Not possible... but my tranny is slipping...

If it is broken and not jambed or broken chain not jambed on the output side/final reduction gear, the wheels can turn. It is like a broken drive shaft to a rear wheel drive.

I believe there is no sensor for the chain, so there will not be any code.

In park, the ICE cannot rev above 1200rpm, otherwise MG1 will over rev. The ECU will limit the ICE.

 

Hi Bob J.,

Thanks for providing more details. It's very weird that no DTC are logged.

I can understand why you are reluctant to spend substantial $ on replacing parts without assurance that you've pinpointed the failed component. Other posters with transaxle problems usually report unusual mechanical noise coincident with the onset of driveability symptoms.

Good luck; please let us know what you decide to do regarding further repairs of your car.

 

Y'know ... I haven't stared at *all* the details here but this is
starting to sound like another burnt Classic MG2. Or maybe it's
MG1 this time. But that's happened to a couple of people with
the older cars, and lurching is one of the symptoms.
.
_H*

 

Hi All,

I agree with Hobbit. But rather than being a shorted winding, as seen in other cars, its sounds more like there is a common open, or high resistance on all the windings. Get some sort of micro-ohmeter and check out the motor windings are to spec. Then again, there was lurching, so maybe its a single open winding.

Is there also some sort of fuse in the inverter? Maybe its half blown, resulting in the OK from the dealar techs, but resulting in a high resistance that wont handle the 100 or so amps.

 

Sorry to hear of your troubles on this Holiday evening. Hope I can help.
I kinda agree with Hobbitt in that you may have a bad CVT but I'm still leaning a little towards a failed Inverter because I can tell when a dealer is lying because his mouth is moving. When a tech say's he's tested an Inverter I'd like to see how. Especially when its 3 phase mains are hooked to what they say is a failed tranny. And the Inverter will not fire without a load.The only true way to test an Inverter is to install a new one.
I don't think you have a mechanical failure of the CVT as it would make quite a ruckus and a broken chain would be catastrophic and scare the sh*t out of you. Like you said no noise. That may well rule out a mechanical failure.
Since you have an excellent background lets get dirty.
With car off.
Pull the cover off the Inverter. Wearing your high potential gloves make sure there is no 200 volts DC anywhere present in the Inverter specifically on the orange connector lugs.Thats hybrid battery in. There should no volts as the main relay from hybrid pack is open. If no voltage present and I would expect none turn the meter to ohms and measure each of the 3 large lugs in the back of the Inverter to ground. These are the 3 phase leads to the motors.Each lug to ground should measure open to meg ohm.
Do that simple test...oh. but first....I read and I never saw your answer to Patricks question. Does the car charge the hybrid battery sitting there with the ICE on?
Also post how the car has been maintained all these years. When was the CVT fluid replaced? Thats important to us.Bob Wilson has done an excellent study of Gen 1 trans fluids and how they affect those cars and possible CVT failure modes. Search his posts its interesting stuff.
Inverter coolant pump ever replaced? Etc.
Good luck.

 

Hi Bob J.,

Since you've got the Toyota repair documentation available to you, that will help. Please let us know whether you find any problems with your multimeter checks. Should you find a problem please take photos and post to help us follow your progress.

If you don't find any obvious problems with your continuity and resistance checks, you don't have logged DTC to narrow down the problem, and you continue to have driveability issues, then it sounds like it is time to start switching out parts. The most likely parts would be the hybrid vehicle ECU, the inverter, and the transaxle.

Since the HV ECU is the easiest and cheapest to replace, it would make sense to replace that first. Since the transaxle is the hardest and most expensive to replace and you don't hear any unusual noise, it would make sense to put that last in the order of parts to change out. Do you have the facilities available so that you can DIY those repairs?

Here's an HV ECU for $200 plus freight (verify part numbers to ensure the replacement part number is same as or greater than the installed unit):
eBay Motors: TOYOTA PRIUS 2001-2003 HV CONTROL ECU 89981-47060 (item 370073098954 end time Dec-26-08 04:04:36 PST)

Inverter for $375 plus shipping:
eBay Motors: 01 02 03 TOYOTA PRIUS HYBIRD INVERTER #WH (item 110307481042 end time Dec-03-08 06:59:43 PST)

Here's a transaxle for $750 plus freight:
eBay Motors: TOYOTA PRIUS 2001-2003 TRANSAXLE TRANSMISSION MG1 MG2 (item 160286203667 end time Dec-22-08 08:29:28 PST)

 

Hi Merc05H,

You will need either a large constant current mode power supply, or a 4 wire resistance meter. Whenever measuring accurate resistance below 1 ohm, 4 wire meters are the best. The extra two wires are sense wires measuring the voltage at the hook up point. So, the voltage drop in the main leads is taken out of the measurement.

