PD Engines - The Problem of Setting Torsion Value

Franko6

Vendor , w/Business number
Joined
May 7, 2005
Location
Sw Missouri
TDI
Jetta, 99, Silver`
Since first working on the PD motors, we found the setting of the torsion value a bit of a mystery. The torsion value is located in VCDS/Engine Module/ block 4/ torsion value. We tended to figure that 0.0 was a good figure and originally let it go at that.

However, early on, we had a Passat Wagon that lost 4 mpg when we set the torsion value to 0.0. The customer had us return the value to it's original setting of -.5 and the mileage came right back.

Now we find that the value can be positive or negative a much greater degree. Most recently, we checked the value on a car that was in for an oil change. The torsion value was set at -5.7 and the owner says her manual transmission car averages 45mpg and on a trip will get over 52mpg!

This is what we think is happening:

The only way to set the engine for TDC is by using a crank lock, such as that sold by Metalnerd. We have been providing their timing tools for several years. However well the tools are made, we find that the hole in the flange that is used to pin the crank sprocket into position has a movement of about 1-2 degrees. There is also a question about the accuracy of the location of the front flange location hole that the crank lock uses. I think the hole location in the flange could allow another 1-2 degrees of error.

The total deviation, given these two variables, could allow the crank setting to be off as much as 4 degrees, +/-. Since the cam position is relative to the crank timing, that can create a wide variation for each engine's correct torsion value. That would account for the reason one engine works well with a +2.5 reading, while another is working very well with a reading that is nearly -6.0.

There is no TDC mark on the flywheel,nor is there any other positioning reference besides the crank lock to locate TDC. That takes away any scientific method of relating the crank position to the cam position.

The only way I know to accurately locate the true TDC mark is to remove the #1 injector and with a dial indicator and find the piston rolling through TDC. I have noticed that in the Passat, there is a hole that is on the side of the block where a mark could be made. This may also be true of the other PD motors. I will look.

But unless there is a method by which you can determine accurately where the TDC actually is, the setting of the cam's torsion value will be experimentation.

Those wanting to optimize their fuel ecomony/ power should move the cam sprocket in relation to the cam shaft. Moving the cam shaft in the direction of rotation moves the cam torsion value in the positive direction. Just like the injection pumps for the ALH engine, the setting is very precise and a small adjustment makes a lot of difference. Movements of a few points and then either taking performance runs or recording mileage should give you an idea if there can be an improvement over the existing torsion value setting.
 

PDJetta

Top Post Dawg
Joined
Nov 6, 2003
Location
Northern Virginia
TDI
'04 Jetta GLS TDI Pumpe Duce Platinum Grey w/ Leather
Thanks Frank. That makes perfect sense.

All I know is that I was maticulous in using the lockdown tools when I replaced my timing belt and when I checked the torsion value, it was positive by a good bit. Possibly +3 or so. I lowered it to +0.5 and my mileage improved by about 3 MPG (I noticed no change in power or engine sound).

--Nate
 

boertje

Veteran Member
Joined
May 24, 2002
Location
Coeur d'Alene, ID
TDI
'01, '01, '03, ‘06 NB - TDIs all.
Thanks Frank, it does make sense. Keep us posted. I'm only getting about 42 with my bug. I lowered my torsion also to 0.5 from 3.5 and mileage increased from 39 to 42. I would like to get more though.
 

JETaah

Vendor , w/Business number
Joined
Jan 18, 2001
Location
mi 48836
TDI
96 B4V, 1999.5 jettaIV,2005 BEW Beetle
Wait a minute...
How does the torsion figure get established?
I know that the cam pulley hub has metal "flags" that pass by the cam position hall sensor and that lets the ECU know which injector should fire and when the cam lobes of #1 cylinder are at the intended TDC position but it has to compare that timing to a reference signal that is created by G28, the engine speed sensor(crank position sensor), no?

From the PD self study guide:

The signal generated by the Engine Speed
Sensor G28 provides both the engine speed
and the exact position of the crankshaft.
The injection point and the injection quantity are calculated using this information.


I would have to think that the crank's encoder wheel is probably more precisely located on the crank than the flange/alignment tool reference...so what you are reading in group 4 as the torque value is based upon the signal from that and the cam pulley hub relationship.
I suppose that there are factory tolerances for the crank's encoder wheel as well as the key way in the cam and hub that could add up to some error, but more than half a degree?? ya think?
 
Last edited:

Mako

Veteran Member
Joined
Apr 4, 2011
Location
Cape Town
TDI
AMFPolo 1.4 TDI 100%Biodiesel, AHF Golf TDI 100% Biodiesel, AHF Golf TDI Wagon 100% Biodiesel, VW Syncro 3CT Kombi 100% Biodiesel, Corsa 1.7D 100% Biodiesel
Purely on cam valve timing.......retarding the cam (-torsion) should equate to an increase in performance at the top end of the rev range with a loss at the bottom. Advancing cam valve timing (+torsion) should increase performance at the bottom with a loss at the top. The above certainly applies to n/a motors. On a PD changing the timing changes the point where the fuel pressure rises and peaks. The actual start of injection (SOI) will not change as the ECU decides when.
I don't doubt Franko6 and his interesting findings, just can't think of an obvious reason. Could it be to do with boost? Maybe later timing limits early boost and reduces losses or could the late timing extract more energy at the bottom end driving the piston more and the turbo less?
 

whitedog

Veteran Member
Joined
Jul 12, 2004
Location
Bend, Oregon
TDI
2004 Jetta that I fill by myself
So it sounds like the thing to do is to record mileage and record "Torsion" then adjust to -some- value and check mileage again. Repeat until what point? Is there some point of diminishing returns?

