IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
> Backwards step... fuel pump doesn't get power. Tips / cross-checking needed!
donutty
post 10th February 2019 18:19
Post #1 | Print


1.4 16v
**

Group: Regular
Posts: 106
Joined: 27th November 2018
Local Time: 23rd July 2019 10:17
Member No.: 13,545
Driving: 145 2.0 '96 Cloverleaf
From: Sheffield, UK



Hi all

Executive summary before the rambling: CODE OK, Inertial fuel cut-off OK, Fuses OK, Relays OK, Camshaft sensor pulses OK, RPM (crankshaft position) sensor OUT OF SPEC



Refitting some internal trim in strong wind with regularly needing to have the driver's side door open, I decided to move the 145 to a more sheltered location. Turn the key. Turning over but no start. Weird.. it was only put here 2 days ago. Fuel was on very low.. must be outta juice. Quick drop from a jerry can, still nothing. Take off fuel pipe to the injector rail, nothing coming out, not priming in key position 1 (0 = off, 2 = start) or when turning over (pos 2)

New fuel pump only installed a month or so ago... has it failed already?! No - multimeter reveals nothing (bar a few millivolts) getting to the pump power connector on top of the tank.

The inertial cut-off switch has been bridged, and there is continuity between the fuel pump connector and chassis ground, so it's not the pesky switch.

All fuses check OK, relays on the BBoB rotated (surely one of them must be good?!). Fuse and relay sockets treated with my expensive 20 quid a can 'if all else fails' imported electrical cleaner (DeOxit).

Is it the key (ALFA CODE) ? Key used was a newly minted clone of the master... and starting is no better with the master. Code light goes out = all OK, right?

Probing the brown/green wire of S52 Camshaft sensor during turn-over shows the expected pulses (ignition off = flat-line, key pos 1 = flat line with a bit of ripple, ignition pos = line goes high then after around 1 sec of turning over dips to 0 for about 1/10th of the cycle then goes high again, and this repeats), so I think this is OK.

[side note: found there is a typo in the service manual where it is describing the cam angle sensor "..supplied at 12V at pin 1.." it is actually 5v (what I measured) and also the description of the ECU says pin 12 (where this goes to) is 5V regulated supply for sensors, contradicting the prior blurb]

Next to be checked is the S31 RPM and Crankshaft Position sensor. The troubleshooting / spec at the back of the Motronic manual says the resistance between pins 1 and 2 should be 486 - 594 ohms. I measure 844 ohms on two different multimeters.

So I all I have to go with is this - the crank / RPM sensor is out of spec due to fault, damp? or damaged wire.



What do those present think?
Anything I missed?
Experienced this issue yourself?

Why does it not do a prime pump..?? This is the thing that bothers me. It primes before even turning over so is it just because the ECU isn't getting a sensible signal even when not turning over, because of the too high resistance, it thinks 'no point even priming' .. But why doesn't stuff like this turn on an engine fault light on the dash??


ANY help appreciated, as the 145 is stuck outside in this nasty weather!! blink.gif Not going to be ready for any 145 meetups at this rate angry.gif


Thanks !!!

This post has been edited by donutty: 11th February 2019 08:10
Go to the top of the page
View Vehicles
+Quote Post
JeremyG
post 10th February 2019 22:53
Post #2 | Print


1.6 8v
***

Group: Regular
Posts: 503
Joined: 19th August 2016
Local Time: 23rd July 2019 10:17
Member No.: 12,797
Driving: 166 3.0 Sportronic, GT 2.0 JTS
From: Oxford



Have you checked the inertial fuel cutoff switch? That caught me out a few weeks ago... it cuts the earth return from the fuel pump, so if you were checking the voltage across the power lead supplying the pump then youd see 0V if the switch has tripped.

The switch is under the passenger seat...
Go to the top of the page
View Vehicles
+Quote Post
donutty
post 11th February 2019 08:08
Post #3 | Print


1.4 16v
**

Group: Regular
Posts: 106
Joined: 27th November 2018
Local Time: 23rd July 2019 10:17
Member No.: 13,545
Driving: 145 2.0 '96 Cloverleaf
From: Sheffield, UK



QUOTE(JeremyG @ 10th February 2019 22:53) *
Have you checked the inertial fuel cutoff switch? That caught me out a few weeks ago... it cuts the earth return from the fuel pump, so if you were checking the voltage across the power lead supplying the pump then you’d see 0V if the switch has tripped.

The switch is under the passenger seat...



Yep, but I forgot to put that in my original post. Since getting the car I have bridged the connector to rule out the switch while the car is being done up, to avoid any accidental switch tripping head scratching moments.

