Yet another mapped points question

[murfie]

You wouldn't split the amount of advance you want by to the weight % splits. The percentage isn't telling us how much we need to do to any table... spark, torque, or SD. It is only telling us the spots in the calibration its getting a final value from. Then if it has snapped to a point or line, it might not even be telling us that accurately. The weight follows the physical cam positions, and their error, not the calibration values being used. If you wanted to add 3*, you would add 3* to every table that had any weight put on it. Just like putting it in the global table. The weight would spread out and all partial percentages would get the increase of MP% multiplied by the +3*, summing together to get the final total +3*. If you split 3* like the mapped point weight, you will get the percentage applied twice, once by you, and again by the ECU, and not get the increase in advance you are looking for.
K44 seems to have described it appropraitely reffering to the cells the weight hits and not applying a weight percentage to what you want to adjust it by.
I think a lot of people try to use the mapped point percentage like its an error percentage or correction factor, and they get no where in fixing issues. Things don't go the way they think it should, or as fast as they think it should, and many give up.
This is a very important concept to grasp when it comes to the load to torque/ torque to load.

The snap to points and lines are the spot calibrators best friends.

[murfie]

I've been able to cut down some of the work by roughly matrixing the mapped points into different driving conditions. For example, on my car there's a 99.9% chance I'll never use MP15 at idle. So I don't tune that area. Then you can focus on the ones that get blended frequently in Best Driveability and Fuel Economy.


In 4cyl world, we start to lose the exhaust cam around 650whp and 7000+ RPM. The valvetrain components definitely have an impact (ie. 85lb springs will do it more quickly). The oil pressure is irrelevant because it only has the responsibility/ability of retarding the exhaust cam. It's the phaser spring that returns the cam position to zero, and which can't overcome the various forces acting against the cam. That's why we see 0% duty cycle from the solenoid while the exhaust cam is moving on it's own (the PCM doesn't tolerate much of this before entering limp mode either). The intake cam doesn't do this because the phaser advances the cam in the same direction as the engine rotation. At 0% DC the cam is already mechanically locked against the backside of the phaser. You could technically run out of solenoid duty cycle if you were trying to advance the cam.

I'm not sure if this is the same in the coyote world. A lockout solves the VCT side until the cam gear itself slips. I've considered pinning both cams along with the crank woodruff key, and locking the exhaust phaser.

1000+whp Coyotes is pretty far off path for the OP, FWIW.

0% duty cycle(off) is the locked position, that could be end or mid phase, they go there passively. When the solenoid is active and the phaser is not in the locked base position, they have oil pressure on both sides of the vane to move it in either direction. The solenoid causes an imbalance in pressure that advances or retards the cam. In theory ~50% duty cycle equalizes and holds the cam stationary at that position. In practice the control loop does a lot of work finding the exact duty cycle that is needed for the appropriate actions.
You can see what I describe in the duty cycle feed forward models, this has been true for both intake and exhaust cam phasers up until 18+ car the exhaust cam where the hold position is a bit lower duty cycle. Exhaust may not be very important to WOT, but for MPGs and emissions it is.
The intake and exhaust respond opposite to duty cycle.

Nearly all of these tables refer to engine oil temperature.
I would say friction, oil pressure, and oil temperature are far more relavent to the camangle error, than the power output of the engine.

You can modify these slopes to be more or less agressive as needed outside the steep parts at 0 deg/s for things like bigger cams and heavier valve springs. With high RPM you could potentially run near 0 or near 100 duty cycle and be at the limit of the phasers.

11 vs 15 vs 18 VCT duty cycle.jpg

[veeefour]

I've found this to be the case on my car as well after nearly a thousand logs.

FINALLY! We know your "only" experience is YOUR car...thank you, come again.

