New inverter pump installed, not working, what could be wrong?

This might sound unrelated to you, but there is a simple test you can do to check to see if your 12v battery is putting out adequate power to boot up the system to start properly.

First, check that all connections to the 12v battery are free of corrosion and tight. Including the ground cable connection at the chassis.

Put the car ro the "acc" position where the dash lights are on but the car is not started. Then measure the voltage of the 12v battery. You can measure it at the battery or at the jumper link in the fuse box under the hood.

If the voltage is 12.1v or less, your battery is insufficient for booting up the system properly.

This can cause numerous issues, such as inaccurate codes and trouble lights being triggered, and ECU dependent components and functions not working properly.

Seriously, do this. It may save you from wild goose chasing problems.

 

I've yet to see anything to convince me of that claim. In the Toyota repair manuals, there will often be a note advising you to make sure it's at least 11 or to charge or replace before proceeding, and that seems to be conservative too (as you would expect them to be); any problems I've ever experienced with "booting up the system properly" have kicked in well below that.

Some of the ECUs in the cars do have trouble codes they are programmed to set when the voltage goes below a threshold, generally 9.5. So if you try to power up and that happens, you will end up with those codes (even though the system may otherwise boot up just fine). Those won't be fake or spurious codes or anything; if you look them up in the manual, they're the exact ones that mean "hey, did you know the voltage went below 9.5 here?" and the ECUs that set them were properly doing just what they're programmed to do.

 

What Chap said. My '02 will boot to "Ready" with less than 12V showing on the ScanGaugeII.

 

It's very unlikely that O2 sensor problems would cause this. And if it's the downstream (sensor 2), then it's absolutely not that.

What kind of code reader are you using? Many generic ones are pretty lousy about pulling codes outside the standard OBD 2 ones. It seems yours is at least picking up the P3191, but I'd be surprised if there aren't more codes stored in there.

For a hoot, unplug your MAF sensor, start it up, and see what happens.

 

Quick way to find out if the code scanner you're using will even show you the trouble code you know you have because the sensor is unplugged.

 

If you still have your OEM cat, you might check the HCAC actuator, especially in Vermont. Maybe it's rusted shut, so it won't open up after your converter warms up. That would trigger the ox sensor codes, I'd think, and mess with the ECU's fuel mixture. That usually throws codes, but I've not heard of your scanner, maybe it's missing them. Also, you don't have to replace the cat to change out the ox sensor.

 

On your pump, you should be able to hear it running. It's right behind the driver's headlight, and it makes the kinda noise you'd think a small electric pump would make. IIRC, it should be running with just the key turned to ACC, it doesn't have to be in Ready mode, so it's easy to hear. If your tech hasn't thrown that old one away, get it back; it's probably still good and you'll have a spare. Or sell it on Ebay pretty fast, one of the easiest Prius parts to sell. Test it with just a hot wire and ground to the connector to make the pump run.

If you can't hear the one in your car, then you need to check the power starting at the pump connector and going back through the fuse box to the ECU, see where you're losing power.

 

Sorry you can't see the obvious general and computer science in my statement. And Yes as you say the system will start down to even below 10 volts.
So will my 19 volt laptop start up at less than 9 volts too. But not without issues and certain functions being compromised for either the Prius system or my laptop.
In fact, low voltages have more of a tendency to cause damage, even over reasonably too high of voltage, though both are insufficient for proper operation.

I have had far too much experience with this voltage issue in computer, ECM, ECU vehicles, where voltages below 12v absolutely can create ghosts and glitches in the diagnostics and even in certain functions.

Oh sure, but they run...

And certain components recover okay once running with the battery receiving 13.5+ volts.
But not all, especially in certain components such as parts of the ECU, ECM, and other computers.

Deep Cycle type batteries such as recommended in the Prius are very voltage resilient, that is why toyota uses then in this hybrid system where they are relatively passively charge from an invertor and not an alternator, and can take considerable discharge at times.
That is why a low battery should be charged then tested before determined bad.
And this is why, mostly otherwise for marketing reasons, Toyota wont list a battery charge as insufficient till below 11 volts. And 9.6 volts is indicative of a low charge under heavy load.