I checked the earth pin of the fuel pump connector to the chassis and there is continuity, so the wire bridge is working.

This post has been edited by donutty: 11th February 2019 08:11
Go to the top of the page
View Vehicles
+Quote Post
donutty
post 11th February 2019 09:11
Post #4 | Print


1.4 16v
**

Group: Regular
Posts: 106
Joined: 27th November 2018
Local Time: 23rd July 2019 10:17
Member No.: 13,545
Driving: 145 2.0 '96 Cloverleaf
From: Sheffield, UK



I've done some research on the crankshaft sensor, and a lot of places are saying resistance is 860 ohms, which is roughly what I am measuring.
Whereas the Alfa Motronic M2.10.3 (55-36A) spec says 486 - 594 ohms.

Can somebody measure their '96 Cloverleaf's crank sensor resistance to see what ball park we are in ?? rolleyes.gif whistle.gif


OK... bit more digging reveals this goldmine:

http://forum.alfa145.com/index.php?showtopic=10113


My VIN ZAR93000002105384 = 10 / 1996 GD Chassis No.: 02105384 Vehicle No.: 03100440 Engine No.: 0362524

ePER says Engine Ignition Systen --> 11. Sending unit is 46411135 (was 60811200 until 10/1994)


So maybe my sensor is OK, but it is out of alignment or there is something interfering with the pickup on the crankshaft teeth ?

This post has been edited by donutty: 11th February 2019 09:40
Go to the top of the page
View Vehicles
+Quote Post
dante giacosa
post 11th February 2019 16:26
Post #5 | Print


1.6 16v
****

Group: Veteran
Posts: 1,049
Joined: 28th March 2017
Local Time: 23rd July 2019 10:17
Member No.: 13,068
Driving: 146 CF2 1.8 T-spark
From: Edinburgh



are you going to remove it then, for a cleanup donutty..?
Go to the top of the page
View Vehicles
+Quote Post
JeremyG
post 12th February 2019 02:08
Post #6 | Print


1.6 8v
***

Group: Regular
Posts: 503
Joined: 19th August 2016
Local Time: 23rd July 2019 10:17
Member No.: 12,797
Driving: 166 3.0 Sportronic, GT 2.0 JTS
From: Oxford



Given your CODE light is working correctly we can assume your engine ECU is working. However - have you checked for any engine ECU errors using MultiECUScan or the like?

(As an aside - are you running Motronic 2.10.3 or 2.10.4? )

In any case, if youre not getting power to the fuel pump, Id go back and double-check the relays on the BBoB - bearing in mind that the power supply to the pump is reliant on a chain of two relays (the main relay, S41, which feeds S12a, the pump relay) - so you may have been unlucky when rotating the relays by keeping a bad relay in the chain.

Look at the Motronic wiring diagrams in the workshop manual section here - you need diagram B at the beginning of the doc (note that Motronic 2.10.3 and 2.10.4 are slightly different). Chase the wiring through to relay S41 then S12a (both in the BBoB - you can only reliably identify them by the wiring colours) and test each relay.

In the absence of any ECU errors, I reckon this is probably where your problem lies. Good luck!

(I went through all this last year trying to resolve a starting issue on my old Spider... also running Motronic 2.10.4. In my case, the ECU had died...)

This post has been edited by JeremyG: 12th February 2019 02:17
Go to the top of the page
View Vehicles
+Quote Post
donutty
post 12th February 2019 11:12
Post #7 | Print


1.4 16v
**

Group: Regular
Posts: 106
Joined: 27th November 2018
Local Time: 23rd July 2019 10:17
Member No.: 13,545
Driving: 145 2.0 '96 Cloverleaf
From: Sheffield, UK



I'll not get chance until the weekend at least, but I will try to remove, clean and re-seat the RPM sensor, ensuring the correct distance to the 'phonic wheel'.

I will also check the relays again for function, and the voltage / continuity at the sockets in BBoB.
Also I will check the fault codes in the ECU (have a pro diag tool with Alfa connector). Its a 2.10.4 ECU. No engine warning lights on the dashboard though...

Could somebody please remind me if I am correct or not.. the Alfa CODE only stops the spark plugs firing if it isn't happy, right? (or should I say, it tells the ECU it isn't happy and the ECU's response is not firing the plugs). The ECU's reaction is not to kill the spark AND the fuel?

The fuel cutoff signalled by the ECU would be for safety (in conjunction with the inertial switch), fuel not required since the engine is not turning, or fuel not required because it thinks that there is enough at the injectors already... ??


