Close

Page 1 of 3 123 LastLast
Results 1 to 25 of 55
  1. #1
    Join Date
    Mar 2010
    Posts
    642
    Rep Points
    97.0
    Mentioned
    51 Post(s)
    Rep Power
    0


    0 out of 2 members liked this post. Yes Reputation No

    Talking Different Approaches on Piggyback Boost Control - Consistency and Safety

    Click here to enlarge Originally Posted by Sticky Click here to enlarge
    Exactly, idea is to equalize the power output in a variety of different conditions, something nice about turbos. However, I don't see why there would be inconsistency if this is a factory parameter and why would it be inconsistent with one piggyback and not another?
    Simple. The jb3 piggybacks off the factory boost control system. To raise boost target, it attenuates the map sensor signal. To help the target be reached, it pulls the wastegate pwm signal to ground (basically extending the pulsewidth. A few problems with this approach:
    1) The jb3 can't read the boost target due to lack of CANbus integration. It can only estimate it.
    2) The factory DME boost target is variable for reasons already mentioned. So if you apply a scalar to a variable input, you get a (more) variable output. You can attempt to limit it by applying secondary compensations but, again, they are just estimates.
    3) The jb3 can only increase DC. It cannot decrease it (to catch spikes for instance).

    Also, you do realize that the DME adjusts boosts in the exact opposite way that you would want it to in a tuned car, right?

    I'll be happy to explain how the Procede does it differently. If that is what you want.

    Shiv

  2. #2
    Join Date
    Jan 2010
    Location
    SoCal
    Posts
    119,478
    Rep Points
    32,154.9
    Mentioned
    2109 Post(s)
    Rep Power
    322


    Yes Reputation No
    Click here to enlarge Originally Posted by shiv@vishnu Click here to enlarge
    Simple. The jb3 piggybacks off the factory boost control system. To raise boost target, it attenuates the map sensor signal. To help the target be reached, it pulls the wastegate pwm signal to ground (basically extending the pulsewidth. A few problems with this approach:
    1) The jb3 can't read the boost target due to lack of CANbus integration. It can only estimate it.
    2) The factory DME boost target is variable for reasons already mentioned. So if you apply a scalar to a variable input, you get a (more) variable output. You can attempt to limit it by applying secondary compensations but, again, they are just estimates.
    3) The jb3 can only increase DC. It cannot decrease it (to catch spikes for instance).

    Also, you do realize that the DME adjusts boosts in the exact opposite way that you would want it to in a tuned car, right?

    I'll be happy to explain how the Procede does it differently. If that is what you want.

    Shiv
    Before this gets out of hand yes, regarding your concluding statment, this is what I would want wand many others just in a different thread. So what I am going to do is move this to a separate discussion and give enrita his thread back.

    Stage 2 or 2.5 E9X M3 S65 V8 supercharger kit for sale
    : http://www.boostaddict.com/showthrea...r-kit-for-sale

  3. #3
    Join Date
    Jan 2010
    Location
    SoCal
    Posts
    119,478
    Rep Points
    32,154.9
    Mentioned
    2109 Post(s)
    Rep Power
    322


    Yes Reputation No
    Click here to enlarge Originally Posted by shiv@vishnu Click here to enlarge
    Simple. The jb3 piggybacks off the factory boost control system. To raise boost target, it attenuates the map sensor signal. To help the target be reached, it pulls the wastegate pwm signal to ground (basically extending the pulsewidth. A few problems with this approach:
    1) The jb3 can't read the boost target due to lack of CANbus integration. It can only estimate it.
    2) The factory DME boost target is variable for reasons already mentioned. So if you apply a scalar to a variable input, you get a (more) variable output. You can attempt to limit it by applying secondary compensations but, again, they are just estimates.
    3) The jb3 can only increase DC. It cannot decrease it (to catch spikes for instance).

    Also, you do realize that the DME adjusts boosts in the exact opposite way that you would want it to in a tuned car, right?

    I'll be happy to explain how the Procede does it differently. If that is what you want.

    Shiv
    Ok, some questions.

