VVT Implementation attempt

[JAM]

Ok, I am planning to implement full VVT for the first time, I did some research on my specific engine, here is what i know about the VVT setup;

From the factory-
Cam phaser: exhaust cam
Average hold position: around 45%, (Home position: 0 % PWM)
More Duty= Retard
Sweep: up to 25°of cam angle (50°of crankshaft angle)
Solenoid: 128Hz PWM, 10 ohm, 4 way valve **2 wires as follows: MS3X VVT pin 27 and 12V+**


i have a few questions:

1) determining a suitable frequency, is there documentation i can read on how to determine what is best? i can select 99Hz or 153Hz are the closest to the factory 128Hz
2) Since VVT on the exhaust cam, any thoughts on if it is worth messing with the "adjust injector timing based on" or just leave it off to start with?

[Matt Cramer]

1) determining a suitable frequency, is there documentation i can read on how to determine what is best? i can select 99Hz or 153Hz are the closest to the factory 128Hz

Try both and see which produces the greatest range of duty cycles you can use.

2) Since VVT on the exhaust cam, any thoughts on if it is worth messing with the "adjust injector timing based on" or just leave it off to start with?

I'd leave it off.

[JAM]

Thanks for the response Matt,

I'd leave it off.

As in indefinitely? no need to mess with it since the phaser is on exhaust cam?

JAM wrote:
1) determining a suitable frequency, is there documentation i can read on how to determine what is best? i can select 99Hz or 153Hz are the closest to the factory 128Hz

Try both and see which produces the greatest range of duty cycles you can use.

So am i looking for the greatest range of duty cycle, will this be reflected in the angle while using the "test section" and watching the VVT angle gauge?

(Also on another note, i encountered an error when i was messing with this last night, to paraphrase the error it said "pwm group B must use same frequency (Fidle VVT boost) Fidle already in use" again this is from memory but do i need to have matching Fidle freq and VVT freq?)

[Matt Cramer]

Correct on both counts.

[JAM]

OK great, so with the info i provided above, does this look like a good place to start?

Capture.JPG

one question; need some clarification on the portion i circled in red, not sure i understand where to set hold duty at or if i need to ? when testing for angles does this need to be 'off' ?


thanks for the help

[Matt Cramer]

Try using a 45% hold duty.

[JAM]

Try using a 45% hold duty.

thanks, should that be turned off when testing? what indicators do you look at to optimize that number?

one other thing, if my settings are exhaust cam, more pwm = retard, then in the 3d map i would just put value 10 if i want 10 degrees retard, i would not put -10 or the absolute angle correct?

but in the settings window the min/max angle would be put in at absolute angles, correct?

[Matt Cramer]

thanks, should that be turned off when testing?

Turn it off when you are trying to establish the range of motion.

what indicators do you look at to optimize that number?

Whatever duty cycle holds the cam in place most of the time -that's what it should be.

one other thing, if my settings are exhaust cam, more pwm = retard, then in the 3d map i would just put value 10 if i want 10 degrees retard, i would not put -10 or the absolute angle correct?

The values in the table are relative to the minimum advance.

but in the settings window the min/max angle would be put in at absolute angles, correct?

Correct.

[jsmcortina]

one other thing, if my settings are exhaust cam, more pwm = retard, then in the 3d map i would just put value 10 if i want 10 degrees retard, i would not put -10 or the absolute angle correct?

There are two factors here.
The PWM direction makes sure the code moves the cam the right way.

You also have the option to use the exhaust table as a table of retard numbers. That's the "Table is Advance/retard" setting. See the tooltip for it.
When the "retard" option is selected the exhaust cam is assumed to default to the advanced position and be commanded to be retarded. (That's how most work apparently.)

James

[JAM]

There are two factors here.
The PWM direction makes sure the code moves the cam the right way.

You also have the option to use the exhaust table as a table of retard numbers. That's the "Table is Advance/retard" setting. See the tooltip for it.
When the "retard" option is selected the exhaust cam is assumed to default to the advanced position and be commanded to be retarded. (That's how most work apparently.

thanks for the info, that makes sense.

Here is what i was finding, i hope to do more testing today or tomorrow:

first step i did not use the VVT function; i simply turned on the generic pwm and at 0 this was the same as not using anything- as expected. I then changed table to 100 and the engine died immediately (cam advanced 25 degrees)

now i shut of the generic and moved to VVT and the settings i posted above. I used test mode and it seemed to be the opposite, at 0 it killed the engine, 100 seemed to be normal. somehow it seemed backwards but changing the other settings did not seem to swap it in the right direction. when it was running cam angle was bouncing between -3 and 31 degrees rapidly and sporadically (was thinking this number could be relative to crank rotation so could be anywhere 0-720 depending on how it catches the sync tooth?)

i will revisit this and try and grab some logs of what is going on

[JAM]

After a rough patch of bad luck I am back on this.
As I stated:

first step i did not use the VVT function; i simply turned on the generic pwm and at 0 this was the same as not using anything (unpluggded)- as expected. I then changed table to 100 and the engine died immediately (cam advanced 25+ degrees)

now i shut of the generic and moved to VVT and the settings i posted above. I used test mode and it seemed to be the opposite, at 0 it killed the engine, 100 seemed to be normal. somehow it seemed backwards but changing the other settings did not seem to swap it in the right direction. when it was running cam angle was bouncing between -3 and 31 degrees rapidly and sporadically

would regular log of this help determine what is going on or a composite log? what should i try from here?

