Intermittent Starting Issue - 2003 Rav4 Gx 5Dr 2.0 Vvti

[LeeADAB]

Ok so I have a 12v feed going through the IG2 fuse (inside the car) and leaves that fuseboard with 12v on the BLK/ORG wire.

That goes into the Engine room board and to the IG2 relay Pin 1 where there is 12v so that wire is good. :)

Point A also has 12v as that uses the same wire as IG2. If I test the voltage across A and B I get 4.76v.

I presume that the ECU changes the GRN/RED wire C1 from a ground to a low power circut (4.76V) which stops the fuel pump from being on constantly. Only when the ecu sends a ground signal will complete the circuit and 12v flows through pins 1 and 2 closing the relay on pins 3 and 5 which then supplies 12v to the fuel pump??

So what could be telling the ECU not to change it's status and run the fuel pump.

Any ideas anyone??

Cheers,
Lee

[LeeADAB]

Would oil leak out if I remove either the Crankshaft/Camshaft Position sensors to inspect them?

I did have the Rocker cover gasket replaced a few years ago as oil was leaking into one of the spark plug chambers. Could oil have fouled either of those sensors which in turn doesn't send a signal to the ECU.

I am nearing all I can test and still no fuel pump priming. Pants!!

I don't like to admit defeat!!

[Guest]

This is one of the best blogs on the t'internet, keep up the good work Lee, we cant help you but we are offering our support using empathy :)

Gus

[anchorman]

If you bypass the relay and run the fuel pump does it all run OK?

I have sent you an email.

[LeeADAB]

Bypassing the relay made the fuel pump run fine and it started fine. I did smell a petrol smell from the wheel arch area below the gas cap.

It's booked in for Monday morning so I still have time to fix this!! Time is ticking!!

[Big Kev]

This is one of the best blogs on the t'internet, keep up the good work Lee, we cant help you but we are offering our support using empathy :)Gus

Spot on Gus.......if Forrest Gump had a Rav, we would all line the streets and applaud his journey.

Why garage, Lee.......ffs min......you could teach THEM surely.....?

[LeeADAB]

Cheers GUS and KEV for the moral support :) - I am nearly a "Toyota Armchair Trained Tecnician" after all this investigation.
I am running thin on ideas and tests now so will hopefully report back with the problem resolved. Fingers Crossed!!
The only other thing I can think of and will test this weekend is all the ground connections in the engine area and around the dash/ECM areas. (now I have a wiring diagram of all the grounds!!)
If the ECM isn't getting an "i'm good to go" signal from a component (ie the Crankshaft) then maybe that is why the ecm isn't giving out a ground to complete the c/opn relay.
I really hope it's not the ECM and to be honest I would have though that more would have gone wrong if the ECM had a fault.

Well it's been a steep learning curve if nothing else. I still have the weekend to find the fault so any more suggestions welcome.

[anchorman]

Did you get my email Lee?

[shcm]

Referring to your previous diagram.

Assuming the fuel pump relay is a 12V relay (and I can't see why it shouldn't be), and you're reporting a voltage across A and B of about 5V, that means point B is sitting @ about 7V (12 minus 5) with respect to ground (chassis).

To run the fuel pump, point B should almost certainly be at ground. The relay coil is probably not getting enough current through it to pull the relay in every time.

If you can, take a piece of wire from point B and connect it to ground. It should do no harm, because the ECU relay coil driver transistor is almost certainly an open drain or open collector.

If the fuel pump runs, using the wire eliminates the relay as being the problem.

If you don't fancy doing the above do this below first then:

That leaves either a high resistance harness from point B to the ECU connector or an iffy high resistance main ground lead on the ECU itself. Although the latter I would expect to cause a whole heap of other problems.

If you're really unlucky, it's the fuel pump relay coil drive circuit in the ECU itself.