[veeefour]

0% duty cycle(off) is the locked position, that could be end or mid phase, they go there passively. When the solenoid is active and the phaser is not in the locked base position, they have oil pressure on both sides of the vane to move it in either direction. The solenoid causes an imbalance in pressure that advances or retards the cam. In theory ~50% duty cycle equalizes and holds the cam stationary at that position. In practice the control loop does a lot of work finding the exact duty cycle that is needed for the appropriate actions.
You can see what I describe in the duty cycle feed forward models, this has been true for both intake and exhaust cam phasers up until 18+ car the exhaust cam where the hold position is a bit lower duty cycle. Exhaust may not be very important to WOT, but for MPGs and emissions it is.
The intake and exhaust respond opposite to duty cycle.

Nearly all of these tables refer to engine oil temperature.
I would say friction, oil pressure, and oil temperature are far more relavent to the camangle error, than the power output of the engine.

You can modify these slopes to be more or less agressive as needed outside the steep parts at 0 deg/s for things like bigger cams and heavier valve springs. With high RPM you could potentially run near 0 or near 100 duty cycle and be at the limit of the phasers.

You have to be more specific:
-1 gen coyote has phasers without locking function
-2 gen coyote has a midlock function on phasers
-3 gen coyote has on midlock on intake only

Sorry but you have never pushed the car that far to know, oil temp and friction has nothing to do with this. How I know? because I build couple +1000hp engines....

The other factor is the cam sprocket is not round but rather has 4 ovals to compensate pulses in valvetrain. This will not help you on
high power level or high rpm applications. You have to address that with additional short chain guide or you will be replacing
guides every 3rd pass. We tested MMR short chains they are fine 2nd year in +1000hp engines but were destroyed in N/A 9000+ rpm engine after 2 months.

How I know that? Because I build +700hp N/A 5.2 engine that chew guides few times above 9000rpm...

Now...go take your theories to the field and find how wrong you are. I will watch...take Mike with you grab a beer sit there and learn.

[junits15]

You wouldn't split the amount of advance you want by to the weight % splits. The percentage isn't telling us how much we need to do to any table... spark, torque, or SD. It is only telling us the spots in the calibration its getting a final value from. Then if it has snapped to a point or line, it might not even be telling us that accurately. The weight follows the physical cam positions, and their error, not the calibration values being used. If you wanted to add 3*, you would add 3* to every table that had any weight put on it. Just like putting it in the global table. The weight would spread out and all partial percentages would get the increase of MP% multiplied by the +3*, summing together to get the final total +3*. If you split 3* like the mapped point weight, you will get the percentage applied twice, once by you, and again by the ECU, and not get the increase in advance you are looking for.
K44 seems to have described it appropraitely reffering to the cells the weight hits and not applying a weight percentage to what you want to adjust it by.
I think a lot of people try to use the mapped point percentage like its an error percentage or correction factor, and they get no where in fixing issues. Things don't go the way they think it should, or as fast as they think it should, and many give up.
This is a very important concept to grasp when it comes to the load to torque/ torque to load.

The snap to points and lines are the spot calibrators best friends.

omg you're completely correct lmfao. Yes the total weight is always 100% I was doing this totally backwards, if we're weighting 14 tables by different amounts the delta needs to be the same in order to yield a fixed outcome.

I was treating it like each table had a fixed contribution to the total and the ECM was just clicking them in or out. This explains why it wasn't adding nearly what I expected to the timing curve.

This is what happens when I attempt to think after work. I swear I know math, I promise lmao

[RobCat030]

0% duty cycle(off) is the locked position, that could be end or mid phase, they go there passively. When the solenoid is active and the phaser is not in the locked base position, they have oil pressure on both sides of the vane to move it in either direction. The solenoid causes an imbalance in pressure that advances or retards the cam. In theory ~50% duty cycle equalizes and holds the cam stationary at that position. In practice the control loop does a lot of work finding the exact duty cycle that is needed for the appropriate actions.
You can see what I describe in the duty cycle feed forward models, this has been true for both intake and exhaust cam phasers up until 18+ car the exhaust cam where the hold position is a bit lower duty cycle. Exhaust may not be very important to WOT, but for MPGs and emissions it is.
The intake and exhaust respond opposite to duty cycle.

Nearly all of these tables refer to engine oil temperature.
I would say friction, oil pressure, and oil temperature are far more relavent to the camangle error, than the power output of the engine.

