'87 Porsche 924s

Our cheap little scope got here early it works okay for what it is but couldn't get a clear focus on the pistons, snaked it around inside the bellhousing (getting it stuck for a while) couldn't see anything conclusive in there either, went into the oil pan and couldn't focus on anything in there either.
A decent amount of my anti seize concoction poured out of the oil pan which makes me wonder, although there is still some in the pistons and it seems I could have filled it enough to come out through the valves and drain back into the pan...so still have no idea.
My Karl's camera with provide better intel.
 
So we were dinking around with it, scoping out the flywheel and oil pan and bores and I grabbed the breaker bar on the crank and my son watched the flywheel. It moved a fraction, back and forth about 6 times and bam, she was free!
So it's not siezed anymore, spun around many times and no drama.
Starting getting ready to light her off, fresh oil, spent a LONG time trying to clear the ATF/acetone out of the bores. Had to try and fire it off; pops and wants to go but didn't quite do it and didn't sound all that great trying to so prudence won out and we called it an evening.
Gotta figure out a good way to clear the bores out. Hand pump didn't want to grab anything, resorted to soaking up as much as I could with paper towels poked into bores and spinning it over plugless. I kinda thought that might not be a bad thing; running the pistons up and down a bit in the ATF and acetone/PB Blaster cocktail. But now they need to be cleared out so we can fire it up.:lol::lol::lol:
 
Ouch!!


Tread carefully.

Sounds like stuck pistons, as Rod said a bent valve will go back the other way.

First the fire and brimstone and then the rainbows and fairys at the end of the post my freind.



The 2.4 litre alloy block is not your run of the mill engine with linners, its Nikasil treated, there are no linners. Nickel Silicon Carbide is flaming hard. How hard? Extremely hard (9.5 on Mohs scale) the natural material Moissanite can also be synthesized in the laboratory = (silicon carbide) . The hardest known natural material is Diamonds 10 on old Mohs scale. We could prattle on about 2p versus 3p valence orbits and old and new Mohs scales but I hope you get the idea here. Hard but its a VERY thin layer, once you break through the layer its danger park ranger.

Porsche cast the blocks from an aluminum alloy referred to as Reynolds 390 and treated the bores with a hardned Nikasil/Alusil layer which is only microns thick by etching away the aluminum in the bores leaving the silicon particles standing proud of the aluminum. This is what the rings ride on and the oil loves to sit on. The downside of the trick with Nikasil bores is that they are very sensitive to stuck rings or pistons and will not tolerate foreign objects very happily. In short the walls can be easily damaged, probably why the PO ran like a skunk.


If you havent, already hit Google and take a look at Nikasil. If the rings got stuck in the bore and you were fortunate to disloge the piston and spun it over you can still save the engine. Drive this car like you stole it and it will die, its just a matter of when. I had a BMW K1100LT and its terminal overheating was exactly that, the bores were toast. Nowdays there are folks that can take your block and re- etch and restore the treatment [its not a coating as such but you can think of it that way].

It really depends upon what your long term plans are for the car, are you are going to flip it or keep it longer term. Long term? teach the young fella how to build a strong 2.4 Litre Porche engine, building and engine together can be a great bonding experience. These engines are not cheap in of themselves, but they are not rocket science projects. New slugs and a quick hone is not going to work as there are no known good coatings for aftermarket pistons that will tolerate riding in the Alusil/Nikasil bores. Attempting to install a non coated aftermarket piston on an Alusil bore will always result in a the death of the engine. I learnt this with the K series bike, she was a great ride but I killed her. There are some clever folks that can insert dry cylinder wall liners into the water-cooled Porsche blocks, but that would probably cost more that you paid for the car, hence why they get parted out. Recoating is cheaper and you could look at higher comp pistons for some extra go.

You can get away with doing what you have done to date with a robust Lampredi SOHC and even run it if the bore was stuck, with luck. Not so with a M28 or 944 block.


Take your borescope and inspect the cyliner walls like Sherlock Holmes and ensure the walls are good. A failed spot of Nikasil will grow, overheat the engine and kill it.

Lets have them collective fingers crossed and pray its the missing bit from the starter jammed against the flywheel. Though how that happened is a head scratcher, is there enough room in the bay to split engine bellhousing to take a look? It would be so satisfying [a relief] to find bits of that starter bush and grinning that the PO had given up so easily for such a small bronze bush.