    1. What does the CANbus integration allow you to do different with the boost target? Aren't you still relying on certain factory parameters? As in, won't the factory ECU still call for more boost in certain situations?

    2. Yes, it is variable. Why is limiting it just an estimate? Seems you are limiting in specific instances, doesn't sound like an estimate?

    3. Why can't it decrease it?

    No, I did not realize the DME adjusts boost in the exact opposite way that you would want it to in a tuned car. Why do you say that?

    Stage 2 or 2.5 E9X M3 S65 V8 supercharger kit for sale
    : http://www.boostaddict.com/showthrea...r-kit-for-sale

  4. #4
    Join Date
    Nov 2009
    Location
    USA
    Posts
    2,079
    Rep Points
    1,142.0
    Mentioned
    11 Post(s)
    Rep Power
    12


    Yes Reputation No
    I'm interested in this too:
    1) The jb3 can't read the boost target due to lack of CANbus integration. It can only estimate it.

    I know the CANbus feature has been explained before, but in terms of reading the boost target, can you go a bit more indepth?

    Does the CANbus feature differentiate from the flashed tunes as well, such as the GIAC/others?

  5. #5
    Join Date
    Nov 2009
    Location
    Miami, FL.
    Posts
    1,888
    Rep Points
    771.0
    Mentioned
    9 Post(s)
    Rep Power
    8


    Yes Reputation No
    interesting thread, subscribed
    Click here to enlarge
    2007 335i Coupe
    Mods: Check the Garage

  6. #6
    Join Date
    Nov 2009
    Location
    Simi Valley, CA
    Posts
    8,129
    Rep Points
    9,106.3
    Mentioned
    644 Post(s)
    Rep Power
    92


    Yes Reputation No
    It's a complex topic and I'm not sure I have the energy to fully discuss it here given the time involved and other projects I'm working on. But a few general comments:

    1) Oxygen density changes with air intake temperature, barometric pressure, etc. Thus the OEM system alters boost pressures to keep your oxygen levels and thus power and tuning consistent. It also varies boost through the curve to keep torque level as VE changes. I agree with what BMW has done in this regard. It makes for a smoother more consistent drive with no performance drawback.
    2) We could implement the counterpart system on the JB3. Which would be just picking a boost target (say 15psi) and letting power fluctuate as temperatures and barometric pressure did. This is done by letting the JB3 compute the attenuation required to keep boost focused on a particular target.
    3) This talk about the V4 method being safer or more spike free is nonsense and Shiv knows it. The fact is many V4 owners experience extreme shift spikes and misfires from the lack of boost control during automatic shifts. And since he dampens the throttle movement the ECU can't even close the throttle to protect the motor during these spikes.

    Here is one of his customer logs showing this major spike in action. Also a JB3 log that had been posted in that same thread showing no shift spike. Even if you force a shift @ peak HP on the JB3 you'll get almost no spike on the street maps. On the race maps we let the system spike a bit more as generally customers want to be a little more aggressive in those cases. But they have full control over that in the tuning interface should they want to dampen it down.

    Regardless it's just two ways to do things. In the future we will allow customers to select either method.
    Attached Images Attached Images   

  7. #7
    Join Date
    Mar 2010
    Posts
    642
    Rep Points
    97.0
    Mentioned
    51 Post(s)
    Rep Power
    0



    Yes Reputation No
    Click here to enlarge Originally Posted by Terry
    It's a complex topic and I'm not sure I have the energy to fully discuss it here given the time involved and other projects I'm working on. But a few general comments:

    1) Oxygen density changes with air intake temperature, barometric pressure, etc. Thus the OEM system alters boost pressures to keep your oxygen levels and thus power and tuning consistent. It also varies boost through the curve to keep torque level as VE changes. I agree with what BMW has done in this regard. It makes for a smoother more consistent drive with no performance drawback.
    2) We could implement the counterpart system on the JB3. Which would be just picking a boost target (say 15psi) and letting power fluctuate as temperatures and barometric pressure did. This is done by letting the JB3 compute the attenuation required to keep boost focused on a particular target.
    3) This talk about the V4 method being safer or more spike free is nonsense and Shiv knows it. The fact is many V4 owners experience extreme shift spikes and misfires from the lack of boost control during automatic shifts. And since he dampens the throttle movement the ECU can't even close the throttle to protect the motor during these spikes.

