[scalbert]

From what I recall reading, the boost control is somewhat different than traditional applications. From what I recall the wastegate are operated off of vacuum; hence the need for the two vacuum reservoirs. But do the wastegates default to open or closed? Is 0% DC on the boost control solenoids equates to the wastegates being open and 100% is fully closed? Or is it the inverse?

I am sure Shiv has figured this part out and can shed some light. And yes, this is leading somewhere.

[e90AW335i]

Originally Posted by scalbert

And yes, this is leading somewhere.

Terry could probably tell you as well....

Regardless, I am interested....

[Terry335]

Originally Posted by scalbert

From what I recall reading, the boost control is somewhat different than traditional applications. From what I recall the wastegate are operated off of vacuum; hence the need for the two vacuum reservoirs. But do the wastegates default to open or closed? Is 0% DC on the boost control solenoids equates to the wastegates being open and 100% is fully closed? Or is it the inverse?

I am sure Shiv has figured this part out and can shed some light. And yes, this is leading somewhere.

It's a vacuum based system and the wastegates default open (no boost). The solenoids are 0-12v, ~1a at full load. 0v is fully open, 12v is fully closed.

[OpenFlash]

He's just too fast for me

[scalbert]

Thanks for the confirmation: I was pretty sure the default state was no boost (my old Typhoon was the opposite; I line blew off the boost control solenoid and I was seeing 30 PSI).

Now for the reason of the question. Everything had been fine for the most part for the last week. I should note I was out of town all of last week so the car was only driven with V2 installed for a few days total. Upon initial testing I was seeing what was expected. Everything was fine so far. Last Friday, the second day I drove the car with V2 installed, I noticed that in the mid range the car was only making a hair over 10 PSI boost. Whereas on Monday (not driven the other days) I would see 12 - 13 PSI in the mid range. This was the prompt for the load based thread I created as it was semi-cool here on Friday morning.

Fast forward to yesterday and today. Yesterday, out of curiosity, and since I know the quality of fuel in the tank, I bumped up the settings to about half way (for each value) to 100%. In other words, if it was 92%, that now become 96%, etc. I ran some logs and timing seemed to be fine although boost was spiking a bit more than I thought it should. A/F ratio also seemed in check. However, today was a different story, at least with boost; no other data was monitored.

Today, I would sometimes see spikes as high as 22 PSI but most of the time it would drop to about 15 PSI. But there was one time it didn't drop so I immediately got off of it. The kicker is that this did not occur 100% of the time; more like about 30%. Also, if I jumped on the throttle in the low revs, about 2000 RPM, and let it build from there, the boost responded correctly. It was only when I jumped fully on the throttle in the 3000 - 4000 RPM range that it would overshoot. Plus, even when slowly rolling on the throttle and no spikes, the peak boost would start climbing to about 17 PSI just before it was pulled back in the 2.0.1 map to below 10 PSI nearing redline. One time it even climbed to near 20 PSI.

I went back home and loaded back in the base 2.0.1 map and went out for some more tests. Although the spikes were less significant, it would still run as high as 20 PSI. And now it would just climb to about 16 PSI when slowly rolling on the throttle. So things were reduced but not where it had been just days before.

Next I dropped the user values by about 5 points across the board and retested. Now things seemed to be getting closer to where it had been. Spikes were down to the 17 PSI range and it would only climb to about 13 - 14 PSI. However, after several minutes things dropped altogether to where I would expect it. Plus, it was not spiking at all now. That is where I left it and planned on monitoring it on the way into the office tomorrow morning.

The below is a summary on the base 2.0.1 map:

- This only occurred about 30% of the time
- Boost spikes were running upwards of 20 PSI
- If I slowly rolled on the throttle, the spikes did not occur
- Boost occasionally climbing to 16 - 17 PSI by 5500 - 6000
- Boost would drop to below 10 PSI near redline, as it should with this map

Now the caveat, power changes were not noticeable. What I mean is that there was no power increase or decrease when the spikes occurred which with other vehicles, would have been noticeable. In fact, I would not know anything was awry unless I was monitoring the gauge. Everything seemed perfectly normal.

