Cold start : cuts out after ~7 seconds. Did this once every few months, but now totally dead.

I posted a thread about an issue I had with my car recently. Got mini VCI today, also realised something important. The problem is quite different from what I thought, hence : I need to make this thread

Background : I bought this car (2003 Prius NWH11) 2019, with a battery fault. I built a battery with NWH20 modules (I had two spare NWH20 batteries, one of which was a bit dubious). The battery worked fine. I lost my Windows install with TechStream, and my OBD2 scanner worked with all cars except this one.

Since then, once every few months, something weird happened on cold start;

- I would start the car, but not immediately start moving. After 7-30 seconds, the engine would die, with a red triangle, yellow engine light,
- I needed to unplug the 12v battery before it would let me try to start again!
- On one occasion it stalled twice in a row like this! Because I didn’t start moving immediately after starting the second time.
- My OBD2 didn’t work, I didn’t have a Windows install, and I assumed it was due to battery being out of balance.

I hadn’t had any issues for months. Then, last week, I was going to the shops. The car had started fine, and drove fine. Halfway through the journey. The car seemed rough, like it was misfiring. I drove home carefully (home was about 5 miles away).

Once home, the car started fine, but was rough. I couldn’t seem to isolate the rough running to one cylinder (disconnecting coils one at a time made it much worse, despite the apparent ‘misfiring’.

Once the car cooled down, it wouldn’t start again. It runs fine for 4-7 seconds, but then starts ‘knocking’ very badly for several seconds, then cuts out. Just before stopping spinning completely, it often has one ‘ignition?’. It speeds up, enough for three more ‘knocks’, then stops spinning for good.

I got a new TechStream today, and pulled the codes, which are:

(Engine and ECT) ECU : Under ‘Codes’, it has the following string : “None.” Above where it would show the codes, it says ‘Triangle; DTC Monitors are N/A’ and also ‘MIL : N/A’
(HV Battery) ECU : None. (live monitoring shows no issues, and very well balanced!)
(Hybrid Control) ECU : P3101 Engine System (this is tagged for Current, History)

Of note : I disconnected the 12v battery last week - it would have been flat today if I hadn’t. This might be what the ‘DTC Monitors N/A’ is about? Maybe?

I realised over the weekend that the rare cold start issues seem related to what has happened now! But I ignored them as they were rare, I didn’t have Windows and one of the NWH20 batteries I used to rebuild this battery was a bit dodgy (some blocks with one cell shorted, for example!).

in retrospect, the rare cold start issue is probably key here, in retrospect, given the very good balance of the battery!

I would be grateful for any ideas where to go from here. I don’t have any pressure gauges, the hot fans worked well on the way home (I’ve checked for other signs of head gasket failure too); oil is normal and correct level. I did fill my fuel tank from 50% to 100% about 6 miles before the engine because ‘rough’. However that fill was also 3 days before the issue arose, so it did manage a cold start (the last ever cold start) using that particular fuel tank, and I can’t see any reports of bad batches of fuel so far.

One idea I had : the throttle control body? The issue is just like when I blew some fuses in my NWH20. The issue was the throttle body not working (blown fuse) and it seemed just like this!

Does the throttle body behave in a certain way when attempting a cold start? If so, I can watch (do a video recording) of the throttle opening during an attempted start, in case that gives any clues?

 

I had come across that TSB yesterday, but that thread gives a lot more info! Something interesting is : this car is a Japanese used import (not sure when it was imported exactly but will check once I find my documents) - however, I bought it from someone north of Auckland in New Zealand (a subtropical area), and drove it a few hundred kilometres south (where we only get a few weeks in winter with nights below freezing); in retrospect, the failed starts probably all happened on cold days (which again made me think a weak battery was the cause); we are going into late? Autumn currently also;

If I have the faulty ECU, I will just live with it (and use the 12v disconnect if it happens - an easy fix, but not something Toyota could suggest to most owners in the past ). It does seem like something new has happened on top of this, so today I plan to check the throttle plate (video camera during starting, and will shut car down after 8 seconds to try to save HV battery charge); will likely check MAF and Throttle Body first (and order MAF cleaner and will only use that!), remove any pooled oil just in case, and clean Throttle Body. Also, only some of my car tools are here . After that will check fuel pump (I really hope it isn’t that), and if it works, will check relay(s) etc;

In the longer term;
- the ‘fuel pump can only be replaced with an entire fuel tank assembly’ is a death knell for these cars, so if I end up replacing it, I will figure out how to use a different fuel pump. That will require a fair amount of research; I envisage the fuel level connectors will remain on the old pump, and the fuel lines will go to the new pump. I also envisage challenges if the existing fuel lines are ‘blocked’ by the old pump - as any fix I do will be done ‘properly’ for obvious reasons!
- once back on the road, I will figure out what was done to bypass the lack of catalytic converter - as there have never been any persistent codes, despite driving for years - I am wondering whether it is a patched ECU, or some kind of sensor simulator?