Frank mentioned "Those wanting to optimize their fuel ecomony/ power". So I am assuming that doing some power runs with VCDS recording may give more immediate feedback than running multiple tanks of fuel. It almost seems like I tried this one time, but I don't recall the results.
 

PDJetta

Top Post Dawg
Joined
Nov 6, 2003
Location
Northern Virginia
TDI
'04 Jetta GLS TDI Pumpe Duce Platinum Grey w/ Leather
I'm getting 50 MPGs in my PD (sometimes better) and I drive in rush hour in the Washington, DC area, so I am happy (although I am really easy on aceleration, brakes, and watch my top speed).

--Nate
 

runonbeer

Maintenance EnthusiastVendor
Joined
Apr 15, 2002
Location
Austin, TX/Chapel Hill, NC
TDI
'00 Golf 02M, '10 Golf 02E, '02 UTE 02M
Has anyone besides me ever noticed the hand written numbers on the timing belt cover. It's formatted the same as the group 4 field 4 number being -/+ digit decimal digit and there is a K and W next to it. German word for crankshaft: Kurbelwelle. German word for Angle: Winkel

Unit specific, factory cam timing spec? Maybe.

Polish people write the numeral 1 so that it looks like a capital italicized A without the dash connecting the slanted parts and that's what the 1's look like on the belt cover.

I looked into this a little deeper about a year ago but didn't uncover much.

I set torsion value more or less to this number (if it says -0.5, that's where I set it. A lot of them say -/+ 1.1 which I can't do. Makes me think they have a higher resolution tool at the factory for the measurement). When the cars come in, the number is typically considerably negative from the value stated on the cover. Belt stretch? Analogous to ALH pump timing dropping across the belt interval.

Just brain farting out loud here.
 
Last edited:

runonbeer

Maintenance EnthusiastVendor
Joined
Apr 15, 2002
Location
Austin, TX/Chapel Hill, NC
TDI
'00 Golf 02M, '10 Golf 02E, '02 UTE 02M
The wording on the cover next to where the number is handwritten translates to: "Attention: with every test and settings adjustment be aware (or note)".

I had a buddy who speaks German translate that for me.

More supporting evidence or not related at all? Who knows?
 

greengeeker

Vendor
Joined
Feb 8, 2006
Location
Cambridge, MN
TDI
2002 Jetta GLS
Polish people write the numeral 1 so that it looks like a capital italicized A without the dash connecting the slanted parts and that's what the 1's look like on the belt cover.
Not sure about the Polish but that's how the German's write a numeral 1 as well.
 

greengeeker

Vendor
Joined
Feb 8, 2006
Location
Cambridge, MN
TDI
2002 Jetta GLS
...so what you are reading in group 4 as the torque value is based upon the signal from that and the cam pulley hub relationship.
This is exactly how I understand it.
Your cam sensor relays it's postion and the crank sensor relays the crank angle (Kurbelwinkel (KW) ;)) and the ECU measures the degress of retard (negative values) or advance (positive values). Yes, the crank case cover and other associated parts have their tolerances but they are only there to ensure you get the cam/crank relationship close enough to not cause damage as well as start. You then have to go back and fine tune this relationship for optimum performance...exactly like the 1z/ahu/alh cars.

FWIW I rarely get a PD to land right on 0.0 the first time, they are usually in the -2-0 range and need to be adjusted. I have better luck with the VE belts landing right where I want them than I do any of the PD belts.
 

runonbeer

Maintenance EnthusiastVendor
Joined
Apr 15, 2002
Location
Austin, TX/Chapel Hill, NC
TDI
'00 Golf 02M, '10 Golf 02E, '02 UTE 02M
Not sure about the Polish but that's how the German's write a numeral 1 as well.
Yeah, it's probably most of Europe. Only reason I know about polish people doing this is because I fish salmon in AK with a dude from Poland. Kurwa! ;)

FWIW I rarely get a PD to land right on 0.0 the first time, they are usually in the -2-0 range and need to be adjusted. I have better luck with the VE belts landing right where I want them than I do any of the PD belts.
Hey me too!
 

PDJetta

Top Post Dawg
Joined
Nov 6, 2003
Location
Northern Virginia
TDI
'04 Jetta GLS TDI Pumpe Duce Platinum Grey w/ Leather
Has anyone besides me ever noticed the hand written numbers on the timing belt cover. It's formatted the same as the group 4 field 4 number being -/+ digit decimal digit and there is a K and W next to it. German word for crankshaft: Kurbelwelle. German word for Angle: Winkel

Unit specific, factory cam timing spec? Maybe.

