Is this transaxle dead?

I was wondering if my diagnosis of a broken transaxle on a Gen 2 Prius is correct! Initial symptoms were as follows (not witnessed by me):
- Issue started as intermittent shunting or jolting motion on the highway with cruise control active
- Progressed to severe jolting after some time highway driving, continuing down to low speed, no warning lights
- Car drove normally from a cold start, issue only occurred after some time on the highway
- Eventually car stopped with the red triangle, the engine would not restart; 12V battery died while waiting for recovery
- Car was towed on flatbed and then dragged off with parking pawl engaged (12V battery dead)
- Car has 220,000 miles and transaxle fluid was never changed as far as I know

Now I came to look at the car:
- I fully charged the 12V battery. It seems OK (holds a charge and maintains >11V with car on)
- The hybrid battery was measuring around 200V; I charged it to about 215V; there seems to be enough charge to start the engine (maybe not initially, I'm not sure; the Dr Prius app was reporting about 15% SoC initially and 50% after I charged it)
- The inverter cooling pump is running; the car turns on but never displays Ready, the shift indicator moves from P to N but never to D or R; the car does not move under electric power, the engine does not start; the car can be pushed when shifted to N
- DTCs are P0A90 (drive motor A performance) and P0335 (crank sensor)
- Checking the different reasons for the P0A90 code (unfortunately I can't read the info codes), I think the most plausible reason for setting these two codes is a mismatch between MG1 RPM and expected engine RPM during start; on clearing the codes, both are set again following a start attempt
- I replaced the crank sensor with a new one anyway; this didn't make any difference; the harness looked OK but I didn't measure continuity to the ECM
- The car tries to start when turned on with brake pedal pressed, after a short delay; there is a horrible noise of rotating machinery (spanners/wrenches in a washing machine) for about 5 seconds; the engine does not rotate (looking at the crank pulley, it twitches a bit but does not rotate); the 12V increases to 14V and the Dr Prius app shows about 5A drawn from the hybrid battery
- At the very first attempt to start, the engine did sound like it was rotating, at much higher than normal speed; it stopped after about 8 seconds. Subsequent attempts gave the horrible rotating machinery noise, gradually reducing in intensity, and no engine rotation
- I can turn the engine with a socket on the crank pulley; this does not turn the wheels
- During the starting process, monitoring the MG1 RPM and the engine RPM with the Dr Prius app, the MG1 RPM increases to about 4000+ RPM and then slowly reduces; the displayed MG1 RPM appears to align with the horrible rotating noise, which disappears when the MG1 RPM falls to 0. The displayed engine RPM is always zero (consistent with the crank pulley not rotating). (MG2 RPM is always 0).

My understanding is that if the car is stationary (MG2 not turning), if MG1 is turning then the engine must also turn; if MG1 is rotating but the engine is not rotating, then there must be.a mechanical breakage inside the transaxle. Does that sound right? In other words, it's toast?

Thank you!

 

Thanks to all for the feedback! I was a bit limited in time when looking at the car - ideally I would be able to spend a bit more time on it. Some answers to the questions:

@saneesh8 - thanks for this link, I did indeed check these different possibilities for P0A90. I focused most on the info codes in document 0a90239.pdf (codes 239, 241 and 602), because these seem to relate to a discrepancy between MG1 and engine RPMs, but I don't know the actual INF code so can't be sure that this is appropriate. The crank sensor I fitted is a Lucas part, so probably at the bottom end of the quality scale, but I can see that the crank pulley is really not turning, so I lost interest a bit in the crank sensor after that.

@BiomedO1 - when I turn the engine by hand it feels like I imagine it should, with a bit of resistance before going over centre at the top of stroke, but it was easier to turn than I thought it would be. But I read somewhere that an Atkinson cycle engine is easy to turn. I didn't drain the fluid or oil, it's a good idea to check for metal fragments.

@ChapmanF, @BiomedO1 - yes, I was thinking that a broken crankshaft is quite unlikely, and it's maybe more likely to be a shaft breakage inside the transaxle (or bellhousing). I couldn't find any mentions on this forum of a broken torque damper so I'm not sure if it's possible for the two halves of this to separate, effectively disconnecting the engine, or if it's more likely for the shaft itself to break on one side or the other (alternatively, I suppose the chain could break, but then MG2 might rotate instead of the engine, when MG1 runs, and this should be visible in the live data, and it always shows 0). I'll try to find the pictures on Priuschat of the breakages in the bellhousing.

@PriusCamper - Interesting point about rough running possibly damaging the shaft - I wonder if this was the case, and there was some misfire or other issue causing the jolting that became progressively more severe. I didn't experience this myself, and I wonder if severe jolting could have broken the shaft. The engine is not even cranking now, let alone running, so there's not much I can do with the diagnosis of this (there were no other ECM codes set, apart from the crank sensor one, though I'm not sure if they are erased when the battery is disconnected). To charge the hybrid battery I built a very basic 230V mains charger similar to the "LED driver" type that's described on this forum, to charge at about 0.2 amps for a few hours - not ideal and not very safe, but really just an emergency measure to eliminate the hybrid battery as being the reason for the non-start. I'm reasonably sure the hybrid battery has enough charge to start the car, and it looks pretty balanced from the DrPrius app diagnostics (SoC 52% and blade voltage difference 0.08V), though I can't be certain about it.

Here are a couple of screenshots from the Hybrid Assistant app, showing the MG1 RPM (4000 +) vs Engine RPM (0), for the crank attempts. I could hear MG1 (or something internal) rotating and I could see that the crank pulley was not turning:

 

Thanks to all for the info!

@MAX2 - the car was never in water, so possibly not a failure of the damper due to corrosion alone, but there was definitely an engine issue giving hesitation and jolting in the drivetrain prior to the complete breakdown. It sounds plausible that the damper springs have broken and the repeated impacts have then led to the input shaft breaking (none of this confirmed, as I haven't opened the transaxle). The wiring to the crank sensor had no visible damage, and the engine is definitely not rotating, so I think the crank sensor DTC is due to the lack of rotation rather than a sensor or wiring issue. The car does not reverse either - the indicated gear position stays in N when the shifter is moved to D or R and the car never gets to Ready (maybe the engine has to run at least once, after a battery disconnection? I don't know.)

@PriusCamper - indeed, in the UK at the time of writing in 2025 a used transaxle is about $200 equivalent on a popular auction site, plus another $75 or so for a damper. Unfortunately the labour cost, plus engine repairs would be more than the value of the car (working examples with 200k miles and fairly poor bodywork go for about $1500, with non-runners at $500 or less).

One other point may be of interest - I was sent a video showing an attempt to start the car after the recovery, following a recharge of the 12V battery. This shows the engine starting and immediately going to very high revs (I guess 4000+) for a few seconds, then stopping. On a subsequent attempt, the engine didn't rotate and there was the horrible rotating mechanical noise from inside the transaxle. When I first saw the car, there was never any engine rotation. I wonder if the throttle plate was sticking, or there is an air leak somewhere, causing the engine to surge (finally breaking the shaft, but without setting any engine codes). Is this a known engine fault? It's academic now because the car is effectively scrap, but I thought it might be of interest.