Follow up to the video "What Voltage Do Toyota Hybrids REALLY Use?"

[bexander]

As there seem to be some discussion regarding this video I'm starting this thread to keep other threads clean.

First some facts:

In a motor, simplified, voltage produces speed and current produces torque. But to produce power you need both.
To be able to run a motor on the bench at high speeds with a specific voltage using field weakening, but with no torque will be no good when put in a vehicle because as soon as you ever so slightly but a load on the motor, the speed will drop.
Also a motor can produce alot of torque at standstill (zero speed) but as soon as the motor starts to spin back emf will reduce driving voltage and therefor reduce current, hence torque drops.

What I expect from any car is that it can accelerate up to a certain speed and then maintain that speed regardless of the inclination of the road.
So I want torque not only at standstill but all the way up to the desired speed, and I also want torque at high speed when I run uphill, otherwise the car will slow down.

My theory of the Toyota Hybrid System, that the video confirms:

The Toyota/Lexus drivetrain is not a battery EV drivetrain but a Hybrid system that "can" be used as a battery vehicle for short distances but really isn't designed for it.
When driving a Toyota Hybrid at normal speeds/accelerations the ICE is running which allows MG1 to produce a higher voltage that then MG2 can put to good use propelling the car together with some more power added directly by the ICE. The buck/boost is mostly there to allow for battery charging while driving and also provide short bursts of power during acceleration, not to power the car.
So even though the battery voltages is fairly low the motors are designed for high voltages.

Some practical testing of this can be found in my project thread:
viewtopic.php?f=11&t=912&start=25#p18420

My conclusion:

Toyota adjusts the voltage put to the motor from what is required. At low speed and/or low acceleration it can run at low voltage (battery), when more power is needed it boosts the voltage either using the buck/boost converter for battery operations but mostly from the ICEs input via MG1.
To use a Toyota motor in a EV you will need resonable high voltages to be able to produce torque at high speed, i.e. high power.

[ascii-art]

Here's a plot of the DC bus and battery voltages as reported by the OBD interface that my wife made from a 1.5h drive in her Gen2 Prius. Interestingly, the DC bus voltage is actually reported to be lower than the battery when the car is driving very slowly. These data are from stop 'n go traffic in downtown Stockholm and rush hour on the motorway. No chance for pedal-to-the-metal here. Is that the only difference to Damien's experimental values or is the OBD lying??

[johu]

Very nice, but are you sure there isn't a scaling error? Voltage after a boost converter can never be lower than before. input -> inductor -> diode -> output.

[ascii-art]

That would make sense. The numbers are straight out of the app Car Scanner (v. 1.67.9 carscanner.info) and the effect is the same when displayed in the app. I'll try to see if we can get some full throttle data to compare the highest voltages observed on the bus directly with the numbers in Damien's video. The battery voltage at least seems realistic.

[bexander]

In the plot, at low speeds the "Battery" and the "HV DC Bus" looks identical in shape.
Max "HV DC Bus" voltage at speed is about 350V in graph and according to Damiens video it is 500V.
If you scale the "HV DC Bus" voltage to the "Battery" voltage at low speed you end up with 500V at high speed, as per 350V (max HV DC Bus) x 200 (battery) / 140 (min HV DC Bus) = 500V.
A scaling error seem very likely.

[Dilbert]

Does this explain why mg1 is 1/2 the size of mg2 and run at a higher rpm / voltage.

So effectively at the higher rpm mg1 is matched to mg2?