Polish people write the numeral 1 so that it looks like a capital italicized A without the dash connecting the slanted parts and that's what the 1's look like on the belt cover.

I looked into this a little deeper about a year ago but didn't uncover much.

I set torsion value more or less to this number (if it says -0.5, that's where I set it. A lot of them say -/+ 1.1 which I can't do. Makes me think they have a higher resolution tool at the factory for the measurement). When the cars come in, the number is typically considerably negative from the value stated on the cover. Belt stretch? Analogous to ALH pump timing dropping across the belt interval.

Just brain farting out loud here.
Here is a picture of the label on my upper timing belt cover. I do not see the reference to the KW setting:



It looks like the number "126" is hand written in the box and someones initials are underneath the box. I also notice the date of September 15, 2003. I have a VERY early PD, manufacture date on the B pillar is October 2003.

--Nate
 
Last edited:

runonbeer

Maintenance EnthusiastVendor
Joined
Apr 15, 2002
Location
Austin, TX/Chapel Hill, NC
TDI
'00 Golf 02M, '10 Golf 02E, '02 UTE 02M
Indeed, no evidence to puzzle over on that one. I just went out and looked at a beetle I've got here and there is nothing ledgible remaining on the sticker at all. Appears to have 'bleached' out.
 

milktree

Veteran Member
Joined
Apr 8, 2011
Location
Eastern Massachusetts
TDI
2015 Golf Sportwagen, 2004 Jetta Wagon (sold!)
What're you talking about?

Please forgive my newbie question, but ... what is a "torsion value"?

From context, it looks like you can set the force on the timing belt tensioner to trim the cam timing by very small amounts by lengthening or shortening the belt's installed length.

But... that doesn't make much sense to me, I can't imagine the timing belt being that stretchy.
 

Tyrel

Veteran Member
Joined
Aug 19, 2011
Location
Peteborough, Ontario
TDI
2004 Jetta TDI PD BEW
So should we set the timing by the torsion value or the injector quality ? By BEW is now set at a torsion value of 1.5, but my injector quality is all over the map, going from -2.6 to 2.6 alternating between injectors.
 

Mako

Veteran Member
Joined
Apr 4, 2011
Location
Cape Town
TDI
AMFPolo 1.4 TDI 100%Biodiesel, AHF Golf TDI 100% Biodiesel, AHF Golf TDI Wagon 100% Biodiesel, VW Syncro 3CT Kombi 100% Biodiesel, Corsa 1.7D 100% Biodiesel
Your PD will change it's IQ all the time. Each injector can inject a different amount and measure the crank speed then adjust to maintain the desired speed.
 

whitedog

Veteran Member
Joined
Jul 12, 2004
Location
Bend, Oregon
TDI
2004 Jetta that I fill by myself
Your PD will change it's IQ all the time. Each injector can inject a different amount and measure the crank speed then adjust to maintain the desired speed.
He was actually talking about the injection Quality, not quantity. I don't recall the exact name, but it has to do with the correction of injection for each cylinder.
 

Mako

Veteran Member
Joined
Apr 4, 2011
Location
Cape Town
TDI
AMFPolo 1.4 TDI 100%Biodiesel, AHF Golf TDI 100% Biodiesel, AHF Golf TDI Wagon 100% Biodiesel, VW Syncro 3CT Kombi 100% Biodiesel, Corsa 1.7D 100% Biodiesel
OK I noticed that but thought Tyrel just messed up his terminology seeing he spells My as By ;-) The PD is capable of self adjusting what VP's need done on the bench.
 
Last edited:

Tyrel

Veteran Member
Joined
Aug 19, 2011
Location
Peteborough, Ontario
TDI
2004 Jetta TDI PD BEW
So your saying i don't need to worry about the IQ? Just trial and error on the torsion value to determine the best setting for fuel economy?

Thanks
 

whitedog

Veteran Member
Joined
Jul 12, 2004
Location
Bend, Oregon
TDI
2004 Jetta that I fill by myself
So your saying i don't need to worry about the IQ? Just trial and error on the torsion value to determine the best setting for fuel economy?

Thanks
You don't need to worry about IQ. Do some Trial and error on the KW and see how things work out,
 

Diveguy

Well-known member
Joined
May 25, 2008
Location
Portland, OR
TDI
04 Golf GLS TDI PD
Can the torsion value be changed with the VCDS? Or must the torsion value be achieved by moving the cam gear on the cam shaft?
 

Growler

Got Soot Vendor
Joined
Nov 24, 2003
Location
Millersport, Ohio
TDI
Schmutz, 2015 Golf Sportwagen DSG & Schnurren, 2001 Golf GL 2 door 5M
truman, I think the whole point of this thread is that it will be different for each car. and it will be a different number after each timing belt job.

all we installers can do is shoot for a good baseline of 0.0 or -0.5 and its up to you as the owner to determine where is the best setting for your car. unless you want to pay your installer for multiple torsion value adjustments spread out over time, to try to find the best setting.
 
Top