[Matt Cramer]

I would like to start with a regular data log and an MSQ. A composite log would be useful if there is evidence of noise creating a false advance reading.

[JAM]

here is the test mode, trying to establish a range of motion- you will have to excuse the way the car was idling, i am in he middle of a complete re-tune, and then i had to change my freq on my idle valve to set the VVT and it was all over the place- again i set VVT freq closest to what the OEM had the solenoid set at (and that did not do well on my idle valve)

First thing- as i said before:

when i simply turned on the generic pwm and set to 0 this was the same as unplugged- as expected. I then changed table to 100 and the engine died immediately (ex.cam advanced 25 degrees) somewhere around 80% things got crazy
shut of the generic and moved to VVT and I used test mode -and it seemed to be the opposite, at 0 it killed the engine, 100 seemed to be normal. somehow it seemed backwards around 20% things got crazy

here are the data logs- the first dropping from 100% duty until the engine died. the second i was able to get it running at a real low duty and worked up. the angle is all over the place, not sure where to go from here, also attached a composite

2014-07-31_20.57.39 ROM 1.msl

2014-07-31_21.15.50 ROM 2.1.msl

2014-07-31_21.13.22 ROM.csv

[JAM]

CurrentTuneROM20140731.msq

MSQ attached

[Matt Cramer]

First time I've seen a GM 7X VVT setup. That is definitely not a stable read on the cam angle. Try switching the ignition input capture to rising edge.

[JAM]

First time I've seen a GM 7X VVT setup. That is definitely not a stable read on the cam angle. Try switching the ignition input capture to rising edge.

thanks for the reply
that could be a problem because when setting this up we went through a bit of an issue because the car would run in sequential up to about 3500 rpm on rising edge and ignition would scatter, so we switched to falling and it would run in wasted spark all the way up but was out of phase so would not run in full seq.

James did make a change earlier this year to swap the phase so that it would run in sequential and it runs great - but i am guessing that a switch would yield a similar result by switching back to rising edge, seems like something to do with the 7x being notch vs teeth- i guess it did run in the lower rpm so i could see what the results are.

[JAM]

First time I've seen a GM 7X VVT setup. That is definitely not a stable read on the cam angle. Try switching the ignition input capture to rising edge.

I tried switching to rising edge - didnt seem to like that. similar to before hard getting a good stable signal.

also set the test mode to 100% duty cycle and probed the harness VVT pigtail- 14.5V (running bat voltage), set to 0% duty and a steady 12.5 Volts. I was expecting 0% to be 0V?

[jsmcortina]

also set the test mode to 100% duty cycle and probed the harness VVT pigtail- 14.5V (running bat voltage), set to 0% duty and a steady 12.5 Volts. I was expecting 0% to be 0V?

This is the way VVT currently works with exhaust cams. I plan to change it in the near future.

James

[JAM]

also set the test mode to 100% duty cycle and probed the harness VVT pigtail- 14.5V (running bat voltage), set to 0% duty and a steady 12.5 Volts. I was expecting 0% to be 0V?

This is the way VVT currently works with exhaust cams. I plan to change it in the near future.

James

thanks James,

so i know this may be an odd trigger application for vvt, what are the chances I will be able to get a stable cam angle signal with the 7x? i dont really know what to try that wont give me a problem with crank signal. however it must pick up cam good enough to sync for seq ??

[JAM]

a couple more thoughts on this-

This is the way VVT currently works with exhaust cams. I plan to change it in the near future.

1.)Out of curiosity is this the way many factory setups have it? full bat voltage is 100% and some lesser voltage is 0% and all the duty cycles are in between?

2.)Knowing both crank and cam sensors work and sync at startup for sequential (both hall), and looking at the composite log and it seems that the patterns seem reasonable (maybe somebody who knows more can verify this for me) The cam angle is not seeing a stable signal on the regular log- if in fact the composite log looks stable as i suspect, what variables can cause a disconnect like this?

If you look at first log (4893 sec) you can see where its picking up (36 to -4) then when the vvt duty1 drops below 21.6% (the engine starts running rough) and the angle begins to bounce from (114-35). it really does act more like on off at that point and then tapers off with decreasing duty cycle-

VVT1.JPG