No. You are not wromg.
Neither am I.

I have demonstrated this and applied my diagnosis of this so many time and very successfully not just for myself, but for customers (before I retired) and for many people that i have cuurently assisted and advised who have come back to me with successful results and not had to do the coded repair that was supposedly needed per the almighty (sic "Flawless") code reader.

 

I can't say for sure on this specific engine - but on many cars that use MAFs, the car will run with the MAF unplugged. So if it's not running with the MAF, but runs with it unplugged, then the MAF is bad. I don't know the specifics, but I'd imagine that without MAF info the computer just uses substitute values based on throttle position. That's a guess.

 

Not to nitpick, but there could also be wiring issues interrupting the signal from the sensor.

 

LOL. Not a not a nitpick @ronlewis . You are indeed correct! Always inspect/check the wiring before condemning a part.

 

Just a general info add-on, and I'm not sure it completely applies to these cars, but it does on my Ford diesels...

When I see multiple codes appear, some of which don't seem to logically relate to others, the first thing to check on my trucks is "reference voltage," or as I think of it, "network voltage." All the truck's sensors are connected like a LAN such that any short/grounding of the LAN's reference voltage can distort/cause/omit sensor voltages connected to the LAN. There's a Vref datapoint on TorquePro for my trucks that should always be +/- 5v. Multiple unrelated codes is a basic red flag for Vref issues, which isn't usually the problem, just the first thing to check, like fuel and charged battery. But, if it is the issue, it can be a PITA to find the short anywhere in the LAN wiring.

Also, TorquePro and my Forscan (Ford) tool both list codes in the chronological order in which they were set, which is usually a good way to try to fix them, but also, you might logically assume that some later ones were actually caused by earlier ones and will disappear when the first is fixed. If you start at the bottom, you could waste time/money.

 

"Multiple unrelated codes" in PriusChat posts often turn out to be, instead, instances of "multiple codes I think are unrelated."

The Prius is a network of a bunch of ECUs, and sometimes people are surprised that the ECUs actively gossip with each other. Some ECUs have specific codes they will set just because some other ECU set some codes.

So, for example, it's pretty common to have the engine control module set some code (about something really happening with the engine), and the HV control ECU set a code of its own like P0A0F (hey, I asked for engine and didn't get it), and the braking and stability control lights come on with C1259 and C1310 (hey, we noticed the HV control ECU set a code).

Then the owner says "this is nuts! I got engine and hybrid and brake and stability lights all at the same time!' but if you just breathe and look up what the codes mean, it all tells one coherent picture.

 

My trucks have known issues with the harness chafing at different stress points around the truck to cause a short. We're lucky that we don't see more wiring issues on these cars, which are the same age. But yeah, not usually the cause of multi-codes, however, it's the basic first step and so easy to do with TP/Forscan.

I mention it because I don't get many datapoints for these cars using TP - RPM, temps, system voltage, is about it. I assume that you get everything with Techstream and maybe some high-end tools for Snap-On, et al. So, I don't even know if there's such thing as Vref on these cars. I'm flying blind most of the time with only whatever codes TP captures.

 

These cars use multiple network technologies, especially Gen 1, pre-CAN and quite a hodge-podge: BEAN, AVC-LAN, and serial links. It would be too much to assume there's a single concept or value like Vref that applies to them all. The manual does have some details of those individual technologies. It doesn't tell you a lot about their physical layer specs, but generally enough to get through troubleshooting.

It's worth addressing this straw man wherever it comes up. There is no "coded repair" "supposedly needed" by a diagnostic trouble code; not that there aren't people too often treating the codes that way, but that's their mistake. The only thing a DTC tells you is that the detection conditions listed in the repair manual for that DTC happened. It is always up to a human to find out why those conditions happened, and what to do about it. People who proceed accordingly when interpreting a DTC get successful results. People who misuse DTCs aren't really the best population to measure some other methodology against.

 

Yep, doesn't look like it applies here. Had to look up the DTC - P0641. Nowhere to be found in this forum or the service manual. Google shows it in GM and Ford products for sure, along with detailed articles on "vehicle reference voltage," calling it "complex." That leaves me out, lol.