I'll say a few for you

Best of luck

Sandy
 
excellent stuff, much appreciated
We'll precede with all caution although proceed we will and trust to good fortune.
Not much to lose as long as we try to be smart about it. Info like that which you provided being of the utmost importance, so thank you.
I don't see the flywheel jam as being the issue. It seems like it was locked up then it wasn't. We were gentle (most of the time).:eek:mg:
Careful inspection of walls will commence, if I can figure out how to fire up Karl's scope!

It was spinning over freely and quickly last night (with the plugs out).
Will hopefully go for fire in the next couple of days.
Can always transplant my Lambredi in it.
 
Last edited:
We're still trying to get it to fire. Next up is injector signal test and if it passes that we'll pull the injectors and send out for cleaning.
 
Sweet

Spinning over nicely on the starter then

Got any Noid lights, perhaps the PO flooded the cylinders n rings stuck with crud and gave up trying to start it up on an electrical isssue. They are not cheap for what they are, but my Noids have saved me many hours of futzzing around over the years

Hows the crank angle pickup signal, if the ECU gives off either no flashing Noid lights (no injector pulse) when cranking or a steady light then loom probably shorted to ground or a dead Engine Control Unit. IIRC there are 2 CAS.

Got a fuel pressure gauge to ensure the injectors have something to spray?

If you get a dodgy signal then like a lot of 80's trusty Bosch Motronic ECU's it can be a dry solder joint. Long before we got System On Chip devices Motortonic [among others] used resistors, capacitors etc like confetti. I have wacked ECU's like 'The Fonz' use to wack the juke box on Happy Days and got them to work, temporarily. I've saved a fair few Euro' cars from expensive dealers that can't use a soldering iron, and a hot melt glue gun.

Intermittent no starts are often dead a CAS, if you get no Noid lights [injector pulse] check the comfuser, under the dash IIRC. A quick look at the transistors usually gave it away. I have a stash of ass't old style round transistors from a closeout if I can help let me know what you have. If you have never soldered at this level there are a few important things to be warned about, but lets cross that bridge if you come to it.

If they all look good at the ECU /CAS, what is the resistance on the coil supposed to be, is it in range?

BTW

How'd the cylinder bores look,


Nice that you haven't decided to part it out. fingers are crossed for you down here.

Best of luck.
 
Funny you should ask.
I made a bushing for the starter nose and it's spinning the engine okay. With no plugs in it turns over nice, with plugs it make a a bit of disconcerting thud during cranking.
The speed and reference sensor ohm out okay.
I used one off my many LED cluster wedge bulbs with some wires soldered on as long leads, confirmed good pulses to injectors. Pulled injectors (took all of 5 minutes, WAY easier in that regard to our engines).
They look terrible and thats just the outside. Hunks of crud all over the tips, both ends, just nasty. I think I will send them off for cleaning, witchhunter if I can't find a local shop. I should go after the cold start one but it's pretty buried in there.
So this may be the smoking gun.
I did use Karl's "good" scope to see what I could see of the bores, they look okay but I can't really see them entirely just too hard to manipulate the camera and focus it down there.
We're hopeful. I don't like my low compression numbers on 1 and 2 but we'll see. Maybe just sticky rings? Could be head gasket between 1 and 2.
 
Last edited:
Decided it's just too easy to clean your own injectors so I'll try that first. With any luck she'll be runn'n tonight.
 
Got 'em spraying pretty good but still no joy. Now suspecting ignition timing is off and that's a reference sensor and a computer so I'm looking into that. Not too exciting about buying $200 sensors on a whim.:hrmph:
 
Got a multi meter?

Not too exciting about buying $200 sensors on a whim.:hrmph:

Then don't, test the output against the specs before shelling out the spondoolies. A lot of late 80's to mid 90's EFI systems came from Germany and are cross referenced to other car makers who also used them. If you can find the Bocsh part number perhaps VW Volvo Peugeot etc will carry that part rather than walking into a Porsche dealership. One of the Porsche boards should be able to provide the spec's for you, did you find a 944 workshop manual it will be in there too.


