AHU - Air in Fuel System

It's a good point, and I was just reading the advice on dieselgeek's website to change the head o-ring before making any other changes that could further complicate troubleshooting. I'm hopeful that the head is just leaking so badly that the pump can't prime enough to get started. Don't know if that's actually possible when running off of a bottle, but it's a cheap and (relatively) easy thing to swap and test.

I suppose I'm due for a timing belt here in the next 6 months, so I think I'll replace the head seal with the pump in place and see if that gets me back up and running. If yes, then I'll monitor for leaks and consider whether to do a complete reseal when I change the TB.

On a related note, what's the consensus on the dieselgeek viton o-ring? Post #227 of the sticky thread (http://forums.tdiclub.com/showthread.php?t=115689&page=16) suggests that the dieselgeek o-ring isn't fat enough to seal properly. I have to think though, that if this was a common problem dieselgeek would be selling a different o-ring...

Viton seal ordered, hopefully here on Thursday.

I'll update the thread once I've installed it, but thanks again for all the input and advice!

I’ve always run the standard seals. Never had an issue with them.

Let’s rule out the 109 relay. Easiest way is to install a jumper wire between the positive battery post and the stop solenoid. This backfeeds the ecu and the car will start with the 109 removed from the relay panel. Remove the jumper and the engine will stall.

Let’s rule out the cps/ess aka g28. It should read between 1-1.5 ohms, between pins 1 and 2. Since you’re seeing fuel move, I doubt this is the issue, but it’s worth checking.

-Todd

When you say standard seals I'm guessing you're referring to size? Or do you mean Buna vs Viton?

I'll test the cps/ess later this evening after work (and after I do my homework to figure out what/where that is). For the 109 relay, I'll wait to test that until I've replaced the head o-ring.

This is all incredibly helpful - really appreciate it!

Whatever comes in the Bosch kit is standard.

I’d go over the IP harness. Abacus always seems to have issues with broken or chafed wires.

-Todd

CPS/ESS did not get checked this evening - looks like that's going to be a bit of a pain to access. I've got a couple friends coming by tomorrow to help me push the car back into the garage - once it's back on level ground I'll be more inclined to tackle it. I'll have to look in the morning, but is it really impossible to remove the ESS without removing the engine mount?

Good suggestion to check the IP harness - I'll take a look at that too.

Resistance is measured at the plug. It’s easily accessible.

The 109 test is a 1 minute affair.

Personally, I like to get things checked off, rather than adding variables to the issue.

-Todd

That's good news - I was afraid I would have to do some heavy lifting just to access the plug.

I may be missing something, but unless the engine starts I don't think I'll have any conclusive evidence for relay 109. If it doesn't, then I still don't know if 109 is good or bad because the head seal might be preventing the engine from starting. Putting a jumper on the solenoid is an easy thing to try, but since the car is sitting outside I've been closing things up every night - I'd have to re-rig the fuel bottle etc.

Speaking of which, is there any reason I need to bypass the filter and go straight back to the tank when I have the fuel bottle rigged up? I understand it's to avoid making a mess with diesel, but couldn't I just plug the filter output? Or would the lack of a complete loop fill up the filter, pop the plug out and make an even bigger mess?

The thermo-tee recirculates fuel until some temp is reached. If you plug that port to the pump and it wants to recirc you might have the problem you postulate.

Could run from the tee back you your bottle but I like the idea of running the return line from the pump to the bottle. Then you have a closed loop.

Remember that the pump body is pressurized by the internal lift pump so it should run with the head seal leaking unless it is really leaking badly.

Since you only seem to be planning to change the head seal that would negate the concern I brought up about the QA position adding a variable.

Thermo-tee...I'm regularly reminded of how much I have yet to learn

Thanks for the explanation - I'll refrain from plugging the filter port but I'll continue to run the return back to the tank, at least for now. I've got clean diesel in the bottle and I don't want to modify the bottle now and risk getting plastic bits into fuel that's bypassing the filter.

Could just run the line from the tee to another bottle of fuel.

there are 2 grounds, one on the round harness next to the coolant glow plugs and another at the 3rd injector harness, if those are not grounded, the car WILL NOT START. check them

KLXD said:

Could just run the line from the tee to another bottle of fuel.

Click to expand...

That's a good idea - might be the best option. I'd rather leave that 21 year old thermo-tee alone as much as possible.

Mongler98 said:

there are 2 grounds, one on the round harness next to the coolant glow plugs and another at the 3rd injector harness, if those are not grounded, the car WILL NOT START. check them

Click to expand...

Thanks - I'll check those when I check the ESS.