You can modify these slopes to be more or less agressive as needed outside the steep parts at 0 deg/s for things like bigger cams and heavier valve springs. With high RPM you could potentially run near 0 or near 100 duty cycle and be at the limit of the phasers.

11 vs 15 vs 18 VCT duty cycle.jpg

I'd have to inspect my ecoboost more carefully to see if its the same in terms of the solenoid allowing pressure to both sides. Friction and oil pressure are obviously a factor in the solenoid operation, we agree there.

The situation I was describing is when the solenoid is at 0% and the exhaust cam still wants to walk due to some other forces acting against the camshaft (exhaust back pressure, upgraded HPFP, valve springs? Not 100% sure). Its not directly power related but there's a loose correlation between the two. In this situation there isn't anything the solenoid can do to "pull" back the exhaust cam to its locked position, it's already at 0. Unless having the duty cycle at 1% somehow creates a massive oil pressure imbalance on the advancing side of the cam. It's a situation I've never been able to calibrate around, which always led me to believe its a mechanical limitation of VCT

[engineermike]

FINALLY! We know your "only" experience is YOUR car...thank you, come again.

Where did I say "only"? Maybe something was lost in translation?

Sorry but you have never pushed the car that far to know, oil temp and friction has nothing to do with this. How I know? because I build couple +1000hp engines....

I can say with 100% certainty that oil temp has an effect on phaser control. I have logs where the oil-actuated phaser loses control worse and worse as oil temp goes up. This was on multiple cars (apparently, I have to specify that). As murphie pointed out, Ford believes this to be true as well since so many cam control tables are a function of oil temp. If you understand how phaser control works and how oil viscosity changes with temperature, it's not hard to understand. The oil has to pass through the front journal bearing (supply and return) to get to the phaser and the phaser itself has internal clearances. As viscosity decreases with temp, there is more leakage from the supply to the return control passages through the journal bearing, bypassing the phaser completely, and also leakage out of the bearing plus more leakage around the vanes inside the phaser. I've also seen this phenomenon get worse as the motor oil ages and loses viscosity due to shear, though sadly I only have one example of this.

Now...go take your theories to the field and find how wrong you are. I will watch...take Mike with you grab a beer sit there and learn.

More childish cut-downs. You may not realize this doesn't give you more credibility, at least in our culture. In fact, it gives you less.

I would like to point out, though, that you do admit that a) cam timing affects knock and b) in some cases you lose control of cams. If you believe both of those to be true, that reinforces the point I made in post 26 to which you took exception and offense.

[engineermike]

I'd have to inspect my ecoboost more carefully to see if its the same in terms of the solenoid allowing pressure to both sides. Friction and oil pressure are obviously a factor in the solenoid operation, we agree there.

The situation I was describing is when the solenoid is at 0% and the exhaust cam still wants to walk due to some other forces acting against the camshaft (exhaust back pressure, upgraded HPFP, valve springs? Not 100% sure). Its not directly power related but there's a loose correlation between the two. In this situation there isn't anything the solenoid can do to "pull" back the exhaust cam to its locked position, it's already at 0. Unless having the duty cycle at 1% somehow creates a massive oil pressure imbalance on the advancing side of the cam. It's a situation I've never been able to calibrate around, which always led me to believe its a mechanical limitation of VCT

On the exhaust side of a Gen3 Coyote, the solenoid valve sends all the oil flow to the advance side of the vanes at 0% duty cycle. At 100%, it sends the oil to the retard side of the vanes. Of course, the spiral spring is supposed to help overcome torque in the advancing direction. The solenoid is an interesting component. At mid-stroke, it sends oil equally to both sides of the vanes as opposed to shutting it off. There has to always be oil flowing to the phaser because this is also the lubricating oil for the front cam journal. When the phaser solenoid moves off of the middle, it opens the supply passage more (be it advance or retard) and it relieves the "return" side oil passage back to the crankcase to allow the oil on the non-pressurized side somewhere to go. I actually bench tested the plunger position as a function of voltage applied and confirmed the Feed-Forward DC vs flow area to either side, then went on to modify the plunger length and shim the spring to give it more span of control in the advancing direction. It helped but didn't solve the problem.

