rough idle, 5-spd

[SBAtdijetta]

Just wanted to show the right place to check....

Vag-com -> 01 Engine -> Measuring blocks -> Block 4 -> Pic below... Right, Ok.

I checked this 2 seconds before and 4th block was -3.4 but then I shut the car down for a second, when I started the car again it had changed and -2.0 as seen below like most others. Good range is (2-5). I would like to go to about 3.7 and will adjust it tomorrow when I have time...

 

[aNUT]

I don't think that arbitrarily advancing your cam timing is a good idea. Remember that 1 cam degree is 2 crank degrees.

A better practice would be to lock the crank and cam;
loosen the three bolts that hold the cam sprocket to the hub;
and adjust the tensioner as necessary to put the tab in the slot.
You'll have to remove the harmonic balancer/serp drive pulley from the crank pulley.

In my experience, by the time these belts are at interval for changing, the belt has stretched sufficiently such that the tensioner arrow is no longer in the window in the backing.

Since there is more belt length on the waterpump (forward) side than on the tensioner side, the timing will retard as the belt stretches.

 

[TDI 4 RD]

When a qualified mechanic (like one of the gurus here) installs a new timing belt, is it standard procedure to check the timing with Vag-Com?

(I have a very slight shudder after running on the highway, and I'm wondering if I can wait until my timing belt will be changed, next Spring).

Thanks.

 

[SuperJ]

Just wanted to report that after a couple a weeks I have not a had a single shuddering idle incident, whereas before the adjustment it was a daily occurance for the last year. I running just over 5 degrees of advance now.

 

[SBAtdijetta]

Just wanted to report that after a couple a weeks I have not a had a single shuddering idle incident, whereas before the adjustment it was a daily occurance for the last year. I running just over 5 degrees of advance now.

Ok just finished getting everything put back together and rechecked with Vag-Com was at -2.0 now at 6.4......range should be (2-5).

Opinions should I go back and retard it some?

Currently my take........It is less out of range then it was originally just advanced rather than retarded. Everything sounds good on idle and a bit revved have not driven it yet though.

 

[velociT]

Anyone in or around central texas that knows how/has the tools to do this?

 

[syntrix]

Ok just finished getting everything put back together and rechecked with Vag-Com was at -2.0 now at 6.4......range should be (2-5).

Opinions should I go back and retard it some?

Currently my take........It is less out of range then it was originally just advanced rather than retarded. Everything sounds good on idle and a bit revved have not driven it yet though.

sba you will smoke a lot!
If its too bad bring it back.

 

[SBAtdijetta]

sba you will smoke a lot!
If its too bad bring it back.

Yea I went and test drove it ie: beat on it good. It was smoking more than normally but not a lot more. But then again I did just put 10 Gallons of B100 in it this afternoon (I don't normally have any Bio in as the Bio is to far away) so I bet it would be smoking more if it was ULSD + stanadyne only.

Power does not seem much if any more than normal with the RC1+ chip, then again the Bio thing.... I am running close to B70 and normaly run ULSD

Tomorrow I will take it all apart again and bring it back some. Hopefully to about (3-4) range. Im not willing to go any higher than that, plus I think that is more optimal, but what do I know anyway .

I really have to remember to only change 1 thing at a time.....

 

[aNUT]

Anyone in or around central texas that knows how/has the tools to do this?

If you'd only asked a week ago... I'm closing the shop to pack Tues. PM. I might be able to squeeze you in Mon. PM.

Back on topic:

It's essentially impossible to 'check timing' on a PD.

With an ALH/AHU external pump car, It's possible to manupilate the injection timing independently of the cam. When you put a belt on one of these cars, the cam and pump are locked, the belt is tensioned, and then the pump is fine tuned such that it can deliver the amount of advance requested by the ECU at all engine loads.

On a PD, the injection timing is computer controlled. The third set of cam lobes deliver the necessary pressure to the unit injectors about when the injection is supposed to occur. The injection itself is controlled by electronic valves immediately upstream of the nozzles.

In theory, if the cam timing is correct, the injector is pressurized for the full duration of the injection, the piezo valve on the nozzle opens for the prescribed duration, and the right amounts of fuel are injected, at the right time.

Based on what ya'll have found, in practice, the pressure in the injector drops before the injecton event has occured. This is most noticable at idle because the injection timing is most retarded at idle 'load'.

What ya'll have found is that by advancing the cam timing to *about where is should be, the pressurized injector phase is shifted such that proper pressure (and therefore fuel metering) is achieved.

The problem with advancing the cam timing without locking and reseting the tension is that proper valve timing is not guaranteed.

This is why it is better practice to lock the cam and crank instead of 'guessing' about the valve timing.

I'm not criticizing ya'll's actions; rather I applaud them. By experimenting with your cars, you have helped all understand the root of the problem. Now let's take that knowledge and solve the idle problem without guessing about something as critical as valve timing.

