Another Crank no Start

[tutor-408]

I inherited a Jetta 2003 TDI manual transmission wagon with 250K miles that had a new timing belt installed but was mistimed according to VCDS. I drove it home about 50 miles. Using this forum I learned how to pull the timing camshaft and reposition the timing belt and injector pump and was able to get the timing corrected and verified with VCDS. It ran fine for about an hour while I warmed it up and kept on adjusting the timing. I did clean the camshaft and gear and then torqued to 40 ft/lbs.

Unfortunately I also decided to change the oil and I think I overfilled by it by at least a quart. When I drove it out of the garage the engine made a "clunk" and died.
I pushed it back into the garage and found that I couldn't turn the engine over by hand. The timing belt had slipped and was now hitting the pistons.

I understand that most of the recommendations for a timing belt slip is to rework the head and lifters, but I thought I'd try to get this running so I can move it into a better location before working on it.

I first drained about a quart of oil.
I removed the camshaft gear and retimed the engine so I could crank the engine to do a compression check.
Here's the psi I got:
#1 420
#2 470
#3 420
#4 405

In order to get the compression through the glow plugs I had to remove the lines to the fuel injectors. So I knew I would need to reprime.

I've now retimed with a new timing belt kit and primed the fuel filter, injector pump and injectors, but I can't get it to start... when I turn the key the
glow plug lite turns on for a few seconds, and then turns off (I'm in California where it's been above 40 degrees), the car cranks fine but it doesn't start.

Here's what I've tried....

1. I've checked the battery voltage and it's at 13.27 volts
2. I've reprimed the fuel system a zillion times using both Whitedogs and MoGolf's procedures.
3. I see no fuel leaking around the fuel filter lines.
4. I've pulled a pint of fuel through the return line to the fuel filter by applying a vacuum to the return line.
5. I used a dial gage and a small stick in the #1 glow plug to verify I was at top-dead-center.
6. I've checked several times that I'm in the right hole for the injection pump pin.
7. I've loosened the three injector pump bots and slightly adjusted to both advance and retard the timing from the "pinned" position.
8. I've checked the EGR valve is in the right position
9. No diagnostics codes from VCDS.
10. Using VCDS I see that the engine RPM is about 250 RPM's when I crank the engine.
11. When I turn the key the voltage applied to the fuel shutoff value is "on" for about 1 to 2 seconds. then is 0 volts.
12. I've checked and applied Whitedogs "car won't start troubleshooting" tips that might apply.
13. I've tested relay 109 and that seems to be working fine.

My only clues as to what could be wrong are:
When I spray starting fluid in the intake, the engine tries to start but then dies. I assume that's shows a lack of fuel ?
I ask VCDS to show the ecm voltage and it reports it's "10.6" volts. Is that normal ?

My wife has been patiently waiting for me to get this car running so she can have her parking spot back.
So I'd appreciate any help with getting this engine started....my questions are:

1. is the compression good enough for the engine to start ?
2. is the ecm voltage reported by VCDS too low ?
3. should the voltage to the fuel shutoff valve be on longer than 2 seconds ?
4. what could I have damaged so much that even though I can turn the engine over it still won't start ?
5. could a stuck/damaged valve give that compression and prevent it from starting ?
6. Should I just move on to rebuilding the head ?
7. what other diagnostics can I do to check that I have the basics of air, fuel, compression ?

I know... I have lots of questions but I'm completely bewildered .... Thanks !

 

[GBaugh]

