Injection Pump Issue - Runaway or Idle Miss

I'm just getting my Jetta back together after about 2 months of being apart. Prior to taking the car apart, it was stock aside from DLC1019 injectors, a malone stage 3 tune, and I put an 11mm pump head into the stock injection pump. It was running great, but of course, looking for more power.

Decided to convert to a VNT-17 turbo off of a BHW Passat. So changed to the BHW turbo/exhaust manifold. The head was rebuilt by Frank06. While it was apart, I had DBW clean/check the injectors, which they said were in great shape. Picked up a stage 2 cam/lifters at the same time. Put a 2.9bar MAP sensor in the ECU with a pre-liminary stage 5 tune from Malone. I don't yet have a downpipe on it (since NOBODY makes one for a BHW turbo on a mk3). I'm going to take it to an exhaust shop to have one made once i could actually drive it on/off a trailer.

In any case, I cranked it a bit with the injection pump unplugged to get oil up to the turbo. Plugged everything in and tried to start bleeding all the air from the lines. Took some cranking, but got it to start up and half way idle/rev up/etc. It had decent throttle response (important). Definitely needing more adjusting and more bleeding to get remaining air out.

Since it wasn't terribly easy to start, I adjusted the pump all the way toward the back of the car and it would start up great, but immediatly start to run away. Hammer modded the QA toward the passenger side of the car and it would run, but have a miss at idle and have zero throttle response. Tried adjusting the QA with the car running and the only options are to have the IQ down around 0.5 where it will miss at idle or it will start to run away. There's no in between.

I asked Owain at Malone for advice and his thought was an injection pump issue of some sort. I don't have an extra AHU pump and am really hesitant to simply pull my pump off and send it off to be rebuilt without knowing for sure what is going on. It's a lot of money to spend on something that may or may not fix the issue.

I checked block 19 and the min/max voltages are 0.75v/4.45v, which is within spec from what I can tell, which makes the think the QA is fine. Visually inspected the wiring to the pump (large and small plugs) and they both look ok. The collet did cause a little bit of chafing on the wire to the shut-off solenoid, but it didn't seem excessive (the collet was removed a while ago). I have not yet checked them with an ohm meter.

Not sure if it is worth mentioning or not, but I did replace the serpentine belt tensioner while everything was apart with one from IDParts. The replacement is larger than the factory tensioner actually touches the wiring to the solenoid underneath the pump head (cold start injector?).

https://www.idparts.com/serpentine-belt-tensioner-a3b4-p-677.html

So... Any ideas?

Anybody...?
Swapping out the ECU with a stock one made no difference.
I double checked to make sure that there wasn't any post MAF air leaks.
Triple checked all the wiring and it all looks fine.
There are no codes stored in the ECU.
While the IQ is low enough for it to idle with a miss, there is throttle response, which would mean the throttle wiring is ok.
Tried another ECU relay for the heck of it.

Not sure if it matters, but I did have the pulley off of the injection pump while I was changing the timing belt. I didn't see anything in the Bentley about there being a special procedure for putting the sprocket on. I can't imagine there'd be anything special anyway, since it's keyed.

I have no idea what's going on.

Are you sure your timing is right? Get your timing spot on physically and make sure its right in vcds, set the IQ to 3.5-4. If you cant get that to run, your difficulty having an issue with the internals of the IP.

What you mean, run away? how did you stop it? i have never heard of a TDI running away from a timing adjustment or hammer mod.

something aint right.

I'm cornfused.

You seem to be saying you made an arbitrary timing adjustment without checking it with VCDS which caused it to run away then you hammered it? I've never heard of a timing issue causing a runaway. This on a previously properly funcfuncfunctioning TC.

Why would a timing adjustment cause a runaway unless you pinched some wires and shorted something?

How are you running a VNT on an AHU? You didn't mention changes to accommodate a vacuum operated TC on a system designed for a TC with a pressure operated wastegate.

How much are you moving the pump when trying to set the IQ ? It's an extremely small adjustment going from choppy idle with low IQ to "run away" which I believe you mean an extreme high idle situation (2-3k rpms)

When I used to adjust IQ, I would loosen and then hand tighten the injector pump bolts, this way when I "adjusted" the top half it would only move a very small amount with each adjustment. Final torque to the injector bolts always moved the IQ about .3 or .5 mg

Ok, let me answer a few questions. I also took a video of what's going on. Checked the IP case pressure relief valve (and changed an 0-ring) and it was fine. Threw a new filter and thermostatic T on it, while I was at it.

Video starts immediately after start up. I never once touch the throttle throughout the video.

Mongler98 said:

Are you sure your timing is right? Get your timing spot on physically and make sure its right in vcds, set the IQ to 3.5-4. If you cant get that to run, your difficulty having an issue with the internals of the IP.

