Part throttle boost (closed loop) tuning for FI

Discussion in 'Engine & Performance Modifications' started by Mr MoPar, Jun 7, 2009.

  1. Quick

    Quick Mgmt. - I can't help you

    Messages:
    7,549
    Likes Received:
    0
    Joined:
    Nov 6, 2007
    Location:
    San Jose, CA
    This is from the service manual and would match the description of the Zirconia sensor.
    Maybe they just cut'n'pasted old text from one manual to the next?
     
  2. Mr MoPar

    Mr MoPar New Member

    Messages:
    88
    Likes Received:
    0
    Joined:
    Apr 11, 2009
    What year manual is that from? Does not match up with the trouble shooting data I have from Chrysler, and not with datalogging data on the O2 sensors, mine hovers around roughly 3.2v to 4.12v signal back to the ECU. A Zirconia sensor can't produce that voltage level. I will have to dig up some additional information.
     
  3. Quick

    Quick Mgmt. - I can't help you

    Messages:
    7,549
    Likes Received:
    0
    Joined:
    Nov 6, 2007
    Location:
    San Jose, CA
    2006 LX 300 Series/SRT-8, Charger, and Magnum Service Manual

    From Chrysler Tech Authority. Part# 81-270-06065CD (I think)
     
  4. lafrad

    lafrad Supporting Vendor

    Messages:
    17
    Likes Received:
    0
    Joined:
    Apr 10, 2008
    First of all, *I*, by no means have behind the scenes knowledge beyond some casual conversations with tuning vendors out there. HEMI's are the only platform (and a few V6's) that I have had experience with, and it may just be that that has encouraged me to find some ways to get these things working sufficiently well for what we are asking them to do. I am very much open for dialog on ANY of the topics, and will try to back up anything I have done with examples... if I can't, its fair game,hit me with all ya got :-D

    I personally do not DRIVE a boosted LX, I have a stroker N/A motor. This has been a handicap for me, and I have fortunately had the luxury to have several people work with me to "back into" safe tunes. Several have been on Stock block setups (with heads and cam) an have been daily drivers for them. This may very well be where I have learned some concepts from, and it also may be part of why this will always be a learning process. Most of my experiance is with Centrifugal Superchargers... with a positive displacement blower or two and a single turbo setup back there.

    I want to correct something: I'm going to try to refer to "more fuel" as "Richer AFR". I'll try to explain in my comments.

    Okay, we can see if I can do this quote thing... I'm not very good at it :p
    Green will be me


    Gonna move outta the quote for the last part...

    Lots of terms here, some mean what you think others are... slightly odd... based on how mopar decided to use the numbers.

    The LX computer is a 100% "Speed Density" system. it takes "Speed" (RPM), and "Density" (Manifold Pressure), combines those KNOWN values with "assumed" values in the tune, to figure out how much fuel is needed for any given operating condition. Its been used for quite awhile, and has some advantages and pitfalls when it comes to busting away from stock parts.

    First: with a good tune, Speed Density is really VERY difficult to beat, when it comes to operating an engine. if you can account for enough of the things that change the combustion process in your engine, it is able to react to changes in throttle, load, gears, RPM, EXTREMELY quickly. MAP sensors are lightning quick for signal, and the computer can figure out the RPM to its maximum accuracy all the way up to redline (6400) and beyond, *FOR EVERY SINGLE COMBUSTION EVENT that you have. very impressive work for just a bunch of transistors and a few sensors.

    Now: "Volumetric Efficiency", in LX Tuning world, Is a measure of "how efficient is *this engine*, at *This PRatio* at *This RPM* at filling each cylinder with combustion gasses.
    To make the explanation a little bit easier, I'm going to replace "Pratio" with "Manifold Pressure". The sensors we get to use for manifold pressure can only sense the pressure that is occurring *right now*, in relation to *PURE VACUUM*. Since we (thankfully) live in a world with a significant air pressure around us, we have to account for the fact that it may not be the same pressure everywhere we go. There is less pressure in Denver, do the elevation, than there is in Boston. a "PRatio" of 1.0 in Denver and a "PRatio" of 1.0 in Boston both mean that the throttle blade is most likely 100% open, but there is definitely a different "weight" of air in the engine. the computer adjusts for this behind the scenes, and its not relevant to our discussion.