Compliments to you for your diagnostic thought process.
1. Those compression results are remarkably good for an engine with 250k. Yes, they are good enough to start. Actually too good to believe.
2. ECM voltage is ok. I like to see sys voltage stay above 10V during cranking.
3. Yes, the fuel shut-off solenoid should be constantly energized during cranking and running. As a test, you could jumper fused batt power directly to shut-off solenoid to force it on during cranking. Usually in a mechanical pump-line-nozzle fuel sys, the shut off solenoid is powered directly from ignition or relay powe. Not familiar with your specific engine though. Is this solenoid powered by an ecm that is not getting a good input like crank or cam position? You had rpm indicated at dash during cranking so I'm skeptical loss of cam or crank signal would be the issue here. Also, these older mechanical fuel systems usually are not ecm controlled. Hopefully someone else with more knowledge with details of this engine can weigh in here.
4.
5. An exhaust valve that was stuck closed would yield good compression but no starting. Remove valve cover and observe movement of valve rocker arms while turning engine over to verify movement of all valve stems
6. I would NOT rebuild head yet because at this point there doesn't seem to be diagnostic evidence to support it. I think that would create a situation with potential for compounding problems which I would avoid.
7. maybe remove fuel nozzles if not too much trouble and stick a borescope in hole to inspect valves, piston and cyl walls. Any metal in oil filter? Sounds like car was running long enough to circulate any debris. I would focus diagnostic efforts on fuel shut-off solenoid circuit. I was also thinking about exhaust restriction but car was running before incident so I'm steering away from that. When you crack the high pressure injection lines to bleed air during cranking, are you seeing any fuel squirting out? Any tailpipe smoke during cranking? Probably not if your fuel shut off solenoid isn't being controlled properly. I'm operating under the assumption that this is an older mechanical fuel system where fuel injection lines need to be bled. High pressure common rail injection lines on later model cars should NEVER be cracked open to bleed air. I'm just throwing that out there to ensure no misunderstanding.

 

[tutor-408]

Thanks GBaugh for sharing your thoughts and experience...
Yes, on this vehicle the fuel system from the fuel filter to the injectors must be primed with fuel to eliminate any air.

The fuel shutoff solenoid is driven by the ECM and I did jumper the fuel cutoff relay to 12V, but that didn't help. The voltage to that solenoid goes to 12v when I turn the key on and then shuts off after 1 or 2 seconds...this is before I've even started cranking the engine.

When I crack the lines going to the fuel injectors they seem to be dry. I don't have a lot of experience with this car, but I thought that when I removed them to do the compression test lots more fuel came out of those lines. When I prime the injection pump and lines to the injectors I get lots of fuel all over the engine, but after I crank the engine and check the lines they seem dry. If the fuel shutoff solenoid wasn't staying on very long then the little fuel in the injector lines would be injected into the cylinders and not replaced by the injector pump which would explain the dry lines.

When I primed the fuel system I hung a bottle of diesel with a clear hose going into the fuel line to the injector pump and then hand turned the crankshaft and watched the fuel go down into the injector pump. I saw fuel seeping out of the top of the lines going into the injectors that I had cracked open so I know I've primed them thoroughly.

No smoke from the exhaust which again points to a lack of fuel.

Tomorrow I'll remove the valve cover and check that nothings stuck and I'll trace out the line to the fuel shutoff solenoid.

Again, Thanks for your suggestions.

 

[tutor-408]

I feel like an idiot....
I took GBaugh's advise and took off the valve cover to check the lifters.
It turns out I had the timing 180 degrees off ! I thought I'd marked what way the camshaft was supposed to go when I did the timing but apparently not.

When I did the very first timing I didn't have to move the belt very much so that worked out fine. However when I did it this time the cam was in a completely different location.

I've retimed and now I'll prime and do the slight adjustment to get the engine started.

I'm more optimistic now.

I guess you might say I'm a cautionary tale about what to be aware of.

Thanks,

 

[AndyBees]

If you drove the car 50 miles, the Timing had to have been extremely close to being correct, including the proper relationship of all the components (Cam at TDC, Flywheel at TDC and the Injection Pump pinned at TDC),

So, to tweak the timing, it was not necessary to completely remove the Cam Cog, thus, no need to clean it or the end of the Cam Shaft (Morse Taper). I've changed numerous Timing Belts on the ALH engine and have never cleaned the Cog or the Morse Taper... totally unnecessary unless there was oil draining down onto it (which I seriously doubt). So, it concerning as to what you may have used to clean it, considering it slipped with 40 ft lbs of torque (Franko6 recommends 45 ft lbs).

I said the above assuming the Cam Cog on the Cam Shaft is what slipped. The Timing Belt doesn't slip. It may Jump Teeth. However, if it jumped teeth, well, the Tensioner was not tensioned properly and highly likely valves did kiss pistons.

