NoVector

Hi everybody –

I’ve been futzing around with a no-start on my '93 928 US spec with a Roger-Box for the past 6 weeks and I’m running out of ideas. I’m pretty sure it’s an over fueling issue but would like a sanity check before I send my LH to Mr. Speake for an ops check.

- The engine cranks but acts like it’s flooded and there’s a definite gas odor out the tailpipe

- Checked for spark on the 1 and 5 spark plug wires with my inductive timing light and both are good when cranking

- Jumper the fuel pump relay pins 30-87 and have good fuel pump “buzz”

- Hooked a noid light to number 5 injector and notice it flashes at a 4-to-1 ratio compared to my timing light. I think this is right as I’ve read each pulse squirts 1/8th fuel so my spark on a plug wire should be once to every 4 pulses on an injector.

- Hooked the noid light to number 2 injector and see the same thing; a 4-to-1 ratio when compared to spark plug. I.e., noid = "blink", "blink", "blink", "blink" then timing light "flash"; infinity...

- When I remove the fuel pump fuse and crank the engine, the car does start; albeit, it dies after a couple seconds…

- Today I temporarily installed a fuel gauge I just received from Roger (thanks Rog ) and I have 60 psi/4 bar with the fuel relay jumped. When I turn the car off or remove the fuel pump fuse, the pressure hold steady for at least an hour (which I took as good news meaning my injectors aren’t leaking). Crank the engine and it will start for a couple seconds and then fuel pressure gauge obviously goes to zero.

The last thing I “fixed” when this happened was replacing radiator overflow tank level sensor with a used one. Ever since then—no start. I’ve since disconnected the sensor but no effect. Wondering if a bad float sensor shorted out my LH ECU(?) although looking at the wiring diagrams, I don’t see how the two could be related. So, I’m guessing, it was just a coincidence. I was also refilling the transmission fluid at the time after a flush so there was a lot of short starting/stopping as the tranny fluid was being sucked up from the pan.

- I also ordered a new RPM sensor, although I don’t know if a defective one would send the bad signal through the EZK as my spark and noid light seem constant and the LH (which is a slave to the EZK for the RPM sensor) is pulsing the injectors. I also ordered a new Temp II sensor but I’m thinking that would only make it run rich or lean—not flood the engine.

FWIW, I’ve also:

- Checked cam gear marks, they line up (glad I put red paint on the notches when I did that job 2 years ago )
- The injectors do not “click”/pulse with just the ignition on, but they do pulse when engine cranked
- Cleaned all of the ground contacts
- Cleaned all wires to the starter
- Purchased/installed new battery-to-frame and engine-to-frame ground straps
- Checked 1 and 5 plugs after cranking, both wet
- Swapped Starter relay (V), Ign Relay (X), EZK relay (XXII), LH relay (XXV) and Fuel Pump relay (XXVI)
- Replaced fuel pump fuse #38

Appreciate any ideas – I’ve mentally hit a wall. Is it time to send it in the LH? 60 psi/4 bar fuel pressure seems high to me, but I see in the “Test Plan” that 3.8 +/- .2 is correct for a jumped fuel relay. Because i have spark, I'm ruling out alarm relay, limp-home, etc. BTW - Thanks Sean for hammering home the idea of buying a noid light - it really has helped me get this far.

soontobered84

Have you pulled the vacuum lines on the fuel dampeners and checked for fuel smell in the vacuum lines? It sounds as if your dampeners have quit dampening and are free flowing fuel excess fuel into your cylinders through the injectors.

NoVector

I did the front one with no smell this evening, but have not checked the rear damper or FPR. I'll check it tomorrow after some more cranking. I'm charging the battery right now - repeated days of testing and cranking is taking a toll on it... Although I replaced the front dampener last year when I noticed it was rusty looking on the inside when I was replacing the fuel lines. The rear dampener and FPR are at least 10 years old.

Herman K

Did you check yr tempII sensor (the one on the PS of the waterbridge)it serves two circuits one that feeds a signal to the EZK and the other feeds the LH

Check the tempII resistance by connecting yr multi-meter (in resistance mode) to an ECU ground pin and the ECU tempII sensor pin

for the LH Ground pin 17 - temp II pin 13
for the EZK Ground pin 18 - tempII pin 19

at 32F you should have 4.4-6.8 K Ohms
at 86F you shoud have 14.4-3.6 K Ohms

FredR

I have not read anything here that immediately points me in the direction of LH failure. You lose fuel pressure on starting and the motor stops- this suggests to me fuel starvation under the dynamic condition [you start under static conditions so fuel gets through initially].

Do you know for sure the in tank pump assembly is intact [i.e. the pipe from the pump to the tank outlet- very prone to failure but usually manifests as loss of power rather than being hard to start in my experience unless there is a blockage caused by a lump of failed rubber pipe in the line. Presumably your fuel filter is OK?

