Toyota Prius Forum banner

1 - 20 of 23 Posts

·
Registered
Joined
·
13 Posts
Discussion Starter #1
From Graham Davies' fine description I prepared a block diagram of the 2004 Prius' power system, which I attach herewith. I'd appreciate comments. In particular, I may be in error in showing two inverters; it's likely there's only one.

Ken Herrick
Oakland, CA

[See the figure in my Oct 6 posting]
 

·
Registered
Joined
·
928 Posts
Looks pretty good. You might add computer control of the DC/DC converter if you're a completionist. Also there's another DC/DC converter from the 201.6 V battery to the 12 V system. You might want to show an arrow back from planets to engine for starting and B mode.

So what are you going to do with this? Any interesting projects/teaching assignments in the works?
 

·
Registered
Joined
·
311 Posts
I don't have anytihing at hand to use to athoritatively IV&V the chart but have some questions/comments"

Are you sure that energy going from the 200 volt battery passes through the DC:DC converter then through the inverter to energize MG2 and power the DIFF/Wheels? Why does the 200VDC need to be "converted" before being "inverted" (made AC).

Might the flow labeled "To planets" (connecting the engine to planets) be bidirectional?

Not knowing what "structured methodology" you employ for your analysis, I'm not sure what the significance is in dashed vs not dashed arrows. Likewise the difference between arrowheads pointing to the periphery of an object vs those penetrating to the center. A legend explaining your symbology would be useful.

I have been both a student of structured analytical methodologies as well as a teacher of some of them but haven't kept up with the field for several years.

A single arrow connecting a couple "boxes" can be deceptively oversimplifying as there are numerous different flows contained in that arrow. Showing one is good as it prevents over cluttering the diagram but a list of flows and their components is useful.

For example, take the case of regenerative braking. If you didn't know how it worked the diagram wouldn't inform you or permit ou to figure it out. Don't misunderstand my comments. I'm not trashing your work. I think it is a fine start and am only trying to be helpful.

Again, I'm struggling in an information vacuum but get the sense that there may be some info flows missing as regards braking. Light braking, only uses regen, no mech brakes. Isn't this accomplished by controlling the bias on an MG? This would neccessitate an info flow from the computer to that MG (or its control electronics depending on the level of detail in your decomposition.)

I better stop before this post needs chapter headings and a table of contents... I say it again, nice start with your chart. I'll be interested in seeing its evolution.

:D Pat :D
 

·
Registered
Joined
·
13 Posts
Discussion Starter #4
My thanks to Robert Snyder and patrickg for their comments. (Tho I didn't know I was engaging in structured methodology*; how about that?) I attach herewith a revised drawing.

As to patrickg's questions, it's my understanding, from Graham Davies' description, that the (main) dc:dc converter acts to boost the ~200 V of the battery to upwards of 500 V to power the inverter(s)--and I'm still not sure about that "(s)", but I do think there are two.

Also, in order to fully understand the drawing, it's best to refer to Davies' write-up.

And finally, Robert, I don't plan to do anything more with my drawing. I'm long-retired & did it just as an interesting exercise to keep the synapses active.

* but I do admit to being somewhat of a "completionist".

[See the figure in my Oct 6 posting]
 

·
Registered
Joined
·
311 Posts
KenHerrick said:
* but I do admit to being somewhat of a "completionist".
Therefore you will be compelled to decompose the "compound" flows into their atomic constituents, in order to, as Hercule Poirot says, "exercise the little gray cells."

:D Pat :D
 

·
Registered
Joined
·
928 Posts
I like the improvements, Ken. As a quick reminder to those who don't immediately recognize the converter between the 200V battery and the inverters, you might label the line between them as 500 V DC, not simply DC as you currently have.

I just remembered that each MG has a position/speed sensor, so there should be a line from them to the computer, but that will probably clutter the diagram too much. Anyway the box should say COMPUTERS since there are many of them.

Since you're retired, you might be in a good position to volunteer to give a lecture or hold a discussion at your local high school physics class. The Prius is a great example of lots of physics topics.
 

·
Registered
Joined
·
27 Posts
Dumb question, how do you access the block diagram enclosure? Copying the graphic has too low a resolution to see it clearly. What am I missing?
 

·
Registered
Joined
·
311 Posts
RSnyder said:
I like the improvements, Ken. ... there should be a line from them to the computer, but that will probably clutter the diagram too much.

Since you're retired, you might be in a good position to volunteer to give a lecture or hold a discussion at your local high school physics class. The Prius is a great example of lots of physics topics.
RSnyder, Most excellent! The physics example and demonstrations/short rides would be a great thing. Kinetic energy, potential energy, drag, roling friction and on and on. Regen braking alone could make a college level physics topic but a "Reader's Digest" version would certainly be appropriate for high school physics.