The Compression is just fine.
However, the Lifters likely have spider web cracks that are typically barely visible. If so, one or more of them will grenade in about 1500 miles or less. That statement is assuming the valves did kiss the pistons.

Typically, the Exhaust Valves are the ones that get smacked by the Pistons....... The Pistons are chasing the Exhaust Valves as they close. Thus, if the Cam Cog slips on the Cam Shaft, the Cam becomes "slow" in time. So, in that case, the pistons catch-up with the closing Exhaust Valves.

I cannot imagine loosening the IP in it's mount to tweak timing. Hmm, this is a new one for me.

As for Volts at the Fuel Shut-off/on valve, well, with the Ignition On (starter not engaged), yes, the volts will drop off to zero rather quickly. However, while the starter is rotating the engine and when it starts, the volts to the valve will be 12+. The engine cannot run with the valve closed, period.

Below are photos I took of the voltage on the 2002 ALH engine in my Vanagon. I was having starting issues. I checked everything humanly possible, including the volts to the Fuel Shut-off/on valve.

Below, engine running

Below, ignition off for about 3 to 5 seconds

Below, ignition off about 8 to 10 seconds

Below, ignition off about 10 to 12 seconds

Since the TDI engine is very fuel efficient, when bleeding the Injector Hard Pipes, the amount of fuel flowing out is not going to be like a river. So, don't be fooled due to the amount of fuel coming out.

I am totally confused as to how you could have got the timing off 180 degrees? What was off 180 degrees. There will be no damage to anything if the Injection Pump is off by 180 degrees, but if the Cam is off by 180 degrees in relation to the Crankshaft, that could be concerning.......... in that, was it exactly 180 degrees off? It can be disastrous if the Timing relationship of the Cam and the Crankshaft are off by two or more teeth (timing belt teeth).

Apparently, you are not using the appropriate tools ..... Cam Lock, Injection Pump Pin and the Crankshaft holding tool.

Have you observed and identified the V mark for TDC on the Crankshaft by viewing through the hole in the Bell Housing? Removing the Air Filter Box makes the viewing much easier.

Keep us posted.

 

[tutor-408]

Thanks AndyBees for the pictures and for all your input....
I'm new to what to expect with this TDI engine (voltages, fuel injections, etc...) so having references helps a lot.

When I initially got the car the timing belt and fuel injector pump (IP) had been replaced but it was hard to start. So I looked at the timing with VCDS and saw that the IP gear needed to be advanced a lot, however it was already as far advanced as it could go. The three bolts were up against the gear and couldn't go any farther.

I ordered the timing tools from Metalnerd, aligned at TDC, locked the camshaft and pinned the injection pump and then pulled the cam shaft so I could rotate the IP gear just one tooth over so I could advance the timing. I saw several threads advising to clean the cam shaft with brake cleaner before putting it back on so that's what I did. Getting to the tensioner to tighten was tight, but the MetalNerd Tool tensioner helped a lot. So I wasn't moving the IP, I was doing the normal adjustment of the IP relative to the gear and belt that was being limited by the three bolts
.
Once I put it all back together I was able to get the timing exactly where it should be according to VCDS. It ran fine for about an hour until I moved it out of the garage. My guess is that I overfilled the oil and that somehow caused the timing belt to slip (?)

For my next bonehead mistake....I was also surprised that I was able to get the camshaft 180 degrees off but still rotate the crankshaft 2 turns. But after thinking about it, the only problem with the cam shaft being 180 degrees off is that the IP is 180 degrees off. There shouldn't be any interference if the cam is 0 or 180 degrees relative to TDC. It's only a problem when it's 180 degrees off from the IP pin. In my mistake the crank was at TDC, the IP was pinned correctly, but the camshaft and valves were 180 degrees off; ie. on the intake stroke. But there would be no interference at 180 degrees....obviously there will be interference it's off by some other value.

After taking off the valve cover I discovered the error of my ways...I've retimed the timing belt and primed the fuel system but it still won't start.