Since we have established how the boost control functions, I have to say it is working. But for whatever reason the boost is not dropping as designed except in other places where the duty cycle would be significantly lower. It is almost as though the solenoids or the driver for them are not responding in certain situations. That, or the TMAP input isn't working correctly 100% of the time. I need to pick up a USB extension cable to get logs from the PROcede its self to see what the I/O are doing. Or another thought, the source for the boost gauge (the diverter supply lines) do not represent the same value as seen at the TMAP.

Having programmed many PID control systems (single loops, cascading, etc.) and even some hydraulic systems, I know a thing or two about what is required for closed loop control. The difficulty here is that it works fine the majority of the time and without watching the gauge; everything feels fine even when it does occur. This might be a fun one to track down and yes, I did check all of the lines as much as possible. I even see a solid 22 inHG vacuum when warmed.

[e90AW335i]

A couple of other people on here have had this problem too, Maybe Shiv will chime in....

[OpenFlash]

Originally Posted by scalbert

Thanks for the confirmation: I was pretty sure the default state was no boost (my old Typhoon was the opposite; I line blew off the boost control solenoid and I was seeing 30 PSI).

Now for the reason of the question. Everything had been fine for the most part for the last week. I should note I was out of town all of last week so the car was only driven with V2 installed for a few days total. Upon initial testing I was seeing what was expected. Everything was fine so far. Last Friday, the second day I drove the car with V2 installed, I noticed that in the mid range the car was only making a hair over 10 PSI boost. Whereas on Monday (not driven the other days) I would see 12 - 13 PSI in the mid range. This was the prompt for the load based thread I created as it was semi-cool here on Friday morning.

Fast forward to yesterday and today. Yesterday, out of curiosity, and since I know the quality of fuel in the tank, I bumped up the settings to about half way (for each value) to 100%. In other words, if it was 92%, that now become 96%, etc. I ran some logs and timing seemed to be fine although boost was spiking a bit more than I thought it should. A/F ratio also seemed in check. However, today was a different story, at least with boost; no other data was monitored.

Today, I would sometimes see spikes as high as 22 PSI but most of the time it would drop to about 15 PSI. But there was one time it didn't drop so I immediately got off of it. The kicker is that this did not occur 100% of the time; more like about 30%. Also, if I jumped on the throttle in the low revs, about 2000 RPM, and let it build from there, the boost responded correctly. It was only when I jumped fully on the throttle in the 3000 - 4000 RPM range that it would overshoot. Plus, even when slowly rolling on the throttle and no spikes, the peak boost would start climbing to about 17 PSI just before it was pulled back in the 2.0.1 map to below 10 PSI nearing redline. One time it even climbed to near 20 PSI.

I went back home and loaded back in the base 2.0.1 map and went out for some more tests. Although the spikes were less significant, it would still run as high as 20 PSI. And now it would just climb to about 16 PSI when slowly rolling on the throttle. So things were reduced but not where it had been just days before.

Next I dropped the user values by about 5 points across the board and retested. Now things seemed to be getting closer to where it had been. Spikes were down to the 17 PSI range and it would only climb to about 13 - 14 PSI. However, after several minutes things dropped altogether to where I would expect it. Plus, it was not spiking at all now. That is where I left it and planned on monitoring it on the way into the office tomorrow morning.

The below is a summary on the base 2.0.1 map:

- This only occurred about 30% of the time
- Boost spikes were running upwards of 20 PSI
- If I slowly rolled on the throttle, the spikes did not occur
- Boost occasionally climbing to 16 - 17 PSI by 5500 - 6000
- Boost would drop to below 10 PSI near redline, as it should with this map

Now the caveat, power changes were not noticeable. What I mean is that there was no power increase or decrease when the spikes occurred which with other vehicles, would have been noticeable. In fact, I would not know anything was awry unless I was monitoring the gauge. Everything seemed perfectly normal.

Since we have established how the boost control functions, I have to say it is working. But for whatever reason the boost is not dropping as designed except in other places where the duty cycle would be significantly lower. It is almost as though the solenoids or the driver for them are not responding in certain situations. That, or the TMAP input isn't working correctly 100% of the time. I need to pick up a USB extension cable to get logs from the PROcede its self to see what the I/O are doing. Or another thought, the source for the boost gauge (the diverter supply lines) do not represent the same value as seen at the TMAP.