Right again regarding clutter, hence the "Data Dictionary" where you define the constituents of the "major named flows." If you number as well as name each flow and then for each numbered flow you list its components you'll have your detail but not clutter the diagram. Of course you don't have to number the flows but numbering rather than naming on the chart will furher reduce clutter and the need for terse labels. The "FULL DESCRIPTIVE NAME is given in the data dictionary along with the "sub flows" which make it up, if any.

If you really get into this sort of analytical depiction you can distinguish between DATA flows, control flows, and material or energy flows. All connections aren't created equal. To communicate with precision in a clear and concise manner you might choose to differentiate between flows such as gasoline in a fuel line and a computer to sensor connection. Further, if flows are unidirectiional show the arrowheads only going in one direction. For example: gas only flows one way in the fuel line since as wonderful as the Prius is, it doesn't manufacture gas and fill its own tank.

Another example: a certain section of house wiring that is used for both 120VAC power distribution AND to cary a "wireless" intercom signal (FM mode RF) and or X-10 computer signals to control appliances etc. Someone might label the flow on a chart AC Power and data signals. Then in the data dictionary the particulars of the AC power and the data signals can be spelled out. (120VAC FUSED FOR 15Amps, X-10 signals to ...)

Really good stuff. Congrats on the diagram and congrats to RS for the physics class suggestion.

:D Pat :D
 

·
Registered
Joined
·
13 Posts
Discussion Starter #10
To arizonapilot: Just click on the diagram. It's a jpeg file & should come up in IE or whatever browser you use.

To RSnyder: I didn't specify 500 V per se because I understand it's varied depending on load, etc. Didn't want to clutter the diagram. I'll refrain from adding the speed-sensor indications for the same reason. "Computers" is no doubt better (my salesman told me there are 26 of them!) but I won't change that in a post untill/unless I have to make a more substantive change.

To RSnyder and patrickg: Teaching Prius...? Ah, you put me on the spot: Being retired, I just don't have the necessary motivation to get out & try to do that. Sloth rules my life, nowadays.

If anyone can shed more light on a) whether 2 MG inverters are really used and b) just how Toyota manages to make power flow--seamlessly--in both directions through inverters and converter, I'm sure we'd all love to hear it. And I could perhaps make the drawing even more accurate.

KCH
 

·
Registered
Joined
·
928 Posts
KenHerrick said:
. . .
If anyone can shed more light on a) whether 2 MG inverters are really used and b) just how Toyota manages to make power flow--seamlessly--in both directions through inverters and converter, I'm sure we'd all love to hear it. And I could perhaps make the drawing even more accurate.

KCH
Oops, forgot to answer the questions :oops:

a) Yes, there are 2 separate inverters for the two MGs. There are 12 huge power transistors in the HSD box taking up the bulk of the space. Each inverter needs 6, 2 for each leg of the 3-phase circuit. There would be no way to run the two MGs at different RPM (frequency) off the same inverter.

b) There's a lot of engineering involved in getting everything seamless, but the main tools are very accurate sensors for motor position and speed information to the computers, pulse width modulation (PWM) at high frequency to get very precise control over power flow in the inverters and converters, and lots of different tuned PID control loops for all the different situations that arise. These tools allow very precise and rapid control of torque on both MGs The other trick is to take advantage of the leverage available in the PSD to use MG1 to control the RPM of the engine. Then they can use the throttle and fuel injector dwell time to control its power output at a much slower rate of change to reduce pollution (engines do the bulk of their polluting during rapid changes in throttle position, RPM and load). MG2 and the battery are there to provide the rapid changes in torque to the wheels expected by the driver.
 

·
Registered
Joined
·
13 Posts
Discussion Starter #13
Thanks to RSnyder for the info on the two inverters. But now... From your explanation, I don't infer enough hardware to effect the reversal of power flow that is to occur when a MG's function changes from that of motor to that of generator. In other words, is each "MG", in fact, used as both a motor and a generator on appropriate occasions? Or is it actually the case that only one is used as a motor and only the other is used as a generator?

In order to reverse the power flow (DC power to AC power to mechanical power vs. mechanical power to AC power to DC power), it would seem to me that even more power transistors would be required to accomplish the necessary switching.

KCH
 

·
Registered
Joined
·
27 Posts
Hey, all we retired guys have to stick together or they'll put us back to work! Speaking of teaching aids- How about a Prius simulator with animation? Sort of like the one on the video screen but with more accurate detail including the power split animation and with joy stick input for accelerator, brake,etc.