    For an ideal Naturally Aspirated 5.7L engine, an example is as such:
    5.7L engine
    x
    1000 RPM
    x
    100% efficient per combustion cycle for Manifold Pressurized air
    _divided by_
    2 revolutions per combustion cycle
    =
    2850L of fresh, atmospheric pressure air/fuel mixture THROUGH the engine

    Our cam/intake manifold, etc is no where NEAR 100% on the volumetric efficiency scale (its more in the area of 60-70% efficient at that RPM).
    Notice I kept the "Manifold pressure" tag on there. 5.7 L of air at current "Manifold pressure" can go through some math to give you the "Air Mass" that is there. a specific mass of air requires a certian amount of fuel for proper, 14.7:1 AFR combustion. this amount of air is what the computer uses to determine how much fuel it needs to add for a good running motor.

    Now, if you start changing that "Manifold Pressure" (you close the throttle blade, or you start building boost), you change how much mass the air actually contains. in high vacuum situations, there is relatively little air mass, in boost, the air gets much heavier. This means, in a vacuum, 100% "Volumetric efficiency" is related to the MANIFOLD pressure, not the air that you and I breathe. 100% VE at 7 psi of vacuum is (ROUGHLY) half of the air mass that is present at no vacuum. 100% of VE at 7psi of BOOST is (roughly) 50% more of the air mass that is present at no boost.

    Volumetric efficiency, when used in a (simplified) relation to manifold pressure like our LX cars are, becomes almost *exclusively* modified by the profile of the camshaft. Does the Camshaft profile let the engine efficiently fill the cylinder at individual RPM and Manifold Pressure data points? Intake manifold and Exhaust can help adjust that, but usually its only in broad sweeps... intake may increase all of it by 5%. Exhaust may increase everything above 3000 rpm by 3%. "larger cams" usually DECREASE the efficiency at low RPM and pressures, as the manifold vacuum is able to draw exhaust back INTO the combustion chamber when both valves open. As soon as RPM increases, overlap and large durations allow for higher efficiencies than stock!

    Where does this leave boost? Well, depending on how you MAKE the boost, you very well can have your N/A tune work *exactly the same* in all the vacuum areas as your boost tune. Superchargers are able to get away with running a *VERY* close to stock tune in the vacuum areas. Now, getting into the boost areas, do the efficiencies go UP? not really. They mostly follow the curve that was set in the N/A areas of the tune, with some exception. Exhaust overlap allows boost to flow through the cylinder and go out the other side. This doesn't make ANY power, but it MUST be accounted for because we are measuring Air Fuel ratios for ALL air going through the system, therefore, raising the volumetric efficiencies at higher levels of boost. Turbo backpressure actually pushes back on the exhaust, and can LOWER the volumetric efficiency due to that fact. In the end, getting the tune "right" is still matching up the "assumed" tables with what actually is coming outta the tailpipe, while taking into account all the gobbly-gook that is provided from the factory.

    After all of this, you throw in the fact that some assumptions can be made: "a given air mass, with proper fueling, will create a specific amount of horsepower" Now you are getting into loading areas, estimated torque, and how the transmission reacts to that. I think we can save that part for another time.
     
  5. lafrad

    lafrad Supporting Vendor

    Messages:
    17
    Likes Received:
    0
    Joined:
    Apr 10, 2008
    In my previous post, I suggested we have an exhaust and catalyst overtemp protection enrichment mode.

    These are VERY interesting. THe LX computers do NOT have EGT sensors to read from, but, apparently someone came up with some rather complex calcuations to ESTIMATE the exhaust and catalyst temperatures over time, given a set of very specific variables.