Turns out the hill is too steep for 3 people, so the car remains outside. And it was raining all yesterday evening so I didn't get a chance to check grounds or the ESS. Forecast doesn't look good for today either, so I may not be able to work on it before Friday. Fingers crossed...

kurtzl said:

That's a good idea - might be the best option. I'd rather leave that 21 year old thermo-tee alone as much as possible.

Click to expand...

Unfortunately the T can become leaky over time just because. There's also conversation from the past about whether the stupid thing actually works. I completely eliminated mine and went back to the Mk2 filters that don't have a thermo coupling.

The T requires that the o-ring supplied with the filter is in good order and not some old stock worn out product. I've received filters in the past that have 1) missing o-ring or 2) cracked and badly stretched o-rings.

It's worth verifying that the T isn't contributing to your full issues, after all, you have to pull it off to replace the filter anyway, just make sure that the hose connections to that T are not dried out and allowing air to enter the system. There's only so many places where the air can get in on these things.

Steve

Steve Addy said:

I completely eliminated mine and went back to the Mk2 filters that don't have a thermo coupling.

Click to expand...

I didn't know the older filters don't have the T. Any discernible difference in power or fuel economy?

No luck with the weather today, hopefully I'll be able to get the seal done tomorrow.

Misspoke in my last post. Meant a line from port you disconnected the feed line from to another bucket. You aren't disconnecting from the tee.

kurtzl said:

I didn't know the older filters don't have the T. Any discernible difference in power or fuel economy?
No luck with the weather today, hopefully I'll be able to get the seal done tomorrow.

Click to expand...

Nope, it's just a different filter, slightly shorter but looks about the same and still fits in the bracket the same. This is what it looks like, it has the small bleeder on top but no attachment for the thermo T.



The thermo T 'fuel preheater' system started as an option with the 86 Golf and Jetta and was continued unchanged for the Mk3 and B4 diesel cars. The problem with the T is that 1) there's a question about whether it really works and 2) it introduces another place for air infiltration.

Steve

Anybody want to buy a Jetta? Needs a little work.

How much and where?

kurtzl said:

Anybody want to buy a Jetta? Needs a little work.

Click to expand...

Sent from my KYOCERA-E6790 using Tapatalk

kurtzl said:

Anybody want to buy a Jetta? Needs a little work.

Click to expand...

Love to have one from Cali that had excellent body. As usual the problem is the 1700 mile distance.

Steve

The update, now that I'm in a better frame of mind...

Friday was clear so I spent a few hours replacing the head seal. Everything went well, no hiccups or issues. Took longer than I expected, mostly because I was being extremely particular about making sure everything was as clean as possible before closing it back up. Reassembled everything, primed the pump, and no more bubbles! All signs point to success with the o-ring replacement.

Feeling optimistic, I decided to try starting the car from the filter. No luck. Reinstalled the suspended bottle, still no luck. Removed relay 109 and jumpered the shutoff solenoid, but that didn't work either.

KLXD, looks like you were right about the seal not being THE problem.

So now I'm back to square one. Well not really, but it feels like it. Thanks to a friends's chain hoist the car is back on level ground in the garage. I'll check the ESS now that I can get underneath the car.

Mongler98, you suggested checking grounds. I checked the ones I think you were referring to, but I'm not sure. I'll have to get a picture and post it to confirm that they're the right ones.

If both of those check out, the only thing I can think is that the timing belt jumped a tooth (scary thought). I cranked the engine around by hand when I was setting the pump position, and again after I was done to make sure everything seemed ok. It doesn't seem like there's any mechanical interference, so I'll have to check and see if everything still lines up. If it doesn't, what's the best way to check for damage? Compression test on each cylinder?

Did you go over the IP harness?

I’d scan the car. It doesn’t need to be running to pull codes. If any are there clear them, and try to start it. See what appears, if any at all.

Compression or leak down test will show valve damage.

When I mentioned pulling the 109 relay, it was more of a “something to try” thing. All you needed to do was add the jumper. The relay could have stayed in place.

-Todd

Mozambiquer said:

How much and where?

Click to expand...

Sorry Mozambiquer, that was the voice of frustration talking. The car is in California, but it's not actually for sale (at least not yet ).

ToddA1 said:

Did you go over the IP harness?

I’d scan the car. It doesn’t need to be running to pull codes. If any are there clear them, and try to start it. See what appears, if any at all.

Compression or leak down test will show valve damage.

When I mentioned pulling the 109 relay, it was more of a “something to try” thing. All you needed to do was add the jumper. The relay could have stayed in place.

-Todd

Click to expand...

