1998 Rav4 starts then immediately shuts off

[yt1300]

My Rav4 has developed a problem where it starts, then immediately shuts off. Once it's done this, if I try cranking it again without turning the key back to the fully off position, it turns over, but never tries to fire. However, if I turn the key back to the off position, then try again, it will start just fine, but still immediately shut off. It also won't start even if I give it a little gas while trying to start it (unlike some other descriptions of similar problems that I have read). Also, when it starts, it sounds like a good clean start, it just won't idle afterwords.

It did this once about a year ago, exactly like what I am describing, then half an hour later, it started fine and this hasn't happened again until a few days ago. When it happened again, it wouldn't start, then a a while later is started just fine. It also started fine the next few times I tried to start it, but it would occasionally do this (stop working, then magically start working later). Now it does it every time and I cannot get the engine to stay on. So this is something that progressively went out over the last week or so.

Someone on Rav4World suggested that it might be the ignition switch and very kindly told me the pattern of continuity that I should see if the switch worked vs if it was bad (see attached picture). There does seem to be an issue with my switch, because I have no continuity between 2-3 when the switch is in the Acc or On position (and I should). However, that does not appear to be the source of the current problem. The correct continuity pattern for the car when running should be 2-3, 2-4, and 6-7 (see attached figure). The engine starts via a connection from 7-8.

Therefore, to bypass the switch, I took the female end  of the plug that connects to the ignition switch, unhooked from the switch, and ran a wire from 2-3, from 2-4, and from 6-7 (thus creating the connectivity for the On position). Then, I hooked the battery back up and quickly touched a wire from 7-8. The car started, but it immediately shut back off, just like before. Further, one of the oddities about this problem is that once the engine shuts off, I have to turn they key back to the off position before it will fire again (otherwise it just keeps turning over but never firing). So, while I had the wires rigged up I played with this. As expected, if I connect 7-8 getting it to fire and stall, then immediately reconnect 7-8, it turns over, but doesn't fire (just like trying to turn the key again without first going to the off position). I tried to figure out which circuit needed to be broken for it to reset and let the engine fire again, and its the 6-7 connection. If I start/stall it then unplug and replug 2-3 or 2-4 and try it again, it turns over, but won't fire. However, if I unplug and replug 6-7 and try again, it starts then quits (just like if I had turned the key to off, then tried again).

This makes me think that whatever is wrong is on the circuit for 6-7. Does anyone know what is on that circuit that could cause this?

Just for reference, here are some other things I've tried:
Checked spark plugs and wires (they were replaced fairly recently, and all seem to be in good shape, correctly gauged, etc.)
Checked for vacuum leaks (I did this one of the times that I got it running).
Cleaned the throttle body and idle air control valve
Replaced the fuel filter
Tried starting it with the MAP sensor disconnected (98's don't have an MAF)
Tried starting it with the green coolant temperature sensor disconnected (it really didn't like that at all).

Note: this is a cross post from Rav4World, and everything there has been and is continuing to be very helpful, but so far no solutions have emerged, so I thought I'd expand the collective brain pool a bit.

Thanks

 

 

I'm at my wits end here, so any help would be appreciated. Thanks.

[trentmeyer23]

I had this exact problem(key turns and all) with a Celica. It was the ignitor not sending a feedback signal to the ECU. Grabbing one from a wreckers will save you some money, if it is the case.

Have you done the diagnostic code check? Code 14 is the one I think it will be if it shows.

You need a short piece of wire, strip both ends, open the diagnostic cover(possibly located near the strut tower or alternator, in the engine bay) and 'bridge' terminals TE1 and E1. Then turn on the ignition and watch the CEL(check engine light- on the dash), it will flash, if it is constant then you have no fault codes if not then count the flashes, there will be a pause and it will start again, keep counting. The idea is if the first sequence flashes once (for example) then write down one, then after the pause if it flashes four times then the code is 14.

[yt1300]

Hi, thanks for the suggestion. I don't have a code reader so I haven't been able to check it's codes. Is there any other way to diagnose it? (the stores in my town really gouge you on things like that, but I don't want to wait a week or more to get one from online).

Also, the igniter is the same thing as an ignition module, right? I had one of those go out in a previous car I owned, but when that happened, it would crank all day, but never generate any spark. I suppose they probably don't all die in exactly the same way though. For yours, was it the igniter itself that was a problem or a relay or something else in between it and the computer? Thanks again.

[trentmeyer23]

If you read my post again, the information is all there.

You do not need a code reader.

This is an ignitor.

[yt1300]

Sorry for the confusion, I misunderstood what you were saying (I've never worked with Toyota electric systems before).

I checked the code, and the only one I get is 99, which as far as I can tell, means that the immobilizer is not talking to the computer (I didn't know this car had an immobilizer). I'm a bit confused by that though, because I was under the impression that when the immobilizer wasn't communicating, the car simply wouldn't fire, whereas mine fires then dies. Is it possible that this code is irrelevant to the actual problem?

My wife has a second key for this car, but she is away for several days, so once she gets back I will try her key and see if the car will start with that key.

[yt1300]

I should also mention that I read that the car should be at normal operating temperature for this test, but since I can't get it to run, I obviously couldn't do that (it is about a million degrees where I live so the engine isn't truly cold, it's just not at operating temps). Is that a problem?

[trentmeyer23]

I could definitely be immobiliser related, given the cut to the power at the ignition switch.