    FORTUNATELY FOR US, I have found that most finely tuned N/A and Boosted cars have ESTIMATED exhaust temps that are VERY VERY close to what an actual EGT sensor would read! with all the mods we can do, a good tune, good injector scale, and some revisions to make sure things are right end up keeping those equations happy.

    I was SHOCKED at this, but experiencing it on my own car was the real kicker.

    With my stroker builds for my car, I have kept MDS active all the way through. MDS is basically running the 4 cylenders that are active, in closed loop, near WOT for extended periods of time on the freeway. On my trip from Milwaukee to Minnesota a couple years back, I MDS was kicking out after extended periods of time... without any real explanation. the road was flat, cruise was on, but you could hear the tone change, and feel the little "bump" that occurs with the high stall converter and MDS operation. I searched through my dash hawk to figure out why... and found that MDS was driving the estimated Exhaust and Cat temps too high, and the "Catalyst overtemp Enrichment" mode automatically disengages MDS while it cools things down.

    After a little research with a EGT probe, I found it was doing its job admirably, and promptly found some midpipes to... solve... the cat overtemp problems. I have since been able to work the tunes out to keep MDS and catalysts happy... at least when you are not beating the crap out of the car... but its all related to spark and tuned VE values.

    the cool part about this all, is that it does a damn good job with boosted vehicles too. If you fine tune the boost areas for VE and Spark, the estimated EGT's will be very close to actual values, and the preprogrammed "overtemp enrichment" programs can do their job. with some fiddling to how much fuel it adds for those programs, you can get quite close on keeping the temps in line without having to have wideband sensors for those "hot spots"
     
  6. Quick

    Quick Mgmt. - I can't help you

    Messages:
    7,549
    Likes Received:
    0
    Joined:
    Nov 6, 2007
    Location:
    San Jose, CA
    This part. I think you're saying that you can change the "desired"/base fuel, but the adaptives will bring it back to stoich.

    A couple of preliminary questions.

    1) Is the base injector pulse width a fixed value for all of closed loop or is it read from a table with cells? If so, a decent number of cells?

    2) Just how fast do the adaptives adjust? few seconds for a large change?

    Let's simplify the example and assume I don't care about adapting to DA changes.

    The short and long term adaptives can adjust up to 25% (each) from the base injector pulse width for a total of 50%. What happens if you set the base injector pulse width to overload the adaptives?

    Just for example let's say the base is 100 which results in 14.7. The adaptives are capable of making a max adjust of 50. If I wanted an afr of about 13.2 could I set the base to be 165? The adaptives would max out reducing it by 50 and I'd be left with a pulse of 115... which should result in about 13.2. It would just be fixed with no adaptation possible but if you had a few cells to do that you would have a sort of fixed tune. (heh, I live in CA and the car is always at sea level).

    No, I'm not suggesting this as a tuning solution. Just curious what would happen. Would that sort of work or would the PCM put you in limp mode as soon as the adaptives maxed out. or something else.
     
  7. 8 UR LS

    8 UR LS New Member

    Messages:
    66
    Likes Received:
    0
    Joined:
    Sep 15, 2008
    Location:
    NOLA,LA
    WOW! :stars: ...this thread is giving me a major headache,but in a very good way!...great knowledge being shared here guys!! this has got to be the most informative threads i have ever read on ne forum!!..and this to me is what public forums is all about,it really is a beautiful thing to see everyone come together and help one another.:thanks:to tim and everyone else that has givin input in this thread! CONTINUE ON
     
  8. Mr MoPar

    Mr MoPar New Member

    Messages:
    88
    Likes Received:
    0
    Joined:
    Apr 11, 2009
    Okay, I did some looking in several manuals (including the one you have), and this is what I came up with.

    You are correct, sort of! Read on.

    The information you posted is correct, but it is the context of what it actually means that can give the wrong impression. The values given are correct, but it does not give an explanation of the values, or how they are derived. A typical Zirconia narrowband sensor produces a voltage based on the amount of un-burnt fuel present. Those normal values mimic the values you posted from the service manual. But, a Zirconia sensor is limited to a 0v to 1v range.