If you can find the your Green/red (or white) wire at the ECU end, disconnect the connector and measure the green/red wire resistance with a meter (on about a 200 ohm range) from point B to the ECU connector end. You'll probably need to extend one of the meter leads with a piece of wire. It should read very low resistance.

Alternatively, it may just be a grotty/dirty connector at the ECU end.

Oh, 12V never flows. It's current that flows not voltage! ;)

[Big Kev]

Ah was just aboot to click on that wee "like this" button at shcm's post, but that implies I unnerstaun WTF it is all about.

Why is there no "don't like this" button......?

[LeeADAB]

Just got home - Thanks for the PDF Don. I have printed it for bedtime reading!!

Thanks for that Shcm, I will have a look tomorrow at those points. Can the connector to ecu be unplugged without disconnecting Battery? Cheers. Will let you know.

[LeeADAB]

The green/red wire has very low resistance which is great when I found the end inside the car where it enters a white connection block. That then changes from green/red to a white cable. All the other cables in that bundle also change to white apart from one/two pale blue.

I cannot work out where it goes but it doesn't seem to go into the ECM as the diagram suggests. When I replaced the factory double din stereo YEARS ago there was a couple of connector blocks that had mainly white wires in it and a Blk/red, Blk/wht, blu/yellow.

These were never reconnected as I thought they would just be to do with the computer side of the double din unit. Speed info...temp etc.

Hoovies post in 2008 which I read today maybe of use. Should those connectors be connected to something and would that being not connected be a problem?

http://www.toyotaownersclub.com/forums/topic/86114-radio-connections-on-rav42/

[LeeADAB]

Shcm, do you think the connections should be terminated? There is an "ignition pulse" wire thanks again to Hoovies updated wiring diagram.

Maybe that connector was grounding the white wires through the radio originally.

If that should have been connected then maybe I haven't heard the fuel pump prime for years?

[LeeADAB]

Reminded me of Kev's fault.

Just check everything is OK with the rear connectors. There's actually two "bits" in one box with some of these units and they communicate through connections on the rear connectors (iebus). Bizzare, but true. If the bus is not connected through the rear, you can get similar problems.

You've probably already checked. Probably not that, but worth a go.

http://www.toyotaownersclub.com/forums/topic/134009-radio-58812-problem/

[shcm]

I think you're barking up the wrong tree with that one. Don't worry, it won't be anything to do with headunit.

I probably need to draw some diagrams to try to explain what it could possibly be, so just bear with me a little while.

[shcm]

I don't have wiring diagrams for the 1AZ-FE from before 07/2003. Let's hope they are the same.

You said you'd measured 4.76V across the fuel pump relay coil ( Across points A & B ) If that's correct, I'm 99% certain that is a fault. The driver transistor in the ECU will almost certainly be set up to act like a switch. See the diagram above.

When the fuel pump is supposed to be on, there should be about 12V across the relay coil (across points A & B ). Or put it another way - point B, measured from ground, should be about 0V. (It will probably be a little higher by a fraction of a volt).

When the pump is supposed to be off, there should be 0V across the the relay coil (no current flows, so A and B should be at the same potential). Or, measured with reference to ground, both point A and B should be at 12V (assuming that the 12V supply (battery) is still being applied to point A. It's supplied via the ignition switch).

As it stands, there is barely enough voltage across the fuel pump relay coil, to give sufficient current to pull the relay switch closed. It's probably just about doing it. Once the switch is closed, most relays require less current to hold the switch closed. However, once the engine starts to crank, the Battery voltage will dip. This may mean the relay switch is fluttering slightly, just barely held closed, with a relatively high switch resistance, so pump motor current would be affected. This would be worse with a cold engine, when the voltage cranking waveform would dip lower than with a warm engine.

The relay control goes to pin 3 on ECU connector E7. Check that the harness from there to 1C ( point B ) is low resistance. Also check that the ECU earths like E01 and E02 E3 pins 21 and 31 are low resistance back to Battery negative.