    Here is one of his customer logs showing this major spike in action. Also a JB3 log that had been posted in that same thread showing no shift spike. Even if you force a shift at peak HP on the JB3 you'll get almost no spike on the street maps. On the race maps we let the system spike a bit more as generally customers want to be a little more aggressive in those cases. But they have full control over that in the tuning interface should they want to dampen it down.

    Regardless it's just two ways to do things. In the future we will allow customers to select either method.
    Cute.

    This will be quick:
    The amount of boost spike that 6ATs experience at the upshift is dependent on two things:
    -how much boost is present prior to the shift (which is proportional to turbine speed-- the more speed the longer it times to decelerate when the next gear is selected)
    -how much throttle closure occurs during the boost spike

    Terry is comparing a log of a 6AT running 14+psi prior to the shift to a log that is running just 11psi at the shift. So duh.
    Also, Terry fails to mention that we have a user adjustable called 6AT Upshift softening which dictates how much throttle closure occurs at the shift. When it is set to 0 (as the log above), you get no throttle closure. When it is set to 50%, you get twice as much as stock (and obviously far less shift spike).

    it's also worth mentioning that the Procede's datalogs have 3-4x the sampling rate than the jb3 datalogging (not to mention channels that are accurate and actually useful to datalog). So things are certainly going to be missed in jb3 logs. Which is why Terry is working feverishly to import CAN data and raise sampling rate.

    Shiv

    PS. I'll answer Sticky's questions later today when I get more spare time.
    PSS. And please stop putting "at signs" in your posts. I have to edit them out when replying since my membership status doesn't allow them to be in my posts Click here to enlarge

  8. #8
    Join Date
    Feb 2010
    Location
    Los Angeles
    Posts
    12
    Rep Points
    4.8
    Mentioned
    0 Post(s)
    Rep Power
    0


    Yes Reputation No
    great stuff!
    Click here to enlarge

  9. #9
    Join Date
    Nov 2009
    Location
    Simi Valley, CA
    Posts
    8,129
    Rep Points
    9,106.3
    Mentioned
    644 Post(s)
    Rep Power
    92


    Yes Reputation No
    A couple logs from our 135i on the way to lunch. Fully modified, map 7, default settings, forced 3->4 shift @ 4800rpm as boost is ramping up. The most difficult situation to manage spikes on. Sampling at 10/sec. Will repeat testing with V4 when time permits on defaults.

    On sample speed, Shiv, how many samples/second is the V4 pulling from the ECU on say timing advance or throttle blade position? (hint: you may be surprised by the answer).

    Click here to enlarge

    Click here to enlarge

  10. #10
    Join Date
    Feb 2010
    Location
    NJ
    Posts
    13,462
    Rep Points
    58.0
    Mentioned
    318 Post(s)
    Rep Power
    0


    Yes Reputation No
    freedom of speech at its best right here ladies and gents

  11. #11
    Join Date
    Apr 2010
    Location
    USA
    Posts
    650
    Rep Points
    731.7
    Mentioned
    16 Post(s)
    Rep Power
    8


    Yes Reputation No
    +1

  12. #12
    Join Date
    Nov 2009
    Location
    USA
    Posts
    2,079
    Rep Points
    1,142.0
    Mentioned
    11 Post(s)
    Rep Power
    12


    Yes Reputation No
    To all members, we're really not here to incite any sort of debate, but rather get as much as information as possible from all tuners.

    That way, the public can choose what they want according to their needs and information, and yes, it is definitely good information!

  13. #13
    Join Date
    Mar 2010
    Posts
    642
    Rep Points
    97.0
    Mentioned
    51 Post(s)
    Rep Power
    0



    Yes Reputation No
    Click here to enlarge Originally Posted by DarkPhantom Click here to enlarge
    I'm interested in this too:
    1) The jb3 can't read the boost target due to lack of CANbus integration. It can only estimate it.