FWIW I know I have lost the values for the sensors a long way back but you can measure to see if the ohm reading is correct. A strong signal is what the old Motronic systems liked. Hot non resarts etc are give aways they are okay cold but the signal drops off when hot [dying so replace], I use a hot air [paint stripper] gun to heat them up to check the values are in spec. Thermometer and multimeter Req'd.

The DIY cleaning of injectors is your call, it can be a cost effective way of ensuring an injector is 'working' but removing hardened shellac - I can't see how. But be aware that even fuel metering to each individual cyliner is important. Can you flow test them against a stop watch into a beaker / test tube or something to ensure one isn't going to be obviously lean. I'm not asking that you concider flow matching or achieving the mythical balancing injector flows as your 924 is a street car that should'nt be expected to see a level of tune so precise as I'm sure it will survive just fine if it is not driven * WOT around the streets.


Just for giggles the final diagnosys of my loss of a Uno Turbo engine came down to one lean injector. Admittedly the old UnoT engine was at a high level of tune and HP for a 1300cc, that's the nature of a hand grenade motor. But I hope it illustrates the point that if the fueling to one of the cylinders is off chances are the engine will have to endure undue stress.


Removing the caps
IMG_20141226_125213.jpg



Rod end bearing failure
IMG_20141226_131613.jpg





Your call and your car/budget but I'd recommend a quick read of an article like this, at the link

https://www.injectorrx.com/diy-fuel-injector-cleaning/


For the new PuntoGT engine I had a new set Dynamic flow tested so I could establish the injector latency [injector dead time] so well set the new fuel maps. Anyone who says they can sell you a set of 'balanced fuel injectors' is uttering balderdash. If they have access to an unlimited supply of injectors and endless time to test them that may be rare but excruciatingly expensive.



Did you ever diagnose the reason why the engine was siezed?

Keep chipping away you sound like you are getting to the bottom of the issues.

Best of luck

Sandy
 
Have to assume a piston was stuck in a bore.
Sensors ohmed out okay and evidence of spark and fuel, just not at the right time. I guess the computer will "make it up" for lack of proper input. Sure sounded like bad ignition timing to me so we went after the sensors, man what an adventure. Good news is after market ones are only $32 so praise the Maker there.
The sensors were stuck fast as I guess they often are, went to take the whole bracket off, found some PO had preceded me as the hole for the bracket bolt that holds the sensors (top of bellhousing, behind everything and nearly invisible) was broken! A spacer/alignment dowel was missing. The bracket was sloppy in it's spot and that most likely rendered the readings useless for the motronic as the sensors need to be .8mm space off the flywheel.
Hope to make a new spacer and reassemble with fresh sensor and cross our fingers.
If she fires we'll take it slow and see if the engines returns to health, suspect stuck rings and still questionable injectors but some well-monitored run time might be the best thing for it.
d3314205-404b-4307-9bf9-09cfc560c4a4_zpsi4wsb7oh.jpg

Thanks again to Karl's scope for helping me see the impossible otherwise.
Took two of us working together to find and extract the bolts while using the camera, one below (car on new lift, which is Godsend!) one above, camera in hand. What a trip so far. Sure hope it runs in the end.
 
It is the least I can do :)

Wish I had more time, I would love to help :(

Hope things start going your way with this.
 
Got a oscilloscope?

Hello Jeff

How's it go'in

That sure sounds like a dead/misbehaving computer/fuel relay lead the PO to mess around with the sensors [& make a mess]. I re read above where you mentioned measuring the sensors [missed that sorry], if the PO had messed around with them and got it wrong it's little wonder it wouldn't start for him, the computer is garbage in garbage out confused mode. IIRC that era of early Motronic used one crank [TDC = timing] position and one engine speed sensor [ = fuel] to provide the information Motronic needed. If the TDC works but the speed sensor can't count teeth then it won't resolve a minimum of 200/300 RPM and the fuel pump won't be powered on. So it must have a good combination of both. I remember some Motronics had a start phase where they would just pump away and after initial cranking period shut down the fuel pump if engine speed falls below ~300 RPM after a set period. Early systems used to let the Tach flicker on cranking before that was filtered out by the marketing department, so if it does flicker than you are getting something the computer can see, the magnitude of the signal is something else. No flicker I used to just whip out the sensors and take a punt with the multi-meter, often metal flake or dirt on the sensor dropped the signal below the floor of what the computer needed to see... I have also seen where the two cables on the loom were ass about because of inattention to assembly / disassembly of a home mechanic - no seriously some manufacturers didn't make the connectors different so they couldn't be inadvertedly swapped.