    The values described in the service manual that you posted are a sum value. It is derived by subtracting the output voltage from the input voltage. As I mentioned, the actual data from the O2 sensors is floating in a 0v to 5v input signal range. The ECU knows the input voltage, and simply subtracts that from the output voltage to derive the sum value which is typically between the .29v and .745v. This is why I mentioned that you can’t use the old narrowband tricks. A fixed simulated input, an applied resistance (again a fixed value), or a repeating tic/tock signal, would all be detected by the ECU in a very short time.

    You can see that for a normal operating sensor, if the ECU lowers the input voltage, the output voltage will also change lower. However, the sum value for a normally operating sensor will not change. This is how it can determine a faulty sensor, or one that has been tampered with.

    The information below is from the diagnostic section. You can clearly see from the voltage checks, that this can not be a Zirconia sensor, but the newer Titania sensor which changes its resistance based on un-burnt fuel, and therefore changes the output voltage correspondingly.

    The values below indicate the normal low input (below 1.5v), high input (above 4.8v), and midpoint output (2.5v) range for the sensor.

    1. O2 SENSOR BELOW 1.5 VOLTS
    Start the engine.
    Allow the engine to reach normal operating temperature.
    With a scan tool, read the 1/1 O2 Sensor voltage.

    Is the voltage below 1.5 volts?

    2. (K41) O2 SENSOR 1/1 SIGNAL CIRCUIT
    Turn the ignition off.
    Disconnect the 1/1 O2 Sensor harness connector.
    Ignition on, engine not running.
    With a scan tool, monitor the O2 Sensor voltage.

    Is the 1/1 O2 Sensor voltage above 4.8 volts?

    3. O2 SENSOR
    Measure the voltage on the (K902) O2 Return Upstream circuit in the
    O2 Sensor harness connector.

    Is the voltage at 2.5 volts?

    Hope this helps.
     
  9. lafrad

    lafrad Supporting Vendor

    Messages:
    17
    Likes Received:
    0
    Joined:
    Apr 10, 2008
    Comments in green again.

     
  10. Flanman

    Flanman New Member

    Messages:
    43
    Likes Received:
    0
    Joined:
    Feb 16, 2008
    This is some great reading. I have the privilege of knowing Mike personally and living close enough to him to have him road tune my car. He's one sharp dude, and there are very few tuners out there that match his ability to tune the LX's. We need to get to the track together as soon as my ride is back up and running.
     
  11. Mr MoPar

    Mr MoPar New Member

    Messages:
    88
    Likes Received:
    0
    Joined:
    Apr 11, 2009
    I thought it would be a good reference to show what part throttle boost looks like, and what fuel correction can look like during closed loop operation so others can compare to what they are seeing.

    The following is datalog information from the very first tuning files on my former 2006 turbocharged 300C 5.7 Hemi. The data shows fuel correction for part throttle boost for the unit I am using. On my 2008 Jeep 5.7 Hemi, I have it set to come in at .25 psi (1/4 lb of boost), and have it running about .5 AFR richer than this datalog shows. The ECU is running an unmodified factory file.

    View below is a zoom detail view providing detail of the transition into part throttle (Closed Loop) low boost operation. Note, the vehicle remained in closed loop at all times and did not transition to Open Loop (WOT). Throttle position never exceeded 35%, and was reduced to a low of 26% just prior to the rpm induced shift point.

    [​IMG]

    At 104 seconds, tip in is started by slowly easing into the pedal from a steady cruise state. Just before 110 seconds, vehicle disengages from 5th overdrive and drops into 4th gear and begins a slow steady increase in rpm.

    Just before 111 seconds, system senses the approach to boost threshold and takes control of the O2 sensors and begins to increase fuel and the AFR. This can be seen by the ripple of the AN2 out (O2) line going to a flat line.

    The AFR line starts to trend downward as fuel is increased.