What you mean, run away? how did you stop it? i have never heard of a TDI running away from a timing adjustment or hammer mod.

something aint right.

Click to expand...

I checked all timing marks again today. Crank at TDC, pump pin slides right in, cam lock slides right in.

Just turn the key off and it stops. I've never let it get above about 4k rpm, so I'm not sure if it would keep going enough to self destruct. I'd prefer not to find out.

KLXD said:

I'm cornfused.

You seem to be saying you made an arbitrary timing adjustment without checking it with VCDS which caused it to run away then you hammered it? I've never heard of a timing issue causing a runaway. This on a previously properly funcfuncfunctioning TC.

Why would a timing adjustment cause a runaway unless you pinched some wires and shorted something?

How are you running a VNT on an AHU? You didn't mention changes to accommodate a vacuum operated TC on a system designed for a TC with a pressure operated wastegate.

Click to expand...

I've tried with the pump throughout the entire range of movement with the same result. It's quite a bit harder to get it to actually start if I move the pump all the way toward the radiator. I'm not sure where it falls on the chart in VCDS, because I can't get the engine to warm up enough to use the timing chart, so I'm not sure where it is on the chart. IQ is at 0.0-0.5. Any higher and the RPMs just start going up.

I am running a VNT on an AHU (or at least, trying). I am using the stock N18 (same part number as a mk4 N75) as an N75, with the vacuum lines run the same was as a mk4 (without the vacuum ball), and a Malone stage 5 tune.

greenskeeper said:

How much are you moving the pump when trying to set the IQ ? It's an extremely small adjustment going from choppy idle with low IQ to "run away" which I believe you mean an extreme high idle situation (2-3k rpms)

When I used to adjust IQ, I would loosen and then hand tighten the injector pump bolts, this way when I "adjusted" the top half it would only move a very small amount with each adjustment. Final torque to the injector bolts always moved the IQ about .3 or .5 mg

Click to expand...

I'm only moving the QA ever so slightly. Anything more than an IQ of 0.5 and the RPMs just start going up. I've always tuned it off before it got any higher than 4k RPM because I'm afraid to let it keep going and see if it stops.

Does the video work...? I can't get it to play. Is that just me?

Stromaluski said:

Does the video work...? I can't get it to play. Is that just me?

Click to expand...

Uploaded the video to YouTube, just in case it doesn't work above.

https://youtu.be/4H6hTS2q46M

That iq is crazy low. I’d try raising it to at least 3, to lean it out, and retard the pump a bit.

-Todd

ToddA1 said:

That iq is crazy low. I’d try raising it to at least 3, to lean it out, and retard the pump a bit.

-Todd

Click to expand...

If I raise it at all, it just runs away as soon as I start it rather than waiting 5 or 10 seconds like it did in the video. Will retarding the pump help with that?

It’s still a VE pump, and in my experience, retarding timing will lower the idle. It’s worth a shot.

Does the Malone tune include what’s needed to run the VNT? I thought it needs to be reprogrammed, in addition to whatever else that you’ve done.

-Todd

Stromaluski said:

If I raise it at all, it just runs away as soon as I start it rather than waiting 5 or 10 seconds like it did in the video. Will retarding the pump help with that?

Click to expand...

Is the quantity control properly seated in the injection pump collar? You'd have to remove the upper part of the pump to confirm.

ToddA1 said:

It’s still a VE pump, and in my experience, retarding timing will lower the idle. It’s worth a shot.

Does the Malone tune include what’s needed to run the VNT? I thought it needs to be reprogrammed, in addition to whatever else that you’ve done.

-Todd

Click to expand...

I'll give it a shot after work today. Can't hurt anything to try.

Malone said the tune should take care of it.

greenskeeper said:

Is the quantity control properly seated in the injection pump collar? You'd have to remove the upper part of the pump to confirm.

Click to expand...

I've probably put 5k miles on the car since the last time I took the QA off of the pump. I can't imagine that I would have been able to do that if it wasn't seated in the collar. And I can't imagine that it could have come out of the collar without it being taken apart. If I can't figure out anything else, though, I'll take the pump apart to check it...

Thanks for the splanation on the VNT ona AHU.

I wouldn't pull the QA off. If it was ok before there is no way for the pin to come out out of the collar. I might remove the cover and try moving the QA shaft to see if it's binding.

I agree with you that a pump that was working previously and hasn't been touched should work now. I've let mine sit for well over two months without running with no problems.

KLXD said:

Thanks for the splanation on the VNT ona AHU.

I wouldn't pull the QA off. If it was ok before there is no way for the pin to come out out of the collar. I might remove the cover and try moving the QA shaft to see if it's binding.

I agree with you that a pump that was working previously and hasn't been touched should work now. I've let mine sit for well over two months without running with no problems.

Click to expand...