I don't know off the top of my head if Stuttgart did this early on in the swap from J-Tronic to the better Motronic brain and this could have been the original POs issue. Despite what the masses will tell you most Motronic boxes will outlast the vehicle that they are installed in by a factor of 2 to >5, so reliable yes - foolproof no. Neanderthals with Mig welders and crappy jumper leads exist, uncontrolled voltage spikes - the natural enemy of electronics.

WRT the sensor bracket all the French deployed Motronic systems I have seen never needed their 'air gap' adjusted and the housing was cast/set at the design stage. Sensors of quality are therefore chosen carefully. Some German models had to have a holder / bracket adjusted to allow the standard sensor to suit, is this how the 924S is resolved? If so perhaps the PO stuffed up and that's all you need to do as you mention that he had mangled it.

The Motronic systems used a dual function relay [Peugeot called it a Tacyometric relay, in Porsche 944’s they called it a DME relay. If it's suss' then you get no fuel. The method of operation is similar in the X. With the key in the ON position it powers the computer, only when it sees more than 200 or 300 rpm will it close the circuit to supply the fuel pump with green steam. A suspect fuel relay will exhibit similar behaviour to a poor sensor signal, I'd personally just whip it out and at least double the amount of solder on the relay board. If it’s the original then I'd count that as preventative maintenance as its joints are old and ready to crack and give your young Fella grief later on. So many folks on this board carry a spare and I have never read of anyone opening a Fiat one up to check for cracked solder joints. Really easy to do and you can be fairly ham fisted with the amount of heat you put into it without killing anything. Of nearly all of the failed relays I have come across about 95% of the failures were due to overheated solder joints in the relays’ printed circuit board. Just resolder the little suckers, don't just throw them out, check their guts.

I'm probably telling you something you have already done [how to suck eggs] but what the heck. You have already [from your post above] Ohm'd out the sensors and determined the resistance is okay or not. A multi-meter from there will only tell you if there is/or not an output signal. What I eluded to previously is the signal strength that’s what a multi-meter cannot correctly measure. The signal magnitude of the output voltage needs to be in the range the computer is expecting to 'see', usually around 2 ~ 2.5 volts. So you need to determine if the signal is of adequate magnitude to be sensed by the Motronic computer. To check the speed and reference sensor signals the ultimate test to show whether the sensors are providing a healthy signal to the computer requires an oscilloscope. To do this you need the pinout list of the Motronics connector and connect to the pair of pin outs of each sensor individually.

Usually goes a bit like this, remove the fuel pump fuse disconnect the plug from the computer, attach probes to the two pins in the plug rail, crank the engine and observe a nice saw tooth wave output with the specified peak-to-peak voltage [~2volt?]. Note if the crank sensor output and speed sensor output voltage is in spec. If you get a fuzzy or inadequate signal voltage its usually a dead, poorly adjusted/dirty sensor. If you get a fat square image then it’s the fuel relay or the computer.

In my limited experience it's rarely the Motronic computer, I will always check the belly of a car for recent Mig welding work on the exhaust system as monkeys that are too lazy to detach the battery earth cable are culprits that shouldn't be allowed near modern machinery. But that’s how it usually happens if a computer dies.

If you get through all that and it still won't run its getting easier, now you are back to leads,dizzy,spark plugs. So as you would have for the X are the plugs in good con'd leads in good order and the right way around in the firing order? Lastly the dizzy is only really to distribute the HT energy when the computer triggers the coil but it still needs to be in good order and facing the right way. Is the cap pitted/cracked - suffering from carbon tracking, rotor button U/S or the dizzy full of moisture debris? Pretty average stuff that as an X owner you'll be familiar with the procedure. Often folks think the computer is somehow 'magic' and these items are overlooked. Some of the 'modern’ guys are used to COP systems and overlook a Dizzy altogether as they are 'ancient' and they are not familiar with them. Worth a quick check, but I guess I'm telling you how to suck eggs again, sorry. What you may have found that stuck with me from a 944 I helped with is the Dizzy is a PITA to work on due to its position - owner had the longest flat thin blade screw driver I had ever seen. I thought at the time he was joking, no just prepared! The similar vintage Motronic French XU9 engine Dizzy cap was child’s play by comparison, I always wonder why the Germans have to make things so darn complicated.