Most of the structured methods analysis, which doesn't have to be extemely accurate or totally complete, has been done to some degree, from what I've seen on this site and the referenced material. Of course Toyota has modeled it to a tee, but that doesn't help us.
Many of the PC flight simulators have developed some much more complex simulations including flight dynamics and etc. I don't think it would be a major project. Unfortunately my skills are not up to it and even in my working days structured methods and some of the later program lanquages were not developed too well since I had the misfortune of being a senior engineering manager when they came to maturity. As they say nothing is too difficult for the onlooker, so--
 

·
Registered
Joined
·
928 Posts
KenHerrick said:
Thanks to RSnyder for the info on the two inverters. But now... From your explanation, I don't infer enough hardware to effect the reversal of power flow that is to occur when a MG's function changes from that of motor to that of generator. In other words, is each "MG", in fact, used as both a motor and a generator on appropriate occasions? Or is it actually the case that only one is used as a motor and only the other is used as a generator?

In order to reverse the power flow (DC power to AC power to mechanical power vs. mechanical power to AC power to DC power), it would seem to me that even more power transistors would be required to accomplish the necessary switching.

KCH
I didn't mention the 6 diodes per inverter (because they don't physically take up nearly as much room as the transistors). These are in parallel with and opposite direction to each of the transistors. They allow the inverter to act as a "boost converter" during generator behavior of the MG. It's a pretty neat trick that Lee Hart explained recently on the Prius_Technical_Stuff Yahoo! group. (See message 9809 and related messages). Basically, using PWM, they allow current to start flowing from the generator to one terminal of the battery for a brief time, then slam it shut. The generator windings act like a huge inductor and boost the voltage higher than the opposite battery terminal through the diode parallel to the other transistor on that phase. This enables effective battery charging even at very low generator speeds.

I'd post Lee's diagram, but I'm not sure how to set constant width font.
 

·
Registered
Joined
·
13 Posts
Discussion Starter #16
Thanks to RSnyder for the reference to the Yahoo group. Lots of good discussion there relative to this, before & after msg 9809.

I've also been wondering how the "converter" handles the battery-charging power flow. Here's a quote from Graham Davies there: "The 2004 Prius has a converter between the MG inverters and the battery. It wouldn't surprise me to learn that this converter is also capable of boosting in the MG to battery direction, although it is usually described as boosting in the other direction." Likely that is true, using the same general technique as in the inverters.

Another thing that confuses me, tho: I read mention of battery voltage as being 200-odd V, 340 V, 274 V and 300 V. Which is it really--at least, in the 2004?

KCH
 

·
Registered
Joined
·
928 Posts
From the "Prius New Car Features 2004" book:

The 2004 Prius battery consists of 168 cells ({1.2V x 6 cells} x 28 modules) with a nominal voltage of DC 201.6V.

. . .

A boost converter has been included in the inverter. This boosts the nominal voltage of DC 201.6 V that is output by the HV battery to a maximum voltage of DC 500 V.
Also, I notice that there's a third inverter for the A/C included in the inverter assembly (I believe they meant to say inverter assembly in the 2nd quoted paragraph since it was in a section about the inverter assembly).
 

·
Registered
Joined
·
311 Posts
Man oh man! As you guys "peel the onion", layer by layer, delving ever deeper into the innards of the Prius I continue to be amazed anew at each subsequent revelation. What a tremendous engineering achievment on the part of the Toyota engineers.

:D Pat :D
 

·
Registered
Joined
·
13 Posts
Discussion Starter #19
Not having seen a generalized overall description of the Prius' operation, I've come up with the following, referring to my attached & updated drawing, PRIUSPWR. I'll appreciate comments.

How the 2004 Toyota Prius works

Ref. Dwg PRIUSPWR rev. Oct 6, 2004

A 4-cylinder internal-combustion engine (lower left) and two electric motor-generators (“MG1” and “MG2”, left & right, respectively) are connected to the wheels of the car by means of certain special gearing plus the conventional differential gearing depicted at the right. The engine and MG1 connect via the special gearing (the planetary gear-set), and MG2 connects directly, to what would be the drive shaft in a conventional car. There is no clutch; all mechanical elements are permanently interconnected.