I've primed the IP by vacuuming from the return fuel line until a get about a pint of diesel. Than I open the lines going to the fuel injector and crank the engine. I get some fuel, but it's not squirting out. I had done this previously and remember it soaking the hood. Not this time...
I then tighten down the injector lines and try to start the engine, but it doesn't start. When I open up the line to the #1 cylinder injector it looks dry !

When I spray a small amount of starter fluid into the intake, there is some ignition but it still doesn't start.

Thanks for your pictures about the voltage. I checked the voltage on my fuel shutoff solenoid and here's what I see:
First with the line still connected to the fuel shutoff relay.

Ignition ON : 2-3 volts for 1 second then 0 volts

Then with the line disconnected:
Ignition on: 10.6 volts for 7 seconds, then 6 volts

According to my schematic the ECM drives the fuel shutoff solenoid directly. It doesn't drive anything else on that line out of the ECM.

So my conclusion at this point is that:

1. I'm not getting fuel to the injectors.
2. The voltage to the fuel shutoff solenoid should be on longer
3. The ECM is probably OK because the Glow plugs lights turn off after 2 seconds and the ECM reports engine rotation when I hook up VCDS.
4. When I was initially able to get the engine started the engine timing seemed to be very touchy about adjustment. It was very hard to start until I was able to get it adjusted according to VCDS.

I'll try disconnecting the shutoff solenoid and jumping it to 12V in order to keep it active while I crank. I hope to at least see that the #1 injector line has lots of fuel. I'll then try to make adjustments to the timing to see if I can find the sweet spot that will get the engine started.

Once I get this started and moved out of the garage I'll address the damage caused by the slipped belt. My questions is should I plan on removing the head or would I be able to get by with just replacing the lifters ?

Thanks and Best Regards,

 

[tutor-408]

When I had the valve cover off I saw that all the lifters were moving as I turned the crankshaft.

I do find it odd that the voltage to the fuel shutoff solenoid drops so quickly when I turn the ignition on and haven't even started started cranking the engine. When the line is disconnected it looks closer to what I would expect...10.6 v for 7 seconds then 6 volts

I disconnected the line to the fuel shutoff solenoid and jumped it to 12V. I cracked open the #1 cylinder injector line and placed a bottle over the line to see how much fuel I was pushing through. I cranked twice for 15 seconds each and got about 1 oz of fuel in the bottle. It looks like enough to fill the line going to the injector.

I cracked open all the injector lines to prime the lines, cranked the engine over twice for 15 seconds each. Tightened down the lines and tried to start. No luck, it wouldn't start.

I tried adjusting the the timing from fully delayed to fully advanced in small increment..no luck it still wouldn't start.

I'm starting to question the laws of physics.

 

[tutor-408]

I decided to believe in the laws of physics...
Since I primed the fuel system so many time I thought there must be some other reason I'm not getting ignition.
So I decided to pull the fuel injectors and they all looked awfully dirty. I'm pretty sure these are the original injectors in this Jetta with 250K miles.

I'll break out the nozzles and clean them up.
I'll soak them in mineral spirits overnight and then put them in an ultrasonic cleaner with seafoam.
Any recommendations ? Should I consider a new set of nozzles ?
Thanks,

 

[GBaugh]

I agree with you that the 10.6V measured at the shut off solenoid with harness disconnected is an improvement but still not as high as it should be. Is the shut off solenoid grounded through the harness or through the solenoid to gnd? In other words, is there one wire or two wires to the shut off solenoid? I suggest checking power and grounds at the ECM. If there is one wire to the shut off solenoid, then check the ground path of the solenoid including the engine block-frame ground strap. If two wires to solenoid, what is measured voltage on gnd wire, measured at solenoid, with reference to batt neg?

It almost sounds like you could have a shorted coil on the shut off solenoid. That would explain why voltage is higher with it disconnected. Is the shut off solenoid hot to the touch after being energized for a bit with your temporary jumper? Do you have a multimeter or inductive amp clamp that you could use to measure current through solenoid?

Is there a shut off solenoid coil winding resistance spec available or better yet an amperage spec?