I did go over the IP harness and couldn't find any evidence of bad connections or frayed wires. It's not impossible I missed something, but it looked good to me.

I pulled codes last weekend when the problem first came up. The only code that came up was N109 shutoff solenoid mechanical failure, but I removed the solenoid and it didn't help. I'll try clearing codes tomorrow and see if I get anything new. I actually think the N109 code was from a couple weeks ago, not new from when the car died. I was having the engine race at start issue that's been linked to the solenoid, but it cleared up on its own and the CEL didn't come on.

I wasn't sure if it would matter if relay 109 was in place or not so I pulled it. Does it invalidate the test if the relay is out when I jumper the solenoid?

I understand! I'm about at that point with my ranger =-O I'm a diesel mechanic, and a tdi aficionado, not a gas vehicle mechanic. :-D
I had a very similar problem on my brother in law's car, his is a 98 jetta with the ahu, so I have a few questions for you to help you figure out what's wrong.
1 you mention changing the fuel filter. What was the filter number. The reason I ask is because it is very easy to put the filter for an alh on there, and it won't seal and pull fuel. That's what our problem was.
2 did you crack open all four metal injector lines at the injectors and then crank it until fuel came out clearly?
3 does it sound uneven when it cranks? I don't know how to describe it, but if there is a loss of compression in a cylinder, it will sound uneven.
4 is it cranking well? A low battery or starter going out can cause it to not start as well.

kurtzl said:

Sorry Mozambiquer, that was the voice of frustration talking. The car is in California, but it's not actually for sale (at least not yet ).

Click to expand...

Sent from my KYOCERA-E6790 using Tapatalk

Here's some kinda photos of the difference between the mk4 and mk3 filters. The return tee is different. The one with the red caps is the mk4, the other one is the mk3. It has the o-rings for the return tee on the filter itself, whereas the mk4 one has them on the tee. They are not interchangeable, unless you change the tee out as well. On my brother in law's car we did a service on it and he had a filter in the trunk, so we stuck it on and filled it up with diesel and then fired it up. It pretty promptly died, and so we tried priming the system, and cracked the injector lines open, but couldn't get any good fuel stream. Finally I got to looking at the filter and realized we had the wrong one. Got the right one and no problem.



Sent from my KYOCERA-E6790 using Tapatalk

One more thing to check. Using vag, check for engine speed while cranking. If there is no engine speed listed, I'd chase down the engine speed sensor, because if the computer doesn't see that the engine is spinning, it won't tell it to start.

Also, you mentioned having the timing belt off? If so, I'd check timing, because maybe the cam to crank is correct, but not the injector pump. Then you'd have fuel, but no fire because it's injecting at the wrong times. I watched your video, and the cranking sounded pretty good to me, at least then.

Sent from my KYOCERA-E6790 using Tapatalk

Relay in or out... it doesn’t matter, if the jumper is there. The relay is being bypassed.

I’m curious what’s happening with the engine racing symptom...

-Todd

I think I may be simultaneously both very lucky and very unlucky...

I pulled things apart again this afternoon to see how things line up - looks like they don't.

With the flywheel at TDC, I tried to insert the IP locking pin - it won't go in. When I rotate the IP pulley so that the pin goes in place, I can't even see the mark on the flywheel anymore. With the flywheel at TDC however, the cam locking plate slides into place with no issues, just like I'd expect. See pics below.







Correct me if I'm wrong, but I think this means there should be no damage to valves/pistons, but would explain why the car won't start.

So what's the correct procedure for getting things back in place? Obviously I'll be replacing timing belt and tensioner, but is it as simple as locking the cam and crank at TDC, then rotating the IP pulley back until the locking pin goes in?

Mozambiquer, thanks for the suggestions.

The filter is the correct A3 filter with the o-ring on the filter (not on the tee). Any filter issues would have been eliminated with the suspended fuel bottle.

The engine cranks well and doesn't sound abnormal, aside from the intermittent snapping sound that I described when I posted the video. I wonder now if that might be a bit of a 'cough' where a cylinder almost fires, but injection timing is too far off for it to get going properly

I cracked open the lines at the injectors, but can't see them when I'm in the car cranking to know how the stream of fuel looks I know i'm getting diesel out at all of them but that's about as much as I can say with confidence.

I didn't have the timing belt off, just the cover so I could do an inspection and see if the IP shaft seal was leaking diesel onto the belt. If the IP pulley is off by a tooth or 2, it happened on its own.

I pulled up measuring blocks while I was trying to start, and I could see data from the ESS (~400 RPM) so I don't think that's the issue either.