Wouldn't it still command a fuel priming squirt, even if the engine is not being turned over though? I thought that it did the priming at key position 1 (ignition on, but not going for start), and since the engine isn't being turned over, how does it know that the cam and crankshafts are not turning, and therefore why bother injecting fuel? Unless the signal from one of these is so out of whack that it thinks 'why bother with fuel, this thing is never going to get started' ??? But then why no warning light??

I know its probably very simple, but there are dozens of things that all need to be synchronised exactly to get the engine running. blink.gif Minefield!


Thanks for your hand holding so far. Don't let go !! blush.gif

This post has been edited by donutty: 12th February 2019 11:15
Attached thumbnail(s)
Attached Image
Attached Image
 
Go to the top of the page
View Vehicles
+Quote Post
dante giacosa
post 12th February 2019 13:50
Post #8 | Print


1.6 16v
****

Group: Veteran
Posts: 1,049
Joined: 28th March 2017
Local Time: 23rd July 2019 10:17
Member No.: 13,068
Driving: 146 CF2 1.8 T-spark
From: Edinburgh



"there are dozens of things that all need to be synchronised exactly to get the engine running"

not quite true

actually- the ONLY two things that will prevent the engine from firing & running (even badly) are the crank-sensor and the fuel pump

Go to the top of the page
View Vehicles
+Quote Post
donutty
post 12th February 2019 14:23
Post #9 | Print


1.4 16v
**

Group: Regular
Posts: 106
Joined: 27th November 2018
Local Time: 23rd July 2019 10:17
Member No.: 13,545
Driving: 145 2.0 '96 Cloverleaf
From: Sheffield, UK



QUOTE(dante giacosa @ 12th February 2019 13:50) *
"there are dozens of things that all need to be synchronised exactly to get the engine running"

not quite true

actually- the ONLY two things that will prevent the engine from firing & running (even badly) are the crank-sensor and the fuel pump


I meant all the things that branch out from there, like the Code, relays etc. hopefully most of which I have ruled out unsure.gif ... I hope.
Maybe not synchronised then, but they have to work in some semblance of order.
Go to the top of the page
View Vehicles
+Quote Post
JeremyG
post 12th February 2019 16:46
Post #10 | Print


1.6 8v
***

Group: Regular
Posts: 503
Joined: 19th August 2016
Local Time: 23rd July 2019 10:17
Member No.: 12,797
Driving: 166 3.0 Sportronic, GT 2.0 JTS
From: Oxford



To get 12V to the fuel pump you need the ECU to provide an earth at pin 36 (which activates the main relay S41) and at pin 3 (which activates the fuel pump relay S12a - incorrectly labelled "variator relay" on the below diagram).

Attached Image


Now - does this mean the ECU will not provide the priming pulse to the fuel pump when you first turn the ignition on unless it's happy that there are no errors on the rest of the system? I don't know. But I'd check for error codes in your diagnostic tool before removing the crank sensor - and if there are none then I'd still suspect the relays S41 and S12a first.

... especially as they are much more accessible than the crank sensor...!

This post has been edited by JeremyG: 12th February 2019 17:25
Go to the top of the page
View Vehicles
+Quote Post
Ganz
post 16th February 2019 22:12
Post #11 | Print


1.6 16v
****

Group: Veteran
Posts: 1,047
Joined: 2nd November 2012
Local Time: 23rd July 2019 10:17
From: East Mids
Member No.: 10,640
Driving: 145 Cloverleaf 2.0L ts
From: uk



QUOTE(donutty @ 11th February 2019 09:11) *
I've done some research on the crankshaft sensor, and a lot of places are saying resistance is 860 ohms, which is roughly what I am measuring.
Whereas the Alfa Motronic M2.10.3 (55-36A) spec says 486 - 594 ohms.

Can somebody measure their '96 Cloverleaf's crank sensor resistance to see what ball park we are in ?? rolleyes.gif whistle.gif
OK... bit more digging reveals this goldmine:

http://forum.alfa145.com/index.php?showtopic=10113
My VIN ZAR93000002105384 = 10 / 1996 GD Chassis No.: 02105384 Vehicle No.: 03100440 Engine No.: 0362524

ePER says Engine Ignition Systen --> 11. Sending unit is 46411135 (was 60811200 until 10/1994)
So maybe my sensor is OK, but it is out of alignment or there is something interfering with the pickup on the crankshaft teeth ?


I'll have a look. Still got my old sensor and the one on the 147 that failed. The 145 did not fail but I thought it prudent to replace when I took my Bella off the road.


