Engine rebuild complete - won't start

[rickkellogg]

My son and I just completed rebuilding our first 944. I guess the good news is that the engine cranks fine with no strange noises. Bad news is that it won't fire. There's fuel pressure (loose fitting spraying a mist of fuel helped prove that). No spark when you pull a plug. Also, tach does not jump when cranking the engine, so I'm messing with the reference sensor. Both sensors measure correct resistance (by Clarks guide). Don't have an O-scope, but used my voltmeter to watch sensor output as engine was cranked. Voltage did jump indicating to me that something is getting through. My thought is that I may not have the sensor close enough to the flywheel, so I was going to adjust. Now I have two questions -

1. Does this seem like a likely problem, and...

2. How do you get the %#&@ bracket adjustment bolt loose? I've tried an allen wrench (L-shape), can get it in, but no room to turn it. I can get an allen socket in, but then not the two universal joints and extensions needed to turn it. It looks like if I remove the starter, I might be able to get to it from the bottom with a couple of extensions and universal joints, but I'd prefer not take things off, since I just finished 9-months of putting it back together

[SoCal Driver]

Usually when one pulls the engine they take off the bracket with the sensors intact. Other wise the sensors get bent to hell and will never be adjustable. Note that sensors from the BMW are cheaper than the sensors for the 944.

Should your sensors not be bent to hell then the adjustment is on the rear sensor to the flywheel teeth. About the thickness of a dime or 0.08mm or 1/32". Also check that you have the respective sensor plugged into it's correct cable. The tach should jump as you crank the engine.

Check all of the remaining connectors too. Make sure the rubber boot/duct is connected between the AFS and the throttle body and all of the vacuum lines are connected.

[Granite 944]

^ ...yep, simple things like reversing the flywheel sensors hookups, and AFS connector not connected, or poor connection. Major vacuum leaks can kill ya here also. Note: on a "rebuild", those sensors on the flywheel SHOULD have been already adjusted to spec. prior to the engine even going back into the car. Would be MUCH easier to have done so. Ohmeter check on those sensors may be in order here also. See Clark's garage, for diagnostic checks, for no start condition, if ya will.
If ya really need to adjust those sensors NOW, then I think Clark's can help here also. Small tools and patience will be required. Good luck!

[rickkellogg]

I actually pulled and rebuilt the engine myself. I did remove the sensors intact before removing the clutch housing. I also adjusted them to spec (but without gluing a washer) when I reinstalled them. I used two pieces of masking tape stuck to the end of the sensor. Both sensors have the correct resistance, but since the tach does not jump, I'm still concerned that they may need to be a "little" closer. I wonder if anyone out there has had to "tweak" their sensors a little closer. Again, sense the tach does not jump and there's no spark, it seems to point to a reference sensor problem... or is there something easier I can check...

[ERAU944]

have you tried switching the wires? if you have them reversed it won't work properly... just a guess...

[rickkellogg]

Thanks - I did try switching the wires. In both cases engine does not fire and tach does not jump. It appears from Clarks garage that the sensors are labeled pretty clearly (BG, DG), but the plugs on the wiring harness are not labeled and you have to go by their position in the small bracket that mounts to the rear of the intake manifold. I'm curious if there is a way to identify which connector is which (other then position). In my case, since I removed the wiring harness when I rebuilt the engine, I'm forced to go by the 'memory" in the wire. One of the female plugs favors the top of the bracket and the other favors the bottom. Any better way to tell which is which on the harness side?

[SoCal Driver]

You need to back up to square one on this. See if there is 12 vdc at the coil's small black wire with the key on. If there is see if there is 12 vdc at any of the injectors. One pin on the injector plug will read battery voltage the other a bit lower.

If there is no voltage at the coil then check the ignition switch. If there is voltage at the coil but not at the injector then check the DME relay. Also check that the red wires at the battery positive terminal are clean and tight.

You should hear the fuel pump run for a half second or so after you stop cranking.

You can bypass the DME relay by using a jumper from the battery positive to the coil's small black wire and putting a jumper from fuse #2 to #3 on the aux fuse block (the one without the cover, count from the drivers door).

If there is voltage at both the coil and the injector check that the HV cable from the dist. to the coil is good. Check that the rotor is secured with both bolts and the cap is on straight.

Once all the above is confirmed it's either the sensors or the DME. Note that the front sensor triggers the spark (through the DME and not directly).