    At 111 seconds, vehicle transitions into low boost, there has been minimal pedal movement and the vehicle is experiencing a slow gradual increase in speed and rpm with a smooth transition from vacuum to a boosted state.

    Data from 110 seconds during the downshift, -7.71 inches of vacuum and AFR 14.67, 35% throttle Position;

    RPM [rpm] - 110.00s ; 2168.4
    MPX [-in_PSI] - 110.00s ; -7.71
    IAT [F] - 110.00s ; 113
    AN2 Out [V] - 110.00s ; 3.95
    WB AFR [C] - 110.00s ; 14.67
    TP [V] - 110.00s ; 1.74 (35% throttle position)

    Data from 111 seconds following the downshift, boost pressure is 2.06 psi and AFR is 13.1, 34% throttle position;

    RPM [rpm] - 111.75s ; 3408.2
    MPX [-in_PSI] - 111.75s ; 2.06
    IAT [F] - 111.75s ; 110.75
    AN2 Out [V] - 111.75s ; 4.28
    WB AFR [C] - 111.75s ; 13.1
    TP [V] - 111.75s ; 1.7 (34% throttle position)


    Data from 113 seconds, boost pressure is 2.8 psi and AFR is 12.27, 30% throttle position;

    RPM [rpm] - 113.50s ; 4400.3
    MPX [-in_PSI] - 113.50s ; 2.8
    IAT [F] - 113.50s ; 115.5
    AN2 Out [V] - 113.50s ; 4.24
    WB AFR [C] - 113.50s ; 12.27
    TP [V] - 113.50s ; 1.5 (30% throttle position)

    Data from 115 seconds, boost pressure is 2.21 psi and AFR is 12.08, 27% throttle position

    RPM [rpm] - 115.30s ; 5328.1
    MPX [-in_PSI] - 115.30s ; 2.21
    IAT [F] - 115.30s ; 120.5
    AN2 Out [V] - 115.30s ; 4.2
    WB AFR [C] - 115.30s ; 12.08
    TP [V] - 115.30s ; 1.33 (27% throttle position)

    Data from 117 seconds, boost pressure is 3.53 psi and AFR is 12.55, 26% throttle position, this is following the up shift, and release of the pedal immediately before rpm and boost falls off back to a vacuum state. The AN2 (O2) sensor voltage immediately goes back to a ripple as the system returns to normal operation.

    RPM [rpm] - 117.00s ; 4046.8
    MPX [-in_PSI] - 117.00s ; 3.53
    IAT [F] - 117.00s ; 125.5
    AN2 Out [V] - 117.00s ; 4.16
    WB AFR [C] - 117.00s ; 12.55
    TP [V] - 117.00s ; 1.29 (26% throttle position)
     
  12. lafrad

    lafrad Supporting Vendor

    Messages:
    17
    Likes Received:
    0
    Joined:
    Apr 10, 2008
    In relation to the datalog, how do the Fuel trims react to the extra fuel being present in the mixture?

    how does the shift logic (on the truck I assume its less sensitive to odd airflow.. that is what I felt when doing some road tuning on them) respond to more power but no observed change?


    I guess I would suggest that you DO tune the whole ECM for handling all boost situations as it see's fit, then put the fuel enrichment piggyback *on top* of that... so it at least can account for the manifold pressure in airflow and spark...
     
  13. Mr MoPar

    Mr MoPar New Member

    Messages:
    88
    Likes Received:
    0
    Joined:
    Apr 11, 2009
    The fuel trims increase some on the short term as would be normal for any lean condition seen by the ECU from the O2 sensors. Neither the car or the Jeep exhibit any odd shift behaviors that are noticable.

    I have been toying with the idea of having a specialized file created in a couple of specific areas, it would be similar to an NA file for the most part, scaling for the larger injectors and possibly as you mentioned some VE update for airflow, but no fuel addition and nothing beyond the stock 1 bar map sensor which would be clamped to prevent over range. I would not want to modify anything in the ECU where it would require the ECU to "try" and interpret a boosted state, that is where I have seen issues pop up. Regarding timing, I have the ability to retard timing whenever boost is present, and right now I retard 3 degrees at 5psi and I never see any knock retard, but I am also spraying water/meth which helps. That would make a basic file to baseline.