--------------------
G A N Z 145

Instagram
ganz145_alfa
Go to the top of the page
View Vehicles
+Quote Post
donutty
post 18th February 2019 09:31
Post #12 | Print


1.4 16v
**

Group: Regular
Posts: 106
Joined: 27th November 2018
Local Time: 23rd July 2019 10:17
Member No.: 13,545
Driving: 145 2.0 '96 Cloverleaf
From: Sheffield, UK



Everybody stand down! coolio.gif It's fixed. laugh.gif

Sensor OK
Relays OK
ECU OK
Code OK

**BODGE MERCHANT** wiring NOT OK


After probing the output of the fuel pump relay in the BBoB and seeing nearly 12v - showing that 3/4 of the electrical chain was working, I suspected the link between the engine bay and the fuel tank to be the crux of the problem.

Half an hour spent looking for the mythical G133b connector that interfaces the engine bay loom to the loom that goes to the rest of the car, and not finding it (the service manual seems to suggest it is somewhere near the forward bulkhead, and a search of this forum has another user suggesting on a Ph.2 it is near the throttle body - mine is Ph1 and it could not be found there), I started looking at all the connectors under the dash near the drop-down fuse board.

Then I found that because this is an Italian car, some spaghetti was also included (picture 1)

Now, this was not the work of the good people at the Pomigliano d'Arco factory... this is East Midlands bodging at it's finest.

The green wire (fuel pump live feed) has been cut and spliced to the purple wire (a thinner gauge and looks like it is well mangled up, with insulation punctured) which after further investigation goes to this bundle of fun hanging out near the clutch pedal (picture 2)

There is a blue crimp flag connector, and above that you can see a lone wire sticking up. That used to go to another crimp connector which was attached to the other side of a 7.5A fuse, the other side going to the still attached blue connector.

I had no idea that this circuit was to do with the fuel pump... it shouldn't be.... but it seems somebody took the live feed from the fuel pump wire to power some gizmo or whatnot (maybe a car phone, or separate amplifier ?). There is also evidence of a high-power ICE install (very high current live wire run along the sills to the boot) so I suspect one previous owner was an enthusiastic but not all that competent connoisseur of things that go thump.

But surely this meant that whatever thing they had wired in here only got power when the fuel pump was running... so on deceleration, or with the engine off, said gizmo would not be functioning? Very strange...

There is also a 12V relay and some other 20 pin rectangular connector, so who knows?!!

I put another connector on this straggling wire and put the fuse back in circuit and he fired up on the first turn of the key!!!
Brum brum again and a couple of pints were consumed to celebrate!!!


Oh, and of course I'm not going to just do a patch up job... this whole wiring shambles needs looking at and where safe and sane to do so, will get removed.

This post has been edited by donutty: 18th February 2019 09:39
Attached thumbnail(s)
Attached Image
Attached Image
 
Go to the top of the page
View Vehicles
+Quote Post
Ganz
post 18th February 2019 10:46
Post #13 | Print


1.6 16v
****

Group: Veteran
Posts: 1,047
Joined: 2nd November 2012
Local Time: 23rd July 2019 10:17
From: East Mids
Member No.: 10,640
Driving: 145 Cloverleaf 2.0L ts
From: uk



Wow. That's a real horror show! I know you don't need the info but I did carry out a check on my RPM sensors. Strangely my resistance values are different from the ECU diagram but they work. Looking at the manual and my last ECU diagnostics using Fiat Multiscan my ECU is a Motronic 10.3 but the RPM resister values I get are are from a Motronic 1.5.5. Don't know if this is an Italian thing..

Anyway my RPM values from my new sensor is around 900. Was 1109 ohms when I actually measured but the engine was still warm; the resistance value goes up when the RPM sensor gets warm. My old sensor (cold) is showing 864 which is actually fine. My 147 1.6TS RPM sensor is showing 990 ohms. All these values fall within the tolerance of the Motronic 1.5.5 specifications even though I have the 10.3 ECU. I've had no issues with the sensor. As Giallo said in an old thread the Bosch rpm sensors all seem to be 860 +/-10% Ohms.

When my 147 sensor failed the engine cut out and an ECU failure message came on the dash. when the sensors are on the way out they usually stop working when the engine gets hot. Once the engine had cooled a bit it would start up again. Replacement job was fairly easy to do.

