How-to: building a GPS-based PSE valve controller for passing track sound checks
#91
Intermediate
Join Date: Feb 2015
Location: Mountain View, CA
Posts: 49
Likes: 0
Received 0 Likes
on
0 Posts
Hmm that's interesting. Would you mind sketching the diagram for your 981? My impression is the 981 is the same wiring as the 997 (switched V- / GND). It's not obvious how it could be different than both the 991 (switched V+) and 997 (switched V-) unless the solenoid itself is different.
#92
Intermediate
Join Date: Feb 2015
Location: Mountain View, CA
Posts: 49
Likes: 0
Received 0 Likes
on
0 Posts
OK, I tried to combine and modify your drawings and here is what I've got. That's pretty much how it's is in my 981GTS (if I had to use a passthru configuration instead of 4-wires). Polarity in ECM side connector is just flipped compared to both 991GT3 and 997GT3.
#93
Nordschleife Master
Thread Starter
Nice modification work! That's really odd, though... that says that the solenoid must be a different type where the polarity is reversed. Can you easily access the solenoid and take a picture of the connector area (with the connector disconnected) to show the (+) or (-) marking on the side?
#94
Intermediate
Join Date: Feb 2015
Location: Mountain View, CA
Posts: 49
Likes: 0
Received 0 Likes
on
0 Posts
Yep, it's odd and I checked few times hoping that I was really missing something... Few pictures I took during the 4-wire installation:
https://www.dropbox.com/sh/fq6s7abni...-MjS_tQ-a?dl=0
The side of the connector that is closest to the rear bumper is connected to ECM V+ (output) and it shows "-" at the same time (stamped on the solenoid).
ECM V+ goes from ~+0.6V to +14.5V (Loud/quiet modes, triggered by exhaust button). Car's battery voltage is ~0.2V higher (approx. +14.7V). 0.2V is most likely a voltage drop out on ECM's switching FET.
Voltages were measured between ECM V+ and ECM V- of the connector and also between ECM V+ and chassis ground and results were the same. ECM V- seems to be hardwired to chassis ground according to Ohmmeter. Also there is zero voltage between ECM V- and chassis GND in all exhaust modes. I assume I'd see some voltage variation if ECM V- was a switched line.
https://www.dropbox.com/sh/fq6s7abni...-MjS_tQ-a?dl=0
The side of the connector that is closest to the rear bumper is connected to ECM V+ (output) and it shows "-" at the same time (stamped on the solenoid).
ECM V+ goes from ~+0.6V to +14.5V (Loud/quiet modes, triggered by exhaust button). Car's battery voltage is ~0.2V higher (approx. +14.7V). 0.2V is most likely a voltage drop out on ECM's switching FET.
Voltages were measured between ECM V+ and ECM V- of the connector and also between ECM V+ and chassis ground and results were the same. ECM V- seems to be hardwired to chassis ground according to Ohmmeter. Also there is zero voltage between ECM V- and chassis GND in all exhaust modes. I assume I'd see some voltage variation if ECM V- was a switched line.
Nice modification work! That's really odd, though... that says that the solenoid must be a different type where the polarity is reversed. Can you easily access the solenoid and take a picture of the connector area (with the connector disconnected) to show the (+) or (-) marking on the side?
#95
Nordschleife Master
Thread Starter
That is really odd... why would the solenoid stamping be incorrect? To confirm, are you taking all of these measurements while the valve is actually connected to the wiring (doing so with an in-line harness that lets you probe while the system is functioning)?
Here is what another Cayman GTS owner measured with such a harness:
PSE state, Black wire (-) voltage, Brown wire (+) voltage
PSE on, 0V, 13V
PSE off, 14V, 13V
If his measurements are correct, then his GTS valve signaling is similar to the 997, except that energizing the valve makes the exhaust loud rather than quiet.
But that is still different than you measure... the only difference being that he sees battery voltage on the (+) side rather than ground, which matches the solenoid labeling.
Unfortunately I haven't confirmed any of this first hand on a 981, but I'd be happy to do some time with my debug harness. I'm not far from your location.
Here is what another Cayman GTS owner measured with such a harness:
PSE state, Black wire (-) voltage, Brown wire (+) voltage
PSE on, 0V, 13V
PSE off, 14V, 13V
If his measurements are correct, then his GTS valve signaling is similar to the 997, except that energizing the valve makes the exhaust loud rather than quiet.
