Tapping is an unusual suggestion but I’m game for anything. I’ll give this a try just for the heck of it.
Yes, the injectors should be spraying while cranking. We used to test CIS spray patterns by jumping the pump relay, laying the injectors out on top of the engine (connected to the fuel lines of course and with a few shop rags padding them) and then lifting the sensor plate a tiny bit with a magnet.
As to sufficient vacuum that’s hard to gauge. To me anyway, it acts like the engine simply can’t suck hard enough through the intake tract to lift the plate against the fuel pressure.
Logic would dictate a clogged exhaust, major air leak in the intake tract, or a worn out engine.
All of that has been checked (more than several dozen times) and the engine is not worn out at 170 PSI. I vaguely remember doing a tune-up on this car roughly a few thousand miles before this problem appeared and a test with the vacuum gauge at that time showed the car had around 17" of vacuum at idle.
For the climate and altitude here that can be considered normal.
The intake tract is pretty long (air cleaner on one side, turbocharger on the other) and if I disconnect the air tube between the air cleaner and turbo, cap it off with my hand while having the throttle plate wedged open, and then crank the engine over by using the underhood test connector the engine should pull my hand tight against the tube. It does not.
This should mean an obvious major air leak in the intake tract tubing but I’ve pulled the entire mess off and gone over it with a fine tooth comb. No loose clamps, no cracked boots, etc.
I’ll see if I can give that vacuum suggestion tapping a try and see what happens. At this point nothing is too weird for me.
It’s not a good thing when the experts here get stumped. I don’t know much about the engine OK4450 and may not be of much help to you on this problem but I was wondering if there could be something wrong with the intake valve train, like a broken cam. Does this engine have a seperate cam for the intake valves? Not having a vacuum while cranking the engine seems to be the best clue to the trouble. I assume if you tried spraying some starter fluid into the intake the of the engine it still won’t fire up.
Edit:
After thinking some more on this I guess you couldn’t get a good compression reading on all the cylinders if something was wrong with the intake valve train. Dang.
Got to thinking after my last post and the vacuum tapping idea won’t work as the problem involves ported vacuum instead of straight manifold vacuum.
To tie them together would involve totally reworking the above the throttle plate intake tracts on 2 vehicles; both the donor vehicle and the problem child one.
Hit the key again this evening and it fired right up and sounded good; for the proverbial 2 seconds. After that, nothing.
Really interesting problem but I expect no less from someone as experienced as OK4450!
Considering all the prior info- and pardon my inexperience with your particular make/model but is it possible that you have a big leak through a defective EGR valve?
On a “normal” fuel injection set up, it will still start and run (poorly) because the injection is not dependant as much on vacuum integrity. But this beast will not work at all if the leak is big enough.
One thought is an internal leak and depending on the presence of such a valve and the way it is designed into the system, it could be stuck open and providing a big leak path bypassing the intake where the ported vacuum is developed for the CIS system.
I wished it were something as easy as an EGR fault but this car is one of the few oddballs that does not even have an EGR valve on it. (Meaning that it was not removed and came from the factory this way.)
Unfortunately, these cars do not use an air temp or coolant temp sensor. They do use a thermo-time switch to activate the cold start valve but at this point the thermo switch and cold start valve appear to be the only thing that is actually working right.
I do appreciate the comment anyway because I’ve beat my head on the wall over this for months and just about all of my reasoning skills left town quite a while back.
I don’t know much about this particular setup but I wonder if the problem isn’t in the O2 sensor. Did you reuse the existing or hook up one attached to the new cat? It’s possible that the O2 is not working correctly and so the computer is leaning out engine immediately causing the stall.
It seems beyond reason that the problem is not related to the new cat. I’d put the old one back in and see what it does. Or trace the wiring around that cat to see if something got unplugged in there. I’d bet dollars to donuts the problem is directly related to the cat replacement. Even swapping out the Ign module won’t solve the problem if the computer is getting the wrong data. Does the car have a “limp” mode that you can manually trigger by pulling a sensor lead or something?
popping in here to say, it is air more than fuel, and most likely warpage is the cause of the leak. take the intake apart, and put it back together with some high temp silicon blue and see if that will plug the leak.
Is it possible to monitor the fuel pressure before and after the distributor while the engine starts, runs for two seconds and then stops?
Can you provide a low air pressure supply (around 1 or 2 psi) to the air intake with the exhaust plugged shut to then check for an intake air leak that has otherwise been undetected?
It’s not the O2. These cars will run without a working O2.
As to the valve cover grommet that was one of the first things I checked and it’s fine.
It does not have an anti-theft system on it, the car has a new fuel pump (with check valve), and both the pump relay and ignition switch have been bypassed as part of the testing methods. Same thing.
With constant shots of aerosol carb cleaner it can be kept gasping and wheezing along for a bit but it’s hard to do.
I do have the special fuel pressure tester that monitors the system pressure and both the warm and cold control pressures. They’re all fine and it also maintains the proper amount of residual fuel pressure. The tester actually connects to the fuel dist. (where it’s been for some months now!)
I have not considered plugging the exhaust and applying air to the air filter housing but a smoke test revealed nothing. Most of the lengthy intake tract consists of metal tubes with rubber boots. Every single piece has been removed and checked for cracks with nothing at all found.
I’ve even smoke tested the throttle body clean through the intake manifold where it mates to the cylinder head and nothing appears.
There’s been a few cars over the years that have given me fits but eventually I got through them. This one just flat has me baffled.
It’s a fuel issue in my opinion, or lack of fuel issue. Apparently (based on what was originally an educated guess that has become Guesses Gone Wild) the air sensor plate is not rising and allowing the fuel slots to become uncovered inside the fuel distributor.
These slots are cut with lasers and allow fuel to flow through into the fuel lines and then into the injectors.
Manually raising the air sensor plate will pop the injectors right off and they spray fine.
Excessively high fuel pressure could cause this but it’s normal.
Air sensor plate is not sticking and is free as can be.
Pop-off pressure on the injectors is normal. (changed them anyway just for the heck of it)
The computer would not cause a no-start on this model although it could affect the way it runs hot, cold, or both. The frequency valve buzzes like a hornet so this means the ECM is operating.
I’m stuck on the lack of intake vacuum. I keep going back to leak sources that could bypass the intake. I see yours is a Turbo model. In looking at the APC setup, it appears there is a solenoid that can provide a leak path back to the intake under certain conditions. Wonder if it’s working right or some other leak path through the Turbo…
Apologies for my amateurish attempt to offer a suggestion to a true expert, but sometimes a wild guess by a relative amateur can get lucky. Here goes.
Is it possible for you to measure the vacuum required to lift the sensor plate by rigging an external vacuum source with a manometer (or diaphragmmatic vacuum gage), measure the actual vacuum being pulled during start up, and compare these numbers with the 17" you were pulling at idle when it was running? I know you said you felt no vacuum at the air intake when using your hand, but I’m wondering of in the process of going after actual readings you’ll stumble on something.
I’m unfamiliar with the physical configuration of this setup, but I’m also wondering of it’s possible to measure the vacuum at each intake port, then perhaps couple them in various arrangements. It sounds crazy, but if you truely have no vacuum at the air intake than it wold ssem logical that the air must somehow be being drawn in via another path through the intake manifold rather than pulling the intake. Normally I’d wonder about the valve timing, but I know you’ve triple checked all that. I know this sounds stupid, but is there any possibility at all that the piston could be pulling an exhaust valve open, due to a busted spring or something, rather than pulling the sensor plate? By checking each individual intake port you might find something wierd like this.