    Throughout the entire operational range, I have complete control of the injectors to adjust their pulse width in .2 milisecond increments. In the closed loop state the larger injectors are scaled by a percent of original fuel pulse plus a secondary +/- offset to handle differences in injector operating times in the idle range. This allows normal operation of the larger injectors to mimic the stock injector flow in the vacuum range.

    When in the open loop state, fuel is added by directly adjusting the injector pulse width in the main fuel table as needed based on wideband data observation.

    As discussed, closed loop fuel addition for boost is handled by taking control of the O2 circuits and allowing the ECU to adjust the trims as it would normally do any other time it would see a lean condition.

    If someone really wanted to, and setup with an AFR probe on each cylinder, one could adjust each cylinder indivdually to provide an offset +/- based on variations in cylinder flow for a specific injector model based on the data obtained on each cylinder.

    This has been done on some Viper projects, but that would be for more high end projects.

    I would be happy to work with someone to experiment in this area to see what the results would be.
     
    Last edited: Jun 11, 2009
  14. lafrad

    lafrad Supporting Vendor

    Messages:
    17
    Likes Received:
    0
    Joined:
    Apr 10, 2008
    Ahhh.. I get it now..

    As soon as you hit boost, the ECM is no long doing anything to control injectors, and the "piggyback" (more like, full engine computer with smart passthrough) is giving the ECM good O2 sensor signals so it stays happy. It has no real idea that its going into boost, and the torque/airflow readings in the stock ECM are oblivious to the additional air and fuel moving through the system.

    That is a pretty slick setup, and should allow you to keep everything "cool" without getting into the overtemp protection tables in the stock programming. And it explains why you are suggesting "pulling timing" in boost. With CMR or SCT< you just write an entirely new part to the table, and just choose the spark you want from the get-go, based on the kPa given to you from the Map sensor. it allows you to put some pretty fine-tuned timing maps in all areas of boost in closed loop and WOT. THis is where I spend a LOT of time on stock boosted motors... pull the timing back enough to lower the calculated exhaust temp, but not too much to raise the calc'd cat temp. I've had good success with this. Now i just need a way to hande a pressurized throttle blade :doh: (don't worry Techco and Kenne bell guys, you don't have to worry about that part)

    I *would suggest* that you monitor trans line pressures closely on jeep/truck transmissions. they adjust based on commanded torque.. and if you are boosting in those zones, without letting the PCM know, it may be easier to toast up a clutch.

    I really don't feel that the computer has any trouble putting fuel or spark into a boosted engine, so long as the map sensor is scaled and the injectors are appropriately sized. with those bits in mind, the only sticky point left is the ETC, which several automotive platforms have had issues with in the aftermarket, not just the LX.

    The only things that it can't do is kick the fueling mode OUT of closed loop, and enrich the fuel mixture, when WOT has not been commanded. (part throttle stuff). At that point it just is a balancing act of making sure the internal engine temps are kept to reasonable levels, with or without boost, in closed loop 14.7:1 operation. you get spark and overtemp enrichment modes to help you out on this.

    I don't think the available "kits" on the market exceed that on stock motors. most limit boost to the 5-8psi range in the consumer kits, and a few have good sized intercoolers to help out too.

    On forged motors with low compression and proper ring gaps, you get a TON more room. Low compression alone reduces peak combustion chamber temps by quite a bit (but it INCREASES EGT's), and good ring gaps DO allow you *much much* more room to work with it. At that point it turns into a "keep the spark reasonable" so you don't melt a hole in a piston.
     
