Bjbpe

This is a request for help that I sent out about six months ago but I did not have much information to share at the time. I now have quite a bit of data so I am asking if people might want to give me some advice. The car has been laid up for just about exactly a year. I spent about 4 months trying to track things down but I am really not properly equipped.

1. The car will not start
2. It will start then die when primed with starting fluid
3. The fuel pump is getting electrical juice
4. The fuel pump provides adequate pressure
5. The brain was rebuilt and tested on another 928; it works fine
6. The air flow meter has been replaced
7. The timing belt has been checked, is adjusted properly and is almost new
8. It has been through two auto shops; one was an auto electric shop, the other was a foreign car shop where it currently resides. No luck so far.

Living in Centennial, WY (25 miles from Laramie in the mountains) I do not have a true 928 expert to fix it but the owner of the foreign car shop is very knowledgeable. I suspect that he would be able to track down the problem if some of the knowledgeable folks on this site could help.

The car is a 1987 S4 with a standard transmission. Any help would be greatly appreciated.

MainePorsche

Barney,
Has your guy evaluated the CPS (flywheel or crank position sensor). It is deep in the back of the 'V' under the MAF. The sensor may fail after time, but the electrical connections to it OFTEN degrade because its location and build up of materials on it. The ECU will not no where the crank is if this sensor is down. Your engine will not start nor run if it managed to.

Bmw635

I assume you checked sparks hence it fires up on starting fluid?

I am not familiar with the 928 but did you check the crank sensor as it may give signal to DME, brain?

Double check the plugs on the fuel pump relay- I am getting intermittent firing that fires up when I wiggle the relay.

worf928

If the car actually runs when starter fluid is dumped down its gullet then it is not the CPS.

Remember No CPS==No Spark. No spark means no run.

The symptoms are consistent with a failed LH. However, the LH has been tested.

How is the pump being tested if the car won't run? (I want to understand the specifics of the procedure being used.)

Let's assume that the pump and the LH are indeed good; there is still a fueling issue if there's spark.

First comfirm that spark does indeed exist during cranking; this is a sanity check. If yes...

... then something is preventing the injectors from being grounded for the firing pulse. Or there is no power to the injectors. The injectors ground through the LH and the LH relay. I don't remember the power path (928s usually have ground not power issues.)

Wiring diagrams are needed. If it has spark, fuel pump provides pressure during cranking and it won't start then the injectors are not firing (or its flooding but you'd smell fuel...) and the problem is wiring somewhere.

Wiring becomes a problem, usually, when it's been touched not when it is just sitting. Who's been touching the wiring?

But, first, does it really have good consistent spark when cranking?

Mrmerlin

I will suggest to replace the 4 running relays,

IV, XVI, XX, XXV these are 53 type relays do one at a time
NOTE it may be time for a fresh set of plugs,
NOTE clean the battery terminals and charge the battery
NOTE inspect the coil wires to verify they are not corroded and the wires dont touch any part of their run.
NOTE Put deoxit 100 spray on the back side of the CE panel connections do this with the battery disconnected.
NOTE add the 100 spray to the computer connections.


https://rennlist.com/forums/attachme...-1987-1988.pdf

davek9

^^^ yep ^^^
OP, I did not see any of the relays mentioned in your post, also if the shop is good w/ electrical than an O scope should be on-hand and they can test the CPS for a good square wave at the CE.

Kevin in Atlanta

If you give me the person to talk to at the shop and their telephone information I will contact them directly on your behalf.

If all is as you say in your first post I would guess that the injectors are not firing and there is a broken injector ground.

Bmw635

Does the 928 fires up , just not running even with dead CPS? BMW will fires up engine but not run with dead CPS. TIA.

worf928

I know its hard to separate truth from fiction when M people are saying something not based upon knowledge of the system and N people are describing something based on knowing what the system is designed to do and you don't know who belongs in M and who belongs in N.

If you can think of a way that the engine would be able to ignite a fuel/air mixture without spark then yes, I guess it will run without a speed signal from the CPS.

Let's assume though that without spark the engine can't ignite anything.

Note that Crank Position Sensor is a poor term for what the sender actually does. It doesn't sense the position of the crank. It senses the speed of rotation of the crank. That's why the shop manuals call it a speed / reference mark sender.