 

[velociT]

What are we finding out the sweet spot is?

 

[aNUT]

What are we finding out the sweet spot is?

I'm not entirely sure what 'sweet spot' you're talking about. Pressurized injector phase perhaps? VAG has no provisions for measuring the dwell of the pressurization of the unit injector.

What I'm suggesting is that:

Belt wear retards cam and thus injection timing.

Lock the cam and crank and reset tension.

This is advancing the timing in a more precise and safe manner.

Let the computer do the work of controlling injection timing, as the system was intended to operate.

 

[SBAtdijetta]

I'm not entirely sure what 'sweet spot' you're talking about. Pressurized injector phase perhaps? VAG has no provisions for measuring the dwell of the pressurization of the unit injector.

What I'm suggesting is that:

Belt wear retards cam and thus injection timing.

Lock the cam and crank and reset tension.

This is advancing the timing in a more precise and safe manner.

Let the computer do the work of controlling injection timing, as the system was intended to operate.

Anuthee,

I think I follow finally, but what belt wear is there on a car with less than 30k anyway?

I do think it should be advanced a bit by one method. My .02.

 

[aNUT]

Anuthee,

I think I follow finally, but what belt wear is there on a car with less than 30k anyway?

Well, the smartass answer is 30%...

If you want to be pessimistic, cam lobe and roller wear also reduce the dwell of injector pressurization.

I couldn't tell you how much the belt stretches. Perhaps it stretches more in the first half of it's life than in the latter half.

I can support that with some anectodatal evidence:

When Robby started doing TBs on ALHs, he had some complaints about lost mileage. Oddly enough, on an ALH, you can be in the graph with the pump requesting as little advance as possible, yet the injection timing is advanced about 1.5 degrees more than requested at idle. (specified .7 BTDC; actual 2.0 BTDC)

His car happened to be at the very top of the graph after he did his belt and he lost about 2 mpg. After about 20Kmi, the timing retarded slightly. (about 1 Y direction mark on the TDI timing graph) His actual timing matched specified, and the mileage came back.

 

[CADtechTDi]

Well, the smartass answer is 30%...

If you want to be pessimistic, cam lobe and roller wear also reduce the dwell of injector pressurization.

I couldn't tell you how much the belt stretches. Perhaps it stretches more in the first half of it's life than in the latter half.

I can support that with some anectodatal evidence:

When Robby started doing TBs on ALHs, he had some complaints about lost mileage. Oddly enough, on an ALH, you can be in the graph with the pump requesting as little advance as possible, yet the injection timing is advanced about 1.5 degrees more than requested at idle. (specified .7 BTDC; actual 2.0 BTDC)

His car happened to be at the very top of the graph after he did his belt and he lost about 2 mpg. After about 20Kmi, the timing retarded slightly. (about 1 Y direction mark on the TDI timing graph) His actual timing matched specified, and the mileage came back.

aNUT - what would be your DIY for a proper repair for this issue, beyond what has been pictured/posted on this thread?
I've read the timing belt change DIY - would the motor mount need to be removed (or half the engine dismantled) to lock the crank/cam shafts, etc.?

I'd really like to get rid of the shakes, but want to do it right!

 

[aNUT]

would the motor mount need to be removed (or half the engine dismantled) to lock the crank/cam shafts, etc.?

You won't have to remove the motor mount, but you will need the crank lock, cam lock pin, and tensioner spanner wrench.

How to: (from memory, thus no pics. If VelociT brings me a car before I skip town, I'll take some.)

Remove the upper intake pipe and upper TB cover to gain access to the belt (already pictured here)

Rotate the motor to TDC.

Lift and safely support the front right corner of the car.

Disconnect the lower intake pipe (turbo to intercooler) at the intercooler. Use some bailing wire to secure it to the control arm or sway bar. This will hold it back out of the way so it's not flopping in your way the whole time. You can also increase access down by the crankshaft pulley by cutting the steering wheel all the way to the right.

Wind the serp belt tensioner back and remove the serp belt.

Remove the harmonic damper/serp drive pulley.

You might have to remove the lower timing guard, but I don't think so.

Recheck that you're still at TDC.

Insert the crank lock.

Loosen the tensioner nut and wind the tensioner back. *your TB should have some slack now.

Insert the cam lock.

Loosen the three bolts securing the cam sprocket to the cam hub.

Tension the belt and snug then tensioner nut. (yeah, I know, it says always replace in the Bentley. Really though, it's the same design as an ALH, which is a reusable nut.)

Tighten the three cam sprocket bolts.

Remove the locks.

Rotate the crank 2X.

Insert the crank lock.

Make sure the CAM lock will go back in. (it might take some struggling...1 cam degree is 2 crank degrees.) You're just trying to verify that the hub isn't moving in relation to the sprocket and you won't smack a valve when you start the thing.

