Heretical Mode from a Different Perspective

A while ago, I was told I was using the term "commutator" in a blasphemous manner. This got me thinking again about my first deviation from accepted doctrine - the heretical mode of the PSD in which power is drawn off by MG2 to force MG1 to motor backwards. Previously, I have tried to explain this mode by analyzing the car's behavior. Not everyone is convinced. So, I have come up with an explanation of heretical mode from a different perspective, that of the designer's motivation. I present this below for your edification and, perhaps, amusement.

Imagine you have the job of designing the control software for the Toyota Hybrid System. Further, imagine that Toyota operates in the same way as the company I work for so that software people have no say in how the hardware is built and know nothing about it until it lands in front of them. You know how conventional cars work, but you don't know what the powertrain designers had in mind, if anything, with this weird arrangement of motor/generators and gears. You just have to make it work.

It's pretty obvious how to use MG2. It can boost the car along using power from the battery, either with the engine or alone. It can slow the car down during braking and coasting and also replenish the battery charge. On the other hand, MG1 and the Power Split Device do not, at first, have an obvious function.

So, begin by ignoring them. Lock MG1 in position somehow. Jeff has proposed driving a screwdriver through the casing for this purpose. The sun gear of the PSD is stationary. Now, if we get the car going (perhaps using MG2) we can drive along fairly happily, but we only have one gear. From the nomograph, at the minimum spin rate at which we can expect the ICE to be comfortable, say 1150 r.p.m., the car will be travelling at 27 m.p.h. At the other end, if we drive at 100 m.p.h. the ICE spin rate will be 4270 r.p.m. So, it looks as if we are in high gear. We can't use the ICE below 27 m.p.h. nor can we get much power from it at moderate speeds. We need lower gears.

What characterizes lower gears? Well, in a conventional car, a lower gear means that the ICE spins faster for a given road speed and the torque applied to the wheels is increased. Let's pull out the screwdriver and let MG1 spin. The ICE certainly spins faster because it is now causing both the wheels and MG1 to spin forwards. So that the car moves, we have to limit the spin rate of MG1 otherwise it will just whizz around like crazy and no torque from the ICE will get to the wheels. Conveniently, we can do this with generator drag, that is, we configure the control electronics so that MG1 acts as a generator and we draw off power. What shall we do with this power? If we configure MG2 as a motor and pass the power to it, we can have it help drive the car. Whoa! That's it - both increased spin rate for the ICE and more torque at the wheels! We have low gear! We're done!

So, the car goes out for road trials and everything's fine except for cruising. At 70 m.p.h. the ICE spins at 2990 r.p.m. The testers all have conventional cars that rev. lower than this and they object to the noise level. They've also gone to the engine people who assure them that the engine can produce sufficient power for cruising at a much lower spin rate and with greater fuel efficiency. Your boss "reminds" you that this is supposed to be a highly efficient car (which he actually forgot to mention originally). Obviously, what you need is a higher gear - overdrive, in fact.

A higher gear is characterized by lower ICE spin rate a given road speed and decreased torque applied to the wheels. Well, if spinning MG1 forwards gave you a lower gear, maybe spinning it backwards will give you a higher one. It will certainly reduce the ICE spin rate (refer to the nomograph). But, the torque from the ICE is acting to spin MG1 forwards. As the planet gears in the PSD rotate around the common axis, they push both the ring gear (and therefore the wheels) and the sun gear (and therefore MG1) in the same direction as they are rotating themselves. This is great when we want MG1 to spin forwards; we control its spin rate by taking out power. But, to make it spin backwards, against the torque from the ICE, we need to configure it as a motor and put in power. Where will we get this power? We can't get it from the battery or it will be discharged by just driving on the highway. The boss isn't looking, so let's try getting it from MG2 by configuring that as a generator. The only problem is, taking off power here means generator drag which will reduce the torque applied to the wheels. What? Reduce the torque at the wheels? We've got it! Overdrive! Lower ICE spin rate but less torque at the wheels!

The ending of the story is the obvious one. The car now drives perfectly and the testers are happy. Your boss finds out how you did it and fires you. Your replacement pretends to make a few changes and points out how well the car works and the boss assumes he is no longer taking power from MG2 to achieve the overdrive effect. Since your boss got his job through the Far-Reaching Network of Incompetents, your career is ruined and you take a job behind the counter in a drug store. Just as your promotion to store manager is about to be approved, someone driving the car you helped design but with a screwdriver through MG1 loses control due to an excess of torque at the wheels and plows into the drug store putting you on permanent disability. Your dog runs away from home. But, you have the pride of knowing that you did a difficult job well. Big deal


Last edited June 24, 2002.  All material Copyright © 2001, 2002 Graham Davies.  No liability accepted.