess energy on the bus. I don't think this is the case though. That energy can be discharge very quickly if a brief period of uphill driving is done, the engine will be in fuel cut, and can operate at highway speeds on MG2 alone. This can bring the SOC of the HV battery down very quickly, as it is only about 1.3 kWh. Anyone who has played with an EV mode button can attest to how fast the HV battery can discharge. Anyhow. It seems that MG1 is used to turn the engine during B position, and it has a small power consumption. But overall, regen bus current goes up significantly during B mode, until the battery is to about 80%, then regen seriously derates to almost nothing. I know I have also experienced loss of engine braking all together with a full battery, which doesn't seem to make much sense if it's purpose is to dispel energy with engine braking. More later, and sorry it took so long to get back with information.

XT1585 ?

There absolutely is no doubt that more regen current is produced during engine braking. The engine would be one place to discharge energy to, but another would be heat dispelled from the power transistors to the heat sink. I haven't heard that any plug braking is done by mg2, but I suppose it is possible.

XT1585 ?

Hey listen. I see that I have quoted myself. I am still getting used to this Tapatalk app that I have been using for all my forums.

XT1585 ?
XT1585 ?