Suggest you carry out the "end of curve" pump test where you route the [open] discharge of the pump to a collection cylinder and measure the amount of fuel pumped in a given time- 1.25 litres in 30 seconds I believe. Last time I did this I connected a length of plastic hose to the fuel rail pressure gauge connection secured by a jubilee clip [note: providing there is no blockage in this pipe the pressure in the fuel rail will be minimal during this test]. If the pump delivers fuel OK at this rate then there will be no blockages.

Having said that a friend of mine here had terrible problems with his S4 starting and then suddenly stopping after a short run- defeated all known logical symptoms for brain failure but it was an LH problem nonetheless and John fixed it.

Regards

Fred

NoVector

Hi Herman – I’ll throw in the new Temp II tomorrow for kicks. I could see it running rich or lean but not resulting in a no start. Dunno… BTW, congrats on the two new sharks!

Hi Fred – What got me going down this path was how much faster the injector pulse compared to the sparkplug. Surprised to see a 4-to-1 ratio… the noid light blinking like it belonged in a disco had me wondering—but I guess that’s normal(?)

I’ve never checked the internal pump and have no idea how much I’m flowing, but I’ll give it a shot after I check the vacuum lines to the rear FPD & FPR and put in the new Temp II. I just replaced the fuel filter last spring, less than 3K miles. More to come... Thanks again / Bruce

Stromius

+2

Been through 3 Temp 2 sensors in 6 years. Each time I had symptoms like you, resistance specs above were way off resulting in rich/over rich. Now I have an extra with my spare relays

FredR

What pushes me in this direction is your comment that the fuel pressure is collapsing- problems with temp sensor do not cause that and it does not take much of a drop in fuel pressure for the engine to cut out.

Do keep an open mind until you nail the problem. Many years ago on my British motorcycle I had a problem I could not trace after a re-build- with one of the [two] cylinders cutting out after a short run-wet plug- turned out is was simply the pilot jet not fully engaged in its housing causing the plug to wet with excess fuel! I have seen my motor wet its plugs for no apparent reason- for sure you re not supposed to touch the pedal before or during starting.

If your plugs are wetting and then failing to sustain fire then I doubt the blockage theory but you need an explanation for the collapsing fuel pressure

Good luck with the "hunt"

Regards

Fred

jcorenman

Bruce,

The injectors are batch-fired, normally once per engine revolution, while the spark for each cylinder fires once every second revolution. So the injectors should be getting pulse twice as fast as the ignition, but I don't get 4x faster. If that is the case then you've got 2x too much fuel but I don't see how that could happen. Unless the LH has gone totally nuts. Is there another car around that you can check it in?

Does it try to start, and then die? Or not fire at all? What happens if you open the throttle while cranking?

Temp-II definitely needs checking. When the sensor tells the LH that the engine is cold (high resistance) then the LH injects more fuel (just like a choke on a carb). So a faulty sensor-- or funky wiring somewhere--can definitely cause flooding. The best check is to measure it at the LH connector, per the WSM as noted.

If the fuel-pressure stays up with engine off, then the damper or regulator diaphragms are fine, if they were leaking then the fuel pressure would drop.

If I understand correctly, when the engine starts and runs briefly-- and the fuel pressure drops -- is when the fuel pump fuse is pulled. With the fuse in and the pump running, it never starts and smells like fuel. Is that correct?

Leon Speed

Bruce, do you have access to a 928 diagnoster? If not you're welcome to borrow my Theo tool. I can send it over so you can at least check the critical components.

risaac928

Speaking of Temp II Sensors - on my 82 928 I noticed the PO had placed a resistor in with the wiring from the Temp II Sensor. I wonder if the PO had the same problem and the Tech placed a resistor to get the car to fire.

NoVector

Yes – That is it exactly. I also currently have the EZK, LH and Fuel Pump relays jumped (30-87) With fuel pump fuse in and I try to start it, it just cranks and acts like a flooded engine. Pull the fuse and it will start after a couple seconds of cranking and then die when it runs out of fuel.

Thanks, Aryan – Although I’m not sure how it works; for me it would be like a pig looking at a wrist watch. And God help me if it’s not in English.

I swapped in the new Temp II sensor this evening with no change – 35 degrees in the garage and have 5.4 ohms when just checking it to the block. Long day at work, I’ll pull the MAF and check the vacuum lines on the rear FPR and FPD tomorrow and check for gas smell.

jcorenman

If you mean 35F then 5.4 K-ohms is right. To be sure, pull the LH plug and measure there, just to make sure there is no broken/corroded wire anywhere. Also remember that the temp-2 sensor grounds through its threads to the block-- so teflon tape or sealing compound can insulate the ground connection. If you measured to the engine block then this would hopefully show, but not a wiring fault.

jcorenman

They were trying to make it richer, not leaner. Adding a resistor in series with the sensor increases the total resistance, which would make the controller think that the engine was colder-- so it would add more fuel. It's a common trick for adding fuel, and everyone knows that more fuel is faster

Leon Speed

It is in English (or French or Dutch or German if you like ). You will need a laptop and install the diagnostic software, after that it is plug and play. It is pretty easy to work with, the interface is some buttons and dummy lights and some output values. Let me know in case you want to try it. Hope you get it solved, sounds like a weird problem.