  15. Quick

    Quick Mgmt. - I can't help you

    Messages:
    7,549
    Likes Received:
    0
    Joined:
    Nov 6, 2007
    Location:
    San Jose, CA
    Tim, could you preface the table references with "ECU" or Unit (or piggyback or something). It's probably just me but I have to stop and think about whether you are referring to tables in the PCM or in your unit. So maybe "in the unit's main fuel table". It's easier when Mike's speaking since everything is in the PCM :D.

    This is what I was trying to get at earlier. You're referring to manual observation and test'n'tune to set the table to get a desired afr. If the wideband data was read dynamically and used for cell selection you could just "dial" an afr. You would gain adaptability in the open loop range -- haha, open the electronic exhaust cutouts while at full throttle and the afr stays the same!

    Seems that you could extend that to closed loop as well. Just seems a waste to only use that wideband for monitoring when it could be supplying an active control input :D
     
  16. Mr MoPar

    Mr MoPar New Member

    Messages:
    88
    Likes Received:
    0
    Joined:
    Apr 11, 2009
    See BOLD below.

     
    Last edited: Jun 11, 2009
  17. Mr MoPar

    Mr MoPar New Member

    Messages:
    88
    Likes Received:
    0
    Joined:
    Apr 11, 2009
    I do not make any changes to the ECU program, but I will try to be specific going forward (since someone already mentioned it, I will just use iEMS3 fuel table, ECU table).

    If I can get the manufacturer to develop a new unit, we could incorporate that function. Unfortunately, the current G3 is maxed on memory and processor capability. But, I know they are working on "stuff", so I will prod them.

    In all actuality, the unit I am using was developed from their stand alone system. But the demands on a unit that works with the factory ECU versus in a stand alone state, requires more elctronics and computations than the stand alone does. This is due to all the high speed interpretations it must perform feeding the information to the ECU.
     
  18. lafrad

    lafrad Supporting Vendor

    Messages:
    17
    Likes Received:
    0
    Joined:
    Apr 10, 2008
    As soon as you replace pistons, you are turning our motor into a boost designed motor ;-) I know for a FACT that the GM 3.8's have no problem in cruise control on boost @ 14.7:1. proper ring gaps, a blower cam and just a touch lower compression bring those temps right down and you are GOLDEN. no worries anymore!

    I can't speak for the mopar engineers, but I would expect that the map sensor scaling is performed before any other systems receive the data.

    From my pokeing and prodding of the system,
    The Cars trans logic is almost exclusively based on commanded torque vs available torque. The trucks are almost exclusively based on output shaft speed vs throttle blade position (with some torque stuff in there too for line pressures and kick down)
    Both computers rely on kPa or Pratio for all the critical fueling and spark tables. they also rely on Pratio & barometric pressure for airflow calculations.

    i guess the transition from the factory ECU to an aftermarket unit would be: at what point does <insert engine combo here> loose its ability to be a reliable driver for all available intake manfiold pressures present in the system. I'm confident that the ECU can handle all fueling and spark to keep a tight reign on ALL closed loop fueling situations, no matter *what* Pratio. I have wideband datalogs with predator logs matching that show closed loop 1.2+ Pratio during cruise, behaving properly bouncing around 14.7:1 with long term trims in the low single digits. Spark is definitely conservative there, in the single digits... as far as I know its a reliable setup with lots of daily driver miles on it.

    I have worked with a few MONSTER centrifugal blower setups that have no problem getting to 1.5-1.6 pratio in closed loop, with little consequence. those are usually purpose built forged, low compression motors, but that very fact allows them to be safe at 14.7:1 afr @ 7 psi of boost. would I advocate that particular situation for a stock ringed 6.1? probably not. if the standalone allows you to do that, then, go for it!
     
  19. Mr MoPar

    Mr MoPar New Member

    Messages:
    88
    Likes Received:
    0
    Joined:
    Apr 11, 2009
    When you really boil it all down to simple terms, all of the various tables working together are to arrive at one thing, the injector fuel pulse, and secondly the correct amount of timing. Back when there were carbs, there were no computations, you jetted it to provide the air fuel ratio you wanted (if you were advanced enough to have a meter), or you read the plugs to detect a lean condition. Timing you advanced till it pinged and backed off a bit. The basics are still the same, to be honest, the majority of stuff we deal with is simply trying to get around the programming in today's vehicles that is there to keep us from doing what we want to do.