Your Green machine is a nice looking example of the 924, it deserves to be resurrected - keep at it. There is a generic book that has been useful over the years – worth the $, there is another but a Bosch book that is a bit expensive.



[FONT=&quot]“Bosch Fuel Injection and Engine Management” : How to Understand, Service and Modify
[/FONT]

[FONT=&quot]SAE[/FONT] ISBN-10: 0837603005




As always best of luck


Sandy
 
Last edited:
Good stuff. You know your stuff because all you've said matches what I have learned!
At this point we have replaced both reference and speed sensors, procured an alignment dowel, gotten decent purchase with pivot bolt despite broken area and set gap.
Engine sounds exactly the same, cranks, squirts gas (noid light confirms injector pulses, plugs are wet)
Have about 38psi fuel pressure at the rail.
Have spark but it doesn't look bright and blue so I pulled the cap, looks fine, rotor looks fine (it only fits the cam shaft one way with a set bolt so can't really get that wrong) confirmed wires are correct as to firing order, then I pulled the coil wire off the dizzy and the terminal inside the wire looked quite roached, corrosion and bits missing.
Went to pull it off the coil and it fought HARD. I won but the wire pulled right off its end, it stayed in the coil then the bit sticking out broke in two.
Made temporary coil wire out of old Fiat plug wire.
Checked coil for resistance, primary winding is .9v should be .4-.6
Secondary was infinite, cleaned the terminal and can occasionally get the probe to tell me 5000ohms which is in spec but hard to get.
Ordered new coil.
I also pulled the DME computer, took it apart and checked solder, all looks spectacular.
I've found the DME harness pin out and am going to attack with multimeter (just like I did to my X 6 years ago).
PO has left a mess of wires and it could be a ground is missing or something and the DME is confused, or the wiring to the sensors is bad.
To my mind something is confusing the DME and it doesn't know when to send spark so it's just guessing wildly.
I should post a video of it cranking. It turns over nice but it "thumps" and "knocks". Due to badly timed spark I would think but I don't know.
I need to recheck compression too. Even if there's a major issue with a cylinder or two, it should still fire on at least one. My X did.
I'll start in on wiring check while waiting for coil to arrive.
I'm not very confident it was "weak" spark holding up the works but possible I guess.
I can tell by the wrong bolt that was holding down the speed sensor, and the broken block, and the missing alignment dowel and the completely disintegrated connectors that some PO had messed with the sensors. That is now all cleaned up. The sensors themselves were never changed, judging by the fact the connectors were extremely brittle (old heat cycled) and were both unbelievably seized up in the bracket. It took hours to get them out. The speed sensor had to be drilled out and slowly chipped out of the bracket's bore. Unbelievealbe. They are supposedly slip fit and "should" be able to be pulled out with relative ease, most seem to cement themselves in there. New ones have lots of anti sieze on them.
There was also a lot of talk of alarms messing things up but I don't think the 924 had 'em.

Holy crap, I just found this image:
PlugWiresOrientation.jpg

If this is correct. PO had wires wrong as I have #1 at 1 O'clock...........interesting

If this is true, and my memory is good (ya, right). Wires are WRONG. I blindly followed the PO on this, I should know better.
Will test this theory at 3:35p.m. EST 5/16/16.......stay tuned, oh the drama.........
 
Last edited:
Waiting here, it's well past time for school to get out...

The suspense

If you got it started and went for a joy ride before telling us, we are going to be mad!
 
alas, it is was not that simple, PO had it right after all.
Will try new coil tonight. I'm going insane over here.:confused::confused:
 
no kidding, I wish one had caught my kid's eye up for cheap instead of this evil thing.
My next project car is going to be old school, carburetor and no sensors, no computers. Way too complicated.
It should be a great little car when sorted. I really like the thing although it's trying my patience.
 
Back
Top