 

OK; I did some testing, and made some progress (a partial fix…)

First, I realised something about the largest relay box; the horn has a 10a fuse and a small black relay; all the important fuses are 10-30 amps, and all the vital relays are the small black ones!

“Horn-Test” : all fuses work, and all the black relays work. This is an easy way to do emergency roadside testing

MAF : removed; it looks clean, shiny. There are two recessed wires with small rectangles on the centre. Disturbing airflow near the MAF while the car is running results in the engine stalling.

Throttle body : fuse and relay checked as above. I can’t shift it when the car is ‘on’, but it shifts and returns freely when the car is off. Very slight dirt on it (will clean later). Small amount of clean engine oil below throttle body.

The engine still fails to start. If I go into reverse immediately after starting… it still cuts out. But : if I am moving forward and hit the accelerator at the right time, it doesn’t stall.

Codes : I cleared most codes with the 12v disconnect, before the successful startup. Interestingly : no triangle, and no check-engine-light, after starting the car successfully a few times.

Also, I recharged the battery via the engine, so that removed one worry.

Fuel pump test : presumably, must be working to some degree, as the car ‘works fine’ (to a point).

Exhaust : smells like there is half-burnt fuel in there (no catalytic converter)

Engine : runs incredibly roughly, as if on 3 cylinders. After that first start, it starts ‘fine’ when using the ignition key (except, 3 cylinders)

Summary :
- ‘something is wrong that doesn’t throw any codes, and makes it run on three cylinders’.
- ‘whatever is wrong happened out of the blue, and once it happened, the car has continued to run roughly’
- ‘I have the original faulty ECU software, as evidenced by occasional winter failure to start.’
- and ‘the faulty ECU software also made a cold start almost impossible once (whatever it is that failed) failed a few days ago’

So, I think the ECU fault exacerbated the (unknown other fault) turning it from ‘running rough’ to ‘almost total failure to even start’.

I’m still not sure what is wrong, but will do a short road test after posting this (driving up and down the road near home, keeping enough battery to get home if needed). I will use the hot fans as a very basic way to monitor undiagnosed head gasket failure, so I can stop the engine before overheating occurs.

So, my plans for the next day or two:
- road test now (gently)
- obtain new spark plugs and tool tomorrow, and fit new plugs. I really hope it is plug. Or maybe a fuel injector?

- I still worry that something is wrong with the VVTi sprocket or chain. I thought maybe the locking pin had failed … then discovered the Prius doesn’t have that particular locking pin in the sprocket (which is also interesting, because when that pin stops working in non-Prius cars, it makes the sprocket rattle around, and shortly after shear off completely, destroying the engine.

- The car sort of sounds like it has thrown a rod when idling (although is fine for the MG1 assisted phase).
- In park, flooring the accelerator results in a slow increase in RPM (as expected), and the engine becomes smooth, Under load (charging the battery) it is still rough.

 

repeat deleted

 

rm -rf deleted repeat post

 

repeat delete

 

Delete

 

delete again

 

Nearly 7! I saved the post and came back to post it now! (PriusChat went down - I couldn’t access it for a while, using 3 different ISPs, and 3 OSes : https://www.priuschat.com wouldn’t even open at all (but must have been receiving posts still!)

I did a cautious ‘road test’ (half a mile);

The issue appears to be in the ignition system, as there is a slight unburnt petrol smell, and also, under heavy load (very brief testing, parked), there is also intermittent detonation, presumably of the cylinder that isn’t burning the fuel;

I will get a new set of plugs and spark plug tool in the morning; it needs plugs anyway, and there may be a clue on the old plugs. Hopefully it is that, but the sudden onset during gentle driving may suggest a coil randomly failing, rather than a plug randomly failing, but I will have to wait and see. I bet the coils are not the same as NWH20 ones :/ but will cross that bridge when I come to it (I could steal an NWH20 coil if they were the same - so will look that up if new plugs doesn’t fix this!)

 

I will do (most of my tools are about 100 mile away … will correct that when fixed - I have a compression tester among other things, so will grab those tools once my car is driveable again (hopefully compression is OK - the detonation I noticed when testing the car parked suggests there must be at least some compression on the bad cylinder)