Does this mechanical injection pump have an internal transfer pump and also, is there a separate lift fuel pump to supply fuel to the inj pump?

If you want to test those fuel nozzles yourself, Mercedes Source, over in WA, sells a modified bottle jack with a pressure gage on it. Its worked for me with testing nozzle opening pressure as well as spray pattern on a variety of nozzles. and for a reasonable price.

Now that you've established correct cam timing, maybe redo your compression testing.
When you're cranking, is there a consistent sound from each cylinder? A variation in the cranking sound might indicate a compression issue. Considering all the carbon accumulation on your nozzle tips I wonder what your valve seats look like, as well as your piston rings.

Wish I could give you answers but at least my questions might trigger some ideas for ya.

 

[AndyBees]

Only one wire to the Solenoid (Shut off/on) valve.
As I demonstrated with photos, the current drops immediately after the Ignition is turned On. If cranking or the engine starts, voltage stays at or above 12 volts.
The ground is via the Injection Pump to the engine block back to the battery.
If the Starter is working, well, the ground to the Solenoid (Shut off/on) valve is just fine.
The Injection Pump also contains a Vane Pump internally, The Vane Pump pulls fuel from the Tank.
There is a "check valve" in/at the exit port (fuel return) to maintain internal pressure for loading the plunger on each stroke of the Injection Pump.
No lift pump unless someone added one...... which I doubt. Lift the back seat and remove the cover to see if the plug is a 2 wire or a 4 wire.

I believe the OP is chasing a Wild Goose worrying about the Solenoid (Shut off/on) valve.

I believe the issue is Timing of the Injection Pump.

 

[GBaugh]

AndyBees, It would make a lot more sense if this were a timing issue because that's how this all started. It seems unlikely that a new problem with the shut off solenoid would abruptly appear after it was running for a hour after initial timing adj, but there's some strong diagnostic evidence pointing us to the shut off solenoid. Definitely agree that inj timing needs to be verified though. Tutor doesn't have 12V at shut off solenoid during cranking which is why that caught my attention as a potential diagnostic direction. When 12V was jumpered directly to shut off solenoid, it still didn't start which would indicate something else going on, maybe inj pump timing.

Is the inj pump drive gear keyed to the shaft or is it tapered? Could the key be sheared?

Those nozzles need to be checked out too. Tutor, is there a fuel inj repair shop near you that could test those nozzles for you? nozzle opening psi and spray pattern as well as leakage. Those nozzles look too wet.

 

[tutor-408]

The solenoid has one wire going to it so It's grounded through the solenoid mounting to the IP.
I physically checked the ECM and there's no obvious damage; also the glow plugs light up briefly and I can read the rpm's through VCDS when I crank the engine. But, it's odd that when the line to the fuel shutoff solenoid is disconnected it seems to act normal; ie. at 10.6 volts for 7 seconds when the ignition is turned on, but is virtually 0 volts when connected to the solenoid and turned on. That sounds like the ECM isn't capable of supplying the necessary current (.4 amps) to the solenoid.

There's about 7 ohms of resistance when I measure across the fuel shutoff solenoid to ground. When I jumper the solenoid to 12V, I get .4 amps (400 milliamps) so the resistance from passive to active isn't constant.

I understand that the ECM has multiple power sources that are turned on in phases. Is there a way to check that the ECM is receiving all it's required power ?

As far as I know the injector pump does not have an internal transfer pump.

Yes, checking the compression will be the next step. I checked before I put on a new belt, but not after my big mistake with installing a new belt.

When cranking, all the cylinders sound the same.

After cleaning up the nozzles and reinstalling them into the injectors I taped the injectors into 4 clear plastic water bottles, I connected the fuel lines to the injectors and IP, primed the IP and the injectors and then watched while I cranked the engine. I was hoping to see some type of spray pattern out of the injectors, but I didn't see anything at all...no spray no fuel at all. Again, when I disconnected the fuel lines to reinstall the injectors they seemed "dry" to me.

So....should I have seen fuel spraying into the water bottles with my "Rube Goldberg" test of the injectors ?