Remove the locks.

Reinstall:
Serp pulley, serp belt, side skirt. belly pan, lower intake pipe, upper timing cover, upper intake pipe,

No motor mount removal required.

 

[syntrix]

I think I finally see the physical concern that aNUT is preaching.

If you aren't experienced, you could BUMP the cam sprocket too much and tap a valve against a piston.

For those that have done this extensively on tdi's, even adjusting injection timing on alh's, a 1/2 to 1 mm bump is nothing out of the ordinary doing it the easy way.

I do wish that you could adjust injection timing separate of valve lift, but you can't (that I know of). I also find it hard to be believe that your timing belts give up that much free play in two direction (think tensioner balancing it out) that you are going to be this much out of spec.... which my though would be 1 degree cam, 2 degrees crank, and the equivalent of 4 degrees crank of wear out of the belt. Maybe it is 2 total, but you get the point.

I think that deviantspeed should do the write up and call it the "bump and soot" mod

 

[Spdmini]

So, after you do this, what happens after the timing belt change.. could one be in situation where it has been advanced it to much?

 

[oldpoopie]

So, after you do this, what happens after the timing belt change.. could one be in situation where it has been advanced it to much?

No chance of that if done properly. Crank and cam are locked and cam pully bolts are loosened in the belt proceedure anyway.

 

[SBAtdijetta]

I must say this is a pain at least for me to get it right were you want it to be. So be forewarned, I stared at -2.0 moved to 6.4, but felt that was too far so moved it back some. Now at 0.0.... lol maybe 3rd try is the charm. I want to see 3.0+/-

All I am saying is a very little goes a long way. Plus it is kind of a pain to take everything apart and put it back together, then check with Vag-Com to see how far you have moved it. Especially when you don't remove the last 90* elbow of the upper IC pipe reason why in below pic. This makes it a pain to get the timing belt cover off

Ok done complaining now time to test drive the snot out of it...

 

[syntrix]

What did you use for your boost gauge tap? Looks interesting!

Here's a hint on the adjustment. Take a pencil and draw around 1 or two of the 13mm's.

As you bump the 18mm cam bolt clockwise, you can see it's relation to the loosened bolts easily.

 

[SBAtdijetta]

Boostgauge tap was drilled w/ 1/8 NPT and taped with VDO gauge included 90* tap fitting, the other one there is the boostvalve line on a seperate tap.

Yea thanks for the tip. Finaly did that with a marker and just finished checking it with Vag-Com, after removing everything again and taking off the last 90* upper IC elbow too. It made it soooo much easier as the belt cover is supposed to come off in that direction.

But what was more helpful was that before I was turning the 18mm bolt with a wrench only by hand. As soon as I broke the force to move it I was too hard on it (to much force) and moved it more than desired. This time I used a bar to tap on the wrench to move it very slow and in very small increments.

Now set at +2.0 from original setting of -2.0 and I am satisfied. Lol done time for one last test drive.

 

[scrichy84]

Soothappens and I just started the car up without the boost pipe on to check the timing.

It is a little loud, but it didn't throw any codes and it ran fine enough to test the timing.

 

[SBAtdijetta]

Soothappens and I just started the car up without the boost pipe on to check the timing.

It is a little loud, but it didn't throw any codes and it ran fine enough to test the timing.

I thought about doing that so I would not have to put it all back together then take it apart again if it needed more adjustment but did not want to risk anything getting sucked in to the motor unfiltered.

Regardless doing that at anything more than idle, I think that would kind of bad. Ie: boost requested = x, actual = none.... plus risk from no filtering.

BTW my car runs great at +2.0, no idle problems, thats where she will stay.

 

[Mach1]

SB/anyone else interested, there is a window with a bunch of 'teeth' that one could use for a 'reference'. I marked mine then moved it by 1 tooth or about 3 degrees.

This helps reference how far you move it.

You can also run the car w/o the cover as well,

I leave the cam bolts just snug, then just tap the cam bolt, then check the mark.

I also greased the CAC pipe to ensure sealant.

Its a worthwild mod.

Just dont bump it too far to begin with..

 

[DeviantSpeed]

Diy

I have a written up a Walkthrough / DIY for this. I should have it hosted and posted shortly.

 

[TDikook]

this is VERY interesting.

 

[TornadoRed]

this is VERY interesting.

This is so interesting that I'm subscribing to this thread, and I don't even have a PD engine.

 

[SuperJ]

SBA, when I was past 6 degrees I would get a bit of jerk when I blipped the throttle at idle. It seemed really touchy (wouldn't shudder though).

 

[SBAtdijetta]

Yea I backed mine off from 6.4 to 2.0 was -2.0 to start with.

 

[Mach1]

SB, you aren't going to take it to 5?

You should be good at it now..