    I know the CANbus feature has been explained before, but in terms of reading the boost target, can you go a bit more indepth?

    Does the CANbus feature differentiate from the flashed tunes as well, such as the GIAC/others?
    Without the ability to read CANbus data, you are forced to "model" the boost target and the wastegate DC required to acheive it. In the real world, your model can ever work on all cars in all consitions. This is fact. So you have some cars that trigger underboost codes and other cars that don't. Same goes for unwanted throttle closures.

    With CANbus integration, you can literally read the boost target that the DME is expecting as well as the actual boost pressure the DME is perceving. No need to estimate anything, model anything, etc,. All you do is manage that delta with a simple PID system. There is no guestimation involed. As a result, boost control will work identically well in any car, regardless of condition, wastegate stiffness or even DME flash (stock or otherwise). With the Procede, you can slap it on a Giac flashed car and it will control boost just as perfectly as if you installed it on a stock DME car.

    I'd be willing to participate in a 3rd party boost control performance comparison if Terry is up to it. Not sure if this is even a question at this point. But if Terry believes otherwise, I'll be happy to participate.

    Shiv

  14. #14
    Join Date
    Mar 2010
    Posts
    642
    Rep Points
    97.0
    Mentioned
    51 Post(s)
    Rep Power
    0



    Yes Reputation No
    Click here to enlarge Originally Posted by Sticky Click here to enlarge
    Ok, some questions.

    1. What does the CANbus integration allow you to do different with the boost target? Aren't you still relying on certain factory parameters? As in, won't the factory ECU still call for more boost in certain situations?
    What the factory ECU calls for is completely irrelevent when you double the boost pressure. DME logic suggests raising the boost when IATs get hot and dropping it when it gets cold. it suggests raising boost 1psi for every 1psi of barometric pressure reduction. Both schemes aren't feasible (or even remotely desirable) when running a turbo size limited sytem at full boost.

    2. Yes, it is variable. Why is limiting it just an estimate? Seems you are limiting in specific instances, doesn't sound like an estimate?
    Because if it wasn't a estimate, you would get rocks solid and consistent boost performance out of a jb3. Which isn't remotely close to being true. Also, the compensations the DME applys don't come from simple 2D maps. There are actaully over a dozen 3D tables used to compensate for baro, temp and fuel trims. You are not going to be able to reverse engineer that at this level.

    3. Why can't it decrease it?
    Because the Jb3 does not isolate the wastegate DC output signal from the input signal. All it can do is apply positive tweaks to it (by pulling it down to ground). So the factory DME's wastegate control DC represents the absolute minimum DC allowable by the jb3 tuned system. You can probably see the shortcomings of this approach. Especially when running upgraded turbos.

    No, I did not realize the DME adjusts boost in the exact opposite way that you would want it to in a tuned car. Why do you say that?
    Already explained above.

    Cheers
    shiv

  15. #15
    Join Date
    Apr 2010
    Posts
    254
    Rep Points
    328.3
    Mentioned
    11 Post(s)
    Rep Power
    4


    1 out of 1 members liked this post. Yes Reputation No
    Click here to enlarge Originally Posted by Terry@BMS Click here to enlarge
    Here is one of his customer logs showing this major spike in action. Also a JB3 log that had been posted in that same thread showing no shift spike. Even if you force a shift @ peak HP on the JB3 you'll get almost no spike on the street maps.

    Regardless it's just two ways to do things. In the future we will allow customers to select either method.
    Click here to enlarge Originally Posted by shiv@vishnu Click here to enlarge
    Cute.

    Terry is comparing a log of a 6AT running 14+psi prior to the shift to a log that is running just 11psi at the shift. So duh.
    Also, Terry fails to mention that we have a user adjustable called 6AT Upshift softening which dictates how much throttle closure occurs at the shift. When it is set to 0 (as the log above), you get no throttle closure. When it is set to 50%, you get twice as much as stock (and obviously far less shift spike).

    I think that is my Procede graph Terry posted.........so I have a couple of comments/questions.