    Part of the simplicity of the iEMS3 I am using, is that it leaves the ECU alone and lets it do it's thing whenever it is not in boost. When it does go into boost, the ECU is seeing it's maximum load that it has always seen because the map sensor is reading 0" of vacuum, just as it would under max load.

    The iEMS3 is providing the injector pulse, it is programmed in milliseconds of pulsewidth, the same sum value of all the myriad of computations from all the various tables trying to figure out what that value should be. It is easy enough to determine, if you want to see a 10.8 AFR in boost, you datalog the run, check for any points above or below in the iEMS3 fuel map, and simply add or subtract milliseconds of pulse from the current value.

    For timing, you monitor for spark knock, and retard the timing a bit in the iEMS3 ignition map. I am running -3 degrees retard at 5psi and I have zero knock retard across the power band.

    It is super easy, and end user adjustable, just like things used to be. Sure, maybe I give up a little efficiency, but that depends on how cell specific you want to play with it.

    As far as reliability (Mike I understand what you are saying and I don't take it as a negative comment), that is on everyone's mind, and rightfully so. I mean that is the main reason I went to the lengths that I did over the last 3+ years. If anything, I have been accused of testing too much.

    My 2006 5.7 300C (man I miss that car), had over 32,000 miles put on it testing the system, and it never went in for anything but routine maintenance. It was running from 7-9 psi all the time whenever in boost, so it was certainly not babied. At 6.5 lbs of boost, it was making 411 RWHP and 500 RWTQ. Oddly enough, for the first 22,000 miles, I intentionally ran the original stock champion plugs that came in the car just because I wanted to deal with the worst conditions possible. I still have them, and they still look new.

    A big part of what attracted me other than the simplicity of it, was the fact that now I had a single unit that was able to control fuel and ignition anytime boost was present, initiated from live map sensor data. It also gave me integrated electronic (a closed loop boost feedback circuit) boost control that was load, rpm, and throttle position controlled so I could develop three dimensional boost maps, it gave me integrated water/methanol spray control, where I could simply input the desired percent of pump output based on load (this could also be used for progressive nitrous control). It has a couple of digital outputs that could be used for shift lights or warnings or whatever. It integrated all of these things in one unit working directly with the factory ECU giving me a system that would function almost exactly like one would from the factory. And since the vacuum range was simply untouched, it retained the stock like drivability.

    I have no doubt that it is possible to scale everything in the ECU so that there is a relatively normal closed loop range running at 14.7 AFR. but, instead of having to manipulate a bunch of other parameters to try and make a closed loop boosted situation safe at a 14.7 AFR, if the option was there in the ECU to run a richer AFR, obviously that method would be used because it would be easier, and add an additional margin of safety.

    As I mentioned, I would really like to hear about the ETC (Electronic Throttle Control) issues. Or maybe I should start a companion thread? Hmmmmm.....
     
    Last edited: Jun 11, 2009
  20. lafrad

    lafrad Supporting Vendor

    Messages:
    17
    Likes Received:
    0
    Joined:
    Apr 10, 2008
    The ETC issue is actually a rather simple problem, made complex by the fact that its ETC>

    Airflow across a throttle body, at specific throttle angles... is VERY VERY preditable if you know the pressure differential from one side to the other. If you change the pressure on one side that you CANNOT MEASURE... you just threw off any assumptions you can make.

    when you pressurize the "outside" of our ETC throttle blade, you change that ratio, and now the rest of the TB calculations are going to be incorrect.

    that gets worse and worse as you get more and more away from stock... auto-drive/run and that throttle hang after a WOT run are caused from this.

    You can play with the ETC airflow tables a bunch and minimize/eliminate that, but it never will have "perfect" and "Stock like" behavior unless you have a good way to figure out how much air is flowing across the TB.