But that is still different than you measure... the only difference being that he sees battery voltage on the (+) side rather than ground, which matches the solenoid labeling.
Unfortunately I haven't confirmed any of this first hand on a 981, but I'd be happy to do some time with my debug harness. I'm not far from your location.
Yep, it's odd and I checked few times hoping that I was really missing something... Few pictures I took during the 4-wire installation:
https://www.dropbox.com/sh/fq6s7abni...-MjS_tQ-a?dl=0
The side of the connector that is closest to the rear bumper is connected to ECM V+ (output) and it shows "-" at the same time (stamped on the solenoid).
ECM V+ goes from ~+0.6V to +14.5V (Loud/quiet modes, triggered by exhaust button). Car's battery voltage is ~0.2V higher (approx. +14.7V). 0.2V is most likely a voltage drop out on ECM's switching FET.
Voltages were measured between ECM V+ and ECM V- of the connector and also between ECM V+ and chassis ground and results were the same. ECM V- seems to be hardwired to chassis ground according to Ohmmeter. Also there is zero voltage between ECM V- and chassis GND in all exhaust modes. I assume I'd see some voltage variation if ECM V- was a switched line.
https://www.dropbox.com/sh/fq6s7abni...-MjS_tQ-a?dl=0
The side of the connector that is closest to the rear bumper is connected to ECM V+ (output) and it shows "-" at the same time (stamped on the solenoid).
ECM V+ goes from ~+0.6V to +14.5V (Loud/quiet modes, triggered by exhaust button). Car's battery voltage is ~0.2V higher (approx. +14.7V). 0.2V is most likely a voltage drop out on ECM's switching FET.
Voltages were measured between ECM V+ and ECM V- of the connector and also between ECM V+ and chassis ground and results were the same. ECM V- seems to be hardwired to chassis ground according to Ohmmeter. Also there is zero voltage between ECM V- and chassis GND in all exhaust modes. I assume I'd see some voltage variation if ECM V- was a switched line.
#96
Intermediate
Join Date: Feb 2015
Location: Mountain View, CA
Posts: 49
Likes: 0
Received 0 Likes
on
0 Posts
I can only guess that maybe factory harness connection was flipped on ECM module side in my car (if there is a connector that allows something like that. But it won't change the fact that ECM V- is hardwired to the GND in my car.
Yes, initially I measured both ways - with and without the load (solenoid) and saw no difference. My first harness had a 3-wire passthrough configuration, so I could do it. I was really puzzled with what I saw and tried to test it extensively. I can check it one more time. Just need to make a small adapter with open contacts (in-line harness) that will allow that or plug in 25 Ohm (that's the DC resistance of the solenoid) directly to the ECM V+/V- connector and measure it this way.
There is another odd thing that I see. I usually do these checks while my car is warming up (idling). If I disconnect a solenoid (easy with 4-wire connection) I can hear exhaust getting louder immediately (that's with PSE switched to QUIET mode) on IDLE. But then I plug it back in and try to use PSE button and hear no difference on IDLE. I can hear the difference when I rev it up though! I assumed that it's also due to engine mapping being changed when PSE button is used. But that's the only thing I can think about.
I'd be happy to meet with you if you have time and desire to do so! A sanity check of what I'm doing definitely won't hurt.
Yes, initially I measured both ways - with and without the load (solenoid) and saw no difference. My first harness had a 3-wire passthrough configuration, so I could do it. I was really puzzled with what I saw and tried to test it extensively. I can check it one more time. Just need to make a small adapter with open contacts (in-line harness) that will allow that or plug in 25 Ohm (that's the DC resistance of the solenoid) directly to the ECM V+/V- connector and measure it this way.
There is another odd thing that I see. I usually do these checks while my car is warming up (idling). If I disconnect a solenoid (easy with 4-wire connection) I can hear exhaust getting louder immediately (that's with PSE switched to QUIET mode) on IDLE. But then I plug it back in and try to use PSE button and hear no difference on IDLE. I can hear the difference when I rev it up though! I assumed that it's also due to engine mapping being changed when PSE button is used. But that's the only thing I can think about.
I'd be happy to meet with you if you have time and desire to do so! A sanity check of what I'm doing definitely won't hurt.
That is really odd... why would the solenoid stamping be incorrect? To confirm, are you taking all of these measurements while the valve is actually connected to the wiring (doing so with an in-line harness that lets you probe while the system is functioning)?