One can rig such a thing out of a large power supply with a constant current mode and a volt meter. Then just use a calculator to come up with the resistance. Constant current mode is a common feature of electronics lab power supplies. Depending on the current rating of the winding, and its rate resistance, the required current capability of the power supply will be determined. I would guess you will want somethng with 10 percent of the rated winding current - 10 amps ? - to get enough voltage across the winding to get a good voltage measurement with the DMM. Another issue will be a proper connector to get good connection resistance. Getting a connector out of a scrap inverter, and wiring it up with 10 gauge wire to the power supply, and 20 gauge leads to banana plugs for the meter probably makes sense.

I would avoid hooking that test power supply to the motor while the motor is connected to the inverter. Also, you might lift a front wheel in case the motor makes a 30 degree movement when you hook up the power supply to a good winding. This way the car wont lurch with you under the hood.

 

Bob Jones,

This is the garage I mentioned to you earlier --
Replacing A First Generation Prius Transaxle

 

This poses an interesting thought, using house power through a current limiting light bulb to test stator coils:

• lift front wheels
• disconnect primary power leads from inverter
• using a 200 W. current limiting light bulb, test pairs of the three power leads
• expected result: a 60 Hz buzzing from the transaxle of equal sound level; also equal AC current for all combinations

If the windings are OK, I would expect all stators to equally rock the respective rotors at 60 Hz. This should be easily heard or felt at the wheels and the intensity should be identical. If one or more stators has failed, I would expect asymmetry in the sound.

The stators are all tied to a common point so as you apply power, you are enabling 'sets of stators.'

It may make sense to take an early transaxle oil sample. The fill hole is located at the front and if you can snake a tube in, you can draw a sample. You'd be looking for discoloration and burn smells.

Of course the ultimate sample is to drain the oil (take a sample) and then drop the pan. The pan may contain interesting debris material. Short of taking the transaxle out (entails engine removal) this is about as invasive as one can get without big problems.

I agree with Patrick on the order of 'swaps' if you go that way. However, I'm not happy with that approach. I would prefer to diagnose the problem(s) and parts swapping is always ugly. The HV ECU is not that bad and you might want to take a quick peek there first. The inverter is a about a 2-4 hour job, not too bad. But swapping the transaxle starts with removing the engine and transaxle from the car. This one is very ugly.

I wish we could get someone with a Graham scanner to visit and look through the codes. I find it all but impossible that there weren't any unless it is the HV ECU that has gone bad.

Bob Wilson

 

Hi Bob Wilson,

Also, the light bulb should glow equally for each three stator windings.

I would also use a GFCI plug , and an isolation transformer in combination with a AC Power auto transformer for the first trial of this technique. The auto-transformer may not be needed, but can't hurt to be able to ramp up the AC voltage gradually, and then take it down gradually. An alternative might be a zero-switched AC relay. This way no nasty spikes are going to occur.

 

Hi Merc05h,

Engine Does Not Start is commonly caused by a dirty throttle assembly. The engine should not have any issues starting before you delve into the more expensive problems. You might also check the plugs with a car of that age. And use some injector cleaner too. Another possibility would be a catalytic converter clog, or fuel pump malfunction. You probably need to take care of this first.

So, the question is, was this the problem from the start, or was there another problem, and the car sitting developed the sticky throttle / fouled plug / other issue?

Now, you are saying you checked the 3 phase leads to casing, and they were open, right ? Did you do that with a megger? If not, you could still have a problem. Also, did you check the winding resistances? On a simple DMM they should all be down under an ohm. And with a 4 wire resistance measurement they should be at the specified milli ohms value.

You are not saying that the resistance of each motor coil is infinity, right ? Because that would be a big problem.

 

From the kind folks I linked to earlier, a description of of how to check for and isolate voltage leaks:

------------
The first steps in diagnosing a P3009 are to clear the code with the scanner, turn the ignition switch to the "ON" position, let the car sit in "PARK" for a minute, then recheck for the P3009. The idea is to divide the circuit -- when the car is in park there is no power leaving the battery pack in the rear and going to the inverter in the front, so if the code resets, the leak is somewhere in the battery pack. If the code does not reset, the problem is somewhere in the cables, inverter, or transmission (or is intermittent).

