1.6TD - wheres my injection pump leaking from? (video)

[ORCoaster]

I would say so.  It sounds like a supply limiting situation.  At this point in your situation I would go back to running it in and out of a bottle of some sort.  A quart or Liter bottle to rule out the rest of the fuel system and verify the engine and all is good to go.  Chase one problem at a time otherwise you may compound the problem.

[TylerDurden]

Any chance you got the IN and OUT bolts mixed up?

I think you meant fuel in/fuel out? I used the hose closer to the driver side of the car which has the banjo vertically mounted against the pump body.

Belated followup: to clarify, the OUT bolt has a tiny orifice to restrict flow and create pressure in the IP for dynamic timing. If the OUT bolt mistakenly gets swapped with the IN bolt, the orifice will cause a greater vacuum in the lift section which can pull air into the IP from the main seal and starve the IP.

[ORCoaster]

Funny that my IP has them stamped as IN and OUT.  I can't mix them up.  Well, unless I wanted to.

[TylerDurden]

Funny that my IP has them stamped as IN and OUT.  I can't mix them up.  Well, unless I wanted to.

Sometimes it's not so easy to see the marking.



I haven't tried swapping them, but I bet the symptoms would be similar.

[ORCoaster]

That is an OUT bolt if the tiny hole on the side also has a screen like thing in the opening in the middle of the bolt itself.  IN bolts have a very noticeable hole drilled all the way across the bolt.  So actually two holes that allow fuel in and down the center. 

I will see if I can do pictures later this evening.  But it has been a tiring day.

[oblique]

Car ran fine from a bottle.

I pulled fuel from the IP supply hose with a little device I rigged up and it ran for a minute or two but starved again and died.

Update: Changed filter, blew out both hoses (supply->filter, filter->pump) and easily drew fuel from the hard steel line behind the engine coming from the tank.

Same story. Started fine and idled for a while then began to get rough and died.

The IN and OUT bolts I think are fine since my supply bolt has the large cross drilled openings.

Is this an issue with my IP?

[ORCoaster]

If you can pull fuel from the tank and through the filter to the IP and you can push fuel back to the tank from the IP and runs fine off a bottle I am really puzzled why it dies after a bit when properly hooked up.  Granted there is a greater vacuum needed at the front of the IP to get fuel there then if you run it off a bottle so that might be our clue here. 

Do you know if any of the four vanes in the ring inside the pump are sticking?  Remind us again if you dismantled this pump down to that point.  I almost want to guess that you are not developing a full lift on the front end of your pump.  Also remind us if you put a lift pump on it at any time during all of this.

Running rough sounds like not enough fuel to the injectors and that happens when the pump runs out of fuel or all the time? 

[oblique]

The pushing of the fuel back to the tank from IP I havent specifically confirmed but when I remove the supply fitting from IP after each run it is dry down there each time.

work done on IP:
shaft seal
cold start seal
high pressure valve seals
main seal (using the in-car method of stretching the seal around the injector line manifold square thing)

Never installed a lift pump in the car. I dont think this 84' Jetta (Candian) came with one from the factory.

I do not know if any vanes were sticking but when fed from a bottle, engine idles smoothly and revs up fine. When its all hooked up she slwoly starts getting rougher and rougher then dies after 30 seconds. Applying throttle has no effect.

Edit: One note. Libby's method changing the main seal with the pump on the car requires the seal to rest on the 4 bolts that hold the injector manifold square thing onto the pump body and backing out each bolt to get the seal past the bolts and onto the large round shaft where the seal's groove is. Its possible the while turning the bolts, the threads could cut into the seal. So if I compromised the seal in this way, would it explain my symptoms? There doesnt appear to be a leak there but could air be getting in past it?

[TylerDurden]

The big seal is not a problem.  Like said above, the supply /lift portions of the systems are suspect if the engine ran fine from the bottle.

I'd give the pump a good soaking in atf or diesel purge for a day and then rev it good while running a quart of the stuff to/from a container. 

[ORCoaster]

I don't think the main seal at the back has been compromised. 

