UpFixenDerPorsche

My '90 GT

Engine simply died while driving, with that instant-off "no spark" feeling, and indeed there is no spark.

So - I've worked thru all possibilities I can think of viz

- disabled the immobiliser
- disabled/bypassed the exhaust temp monitor
- changed out LH, EZK, fuel pump relays.
- changed out the LH ecu and EZK units
- checked resistance of CPS from the EZK connector: approx 990 ohms.

As I don't have a CRO available I haven't checked the CPS output waveform.

Now the plot thickens:

This has happened 7-8 times in the last few weeks.

- first time I did the standard thing: checked the battery terminals and power feeds to the CE. All seemed ok. Engine re-started and all seemed good, but I couldn't put my finger on anything definite.

- next time, I did the same checks, thinking there must be an intermittent open circuit in a CE feed. And I did the same for each subsequent stop, except the last stop, which was 'terminal'. ahh haha :-).

Now the intriguing part: all stops occurred while on uphill grades. Sometimes just after the bottom; sometimes halfway up or near the top. Never downhill or on the flat.

Always before the engine had reached operating temp. Never occurred once the engine was hot. Then after 5-10 minutes it would re-start.

The problem has also coincided with the first patch of cold winter weather here.



Over to you.


Cheers

FredR

Sounds as though you need either a scope, a new CPS or most likely both.

Rgds

Fred

the flyin' scotsman

if you remove the MAF you can see the CPS connector........they do get baked over time.

StratfordShark

Ignition switch not in your list of tests. Maybe try wiggling it?

Mrmerlin

how many miles are on the engine and has the intake refresh ever been done?

Its a safe bet that if you have over 50K miles and the intake is original then its time to refresh all of the parts under the intake ,
this would include the CPS, the knock sensors ,the Hall sensor, the flappy valve, the ISV and the TPS.

That said inspect the grounds and the CPS connector at the rear of the block ,
also look for a damaged harness as it crosses from the engine head to the firewall.

docmirror

Well, I have to say I'm a bit -- gobsmacked by Stan's list of todo's, most of which have nothing to do with a completely missing spark. Now, if you want to start throwing money, certainly all those items should be replaced but except for the CPS, none of them are going to cause the complete ignition failure.

The S4/GT has two independent ignition circuits, with a few exceptions. I think it's critical to eliminate those items which would affect one 'bank' (I know they are intermixed) or the other. Here's a list I can come up with: Plugs, wires, cap, rotor, coil, final ign amp, ign monitoring relays(cut fuel, not spark), and wiring to EZK.

More important, you want to find, and test common items: Battery, current supply to LH and EZK, ign switch, EZK, LH, CPS, ground(s). Maybe I missed something but that should give a decent list to start. IIRC, the immobilizer cuts fuel, not spark, but I haven't made a detailed examination of it.

So, that leaves battery, supply to the LH and EZK, ign switch, CPS, and the ground associated with the EZK. Check your voltage, right at the batt and make sure you have good voltage. I like to start the car, and watch the meter rise while running to prove that the charging circuit is working. Check the power and ground to the EZK. Check the CPS waveform with an o-scope, and test the ign switch. Alternately, you can replace the CPS without taking out the intake, although it's a messy job. The Ign switch can be replaced in a few hours, and there are cases where the ign switch has caused this type of symptom. I wouldn't waste time on the parts that only affect one bank yet.

Hilton

After the EZK, the ignition is two separate banks (pins 32 and 15 off the EZK connector), so loss of spark on all 8 cylinders has to be somewhere before that connector.

CPS signal goes to the EZK directly (EZK feeds it to the LH to get fuel pump running). So IMO your problem is either the Crank sensor on top of the bell housing, its harness connector (most common failure), the EZK itself, or the power to the EZK (via a relay.. dun dun dunnnnn).

First things I'd try in order to troubleshoot, re-testing after each step for change in symptoms:

1. Swap the EZK relay (its a type 53 one)
2. swap the EZK with the one in your S4
3. remove the airbox and MAF, and check the connector for the CPS isn't cracked to pieces (try unplugging and re-plugging it.. if re-plugging is a word).
4. buy a scope, or just replace the CPS

Oh.. one other point..