This post has been edited by Ganz: 18th February 2019 11:01


--------------------
G A N Z 145

Instagram
ganz145_alfa
Go to the top of the page
View Vehicles
+Quote Post
dante giacosa
post 18th February 2019 13:34
Post #14 | Print


1.6 16v
****

Group: Veteran
Posts: 1,049
Joined: 28th March 2017
Local Time: 23rd July 2019 10:17
Member No.: 13,068
Driving: 146 CF2 1.8 T-spark
From: Edinburgh



eventually;

you see-it-all, don't you...?


glad to hear that's worked out!
Go to the top of the page
View Vehicles
+Quote Post
sanitary
post 18th February 2019 19:26
Post #15 | Print


1.4 16v
**

Group: Regular
Posts: 201
Joined: 12th December 2004
Local Time: 23rd July 2019 10:17
Member No.: 662
Driving: in front of you
From: Lardtown, 'ulll



20 pin connector looks like it is out of a computer power supply unsure.gif




san
Go to the top of the page
View Vehicles
+Quote Post
donutty
post 19th February 2019 12:55
Post #16 | Print


1.4 16v
**

Group: Regular
Posts: 106
Joined: 27th November 2018
Local Time: 23rd July 2019 10:17
Member No.: 13,545
Driving: 145 2.0 '96 Cloverleaf
From: Sheffield, UK



QUOTE(sanitary @ 18th February 2019 19:26) *
20 pin connector looks like it is out of a computer power supply unsure.gif
san



Yeah I know, an ATX power supply.

Where's the motherboard?? blink.gif

I'm guess it was for a phone or amplifier. There is evidence of just a little more than a stock radio having been fitted in the car's past...
Go to the top of the page
View Vehicles
+Quote Post
dante giacosa
post 19th February 2019 14:21
Post #17 | Print


1.6 16v
****

Group: Veteran
Posts: 1,049
Joined: 28th March 2017
Local Time: 23rd July 2019 10:17
Member No.: 13,068
Driving: 146 CF2 1.8 T-spark
From: Edinburgh



do you still have a working alarm system, and remote-locking donutty?
Go to the top of the page
View Vehicles
+Quote Post
JeremyG
post 20th February 2019 06:30
Post #18 | Print


1.6 8v
***

Group: Regular
Posts: 503
Joined: 19th August 2016
Local Time: 23rd July 2019 10:17
Member No.: 12,797
Driving: 166 3.0 Sportronic, GT 2.0 JTS
From: Oxford



Excellent - glad you found the problem.

G133a might be somewhere under the centre console - certainly this is the case with the Phase 1 Spider with the same ECU. Its a 6-way connector in black plastic.

(I have a photo of this somewhere - will post later this week!)

Go to the top of the page
View Vehicles
+Quote Post
donutty
post 20th February 2019 09:22
Post #19 | Print


1.4 16v
**

Group: Regular
Posts: 106
Joined: 27th November 2018
Local Time: 23rd July 2019 10:17
Member No.: 13,545
Driving: 145 2.0 '96 Cloverleaf
From: Sheffield, UK



Thankfully I don't think I need to find G133 a / b just yet, and it may have even been chopped off and became that mess that you see!!!
I'm going to do a proper tidy up job of that wiring but it will never be back to original... unsure.gif

No, there is no alarm or remote locking.

I de-installed an after-market alarm with ultrasonic sensors and a pin-pad in the cabin. There was a sounder in the boot and the engine bay.

I'm not even sure if there is any (non-remote) central locking working as the driver side door lock doesn't function properly.

Here's another picture of that weird connector, and the fuse responsible for my no-start, taken a while back. Maybe it is something to do with the alarm or it acts as an immobiliser - it is connected to the fuel pump, and there is a relay - but when it is left alone (i.e. fuse not dislodged), it doesn't seem to be doing any immobilising...

And yes... it really does look like an ATX power supply connector that somebody has re-purposed!!!

This post has been edited by donutty: 20th February 2019 09:27
Attached thumbnail(s)
Attached Image
 
Go to the top of the page
View Vehicles
+Quote Post
sanitary
post 20th February 2019 16:13
Post #20 | Print


1.4 16v
**

Group: Regular
Posts: 201
Joined: 12th December 2004
Local Time: 23rd July 2019 10:17
Member No.: 662
Driving: in front of you
From: Lardtown, 'ulll



I think that it is part of a left over alarm/ immobilizer system

I know you say you removed one but it could be a previous installation

some people are not too bothered about the state of the electrics as long as it works for them

I've removed a few alarms etc. and some of the work makes you wonder how the car didn't burst into flames




san

Go to the top of the page
View Vehicles
+Quote Post

Reply to this topicStart new topic

 



- Lo-Fi Version Time is now: 23 Jul 2019 09:17