    First off.....Shiv is right. If you look closely I was shifting at 6000 rpm before any boost taper had taken place, so it wasn't fair to compare to a JB graph that already had about 4 psi of taper.

    But even so....both tunes are very capable of hitting 16 psi - and 16 psi is 16 psi to the engine (as shown in Terry's second set of graphs), so why is Terry inferring the engine is seeing the spike as something harmful when his Map 7 log is hitting the same psi. Click here to enlarge

    Unfortunately, I didn't have ignition timing on that particular log, but I think it's correct to assume that the drop of ignition timing to zero degress at the shift point serves to dampen the impact of the spike. Why is that not mentioned? I thought that was part of the shift mechanism. Seems kind of deceptive to leave that part out - but I could be wrong here.

    OK....so now for something Shiv is saying that clashes with my thinking.

    It is true that I run zero upshift softening.......so the throttle blade stays wide open during the shift.

    But I don't see how using more upshift softening would serve to reduce the spike on the TMAP side of things. Given that the TMAP sensor is in front of the throtle plate, then wouldn't any amount of throttle closure only serve to create more of a spike measurement?

    Inducing some Throttle closure would reduce the spike the engine sees for sure, but I think the exact opposite would happen on the other side of the throttle plate.

    Unfortunately, it's that side which shows up on the logs.

    Bottom line, I dont feel that 16psi with zero ignition timing is going to do anything harmful to the engine.......the upside is that boost stays high between shifts and recovers very quickly.

    If I was worried about a spike getting too high, I would just do what Shiv suggest.....inlcude some upshift softening so that the throttle plate closes between shifts and diverts the pressure to the diverter valves.

    It's nice to have that option.Click here to enlarge
    Last edited by DCAFS; 06-19-2010 at 08:03 PM.

  16. #16
    Join Date
    Nov 2009
    Location
    USA
    Posts
    2,079
    Rep Points
    1,142.0
    Mentioned
    11 Post(s)
    Rep Power
    12


    Yes Reputation No
    If I was worried about a spike getting too high, I would just do what Shiv suggest.....inlcude some upshift softening so that the throttle plate closes between shifts and diverts the pressure to the diverter valves.
    Do you think the issue would be alleviated with a BOV?

  17. #17
    Join Date
    Nov 2009
    Location
    Simi Valley, CA
    Posts
    8,129
    Rep Points
    9,106.3
    Mentioned
    644 Post(s)
    Rep Power
    92


    Yes Reputation No
    Right, that was your log from the thread on n54tech a couple weeks ago. Where I discussed the various items I was experiencing with V4 automatic maps.

    As I've said before there are benefits and drawbacks to integrated boost control and benefits and drawbacks to isolated boost control. In laying out a new PCB (For N55, more advanced N54 users, etc) we've actually put accommodations for both with a jumper to move between them.

    Isolated boost control can't work with the throttle body for boost management as effectively and I'm afraid there might be no good solution to solving the low load automatic shifting and misfire problems that plague it. We continue to get a few JB3 converts per month for those reasons alone. On the other hand it has some advantages in manuals and high boost applications.

    Also as I've said before I'm not convinced the spikes are a big deal either. But they are there and since this was a thread about boost control it makes sense to discuss. Sure you could close the throttle body more during the shift but that is easier said than done with isolated control. Especially on forced unexpected shifts.

  18. #18
    Join Date
    Apr 2010
    Posts
    254
    Rep Points
    328.3
    Mentioned
    11 Post(s)
    Rep Power
    4


    Yes Reputation No
    Click here to enlarge Originally Posted by Terry@BMS Click here to enlarge

    Also as I've said before I'm not convinced the spikes are a big deal either. But they are there and since this was a thread about boost control it makes sense to discuss. Sure you could close the throttle body more during the shift but that is easier said than done with isolated control. Especially on forced unexpected shifts.
    Agreed.....the forced shifts before redline are a bit problematic.