One other point is that the zero-lock plunger in the phaser can't help us any. That pin is pushed out of its hole as soon as it sees oil pressure, so it can't move back into the lock position until the engine has no oil pressure.

Interesting you note the upgraded HPFP. The problem is much worse when an XDI pump is used. I e-mailed XDI about the issue and they did not respond.

I've worked quite a bit to calibrate around it as well, but once the phaser solenoid goes to 0% duty cycle, there's nothing else the calibration can do. The most effective thing I've found so far is to increase high temp oil viscosity.

[veeefour]

I can say with 100% certainty that oil temp has an effect on phaser control. I have logs where the oil-actuated phaser loses control worse and worse as oil temp goes up. This was on multiple cars (apparently, I have to specify that). As murphie pointed out, Ford believes this to be true as well since so many cam control tables are a function of oil temp. If you understand how phaser control works and how oil viscosity changes with temperature, it's not hard to understand. The oil has to pass through the front journal bearing (supply and return) to get to the phaser and the phaser itself has internal clearances. As viscosity decreases with temp, there is more leakage from the supply to the return control passages through the journal bearing, bypassing the phaser completely, and also leakage out of the bearing plus more leakage around the vanes inside the phaser. I've also seen this phenomenon get worse as the motor oil ages and loses viscosity due to shear, though sadly I only have one example of this.

Never said oil has no effect on VCT control, of course it has you have a bunch of tables to control phasers depending on oil temp.

What you say is that you should tune out mechanical problems should have VCT phaser control problems...wired flex but ok.
When you oil becomes "water" just change it...we had 666 cases +2018 coyote owners coming in the shop with stalling issues
the were cured with simple oil change...in maybe 5 out of those 666 we replaced the phaser as it was the culprit.
But took the phaser apart, cleaned it and it was working like a new. 3 gen has no filtering screen for exhaust phaser which 1-2 gen had.


You like to blah blah blah your way into sounding smart oh you do...

[engineermike]

Never said oil has no effect on VCT control...

You literally said in post 44:

oil temp and friction has nothing to do with this.

What you say is that you should tune out mechanical problems should have VCT phaser control problems...wired flex but ok.

I thought the job of the calibrator was to calibrate the controls to match altered physical characteristics of the engine or driver preference. What you are saying is akin to saying you shouldn't have to calibrate for a supercharger, and the better solution is to remove the supercharger. If you add, say, an XDi HPFP that takes more cam torque to drive, then it is not so surprising that cam control needs to be re-calibrated to match the new physical characteristic.

When you oil becomes "water" just change it...

If only it were this simple. Not all loss of cam control is due to oil degradation. Oil selection and change interval must match the application and varies widely...not as simple as "just change it".

You like to blah blah blah your way into sounding smart oh you do...

Did I post anything technically incorrect? Are you not capable of keeping a post technical and acting like an adult?

[murfie]

I'd have to inspect my ecoboost more carefully to see if its the same in terms of the solenoid allowing pressure to both sides. Friction and oil pressure are obviously a factor in the solenoid operation, we agree there.

The situation I was describing is when the solenoid is at 0% and the exhaust cam still wants to walk due to some other forces acting against the camshaft (exhaust back pressure, upgraded HPFP, valve springs? Not 100% sure). Its not directly power related but there's a loose correlation between the two. In this situation there isn't anything the solenoid can do to "pull" back the exhaust cam to its locked position, it's already at 0. Unless having the duty cycle at 1% somehow creates a massive oil pressure imbalance on the advancing side of the cam. It's a situation I've never been able to calibrate around, which always led me to believe its a mechanical limitation of VCT

The springs are there to return the cam to the base position when the engine shuts off.
Its the hydralics that move the phase postion of the cams and they are very just as robust as any other hydralics.
The forces acting against the oil pressures in the vanes that are holding the cams position are at a severe leverage disadvantage. It would need to be a constant very extreme force.