Something is forcing the voltage on point B up, when it should be low impedance. It may be a faulty ECU, it may be quite a simple harness problem. It's probably quite simple, but much harder to diagnose via posts on a forum.

[anchorman]

Whats to stop him working it with a switch just to get going and test the starting? The excess fuel will just run back to the tank like it does all the time the engine is running.

[LeeADAB]

Quick update as in middle of testing...

Point B to pin 3 at ecu is a perfect connection.

Now I cannot test the two earth pins on 21 & 31 as all my security torx bits are 6 star and not the unusual 5point. So all I can now check is the ground on intake side of cylinder block.

The two ground wires from 21&31 are looking "wavy" and feel bumpy (very technical description) is that an odd sign?

[shcm]

Nothing stopping using a switch. Which is why I was suggesting pulling the relay coil ( point B ) to ground in an earlier post, with a piece of wire.

If the voltage @ E7 pin3 really is that high, I'm surprised the ECU isn't seeing it as a short circuit detect and shutting down the drive. Perhaps it doesn't have that capability.

EDIT: humour me. Remove the fuel pump relay. Turn the ignition switch to position 2, so that everything is powered up, but don't try to start (well, it won't anyway). Measure the voltage between point B (on the relay socket) and ground (chassis). Report back please.

While the relay is out, can you take the relay and measure its coil resistance, ( meter between the pins on the relay itself, that would plug into points A & B ).

[LeeADAB]

I cannot get to the other end of the ground wires on the cylinder block. I can see them and just about feel the bolt they are attached to. Will have to leave that to tomorrow.

I haven't grounded the relay by switch as I have tested the relays with a 12v Battery and also swapped the relays round. The 12v Battery made the relay click and continuity was achieved on the respective pins. That's when I jumped the pins to the fuel pump to simulate the relay working and it runs fine. If I ground the relay from point b it will just run the fuel pump continuously while the car is turned on.

The car did run previously without the c/open relay removed but ran rough as if it was going to die. That was part of the procedure for de-pressurising the fuel tank.

Will test the voltages as requested and the resistance of the relay. Unless the are all bad then I don't think relay. I have tried quite a few while swapping them round.

Thanks again,

Lee

[LeeADAB]

Relay resistance across coil is 108.8 and the others are of similar resistance.

With the jumper attached I hear that the fuel pump runs as soon as I insert the key.

I turned on the car with no c/opn relay and it ran for a few seconds then cut off due to no fuel. It then wouldn't restart and needed the jumper attached to pump the fuel through. Once primed she fired up perfectly.

The voltage is nil at point B when key is inserted and when turned to second position.

[shcm]

Relay resistance across coil is 108.8 and the others are of similar resistance.

OK thanks.

The voltage is nil at point B when key is inserted and when turned to second position.

Good in one sense, It's what I was expecting. At least nothing else is holding that point up. Sorry, but based on your previous statements & measurements, it's beginning to sound like the FPR coil driver stage in the ECU is giving notice.

Could double check, by measuring the voltage at point B with respect to ground, with the relay back in and engine running. Based on figures from your previous posts, I'm expecting it to be around 7V-ish. I really think it should be 0V-ish.

[LeeADAB]

Ok more results......

Tested voltage from inside the car from the end of the green/red wire (point B ) on relay. Relay connected as normal.

Ignition OFF = 0v

Key in but not turned = 0.01v ~ 0.02v

Key in ACC position = 0v

Key in ON position = 12.22v

Engine started and running = 0.05v

[shcm]

Good, that's as it should be.

Point A also has 12v as that uses the same wire as IG2. If I test the voltage across A and B I get 4.76v.

Above statement must have been a measurement error. You can't have both conditions being true.

(The above measurements and the last post measurements).

Unless the ECU FPR coil drive is intermittent.

Did the engine start normally without the problem?

[LeeADAB]

Engine was started normally. The 4.76v was with the engine running and no relay inserted tested at the relay end. I'm sure that is the case still.

There is no two second fuel pump priming noise ever at the moment.