Since the code does not reset if the problem is in the transaxle, our next move was to disconnect the two high voltage cables from the rear of the car at the battery pack, and at the front of the car at the inverter, then check for cable insulation leaks with a megger.

A megger, also called a megohm meter, is a tool used for checking insulation on motor windings or high voltage cables. It's like a regular ohmmeter, except it uses a higher voltage. I checked the output of our megger out of curiosity, and found that in the 500V range, it puts out 529V at 1.74 mA (about 9 watts). My Fluke 88 ohmmeter puts out 2.3V at .29 mA (about .01 watts). I'm not sure if you can zap yourself with a megger, or whether that zap might be painful or even be dangerous. I'm too much of a chicken to find out, so I keep my hands away from the probe tips just to be safe. A megger can be used to test the resistance of each winding of a 3 phase motor. If there are no shorted windings, each winding ought to have similar resistance. Before hybrid cars, meggers were most often used to test large electric motors on pumps, trains, boats, etc.. Our mechanic Enzo says the old megger he used while working as a tug boat mechanic had a hand crank to generate it's test current.

By the way, if the P3009 *had* reset while sitting in PARK with the key on, the next step would have been to remove the battery cover to check for shorts to the cover, etc. We pulled this one off because we had a codes P3030 (high voltage line snapped) and P3001(HV battery ECU malfunction), but that's another story.

Once we confirmed the HV leak was not under the battery cover (with the key on test), and that the HV cables from the battery pack to the inverter had good insulation (with the megger), we then start to check MG1 (Motor/Generator 1) and MG2 (Motor/Generator 2) with the megger. MG1 & MG2 are 3 phase motors, and therefore have 3 cables each. The cables come from the inverter, and must be disconnected from the inverter before testing the motor windings and cables with the megger. Notice that there is a small jumper connector on the square MG2 cover and another on the round MG1 cover? The jumper completes a circuit that tells the HV ECU (I think) that the cover is in place. If you were doing something silly, like running the car in gear on a rack, and tried to remove the cover to poke your fingers in there, the power would be shut off and your suicide attempt would be thwarted. This is one of the many interlocks that makes it hard to hurt yourself with a Prius. Anyway, we checked between each of the 6 cables with the transaxle case with the megger and found that all 3 MG2 cables had continuity to ground, meaning there is a short to ground in one of the windings. All 3 of the MG1 cables had the proper resistance, specified in the manual. This was the same on both transaxles we diagnosed; only MG2 was bad. You may notice that I'm not wearing lineman's gloves in this picture. That's because I'm sure there is no voltage. The linesman's gloves are only necessary until you've shut the power off and tested to make sure it's off. Once you've done that, there's no reason to worry about the orange cables and cumbersome gloves.

If you're like me, you may be wondering why a car with a "high voltage leak" is still driving around with no other symptoms other than the warning triangle. Shouldn't there be smoke, fire, melted cables or at least a few blown fuses? The Prius relies solely on MG2 for motive force in reverse. If MG2 were shorted, wouldn't there be no reverse? Yet all three Priuses with P3009s drove, and backed up, just fine. Here's what I think is happening. Auto mechanics are accustomed to thinking of electrical systems as grounded to the chassis. A typical circuit would start at the battery, go through a main fuse, through the ignition switch, through a circuit fuse, to the component being powered (the radio for instance), and then to the body of the car where the power would travel through the metal in the body to the negative battery cable and back into the battery. So when we think of a high voltage leak, we think of high voltage shorting to ground (the body) before the load, which would cause bad stuff to happen if there were no fuse in the circuit. However, the Prius three phase motor is not grounded to the body. The power does not flow from the inverter, through a winding, through the transmission case and body back to the inverter. Instead power pulsed between two windings in sequence. For instance, winding 1 & 2 might be energized first, then 2 &3, then 3 &1, the 1 & 2 again and so on, pushing the permanent magnet in the motor around and around, making the motor spin. Power is applied to one winding and grounded through another, not through the body of the car. The HV battery, cables, inverter, and transaxle do not push any high voltage through the body of the car; they are completely isolated. So if there is a high voltage leak to the chassis, it does not complete a circuit, but can be detected by the computer, and set a code P3009. This would mean than a voltmeter with one lead touching one of 3 phase cables and the other touching the transmission case, should read 0V, provided everything is working normally. I'll be testing this theory next time I have an opportunity.