0. Check the small wires off the battery positive terminal are tight


edit: If you do find the sensor is at fault, if it doesn't come out easily from the top side, drop the clutch and push it up from beneath. Much less chance of screw-ups, and dropping the clutch is actually pretty easy in a 928.

edit2: I see you've already done most of the above steps (now that I actually read the post I'm replying to heh). Only thing left to do is dismantle the car and sell it for parts Dibs on your clock, and the input shaft from your transaxle.

worf928

I agree with Stan, assuming the same answer that Stan does to the question that he asks in the first line of his post: if the intake has never been done it's time to do it. And, unplugging and re-plugging the CPS is usually 90% of the effort required to replace it if the connector is still in the bracket. If it isn't in the bracket then you know someone has been messing about in there.

UpFixenDerPorsche

My thanks to everyone (with technical input) who has replied.

I'm sorry to burst bubbles but I have exhaustively tested, swapped, disabled, bypassed, replaced, measured everything but the front left tyre pressure on International Something-or-Other day. LOL.

It's since my initial post that I've heard of the apparent significance of the ground connections at the rear of the block, and the operation of the CPS, c/- John Gill (Rennlist). According to John G's experience, unless these grounds are 'true' then the CPS does not produce an output.

Two jigsaw pieces are now beginning to fit:

- occurs only on uphill grades: what in the car changes relative position when changing road grade? The engine on its flexible mounts, rotating slightly rearward, and perhaps stressing an electrical connection between the engine block and the chassis earth.

- occurs while the engine is below operating temp AND on an uphill grade (and not on a level grade).

Should there be a poor quality earth connection between the block and chassis earth (exacerbated by accumulated corrosion between each mating surface), then it's not unreasonable to expect that the combination of inertial engine change of position (to a gentle uphill grade) and the resulting relative movement between loose/corroded ground/active connections will affect the performance of electronic sensors such as the CPS.

- overlay all the above with the effects of the contraction between electrical contacts at lower than normal temperatures.

Anyway this is my theory and I'm sticking to it. LOL.

So tomorrow it's the 'Manifold Out and Fix Earths' day (anyone got a supply of hot needles to poke into my eyes - such is my love of this procedure LOL).

Watch this space.

John Speake

AFAIK you don't need to remove the intake to get at the two main ground points...

docmirror

Hmmm, I thought he was talking about the big heavy gauge cable from the lower rear of the engine block over to some point on the chassis? either way, the intake doesn't need to come off. There's basically nothing under the intake that will stop the car dead from no ignition. The MAF will cause it to stumble, and run like crap, but not stop ignition. Nor does the intake need to come off to change the CPS, nor the ignition switch.

Sigh, I might as well not bother. He's got the bit in his teeth, and it's all systems forward.

StratfordShark

I recently completed timing belt replacement so got a good look at the Hall sensor, since I had to back out the main harness from the 14 pin connector from behind the cam cover.

As far as I can tell from using JDS Spanner my hall sensor (76000 miles) is OK, but if it had been easy to replace it I might have done so (it's a semiconductor device so any problems more likely in the connector that sits behind cam plate).

But even with the harness out of the way, it looked impossible to get fingers onto the connector with the engine in place! How is the sensor replaced, and what difference would it make whether the intake was in place or not given it sits behind cam?

UpFixenDerPorsche

\\

Ya think?

Anything but the teeth in my bit to avoid removing the intake manifold just to get at these "ground chassis points".

The more I prevaricate the more is my accumulated stress, to the point where I say "ALRIGHT. ENOUGH OK" and then get on with removing the intake manifold while wondering why I/We thought it seemed so difficult in the first place.

Chalkboss

It is possible... but not a very nice job to do, to say the least. I did it with everything in place, and used a combination of an allen key and a 1/4 drive with varying length extensions and socket. The key is to use at least a 12 inch extension, as this gives you enough angle to get in there. Then getting the connector on and situated is very fiddly. It took me a few hours, if I recall.

the flyin' scotsman

The hall sensor is 'easily' replaced when doing a tbelt replace project by removing a few more items

Remove the RH (passenger side in North American cars) cam gear then the back plate. Theres then lots of room to disconnect the HS connector, replace and reconnect.