Having programmed many PID control systems (single loops, cascading, etc.) and even some hydraulic systems, I know a thing or two about what is required for closed loop control. The difficulty here is that it works fine the majority of the time and without watching the gauge; everything feels fine even when it does occur. This might be a fun one to track down and yes, I did check all of the lines as much as possible. I even see a solid 22 inHG vacuum when warmed.

Seeing a bit over 10psi at times in the midrange with the baseline v2.0.1 map is completely normal. And it's also normal for it to vary a couple of psi in the midrange with changes in temp and boost decay parameter adjustments. The overboost feature is also active in 2.0.1. Which means that if you have not been on boost for some time you may see a 1-1.5psi jump in boost pressure for a very short period of time (up to 1 second).

From you post, it reads like the spiking occurred after you raised the user torque settings. Even though you said otherwise in the summary. Also, do you have any other modifications?

shiv

[scalbert]

Originally Posted by shiv@vishnu

From you post, it reads like the spiking occurred after you raised the user torque settings. Even though you said otherwise in the summary. Also, do you have any other modifications?

To be honest, I had not been watching the gauge except for last Friday morning when things seemed relatively nornal. It wasn't until today that I saw it. But yes, the very high spikes did occur after raising the values but still occured after re-loading the base 2.0.1 map values. I even saw 16 - 17 PSI spikes just after the intial install but wwere very intermitant and to me, less than alarming.

And no, there is nothign else on the car.

I have a sneaking suspicion I will see lower values in the morning...

[OpenFlash]

Originally Posted by scalbert

To be honest, I had not been watching the gauge except for last Friday morning when things seemed relatively nornal. It wasn't until today that I saw it. But yes, the very high spikes did occur after raising the values but still occured after re-loading the base 2.0.1 map values. I even saw 16 - 17 PSI spikes just after the intial install but wwere very intermitant and to me, less than alarming.

And no, there is nothign else on the car.

I have a sneaking suspicion I will see lower values in the morning...

Sometimes adjusting the user torque values requires some learning time for the factory ECU. Since you are basically adjusting the boost targets, the factory ECU needs to revise it's initial duty cycle target. Until it does so, it may overboost a tick. And when adjusting the user torque values, it's best to adjust them only 1-2% at a time, not the 4% you did These are all the things I'm covering in the v2.0.2 documentation, btw.

Shiv

[scalbert]

Originally Posted by shiv@vishnu

Sometimes adjusting the user torque values requires some learning time for the factory ECU. Since you are basically adjusting the boost targets, the factory ECU needs to revise it's initial duty cycle target. Until it does so, it may overboost a tick. And when adjusting the user torque values, it's best to adjust them only 1-2% at a time. These are all the things I'm covering in the v2.0.2 documentation, btw

Thanks, I will keep all of that in mind. I am not really concerned as I will be monitoring it and seeing what occurs as well as control my foot. I may just re-load the base values back in and see what it occuring in the morning. This was just an observation which will be monitored and wanted some input.

But I would really like to see what the 2.0.2 map does.

And you really think any of us will read that documentation; no one reads manuals.

[Terry335]

Originally Posted by scalbert

Today, I would sometimes see spikes as high as 22 PSI but most of the time it would drop to about 15 PSI. But there was one time it didn't drop so I immediately got off of it. The kicker is that this did not occur 100% of the time; more like about 30%.

Honestly, tuner politics aside, that is a pretty serious issue. I would not run the car again until you positively identified and corrected the cause. If you must do R&D work where boost might spike like that, at least run straight 100 octane to mitigate your risk.

[scalbert]

Originally Posted by Terry335

Honestly, tuner politics aside, that is a pretty serious issue. I would not run the car again until you positively identified and corrected the cause. If you must do R&D work where boost might spike like that, at least run straight 100 octane to mitigate your risk.

What I will do is control my right foot and keep monitoring things. Once it is under control, I will cease the monitoring and the self induced throttle restrictions.

[Terry335]

Originally Posted by scalbert

What I will do is control my right foot and keep monitoring things. Once it is under control, I will cease the monitoring and the self induced throttle restrictions.

Can you imagine if you didn't have a boost gauge and the first hint of trouble was the sweet smell of coolant in the air?

[scalbert]

Originally Posted by Terry335

Can you imagine if you didn't have a boost gauge and the first hint of trouble was the sweet smell of coolant in the air?

That is why I always have a guage in a forced induction vehicle.