    They just happen to be the ones I use most often. Click here to enlarge

  19. #19
    Join Date
    Apr 2010
    Posts
    254
    Rep Points
    328.3
    Mentioned
    11 Post(s)
    Rep Power
    4


    Yes Reputation No
    Click here to enlarge Originally Posted by DarkPhantom Click here to enlarge
    Do you think the issue would be alleviated with a BOV?
    I don't know.....but I am not a fan of them sound wise.

    However, I have to be fully transparent here and state that I do run the forge dv's with yellow springs.....so I am pretty sure there is going to be a bit more pressure building up in the charge pipe before they dump.

  20. #20
    Join Date
    Mar 2010
    Posts
    642
    Rep Points
    97.0
    Mentioned
    51 Post(s)
    Rep Power
    0



    Yes Reputation No
    Click here to enlarge Originally Posted by DCAFS Click here to enlarge
    I think that is my Procede graph Terry posted.........so I have a couple of comments/questions.

    First off.....Shiv is right. If you look closely I was shifting at 6000 rpm before any boost taper had taken place, so it wasn't fair to compare to a JB graph that already had about 4 psi of taper.

    But even so....both tunes are very capable of hitting 16 psi - and 16 psi is 16 psi to the engine (as shown in Terry's second set of graphs), so why is Terry inferring the engine is seeing the spike as something harmful when his Map 7 log is hitting the same psi. Click here to enlarge

    Unfortunately, I didn't have ignition timing on that particular log, but I think it's correct to assume that the drop of ignition timing to zero degress at the shift point serves to dampen the impact of the spike. Why is that not mentioned? I thought that was part of the shift mechanism. Seems kind of deceptive to leave that part out - but I could be wrong here.
    You are 100% correct. It actually goes to negative advance during the shift event. But the Procede is only scaled to read positive advance at this time.

    OK....so now for something Shiv is saying that clashes with my thinking.

    It is true that I run zero upshift softening.......so the throttle blade stays wide open during the shift.

    But I don't see how using more upshift softening would serve to reduce the spike on the TMAP side of things. Given that the TMAP sensor is in front of the throtle plate, then wouldn't any amount of throttle closure only serve to create more of a spike measurement?
    Again, 100% correct. You would see more positive spike in the TMAP (pre-throttle blade). But with the thorttle partially (or fully) closed, you wouldn't see it at the engine.

    Inducing some Throttle closure would reduce the spike the engine sees for sure, but I think the exact opposite would happen on the other side of the throttle plate.
    Yep!

    Unfortunately, it's that side which shows up on the logs.

    Bottom line, I dont feel that 16psi with zero ignition timing is going to do anything harmful to the engine.......the upside is that boost stays high between shifts and recovers very quickly.
    Yep!

    If I was worried about a spike getting too high, I would just do what Shiv suggest.....inlcude some upshift softening so that the throttle plate closes between shifts and diverts the pressure to the diverter valves.

    It's nice to have that option.Click here to enlarge
    Yep!
    btw, you will different spike behavior depending on what bypass valve/spring rate you are running. The stiffer the spring, the more spike pressure is needed to lift the piston and relieve the excess charge pressure. Totally normal.

    Shiv

  21. #21
    Join Date
    Jan 2010
    Location
    SoCal
    Posts
    119,478
    Rep Points
    32,154.9
    Mentioned
    2109 Post(s)
    Rep Power
    322


    Yes Reputation No
    Click here to enlarge Originally Posted by shiv@vishnu Click here to enlarge
    Cute.

    This will be quick:
    The amount of boost spike that 6ATs experience at the upshift is dependent on two things:
    -how much boost is present prior to the shift (which is proportional to turbine speed-- the more speed the longer it times to decelerate when the next gear is selected)
    -how much throttle closure occurs during the boost spike

    Terry is comparing a log of a 6AT running 14+psi prior to the shift to a log that is running just 11psi at the shift. So duh.
    Also, Terry fails to mention that we have a user adjustable called 6AT Upshift softening which dictates how much throttle closure occurs at the shift. When it is set to 0 (as the log above), you get no throttle closure. When it is set to 50%, you get twice as much as stock (and obviously far less shift spike).

    it's also worth mentioning that the Procede's datalogs have 3-4x the sampling rate than the jb3 datalogging (not to mention channels that are accurate and actually useful to datalog). So things are certainly going to be missed in jb3 logs. Which is why Terry is working feverishly to import CAN data and raise sampling rate.