Even fully accelerating a free reving engine that is capable of 1000+hp, under normal operations I just don't see the cam shafts becoming that hard to slightly speed up/ slow down. Not for a hydralic system.
I have not heard from anyone reputatable, a stock gen 2/3 coyote with stock cams and phasers hitting an engine power output that limited their control. In fact many brag about making xxxx and still having the full VCT control.
Now if you increased their duration a ton, and don't want to allow them in all the available phase range or any at all, I could see limiting or locking them physically and not trusting that job to the hydralic control.

None of this has anything to do with the work the PID feed back loop has to do to control the cams with the proportioning solenoid, which is active and working hard on a stock engine at light acceleration.

[RobCat030]

The springs are there to return the cam to the base position when the engine shuts off.
Its the hydralics that move the phase postion of the cams and they are very just as robust as any other hydralics.
The forces acting against the oil pressures in the vanes that are holding the cams position are at a severe leverage disadvantage. It would need to be a constant very extreme force.

Even fully accelerating a free reving engine that is capable of 1000+hp, under normal operations I just don't see the cam shafts becoming that hard to slightly speed up/ slow down. Not for a hydralic system.
I have not heard from anyone reputatable, a stock gen 2/3 coyote with stock cams and phasers hitting an engine power output that limited their control. In fact many brag about making xxxx and still having the full VCT control.
Now if you increased their duration a ton, and don't want to allow them in all the available phase range or any at all, I could see limiting or locking them physically and not trusting that job to the hydralic control.

None of this has anything to do with the work the PID feed back loop has to do to control the cams with the proportioning solenoid, which is active and working hard on a stock engine at light acceleration.

Agreed on the last point, we drifted from the original topic about 20 posts ago when the appendage measuring contest began. No harm in exploring the idea now that we're here though.

I'm not a tuner of 1000+ Coyotes so I certainly can't speak to that. But the issue has been very real on the 4cylinder Ecoboost/Duratec for me. I guess I still feel a little lost in the weeds as well. If the solenoid is at 0%, then there's nothing the calibration can do to prevent the unwanted/uncommanded phasing of the exhaust cam (that's my understanding, but I'm having trouble understanding if you agree with that statement). I have logs that show this; the only dispute I could see is some type of logging error.

Not my tune, but a log sent to me to verify this exact issue:
https://datazap.me/u/robcat030/log-1...a=4-7-23-24-26

[murfie]

Agreed on the last point, we drifted from the original topic about 20 posts ago when the appendage measuring contest began. No harm in exploring the idea now that we're here though.

I'm not a tuner of 1000+ Coyotes so I certainly can't speak to that. But the issue has been very real on the 4cylinder Ecoboost/Duratec for me. I guess I still feel a little lost in the weeds as well. If the solenoid is at 0%, then there's nothing the calibration can do to prevent the unwanted/uncommanded phasing of the exhaust cam (that's my understanding, but I'm having trouble understanding if you agree with that statement). I have logs that show this; the only dispute I could see is some type of logging error.

Not my tune, but a log sent to me to verify this exact issue:
https://datazap.me/u/robcat030/log-1...a=4-7-23-24-26

I'll see what's in that datazap later.

Sorry if I'm being to general here, but it's information that I think applies to many applications and cam phasers.
If the solenoid is at 0% it's commanding a change in the cam position and has not achieved it for a good bit of time. The same goes for 100% but the commanded position is the opposite direction. Something is probably physically wrong with either the position sensor or the phaser if it's having to do this. An example would be a lockout or limited kit being installed, and commanding angles the phaser can no longer get to.

If it achieved the position it would target 50% because that's equal pressure on both sides of the vane meaning no movement. Even if the commanded position was one of the limits of the phaser, it's not going to hold position with anything but 50%.

I feel like I shouldn't have to say it, but this is all subject to the PID feedback loop, so when I say 50% it might really be in a range 45-55%, dealing with the small accelerations and decelerations the engine RPM has or other variances like oil pressure.

The cam, the oil, the sprocket all rotate together so the force to phase isn't actually that much. It doesn't change much either. Under normal operating temperature with bearings properly being fed oil, the amount of friction is much less than you would expect. That's why using hydraulic oil pressure of 20-80psi is adequate, and it doesn't need to be crazy high like you see in heavy machine rams.
Trust me or don't, if it changed significantly with torque output of the engine, there would be some correction table for the feed forward tables based on that. There would be some mention of it in at least one of the numerous patents on phasing cams.