The planetary gear-set can be thought of as being like a conventional differential gear-set in that mechanical power is applied to (or taken from) an assembly of “planet” gears acting together (analogous to one of the wheels in the differential system) and a “sun” gear (analogous to the other wheel) while a resultant power appears at a “ring” gear (analogous to the drive-shaft). Depending upon the occasion, engine power may flow from planet-gear assembly to ring gear and thence to the wheels, provided that MG1’s torque sufficiently impedes the turning of the sun gear. MG1 power may flow from sun gear to planet-gear assembly and thence to the engine (for starting the engine) when torque from the wheels or from MG2 sufficiently impedes the turning of the ring gear. MG1 power may also flow from sun gear to ring gear at the same time as the engine-power flow. And excess wheel power (as when going downhill, for instance) may flow from ring gear to sun gear and thence to MG1, given sufficient resistance-to-turning of the planet-gear assembly (as, for instance, when the engine is off). Excess wheel power may also flow into MG2 as will be shown.

Both MG1 and MG2 are electrically connected, ultimately, to a 200 volt nickel-metal-hydride storage battery (upper left) through respective electronic power-inverters, which act to change between ac (alternating-current) and dc (direct-current) power, and a converter which acts to change between the fixed 200 volts of the battery and a variable dc level of up to 500 volts. The variable level connects to the inverters.

MG1 and MG2 are what are called step-motors. Each consists of a rotating element (the rotor), incorporating a powerful permanent magnet, and a stationary element (the stator) incorporating three rotationally-spaced coils of wire. Under computer control, rapid trains of electric-current pulses are applied from the inverters to the coils so as to produce rotating magnetic fields that pass through the rotors. Those fields interact with the fields of the permanent magnets to produce rotational torques on the rotors. The torques produce the mechanical power outputs when the rotors turn.

But MG1 and MG2 can also act as electric generators. While the trains of pulses are applied to them, and should the magnetic fields of the rotors’ magnets be caused—by external forces from, say, excess wheel power—to “get ahead” of the stators’ rotating fields, that excess mechanical power will be turned into added electrical power within the motor-generators. That added power will then automatically flow back through the inverters and the converter into the battery. In that way, power taken from the battery to produce pulse trains can be replenished in a recharging process. And charge-replenishment can come about not only from excess wheel power but also from the power of the engine itself.

The motor-generators MG1 and MG2 can turn freely when no electric pulses are applied to their stator coils. This condition occurs when the accelerator pedal is not depressed (neglecting a small excitation that is applied to simulate the “creep” action of conventional transmissions).

The engine runs essentially “on” or “off”, with its throttle set to run close to “wide open” for best efficiency. Because of that, there must be some way to vary the power that it delivers to the wheels. That is the primary function of MG1. MG1 acts, in effect, as clutch-and-throttle for the engine; as the means by which the relatively-constant rotation rate of the engine’s output shaft is coupled to the varying rotation rate of the car’s wheels, through the planetary gear assembly and the conventional differential. As noted, with no electrical excitation at all to MG1 its rotor will spin freely. In that case, rotation of the planet-gear assembly, caused by the engine, will impart no motion to the ring gear but rather will merely spin the sun gear and thus freely spin MG1’s rotor. Only when MG1’s inverters deliver the proper pulse trains to its stator coils, from the battery and via the voltage-converter, will MG1’s rotation become impeded, allowing engine power to flow from planet gears to ring gear and thence to the wheels. MG2, on the other hand, acts primarily in delivering power derived from the battery directly to the wheels, to augment engine power when required.

While MG1 is doing its throttling function, it is actually receiving some mechanical power from the engine—because it is impeding the rotation of the sun gear. That mechanical power becomes converted, in the manner described, into electrical power that finds its way back into the battery. Thus not much power is wasted in implementing its power-control function.

The engine is shut off at every opportunity. Whenever it is to be run, it is “spun up”, almost unnoticeably, by power from MG1. During reverse operation it does not run; only the electric motors propel the car.

There are numerous computers that implement the various control tasks. In addition to the control functions described here, accelerator-pedal position and brake-pedal force are monitored so as to control the engine and the motor-generators, and also to modulate brake-pressure. The latter action is done as a function of the “dynamic braking” that occurs whenever possible. In that action, excess wheel power flows through MG1 and MG2 as recharging power into the battery, helping to slow the car and also to diminish brake wear.

An additional voltage-converter and a small 12 volt lead-acid battery provide for auxiliary power, and a separate inverter powers the air conditioning system from the 200 volt battery.

Ken Herrick Rev. October 9, 2004
 

·
Registered
Joined
·
311 Posts
Ken, It would be most helpful if you could describe/define the content of the arrows connecting the boxes. I would be pleased to play devil's advocate and provide some feedback but as the chart is sort of general and notional I'm afraid I'll not be nearly as effective in providing useful feedback as I could be if you supplied more detail as to the content of the arrows. It is a good start, useful as far as it goes. I'm sure others would be able to help better also if you suppied more detail.

:D Pat :D
 
1 - 20 of 23 Posts
Top