It is this speed signal that is sent to the EZK. Once the EZK has separated the speed pulse from the noise in this signal (that's why it is a shielded branch of the harness: to reduce EM noise from the engine that would otherwise drown-out a weak signal) the EZK begins commanding the coil amplifiers to energize the coils to generate spark.

At the same time the EZK sends the speed signal to the LH. The LH then turns on the FP relay and begins grounding the fuel injectors periodically. That causes fuel to be injected into the cylinders: the periodic connection to ground.

That's the way it works on an LH-2.3-equipped 928.

I have no idea what a BMW does.

No CPS == no Spark. The EZK will NOT command the plugs to fire without a speed signal. And without a speed signal the LH will not activate the FP relay nor attempt to pulse the ground on the injectors.

If you have spark then the CPS is good. If you have spark on any plug for 1,4,6,7 and any plug for 2,3,5,8 then the coil amplifiers are good.

If you have spark on all cylinders and there's no hint of firing then there's no fuel pulse. Or there's no fuel. (Or, the LH has failed and is grounding the injectors continuously thereby filling the cylinders with fuel. )

If the LH tests good and there's fuel pressure at the rails and all three relays (LH, EZK, FP) and relevant fuses are fresh then it's a wiring issue. Most likely wires at one of the injector connectors are shorted together. Less likely is a loose ground at the back of the block under the fuel pressure regulator/dampeners.

M? or N?

Bjbpe

Thanks for the help, Dave:

At this point in elusive hunt we have established that the fuel pump is getting fuel and is putting out sufficient pressure to satisfy the requirements of the engine. The shop had a pressure gauge to check the supply at the engine.

The engine has enough spark to start when starter fluid is sprayed into the air cleaner but, of course, will not sustain itself after the spray is consumed. From this it would seem that the fuel injectors are the culprit. Both the "brain" and the air flow meter are good. They are both new and have been checked out on a running 928.

As an 80 year old aerospace engineer, I probably have sufficient grey matter but insufficient analytical tools to handle this thing. The owner of the shop where the car is now located probably has both the grey matter and the tools but is simply not fully interested in making a career out of starting 1987 928s.

Any other inspiration on your part greatly appreciated.

Barney

Hilton

Barney, I believe what Dave was asking was..

Was the fuel pressure tested by just cranking the car and trying to start it, or was it tested by jumpering the fuel pump relay position (or just applying voltage directly to the pump) to make it run?

If it was measured by just cranking the car over and trying to start it, then that means the fuel pump circuit is being energized by the LH, which in turn means its getting a CPS signal from the EZ-K, which narrows down the possible culprits.

However, the fact that it starts with fluid indicates you have spark.. which means its a fuel issue somewhere. Being batch-fired, the injectors can get taken out by a single short in one injector boot.

Kevin in Atlanta

Barney wrote this in the original post and it needs to be considered in context of this current thread.

...............
Some time back I was traveling over a rough rural road (we have a lot of them in Wyoming) and I experienced a sudden loss of power. Fearing that a shut-down might result in a no-start I drove immediately to the shop that usually takes care of the car. It turned out that one of the two ignition coils had broken (from the rough road) and, after replacement, I was back on the road.

I had been having occasional non-start problems before the above occurrence and they continued after the coil repair. However, I was able to start the car and drive it home; however, upon arriving home and shutting the engine off I could not restart; it cranks but does not start.
..............

dr bob

There's a really handy and inexpensive tool referred to as a "'noid light" that fits between and injector and the plug for that injector. It has an LED indicator that lights when power is passed through to an injector. From a local parts store. The injector plugs are accessed by removing the cover on the injector rail, that's the cover with a couple allen capscrews right next to the mainifold. There's a little wire 'bail' that holds the connectors to gether. I usually pull that part out. Then usually need a magnet pickup tool to recover it. So plug the areas around where you are working to catch the wire if it comes loose unexpectedly. After that, pull the connector and put it on the back of the 'noid light. Then have your assistant crank the car on the starter motor. If the LED blinks as the engine is cranked, the injectors are being energized. A short or a break in the wiring will give you no blinks.

Replace the fuel pump relay if you haven't already. The horn relay will take its place for testing purposes.