[murfie]

Instead of commanding 0, try commanding 3-5* and never 0 anywhere.

I would be confident the phaser DC doesn't get pegged at 0% doing that. If this is an eco boost their feed forward tables are around 40% for the hold position.
37-39% seems to be where your phaser is able to hold the cams position, and it does that fine until you command less than 3-5* and that error is there for a few seconds not being corrected. It is still there even on decel.
If this is a coyote and its holding position at 37-39%, and not near 50, thats an indication of an issue.

as you stab the throttle:
Shooting up into the 50's it moves it oneway
Slightly after the pull starts:
Shooting down into the 30's it moves it the other.
The lag between desired and actual, thats all normal.
making the FF tables more aggressive tightens this lag up.

Why can't you command 0 and achieve that position ever? idk some physical problem with the phaser or position sensor. Maybe the exhaust timing is off a tooth.
Phaser duty cycle and error.jpg

When you start this car up, is the exhaust cam at 0*? When the engine is off does it show 0*?

You have other logs, different cars, that show exhaust cams reaching 0. not logged was desired, duty cycle, or error, so I cant show those and what they should be doing. What you do see is its not exactly pegged at 0, its holding it there with small variations just like other positions.
exhaust angle at 0.jpg

[RobCat030]

Instead of commanding 0, try commanding 3-5* and never 0 anywhere.

I would be confident the phaser DC doesn't get pegged at 0% doing that. If this is an eco boost their feed forward tables are around 40% for the hold position.
37-39% seems to be where your phaser is able to hold the cams position, and it does that fine until you command less than 3-5* and that error is there for a few seconds not being corrected. It is still there even on decel.
If this is a coyote and its holding position at 37-39%, and not near 50, thats an indication of an issue.

as you stab the throttle:
Shooting up into the 50's it moves it oneway
Slightly after the pull starts:
Shooting down into the 30's it moves it the other.
The lag between desired and actual, thats all normal.
making the FF tables more aggressive tightens this lag up.

Why can't you command 0 and achieve that position ever? idk some physical problem with the phaser or position sensor. Maybe the exhaust timing is off a tooth.
Phaser duty cycle and error.jpg

When you start this car up, is the exhaust cam at 0*? When the engine is off does it show 0*?

You have other logs, different cars, that show exhaust cams reaching 0. not logged was desired, duty cycle, or error, so I cant show those and what they should be doing. What you do see is its not exactly pegged at 0, its holding it there with small variations just like other positions.
exhaust angle at 0.jpg

The first suspicion with that car was that the friction washer gave out. So it was replaced and re-timed quite a few times. We also verified that it was only really happening as the car approached 65lbs/min. My datazap account is poorly organized but I will try to find those other logs. But if the timing slipped by 2-3* that still isn't a great explanation of the phaser moving almost 10* at the top of the pull.

I do understand your explanation of the ideal operation though, thank you taking the time to say that a few different ways. 50 holds it, >50 pushes it, <50 pulls it, PID loop picks up the slack. I'm buttoning up another ecoboost here soon, so I may try to key the cam (without a phaser lockout) to rule out mechanical timing and see if I can get 3-5 degrees work. I'd love to maintain VCT if possible.

[murfie]

The first suspicion with that car was that the friction washer gave out. So it was replaced and re-timed quite a few times. We also verified that it was only really happening as the car approached 65lbs/min. My datazap account is poorly organized but I will try to find those other logs. But if the timing slipped by 2-3* that still isn't a great explanation of the phaser moving almost 10* at the top of the pull.

I do understand your explanation of the ideal operation though, thank you taking the time to say that a few different ways. 50 holds it, >50 pushes it, <50 pulls it, PID loop picks up the slack. I'm buttoning up another ecoboost here soon, so I may try to key the cam (without a phaser lockout) to rule out mechanical timing and see if I can get 3-5 degrees work. I'd love to maintain VCT if possible.