[Terry335]

Originally Posted by scalbert

That is why I always have a guage in a forced induction vehicle.

[OpenFlash]

scalbert-- Scare tactics aside, just go ahead and put back in the baseline map. You should see boost no greater than 15psi in the upper midrange. Maybe 16 if the overboost feature kicks in. If it's any higher than that (and it shouldn't be), just drop the user torque value down 1-2% in that RPM. It's totally normal to see only 10-13psi in the midrange at time. No need to raise the user Torque values at all if you see this boost. This boost reduction (during low rpm dyno-like single gear pulls) was just an attempt to limit torque to the sub 380lbft level.

Shiv

[scalbert]

Originally Posted by shiv@vishnu

scalbert-- Scare tactics aside, just go ahead and put back in the baseline map. You should see boost no greater than 15psi in the upper midrange. Maybe 16 if the overboost feature kicks in. If it's any higher than that (and it shouldn't be), just drop the user torque value down 1-2% in that RPM. It's totally normal to see only 10-13psi in the midrange at time. This boost reduction (during low rpm dyno-like single gear pulls) was just an attempt to limit torque to sub 380lbft.

Already done a few minutes ago...

Can you elaborate on the overboost feature. This may be what I occaisionaly see with the 16 - 17 PSI climb. Please forgive my inquisition. If I see soemthing occuring I always like to know why; force of habit in my field.

[OpenFlash]

Originally Posted by scalbert

Already done a few minutes ago...

Can you elaborate on the overboost feature. This may be what I occaisionaly see with the 16 - 17 PSI climb. Please forgive my inquisition. If I see soemthing occuring I always like to know why; force of habit in my field.

No problem. The overboost feature allows for a 1-1.5psi jump in boost pressure for a short period of time. When this feature is activated and the duration of the overboost is dependent upon how hard (a function of average boost and average RPM) the car has been driven in the last 4 minutes. Basically allows for maximum safe power when conditions are right. But the extra boost fades away as components (IC, turbos, inlet charge, etc,) get hotter. And a bit after that, the boost decay feature kicks in (power drops below normal levels) when load is sustained (ie, track day hot-lapping). Pretty nifty stuff

Shiv

[sajakh]

Originally Posted by shiv@vishnu

No problem. The overboost feature allows for a 1-1.5psi jump in boost pressure for a short period of time. When this feature is activated and the duration of the overboost is dependent upon how hard (a function of average boost and average RPM) the car has been driven in the last 4 minutes. Basically allows for maximum safe power when conditions are right. But the extra boost fades away as components (IC, turbos, inlet charge, etc,) get hotter. And a bit after that, the boost decay feature kicks in (power drops below normal levels) when load is sustained (ie, track day hot-lapping). Pretty nifty stuff

Shiv

When is this overboost feature activated and is it in certain rpm ranges?

I was comparing it to the 997 911 turbo overboost feature where upon pressing a button gives you 10 sec of 2.9 psi increase (about 60 N-m increase in torque. How is the Procede feature different/similar?

By the way, awesome dyno comparo in the other thread.

[DrDomer]

Do the recommendations to change the torque values only 1-2% apply for both going up and down? For instance, say I want to run the Procede V2.0.1 map and drop settings to around 70%, do I have to do this in several increments if I started at the standard settings? How about if I upload the changes from the get go? Thanks.

[scalbert]

Originally Posted by sajakh

When is this overboost feature activated and is it in certain rpm ranges?

From the sound of it, it will automatically go into overboost if you have not really gotten into it for the last four minutes. Which may be why my car seemed to settle down after some tests. Albeit, originally higher than it shoudl have been.

[scalbert]

Originally Posted by DrDomer

Do the recommendations to change the torque values only 1-2% apply for both going up and down? For instance, say I want to run the Procede V2.0.1 map and drop settings to around 70%, do I have to do this in several increments if I started at the standard settings? How about if I upload the changes from the get go? Thanks.

I would guess that this would also apply to some extent. However, the more I thought about it the more I am curious as to how the DME is involved. The PROcede does receive the boost control solenoid signals and then has outputs specifically for those valves. It then must appear that there is a significant influence in what the stock duty cycle is doing and how the PROcede responds.

Initially I thought the PROcede used the signals as a reference as to major changes in the target boost but functioned solely independent and closed loop controlled. But it seems to be much more integrated than this based on what has been presented.