It would be leaking if it was, as at that point there is plenty of pressure against it and no vacuum.

Isn't the process of replacing that O ring to pull the lines off the back of the pump manifold, loosen the four bolts a bit at a time to expose the old O ring, cut it out and then roll the new ring over the square part to the round part of the piece? 

Not much chance of cutting it or nicking it until you start bolt tightening sequence. 

Did you lube it up good with Vaseline or something before running the bolts home?

[oblique]

Yes, thats the process I used. You are better at describing it than I am

Lubed it with some wd40 before tightening the bolts.

So is my pump shot? Where do I go from here?

[ORCoaster]

Man what a head scratcher! 

I was re reading the thread and I picked up on this question you asked and as yet no one answered:

edit: is there a way independently verify crankshaft position?
« Last Edit: February 15, 2017, 10:27:15 by oblique »

 To do this without question I would be likely to just pull the #1 injector out and remove the heat shield.  I would rotate the engine to when the pointer and the flywheel mark are showing in the peek hole on the bell housing.  Then take a piece of wire, say something like a 14 gauge thickness and poke it down to the top of the piston in the cylinder.  Then slowly rotate the engine back and forth a bit to verify that the TDC of the #1 cylinder is aligned with the timing mark you are seeing in the transmission peak hole.  When the wire stops going up and starts going down you should be at TDC. Others here have used such a technique in the past.

What I can't understand is how it is loosing fuel other than by use and not getting it replaced.  Even off a bottle it should take it out, suck it into the IP, spray a little to the injectors and send the rest back to the bottle.  A quart or liter would run the car for a long time at idle.  Clear hose on the tank side would show you if air is being pushed out of the pump while it is running.  Have we already gone down that route? 

You have had electrical problems and seals that leaked on this pump from what I can tell.  When the engine is running for that brief amount of time is there still a good solid 12 VDC on that fuel solenoid?  If it drops off then it is like you hit the key. 

Then there was the replacement of the throttle bushing too was there not?  I wonder now if you have the alignment of that mechanism out of spec now.  So that even though you are moving the lever it isn't moving the rest of the collar and such inside the pump.  Taking off the throttle arm after you make a mark where it is at right now and shifting the whole thing backwards, counter clockwise one of those notches on the shaft, that might bring the throttle back to life.  I have had to do this on a couple of pumps in the past and just takes a little time.  Getting those springs to go back to where they were is the bugger for me.  I found using a pair of needle nosed vice grips was the way to grab them and pull them around to lock them in.

I can't believe a running pump has gone bad with the few things you have done so far. 

[TylerDurden]

The rig runs ok off a bottle...  supply or lift section, methinks.

[oblique]

ORCoaster, I think you hit the nail on the head early on with fuel supply. The motor runs (and is willing to rev) with the bottle however with everything hooked up and bled the the idle slowly but surely gets rougher and slower until it dies. Removing the fuel supply hose off the pump reveals the fuel cavity down there is empty.

As side notes to clarify some things, I never touched any of the throttle components at all. Those seals/gaskets are still original.
I was asking about TDC because I wanted to know how the heck I buggered up the pump sprocket by 90 degrees.

Tyler/ORC:
Could I have damaged the IP internally (or otherwise) by setting the sprocket 90degrees behind the first time? (still not sure how it ran like this)
I've rebuilt carburetors and a few motorbike engines before but those all had manuals so getting into the IP seems scary - will I benefit anything by opening up the pump? Assuming its not the fuel solenoid, what do I look for?

[TylerDurden]

I don't think it's possible to harm the IP by having it out of time, folks have run engines 180^o out of time.

Because it runs ok from the bottle,  I'm not convinced there isn't a restriction in the supply.  I would try running it off a bottle with the bottle 6-8 feet away on the ground... if it can pull fuel that far/high, the IP is likely fine, but the supply needs work.

Before I would open the IP, I'd try some cleaner (Diesel Purge, ATF/Naptha) like I mentioned before. If the vanes in the lift section got sticky, solvent and revving might get them freed up.