Here is what another Cayman GTS owner measured with such a harness:
PSE state, Black wire (-) voltage, Brown wire (+) voltage
PSE on, 0V, 13V
PSE off, 14V, 13V
If his measurements are correct, then his GTS valve signaling is similar to the 997, except that energizing the valve makes the exhaust loud rather than quiet.
But that is still different than you measure... the only difference being that he sees battery voltage on the (+) side rather than ground, which matches the solenoid labeling.
Unfortunately I haven't confirmed any of this first hand on a 981, but I'd be happy to do some time with my debug harness. I'm not far from your location.
Here is what another Cayman GTS owner measured with such a harness:
PSE state, Black wire (-) voltage, Brown wire (+) voltage
PSE on, 0V, 13V
PSE off, 14V, 13V
If his measurements are correct, then his GTS valve signaling is similar to the 997, except that energizing the valve makes the exhaust loud rather than quiet.
But that is still different than you measure... the only difference being that he sees battery voltage on the (+) side rather than ground, which matches the solenoid labeling.
Unfortunately I haven't confirmed any of this first hand on a 981, but I'd be happy to do some time with my debug harness. I'm not far from your location.
#97
Nordschleife Master
Thread Starter
I can only guess that maybe factory harness connection was flipped on ECM module side in my car (if there is a connector that allows something like that. But it won't change the fact that ECM V- is hardwired to the GND in my car.
Yes, initially I measured both ways - with and without the load (solenoid) and saw no difference. My first harness had a 3-wire passthrough configuration, so I could do it. I was really puzzled with what I saw and tried to test it extensively. I can check it one more time. Just need to make a small adapter with open contacts (in-line harness) that will allow that or plug in 25 Ohm (that's the DC resistance of the solenoid) directly to the ECM V+/V- connector and measure it this way.
There is another odd thing that I see. I usually do these checks while my car is warming up (idling). If I disconnect a solenoid (easy with 4-wire connection) I can hear exhaust getting louder immediately (that's with PSE switched to QUIET mode) on IDLE. But then I plug it back in and try to use PSE button and hear no difference on IDLE. I can hear the difference when I rev it up though! I assumed that it's also due to engine mapping being changed when PSE button is used. But that's the only thing I can think about.
I'd be happy to meet with you if you have time and desire to do so! A sanity check of what I'm doing definitely won't hurt.
Yes, initially I measured both ways - with and without the load (solenoid) and saw no difference. My first harness had a 3-wire passthrough configuration, so I could do it. I was really puzzled with what I saw and tried to test it extensively. I can check it one more time. Just need to make a small adapter with open contacts (in-line harness) that will allow that or plug in 25 Ohm (that's the DC resistance of the solenoid) directly to the ECM V+/V- connector and measure it this way.
There is another odd thing that I see. I usually do these checks while my car is warming up (idling). If I disconnect a solenoid (easy with 4-wire connection) I can hear exhaust getting louder immediately (that's with PSE switched to QUIET mode) on IDLE. But then I plug it back in and try to use PSE button and hear no difference on IDLE. I can hear the difference when I rev it up though! I assumed that it's also due to engine mapping being changed when PSE button is used. But that's the only thing I can think about.
I'd be happy to meet with you if you have time and desire to do so! A sanity check of what I'm doing definitely won't hurt.
#98
Nordschleife Master
Thread Starter
Yep, it's odd and I checked few times hoping that I was really missing something... Few pictures I took during the 4-wire installation:
https://www.dropbox.com/sh/fq6s7abni...-MjS_tQ-a?dl=0
The side of the connector that is closest to the rear bumper is connected to ECM V+ (output) and it shows "-" at the same time (stamped on the solenoid).
ECM V+ goes from ~+0.6V to +14.5V (Loud/quiet modes, triggered by exhaust button). Car's battery voltage is ~0.2V higher (approx. +14.7V). 0.2V is most likely a voltage drop out on ECM's switching FET.
Voltages were measured between ECM V+ and ECM V- of the connector and also between ECM V+ and chassis ground and results were the same. ECM V- seems to be hardwired to chassis ground according to Ohmmeter. Also there is zero voltage between ECM V- and chassis GND in all exhaust modes. I assume I'd see some voltage variation if ECM V- was a switched line.
https://www.dropbox.com/sh/fq6s7abni...-MjS_tQ-a?dl=0
The side of the connector that is closest to the rear bumper is connected to ECM V+ (output) and it shows "-" at the same time (stamped on the solenoid).