    Shiv

    PS. I'll answer Sticky's questions later today when I get more spare time.
    PSS. And please stop putting "at signs" in your posts. I have to edit them out when replying since my membership status doesn't allow them to be in my posts Click here to enlarge
    What I gather from this is that there will be inconsistency with any piggyback as the factory ECU will still cal for more boost or different boost depending on different conditions. The piggyback should take this into account but even a V4 will spike.

    The sampling rate is interesting and I don't doubt the V4 has a higher rate. The sampling per dollar is in the favor of the JB3 though Click here to enlarge There are tradeoffs, but what exactly would a JB3 log miss because I haven't ever heard of significant data being left out or glossed over. Yes, BMS is working to add the CAN data functionality, clearly it is a good thing. It will be interesting to see what price point it comes in at.

    Stage 2 or 2.5 E9X M3 S65 V8 supercharger kit for sale
    : http://www.boostaddict.com/showthrea...r-kit-for-sale

  22. #22
    Join Date
    Jan 2010
    Location
    SoCal
    Posts
    119,478
    Rep Points
    32,154.9
    Mentioned
    2109 Post(s)
    Rep Power
    322


    Yes Reputation No
    Click here to enlarge Originally Posted by shiv@vishnu Click here to enlarge
    What the factory ECU calls for is completely irrelevent when you double the boost pressure. DME logic suggests raising the boost when IATs get hot and dropping it when it gets cold. it suggests raising boost 1psi for every 1psi of barometric pressure reduction. Both schemes aren't feasible (or even remotely desirable) when running a turbo size limited sytem at full boost.



    Because if it wasn't a estimate, you would get rocks solid and consistent boost performance out of a jb3. Which isn't remotely close to being true. Also, the compensations the DME applys don't come from simple 2D maps. There are actaully over a dozen 3D tables used to compensate for baro, temp and fuel trims. You are not going to be able to reverse engineer that at this level.



    Because the Jb3 does not isolate the wastegate DC output signal from the input signal. All it can do is apply positive tweaks to it (by pulling it down to ground). So the factory DME's wastegate control DC represents the absolute minimum DC allowable by the jb3 tuned system. You can probably see the shortcomings of this approach. Especially when running upgraded turbos.



    Already explained above.

    Cheers
    shiv
    You state what the factory ECU calls for is irrelevant. When you double or raise the boost, won't the ECU still suggest raising or lowering boost in certain situations leading to fluctuations? Why would this factory aspect just suddenly stop, isn't it taken into account? Is it disabled? This question is for both the V4 and JB3.

    I don't think you can attribute rock solid and consistent boost performance to any piggyback as a flash will be more consistent, won't it? You attributing not being able to reverse engineer all those tables (which I definitely agree with) to all piggybacks, correct?

    The V4 does isolate the wastegate DC output signal from the input signal? Shortcomings may not be the term I would use as I'm not convinced the wastegate control is not enough to run factory style upgraded turbos (modified OEM turbos). I'm just not sure what to think on this one yet although greater control is preferable, definitely.

    Stage 2 or 2.5 E9X M3 S65 V8 supercharger kit for sale
    : http://www.boostaddict.com/showthrea...r-kit-for-sale

  23. #23
    Join Date
    Mar 2010
    Posts
    642
    Rep Points
    97.0
    Mentioned
    51 Post(s)
    Rep Power
    0



    Yes Reputation No
    Click here to enlarge Originally Posted by Sticky Click here to enlarge
    What I gather from this is that there will be inconsistency with any piggyback as the factory ECU will still cal for more boost or different boost depending on different conditions. The piggyback should take this into account but even a V4 will spike.
    You're not following. The Procede is not a piggyback as far as boost control goes. Unlike the jb3, it has absolute control over the boost control solenoids. The Procede defines its own setpoint which is mapped as clear as black and white. And it does a ridiculously good job of hitting that set point, time and time again. Regardless of mods, wastegate stiffness, or even turbo (stock up upgraded). And this can be proven by any customer running the Procede. Not just me with my car.