Disconnect the battery for a few minutes and then try again. See if you have the same issue/code.

[yt1300]

I let the car sit for a few hours with the battery unhooked, then tried starting it, and it started right up. I shut it down and tried again, and it was still starting 5 tries later (I stopped after 5 tries). I imagine that it isn't fixed, but what would magically start working again after leaving the battery unhooked for an afternoon?

Also, while it was running, it sounded absolutely terrible. The engine was making all sorts of terrifying clunking sounds, and the idle RMPs very gradually increased while it was on (I only ran it for about 60 seconds before shutting it back down). I'm guessing that I screwed something else up in my attempts to fix the main problem, but I thought I'd share in case that info is somehow useful (my first guess is that I messed something up when cleaning the IAC).
 

[campbeam]

8 hours ago, yt1300 said:

what would magically start working again after leaving the battery unhooked for an afternoon?

This is the "magic" to reset error codes relating to the electronic engine management system.  

[trentmeyer23]

The higher RPMs are probably due to the ECU recalibrating.

[yt1300]

That makes sense I suppose, but what does this mean as far as what is actually wrong with my car?

[yt1300]

Update: after letting it sit all night with the battery connected, I checked the codes this morning, and I got the continual flashing that indicates a lack of codes. Then I tried to start it and, unsurprisingly, it started then crapped out again. I checked the codes again, and as expected, code 99 is back.

I'm guessing that is a pretty clear sign that it is the immobilizer, but I'd appreciate second opinions. Also, is there anything that I can do except wait for my wife to return and hope that her key works (indicating that my key is bad, rather than the ecu being bad)?

Thanks again everyone.

[yt1300]

My wife came back today so I tried her key. Unfortunately her key won't work other.

Also, I tried to replicate being able to start it after leaving the battery unhooked for several hours, but this time it did not work. I also realized that when it worked before, I not only left it without a battery for hours, but I also changed the fuel filter. Technically, this was changing it back to the original, because the aftermarket wouldn't stop leaking, and I remembered that when I put the aftermarket on in the first place, it started after that as well.

So something about changing the fuel filter seems to reset it, not sure why. Both times, to depressurize the line, I unplugged the pump and tried to start it once before unscrewing the filter. So I tried just unhooking the pump, starting it once, then hooking the pump back up and trying again, but it did its usually start then stop.

I'm really stumped.

[yt1300]

A mechanic who is a friend of a friend sent me this flow chart and suggested that it might be the transponder key coil (the thing that actually reads the key).

The flow chart seems off to me, because it says that if two keys both fail, then the key is the problem, but surely that is backwards, and two keys failing indicate that they keys aren't the problem, right? Or am I missing something?

I'll put the center panels off tomorrow to get access to the ECU and check the grounding, but it sounds like I may be seriously screwed.
 

[yt1300]

I thought you all might like to know the final conclusion of my saga. It was in fact the immobilizer unit. It cost me a pretty penny to get a replacement but the RAV4 starts and runs fine now. The engine noise was just that all the stop/stalls had drained the oil out of the top end of the engine. Once it ran for a bit, warmed up, and got the oil circulating again, the noise completely went away.

Thanks again for all the advice.
 

[Bettz]

Hi Donald.  I have a rav4 1998 and im having the same exact problem as your vehicle,  Cranking and starting nut immediately dies.  Cant get the damn thing going, even had an auto electrician over to look at it and check everything which all seems fine.  I get codes 21 and 99.  i just want to pick your brains on your vehicle. i know it was many moons ago but you may still be able to shed some light.  Or anyone thats had the same issue. Model is SXA11R, Engine 3S-FE. I dont know if this has an immobilizer in it.  Cheers

Edited March 3, 2020 by Bettz

[deanosse]

Mine was Immobilizer (1997)

got a secondhand one and local locksmith cut 2 new keys and programed it was around $200 + $80 for immobilizer 

[Manmodu]

On 3/3/2020 at 12:37 PM, Bettz said:

Hi Donald.  I have a rav4 1998 and im having the same exact problem as your vehicle,  Cranking and starting nut immediately dies.  Cant get the damn thing going, even had an auto electrician over to look at it and check everything which all seems fine.  I get codes 21 and 99.  i just want to pick your brains on your vehicle. i know it was many moons ago but you may still be able to shed some light.  Or anyone thats had the same issue. Model is SXA11R, Engine 3S-FE. I dont know if this has an immobilizer in it.  Cheers

Hi Bettz, I'm in a very similar situation to you. I have exactly the same year RAV4, Car cranks but won't start. It's showing the codes 21 and 99. Some research on Google shows that 21- relates to O2 sensor and 99 - relates to immobiliser. 

I have gotten a new starter as the car didn't crank before due to starter failure.

I have checked that sparks are happening.

I have a new battery.

I put some fuel into air intake and it seemed to start but was very rough and died quickly.

I have ordered an O2 sensor and hope that it will clear the code 21.

Any updates on how to clear the code 99?

On 3/3/2020 at 12:37 PM, Bettz said:

 

I should mention that the inside of the car had flooded to the level of the pedals, also the rad fans switch on when I turn the key to 'ON'. I have removed the interior panels, seats and carpets for better access and cleaning. 

Any info on how to fix this code would be appreciated. 

 

blockquote widgetblockquote widgetblockquote widget

Edited January 28, 2022 by Manmodu