ECM V+ goes from ~+0.6V to +14.5V (Loud/quiet modes, triggered by exhaust button). Car's battery voltage is ~0.2V higher (approx. +14.7V). 0.2V is most likely a voltage drop out on ECM's switching FET.
Voltages were measured between ECM V+ and ECM V- of the connector and also between ECM V+ and chassis ground and results were the same. ECM V- seems to be hardwired to chassis ground according to Ohmmeter. Also there is zero voltage between ECM V- and chassis GND in all exhaust modes. I assume I'd see some voltage variation if ECM V- was a switched line.
On the 991, the 3-port valve switches the common connection (the exhaust valves) between a vacuum source, or an atmosphere source, depending on solenoid state (energized vs. not).
If the 3rd port is plugged on the Cayman, then the system must operate differently and all bets are off until I understand more about the vacuum plumbing of the system...
#99
Intermediate
Join Date: Feb 2015
Location: Mountain View, CA
Posts: 49
Likes: 0
Received 0 Likes
on
0 Posts
Actually I need to retract what I said about sound not changing at IDLE when using PSE button. It does. Looks like this button just doesn't work when car is cold started. It's locked in "quiet" mode for a few moments. After that it works normally and I hear sound changing same way as when I just unplug the solenoid.
#100
Intermediate
Join Date: Feb 2015
Location: Mountain View, CA
Posts: 49
Likes: 0
Received 0 Likes
on
0 Posts
Yeah, looks like the is a cap there.
But it works - sound gets changed and I actually used this system at Laguna Seca once with positive result.
I made an in-line harness adapter to measure volatges between ECM V+ and chassis GND and between ECM V- and chassis GND. Attaching couple of pictures.
Voltage measurement results:
PSE state, ECM V -, ECM V+
PSE on, 0.068V, 14.34V
PSE off, 0.01V, 0.01V
So, it's the same with loaded harness. Maybe Porsche just changed the "revision" of the ECM in my car?
#101
Rennlist Member
Join Date: May 2014
Location: Bay Area
Posts: 588
Likes: 0
Received 0 Likes
on
0 Posts
I haven't traced out vacuum lines as I knew that this solution works on 981GTS. I think this solenoid looks the same as on doborder's puctures: https://rennlist.com/forums/987-981-...e-control.html
Yeah, looks like the is a cap there.
But it works - sound gets changed and I actually used this system at Laguna Seca once with positive result.
I made an in-line harness adapter to measure volatges between ECM V+ and chassis GND and between ECM V- and chassis GND. Attaching couple of pictures.
Voltage measurement results:
PSE state, ECM V -, ECM V+
PSE on, 0.068V, 14.34V
PSE off, 0.01V, 0.01V
So, it's the same with loaded harness. Maybe Porsche just changed the "revision" of the ECM in my car?
Yeah, looks like the is a cap there.
But it works - sound gets changed and I actually used this system at Laguna Seca once with positive result.
I made an in-line harness adapter to measure volatges between ECM V+ and chassis GND and between ECM V- and chassis GND. Attaching couple of pictures.
Voltage measurement results:
PSE state, ECM V -, ECM V+
PSE on, 0.068V, 14.34V
PSE off, 0.01V, 0.01V
So, it's the same with loaded harness. Maybe Porsche just changed the "revision" of the ECM in my car?
Did you all figure this out yet?
#102
Nordschleife Master
Thread Starter
Update: I worked with a custom acrylic case designer to have a case fabricated that better fits this taller board stack at a reasonable cost. You can order it in a variety of translucent colors here for $10 shipped: https://www.tindie.com/products/mjri...losure-duo-ii/
#103
Nordschleife Master
Originally Posted by Mech33
Update: I worked with a custom acrylic case designer to have a case fabricated that better fits this taller board stack at a reasonable cost. You can order it in a variety of translucent colors here for $10 shipped: https://www.tindie.com/products/mjri...losure-duo-ii/
#104
Rennlist Member
Join Date: May 2014
Location: Bay Area
Posts: 588
Likes: 0
Received 0 Likes
on
0 Posts
Update: I worked with a custom acrylic case designer to have a case fabricated that better fits this taller board stack at a reasonable cost. You can order it in a variety of translucent colors here for $10 shipped: https://www.tindie.com/products/mjri...losure-duo-ii/
#105
Race Director
Okay, I am impressed.