Think about my explanation and from the oil pressure needing to be balanced to hold the position. When the dc goes to 0 failing to achieve the desired position, it traps the available pressure at that moment on one side. As RPM increase the oil pressure changes on the other side and only on that side.

The oil goes out the same passageway it goes in to the vanes.

[engineermike]

Murfie, in contemplating your posts above, you're saying that the hydraulic pressure should easily overcome the cam torque and be able to reach target cam angle. The only forces it has to overcome are friction and GDI pump drive, as the valve-spring energy is absorbed but then deposited right back into the cam so you break-even. According to BW, the spiral spring is there to work against "retard bias torque" and they also refer to it as a "torsional assist". I actually took the time to measure the amount of moment the spring applies, calculate the drive power of the GDI pump, and calculate the hydraulic torsion the phaser can apply and I came to the same conclusion that you did. The hydraulic torsion plus the spiral spring torsion should be multiples more than you need to prevent the exhaust cam from falling out of control and retarding, even with an upgraded GDI pump.

Once I determined the oil temp correlation to the problem and found success in increasing viscosity, the only plausible conclusion I could come up with is that the phaser vanes aren't actually getting full oil pressure on one side and zero on the other. If you follow the path of the oil and how it travels through the journal bearing to get into the camshaft passages, it's not too hard to imagine that you get leakage and cross-talk along the way. Plus, the vanes themselves aren't positively sealed, so some oil leaks around the vanes from the high pressure to the low pressure side.

What do you mean by this statement?: "When the dc goes to 0 failing to achieve the desired position, it traps the available pressure at that moment on one side."

I have plenty of logs where the phaser DC goes to zero, forced there by the PID control loop, and the cam retards as much as 30 deg. This, as I said, is drastically worse with low viscosity (5W20-5W30) and/or hot oil.

[murfie]

Murfie, in contemplating your posts above, you're saying that the hydraulic pressure should easily overcome the cam torque and be able to reach target cam angle. The only forces it has to overcome are friction and GDI pump drive, as the valve-spring energy is absorbed but then deposited right back into the cam so you break-even. According to BW, the spiral spring is there to work against "retard bias torque" and they also refer to it as a "torsional assist". I actually took the time to measure the amount of moment the spring applies, calculate the drive power of the GDI pump, and calculate the hydraulic torsion the phaser can apply and I came to the same conclusion that you did. The hydraulic torsion plus the spiral spring torsion should be multiples more than you need to prevent the exhaust cam from falling out of control and retarding, even with an upgraded GDI pump.

Once I determined the oil temp correlation to the problem and found success in increasing viscosity, the only plausible conclusion I could come up with is that the phaser vanes aren't actually getting full oil pressure on one side and zero on the other. If you follow the path of the oil and how it travels through the journal bearing to get into the camshaft passages, it's not too hard to imagine that you get leakage and cross-talk along the way. Plus, the vanes themselves aren't positively sealed, so some oil leaks around the vanes from the high pressure to the low pressure side.

What do you mean by this statement?: "When the dc goes to 0 failing to achieve the desired position, it traps the available pressure at that moment on one side."

I have plenty of logs where the phaser DC goes to zero, forced there by the PID control loop, and the cam retards as much as 30 deg. This, as I said, is drastically worse with low viscosity (5W20-5W30) and/or hot oil.

Two possibilities of not achieving desired phase position.

1. It's being physically stopped
2. The oil is being trapped and not released

There may be others like leaking too much that just aren't coming to mind.

These pictures can show it goes in to the vanes same passage it comes out of better than I can explain how it can get trapped.

Phaser ret..PNG

Phaser hold.PNG

Phaser advanced.PNG

[engineermike]

Ah, I see the disconnect. I was referring to the oil actuated phaser on the exhaust cam of the gen3. They are much simpler but the control oil has to follow a more tortuous path from the solenoid valve to the actuator.

Robcat030, is your issue on cam torque or oil pressure actuated phasers?

[K44]

Where are you finding this phaser pid? Is this limited to gen3?

I'm assuming you would increase the degrees/second in the FF table for increased control.