So I reinstalled the injectors into the engine and then reprimed...still no start. I even disconnected the line to the fuel shutoff solenoid and jumped it to 12V in order to make sure it was active. Still no start...

I'll try priming the fuel lines again and then disconnect them to see if they still seem dry even before I start cranking. When I prime the lines I see fuel dripping out of the slightly cracked open lines, but perhaps I should wait for more fuel to come out.

I'm inexperienced and ignorant about this engine, but another approach I've been wondering about is rather than ask "why doesn't this engine start", is to ask "what shuts this engine off ?" If the ECM is turned off by the ignition switch, then what does that cause to happen ? Is the fuel shut off solenoid turned off so that fuel is no longer being delivered to the injectors ? Wouldn't that drain the fuel in the lines which would require them to be reprimed ?

I appreciate all the advice and suggestions! I'm sorry if I didn't get back with some answers right away. After dropping nuts and bolts, fixing broken multimeters and getting ready for the family Easter egg hunt, it just seems to take me a long time to try things out.

Thanks,

 

[AndyBees]

Without all the details, all I'll say is, do not touch the Injection Pump shaft. (Do not loosen the big 22mm Nut, you will regret it.)

Considering that the Cam Shaft was supposedly off by 180 degrees, who knows where the Injection Pump timing really might be. Has the Cam Shaft and Crankshaft "timing relationship" been verified by someone familiar with an ALH TDI engine? Once that relationship has been confirmed, the mechanical timing of the Injection Pump can be set with the Pin. If this stuff is not verified and set, you are pissing in the wind.

Back to Post #1 or #2, if the Pistons did kiss the Exhaust Valves (or any of them), those valves are compromised as well as the Lifters. They may appear okay, they may go up and down when the engine is rotated, but trust me, they are compromised.

Below, take a look at this head (more photos in my photo gallery)

Below, take a look at the piston that tried to eat that Valve Head

Lastly, over full by one quart of oil never caused any of the problems with the OPs engine. I've driven my formerly owned 2000 Jetta over full by one quart numerous times. It is at 401k miles and doing just fine.

Oh, and again, the Injection Pump does have an Internal Vane Pump. That Vane Pump is what pulls fuel from the tank and it maintains internal pressure.

Keep us posted.

 

[tutor-408]

I did verify the crankshaft and camshaft relationship using the MetalNerd cam locking tool and a dial gauge for TDC. I verified that I was in the right injector pump pin hole. The IP is marked with orange paint on the injector pump shaft/nut and other locations and it still shows that the nut and shaft haven't moved.

I did initially time and adjust it correctly according to VCDS using the MetalNerd Cam Locking tool.

I've done the compression test again through the glow plugs and here's what I get after cranking about 7 seconds:
#1 - 350 psi
#2 - 350 psi
#3 - 350 psi
#4 - 380 psi

This seems much lower than my initial compression test, but is it high enough for ignition ?

In order to get clear access to the glow plug ports I had to remove the fuel lines. I couldn't cap the injector pump outlet ports so there was lots of fuel coming out of the IP, which is probably a good thing since it shows that I'm getting fuel through the IP.

I'll look into getting the injectors pressure tested.

The initial low voltage to the fuel cutoff solenoid still bothers me. Since that is driven directly from the ECM I'll check some of the other lines coming out of the ECM as well as all the input power to the ECM. Since VCDS reports that the ECM voltage is 10.6 instead of 12.4 volts, is that an indication of a problem with the power to the ECM ?

Thanks,

 

[AndyBees]

If you had disconnected the wire to the Shut off/on Solenoid Valve there wouldn't have been fuel everywhere.
Obviously there is 12 or more volts to it while the engine was being cranked, otherwise it would not have allowed fuel to go everywhere.

Anyway, it appears there may be warped valves. Why would the compression be down? Below 400 PSI is considered low but still within spec. However, those three at 350 are concerning. Considering the Cam Cog slipped on the shaft (if that's what happened), well, there's little doubt the pistons did kiss exhaust valves as I have previously reasoned.

How did you confirm where the Timing is using VCDS?

EDIT: For what it is worth, all of the Injectors wouldn't have gone bad at the same time ...... never heard of it.