[rickkellogg]

SoCal - Thanks for the detailed steps, so far no good news. There is voltage at the coil. The voltage on the injector plug reads identical (battery voltage) on both terminals. I was able to adjust the reference sensor gap as well. Fuel pump is brand new and pretty quiet. With my hand on it I can feel it running when the engine cranks. If I disconnect the number one cylinder wire and attach it to a spare plug, the plug sparks once right when the key is turned on (before cranking). When the engine is cranked, still no spark. Based on what you say above, I've narrowed it down to the sensors or the DMS. Is it true that only the reference sensor need work for the engine to fire, or do both sensors have to work? Is it true that the DME relay is mainly for the fuel pump and not the ignition?

[SoCal Driver]

The DME relay has two contacts in it. The first one closes when the key is turned on. This provides the voltage for the injectors and to power the DME. The circuit that provides current/voltage to the coil is also the circuit that goes to the first set of contacts in the DME relay.

The second set of DME relay contacts are for the fuel pump. These close under two conditions; when cranking the engine (the circuit to the starter solenoid is split with one leg going to the DME) and when the engine is running over 400 rpm (the speed reference sensor is what tells the DME to keep the fuel pump running).

I don't like that there is spark as you turn the key on. This means the DME circuit that fires the coil is failing. This could be that the DME needs a bit of resoldering. Could be a really bad DME. Could be that the wires to the coil are connected wrong. The black wire goes to terminal #15. The green wire from the DME goes to terminal #1.

There has also been reports of the pin on the flywheel breaking off or coming out. It's really a set screw.

[todwic]

Originally posted by SoCal Driver


There has also been reports of the pin on the flywheel breaking off or coming out. It's really a set screw.

I can confirm that some idiot did mal-adjust his sensor bracket and sheared the set screw on the flywheel. I'll willing to bet that if you are savvy enough to rebuild an engine, you could tell that the coil wires are two diff. sizes.

A multi meter with an "RPM" setting is what I use to tell if my sensors are firingh. Hook it up as per the manual. No or unsteady RPMs indicate something afoot. I measure the signal pulse that the dme shoots the coil.

I cut off and bent a allen wrench to loosten the bracket bolts. (search "tools I made" for photos of same.

[rickkellogg]

OK... Confirmed coil wiring and picked up a multimeter with an RPM setting. Also did something I should have done earlier, check the resistance of the coil primary and secondary. Based on the coil measurements, the coil appears to be bad (open secondary). I'll order a new one tonight, but just not thinking I'm that lucky. Seems like the coil should still get the fire signal from the DME whether the coil is bad or not. I hooked the multimeter up in RPM mode, attaching it to ground and the green wire on the coil, and I get nothing -- 0 RPM when cranking the engine. Just for grins I used the AC volt setting to monitor the output of the speed and reference sensor. I get a reading from both when the engine is cranked indicatinig (to me at least) that metal is passing close enough to the front of the sensor to generate a small voltage. Interesting that the speed sensor gave a significantly higher reading then the reference sensor. My guess would be that this is because it's generating a pulse at every tooth on the ring gear vice once per rotation for the reference sensor. I'll replace the coil this week and try again before hunting for another DME.

[cchyper]

This may be irrelevant for you, however, i built a motor and had issues starting it as well. The problem was a single connection near the driver side firewall. It was flipped... a buddy of mine had connected it backwards (red - blue, blue to red). Its a white connector with one red and one blue wire coming out of each side. This is unlikely, but thought i would tell ya just in case.

Check this connection

[tnporsche]

Just to be sure,

with the ref sensor bracket removed have someone rotate the eng and watch for the ref screw. Check its condition.

(I had the exact same problem with exactly the same symptoms, it was a sheared off screw)

[Zero10]

the red/blue connector would cause a no crank situation not a no spark
As for the RPM meter, it should read even if the coil is bad, you are correct.
Don't rule the coil out yet, make sure that you are getting pulsed power to the coil beore you decide it's dead.

[rickkellogg]

Thanks everyone for the help - I bought a new computer on ebay and put it in and the car starated on the first crank. Oil pressure is good, but the valves are a little noisy. Have to admit, I've never heard a porsche 944 run as this one wasn't running when I bought it. Is it normal for the valves to be a little "clacky" until oil really gets pumped into the lifters. How long does that take?

[EMBPilot]

it may be the injectors "ticking"

[Techno Duck]

Ticking can be the injectors, they are loud on these cars.

Clacking is probably the hydraulic lifters. Did you have them soaking in oil while they were out of the car? If, not..in my experience they can take up to 15-30 seconds to pump up. This also is dependent on the oil weight you are using. They take even longer..as i had actually injected oil into the lifters with a syringe before reinstalling them.

As long as you have good oil pressure, it should go away.

[rickkellogg]

John - Thanks the clacking did go away after a few minutes - Can't believe it's running!

[Razorback1980]

Feels great, doesn't it? Good work.

[SoCal Driver]

Oil pressure?