I checked box 19 in VCDS and the max/min voltages are fine, meaning that the QA is sweeping correctly when the key is turned on. This morning, I did pull the top cover off and manually rotate the QA shaft to check, and it feels great through the entire range of motion.

Took a few days off from the TDI to work on a few other things around the house.

Checked for continuity between the Injection Pump and the ECU for all three wires on the small plug, and they're all fine.

Had an extra set of complete injectors that I pulled out of a car at the junkyard, so I threw them in. Injectors aren't the issue.

Pulled the injection pump all the way toward the radiator and tried raising the IQ a bit to compensate. It was harder to start, but otherwise no different.

I'm really hesitant to spend hundreds of dollars on another injection pump when I don't understand what could have gone wrong with the one I have simply from sitting for 2 months.

Just saw the end to your first post...Is the wiring to the cold start injector damaged from the incorrect pulley? That's the only thing listed that has "changed" since you last ran the car 2 months ago, and it ran fine then correct? If that N109 (?) valve is getting a funky reading it could mess with the idle and rpm quality/control

greenskeeper said:

Just saw the end to your first post...Is the wiring to the cold start injector damaged from the incorrect pulley? That's the only thing listed that has "changed" since you last ran the car 2 months ago, and it ran fine then correct? If that N109 (?) valve is getting a funky reading it could mess with the idle and rpm quality/control

Click to expand...

The wiring looks fine, it's just slightly touching the tensioner. Wish I hadn't thrown away my factory one before getting everything back together.

Judging by this thread, it seems like if the N108 is bad, it won't cause the issues I'm having, but for kicks, I went and checked the resistance of mine. 14.9 ohms, which is within spec.
https://forums.tdiclub.com/showthread.php?p=4851504

Forgot to mention, I also pulled off the lower intercooler hose just to make sure there wasn't oil in the intake; and there wasn't. Looked great.

Another thought. Is it possible I did something wrong putting it back together? This is the first time I've gone this far into a TDI, though I have built half a dozen gas motors from the ground up. The engine came with a 2 hole headgasket on it, and that's what I replaced it with. I didn't measure the piston protrusion (like I should have), and instead just went with the idea of replacing what was there.

Made sure everything the bottom end was at TDC and both #1 lobes of the cam were pointing up with the cam lock in place before setting the head back on. Used ARP head studs torqued in progression to 125 ft lbs. like the instructions from ARP say to do.

I can't imagine there's much that I could have done wrong, but just throwing it out there as a possibility.

I’m still thinking it’s fuel related. Are the plugs seated 100%? I’m guessing you don’t have a spare pump, to toss on and test with?

-Todd

ToddA1 said:

I’m still thinking it’s fuel related. Are the plugs seated 100%? I’m guessing you don’t have a spare pump, to toss on and test with?

-Todd

Click to expand...

The glow plugs? Torque spec on them is only 11 ft lbs, isn't it? I rechecked them while I was changing injectors and they're still tight.

I do not have another injection pump, no, which was part of the reason for starting this thread. To try to see if somebody could tell me what was wrong with mine before I spend a few hundred dollars or more to buy another one.

No, the electrical plugs, although it’s a long shot.

-Todd

Oh, yeah, they're both seated completely. I've had both of them unplugged probably a dozen times while checking things.

At this point, I’d probably open the pump to verify everything looks normal.

-Todd

If the pump ran fine prior, then the pump isn't the problem. It's something with the install (pulley hitting the N109 wiring).

ToddA1 said:

At this point, I’d probably open the pump to verify everything looks normal.

-Todd

Click to expand...

I'm just about to this point of taking the pump off and beginning to disassemble.

greenskeeper said:

If the pump ran fine prior, then the pump isn't the problem. It's something with the install (pulley hitting the N109 wiring).

Click to expand...

The wiring to the N108/N109 seems fine, though. N108 Resistance measures correctly; wiring all visually looks fine, and has continuity. Not sure how else to check it?

Stromaluski said:

The wiring to the N108/N109 seems fine, though. N108 Resistance measures correctly; wiring all visually looks fine, and has continuity. Not sure how else to check it?

Click to expand...

It may not be that. My reasoning is that you stated the pump was fine prior to sitting for 2 months, so the pump is not the problem but something related to putting the car back together.

greenskeeper said:

It may not be that. My reasoning is that you stated the pump was fine prior to sitting for 2 months, so the pump is not the problem but something related to putting the car back together.

Click to expand...

I very much agree that the pump should be fine after sitting for 2 months, but I've checked everything that I did mess with while it was apart. Timing is good. Had the same problem after swapping both ecu and injectors. Checked for oil in the intercooler piping, and that was fine. I've pretty much run out of other things to check.

Pulled the pump and disassembled. Bore is wiped out and the advance mechanism felt terrible because of it. Not sure if that's the cause of this problem, but it's definitely a problem.