    The sampling rate is interesting and I don't doubt the V4 has a higher rate. The sampling per dollar is in the favor of the JB3 though Click here to enlarge There are tradeoffs, but what exactly would a JB3 log miss because I haven't ever heard of significant data being left out or glossed over. Yes, BMS is working to add the CAN data functionality, clearly it is a good thing. It will be interesting to see what price point it comes in at.
    There are two sampling rates as far as I'm concerned: Adequate and inadequate. It doesn't matter if you have 30 samples per sec or 300 samples per sec. Both are perfectly fine and present enough info to see what the engine is doing. At the jb3's 10 samples per sec, you miss any even that is less than .1s long (like a derivitive correction from a pid boost controls system for instace). And you miss 1/2 of a .2 sec long 6AT boost spike.

    Shiv

  24. #24
    Join Date
    Jan 2010
    Location
    SoCal
    Posts
    119,478
    Rep Points
    32,154.9
    Mentioned
    2109 Post(s)
    Rep Power
    322


    Yes Reputation No
    Click here to enlarge Originally Posted by shiv@vishnu Click here to enlarge
    You're not following. The Procede is not a piggyback as far as boost control goes. Unlike the jb3, it has absolute control over the boost control solenoids. The Procede defines its own setpoint which is mapped as clear as black and white. And it does a ridiculously good job of hitting that set point, time and time again. Regardless of mods, wastegate stiffness, or even turbo (stock up upgraded). And this can be proven by any customer running the Procede. Not just me with my car.



    There are two sampling rates as far as I'm concerned: Adequate and inadequate. It doesn't matter if you have 30 samples per sec or 300 samples per sec. Both are perfectly fine and present enough info to see what the engine is doing. At the jb3's 10 samples per sec, you miss any even that is less than .1s long (like a derivitive correction from a pid boost controls system for instace). And you miss 1/2 of a .2 sec long 6AT boost spike.

    Shiv
    In that case, correct, I was not following. I consider/considered the Procede a piggyback. So what I gather is that the Procede overrules any of these factory settings and simply straight dictates the boost regardless of conditions. However, what does the Procede do when the intake temps get too hot or low? Does it compensate?

    Wasn't there a log shown that displayed spikes with the V4?

    Well, as far as you are concerned. I will take a wild guess as to which sampling rate you consider adequate or inadequate Click here to enlarge Before I comment further, you state the JB3's current sampling rate as 10 samples per second, what is the V4's?

    Stage 2 or 2.5 E9X M3 S65 V8 supercharger kit for sale
    : http://www.boostaddict.com/showthrea...r-kit-for-sale

  25. #25
    Join Date
    Mar 2010
    Posts
    642
    Rep Points
    97.0
    Mentioned
    51 Post(s)
    Rep Power
    0



    Yes Reputation No
    Click here to enlarge Originally Posted by Sticky Click here to enlarge
    In that case, correct, I was not following. I consider/considered the Procede a piggyback. So what I gather is that the Procede overrules any of these factory settings and simply straight dictates the boost regardless of conditions. However, what does the Procede do when the intake temps get too hot or low? Does it compensate?
    Yes, it has both intake temp and baro compensations. And as of now, it has fuel octane/aggression compensations as well. All of which are always active.

    Wasn't there a log shown that displayed spikes with the V4?
    If you are referring to the spike that occurs during shifts, sure. That's a function of physics. Not of boost control shortcomings.

    Well, as far as you are concerned. I will take a wild guess as to which sampling rate you consider adequate or inadequate Click here to enlarge Before I comment further, you state the JB3's current sampling rate as 10 samples per second, what is the V4's?
    Procede samples all channels (for datalogging) at 30-35/sec. There is nothing stopping us from doubling that. We just stopped there because it was fast enough to catch a derivitive boost control correction which is just under .05 second in duration.

    Shiv

Page 1 of 3 123 LastLast

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may post replies
  • You may not post attachments
  • You may not edit your posts
  •