Double-check the connections at the positive battery terminal. There are several smaller wires that connect at the positive cable end, including fuel pump, LH/injection power, and power for the cooling fans. A loose or dirty feed to the LH and/or the fuel pump feeder will cause your symptoms. If the battery isn't very secure in the battery well, it would take only a few bounces to loosen those wires. Clean and shiny (I use a small brass brush) on the connections, bolt snugged to 5-7 lbs/ft. I clean the whole mess as part of an annual ritual, and spread a thin coat of Vaseline over the connection block/battery end after everything is tightened up again. Vaseline protects the connections from corrosive vapors from the battery while charging. It also washes off of hands and clothing with a little soap and water, something the more specialized products don't do. Cheap and does the job just perfectly.

Kevin in Atlanta

Left a message at Barney's independent - the Noid is already on the top of my list. :-)

worf928

There's a better one that does not require disconnecting the injectors.


This is one of my pet peeves. For the '87s and '88s those spring clips are easy to pull off. Once they're off, assuming you didn't drop any, you can't get them back on, securely, unless you melt a 'plug' on the connector body over them with a soldering iron.

If you don't put them back on then you will likely get a loose injector connection and intermittent misfires. I've 'cured' several 928 misfires by putting those clips back on to ensure a good connection to all injectors.

Don't pull the clips off.

If you can't get the connectors off with your fingers buy the right tool:

https://rennlist.com/forums/928-foru...l#post11450239

Yup. I asked the question the way I did because valid observations with the wrong measurement technique can often provide important clues or point to the next diagnostic. For example (but not appropos here): Fuel pressure gauge shows pressure when the ignition key is turned to run. Note: run. Not start. This indicates a failed LH.

Correct. Though, I would still prefer to have spark tested with a timing light while cranking. Any random spark might cause the engine to run with starter fluid. But, a timing light will tell you - better - if the spark(s) happen more-or-less where it/they should.

If the CPS is in it's super-uncommon failure mode from which the EZK extracts the wrong speed, a timing light might, maybe, give you a clue. An 'oscope on the EZK pin is better.

Bumps can cause something to fail. No doubt. The question here - it seems - is what could a 'bump' break that impacts fuel delivery that is not the fuel pump?

As per Hilton, do you know how they established this?

Or, specifically, that the injectors are not firing because they are not getting pulsed.

A new LH is still available from Porsche. Although they look - plausibly - like they have actually been expertly remanufactured. Yes, I have seen a 'new' LH purchased from Porsche a couple of years ago.

However, Mass-Air Sensors have not been available new - as far as I know - for more than 10 years. Used or rebuilt are our choices. Again: as far as I know. Roger? Oh Roger....

The first easy things to do are:

- double check the fuel measurement procedure as per Hilton's post.
- check the spark with a timing light on both coil wires and a couple of plug wires.

Do these things to make absolutely certain that you have good spark and good fuel 'cause it gets harder and you don't want to proceed with the hard stuff only to find that the fuel or spark diagnosis was misleading.

Next: get that EFI Quick Probe, or Dr B's harder-to-use noid lights.

If you get no pulses on all injectors then disconnect the injectors one at a time, repeating the test until you have only one injector connected. So, in other words, do the test with all 8 connected, then 7, 6, 5, 4, 3, 2. Or, speeding it up: 8, 6, 4, 1

If, when you get to only two or three or one injector connected and THEN you get a pulse: you have an intermittent ground on the harness or LH.

After that, you have to start going through the wiring methodically.

There's a super long thread of Andrew's where he chased-down injection harness issues and all the relevant details are in that thread. I'm unwilling to provide the technical details extemporaneously from memory.

I believe that this is it:

https://rennlist.com/forums/928-foru...light=injector

Actually, it isn't, but, it has more info and some pictures. AO's issue was something else.

But, bottom line, if you find that
- you get no pulses to even a single injector,
- you have fuel flow when the starter is cranking
- spark is happening when it should

Then you very likely have wiring issues.

An close examination of the first inch of harness wiring after the injector connectors might prove fruitful. A short is most likely going to be under the connector boot or where the wiring bends 90 or 180 right after the boot.

If you get this far and nothing obvious presents itself you may be at a cliff based upon your post unless you or your chosen mechanic wants to get serious with a meter and the wiring diagrams and can dive into the harness to fix it.

Sean's making new LH harnesses. They aren't cheap. Throwing a new harness at the problem is not a good diagnostic. However, if all diagnostics - double-checked - point to wiring issues in the harness then that might be your solution if you or your mechanic can't dive into the harness.

But, again, not cheap. So, you want to be SURE.