Page 1 of 9 12345 ... LastLast
Results 1 to 20 of 170

Thread: LNF PID Tables

  1. #1
    Advanced Tuner
    Join Date
    Sep 2007
    Posts
    471

    LNF PID Tables

    CodeNameBill should be able to shed some extra light on anything I might miss or any misinformation I might have.
    Last edited by MikeM173; 06-16-2015 at 11:28 PM.

  2. #2
    Thanks, MikeM, for this post. I'll try to do the other tables on a one-by-one thread instead of a single mass post.

    Prop max is a limiting table. It should have no affect if it were only a limiting table, but this doesn't seem to be the case. It appears to be used as reference elsewhere with minor effects as you raise it. I set it to 327% to eliminate any possible interference or extraneous effects.

    I set each table to a single value to give the pedal a linear feel when part throttle. I've tried ramping all three tables. This is not a good idea as, part throttle, it has different variances of "urgency". Treat these like urgency tables. The value of the table sets the level of "urgency" based on boost delta. If this is set to a single value, your throttle pedal will have a much more linear feel regardless of boost delta. Ramp these tables and it will feel more exponential. This makes the transition much harder to control part throttle.

    This linear approach also goes in conjunction to my other changes I will post up later for a cable-driven versus drive-by-wire or exponential feel.


    Stock KO4 Turbo Effects Of the Gain Tables

    Prop Gain Steady-State - This won't need to be touched by most people. It affects the transition between WOT boost and part-throttle cruising in the ecu. Those who road race or drive the hills will notice its effects.

    When you demand high boost at WOT, this creates a big delta between that mode and steady-state. When you're at higher rpm and boost but feathering the throttle, you will notice stuttering. Theorizing, the ecu is fighting trying to establish a low steady-state prop value with a steady-state boost condition already beyond its limiting value. Raising this value will eliminate the part-throttle stuttering. How high to raise? If you do a single-value prop gain table like I do, keep it enough below the prop gain table that the ecu can establish two specific conditions (steady-state and WOT). IOW, keep raising it until the part-throttle stutter at high rpm and boost levels is gone.

    Prop Gain - Like Mike said, this is boost ramp-up. Maybe because the stock KO4 is a twin scroll, I don't experience issues of sensitivity as he has tuning. With that said, the limits of the KO4 spooling up is around .48-.50sec above 3k rpm when flooring it. ALTHOUGH PROP MAX TABLE GOES TO 327%, THERE IS ABSOLUTELY NO NEED TO DO THIS WITH THE PROP GAIN TABLE! YOU SHOULD BE NOWHERE NEAR THE MAX OF THE PROP GAIN TABLE.

    Going too high doesn't make it spool up any quicker. It exceeds the effectiveness of this table causing the ecm to try to overcompensate. This will make your part-throttle transition into an on/off switch. I've experimented with some extreme values. Track tested, across the board for a stock turbo you don't need to raise this table too high to hit spool-up times around 0.5sec. I won't give you my values, but that should give you an idea.

    Integral Gain - Mike has a good explanation of this table. It controls throttle closing and steady-state. When it's too low, you get boost oscillation from an uncontrolled algorithm, kind of like an undamped spring or your mom's 72 Cadillac going down the freeway at 85mph. Too high, and the table is fighting against the Prop Gain table attempting to shut boost down at WOT. This also will cause oscillation in your boost but for a different reason. It's overdamping the system trying to shut down boost.

    Throttle blade response will be to hang wide open after you lift off the pedal if it's underdamped (values too low). Values too high will cause the throttle blade to close completely before the pedal hits 0%. It will close to 3%/5% whatever for idle at your rpm in less than 0.070sec.

    Derivative Gain - Like Mike said, this controls your overboost/overshoot when ramping up boost. You can fine-tune it to get your boost to ramp up perfectly with zero or slight overshoot depending on your preference. Hint: look at the values in the table. Notice how the highest pressure delta columns are all zero? That's why you have overshoot on ramp-up.

    Go too high on this table and it will overrule the Prop Gain table. You gain raise it so high that you never hit full boost by 6k rpm.


    VERY IMPORTANT SIDE NOTE ABOUT THE RELATIONSHIP BETWEEN THROTTLE BLADE, TIMING, AND BOOST

    Stop trying to make the throttle blade 1:1 with the pedal. It's completely irrelevant. If you want a linear drive-by-cable feel, set each table to a single value along with other changes I will post later.

    Why is it irrelevant? Because they're interrelated. The ecm uses throttle blade as boost control. Whenever it wants to control boost, it holds the throttle blade in position from opening fully. Take a look at your logs when you floor it. At about 50% pedal when you mash it, the throttle blade probably holds for a split second. This is usually accompanied by a drop in boost. Same with in between shifts. This drop in boost tells the ecm that lower boost values mean more timing. Subsequently, there is a simultaneous jump up in timing. They are all related. When properly tuned, the throttle blade will lose that flat step and ramp up smoothly to 100% (usually in 0.7-0.9sec). Controlling this is irrelevant. You're controlling boost, not the blade.

    Likewise, full boost does not mean full power when properly tuned. With my current tune, my pedal is completely linear. At 32% pedal, for example, the ecm may have the throttle blade wide open or be boosting at 15+ psi depending on rpm, but that does not mean I'm getting all the torque that can be produced at those numbers. It uses other limiting factors to keep my power linear relative to the rpm and pedal position. This has all been tested at several road racing test and tune days.

    More to come.
    Last edited by codename Bil Doe; 04-23-2014 at 01:10 PM.

  3. #3
    Senior Tuner cobaltssoverbooster's Avatar
    Join Date
    Dec 2008
    Location
    Nevada
    Posts
    4,452
    great stuff guys!
    2000 Ford Mustang - Top Sportsman

  4. #4
    Potential Tuner DrJones_CMR's Avatar
    Join Date
    Jul 2010
    Posts
    9
    Ok, lets talk theoretical PID stuff here...

    I'm not sure of the exact workings or the correct tables so feel free to comment and clarify.

    Say the driver pushes the pedal to WOT.
    The ECU now has a target of maximum power (which is set by which tables? DAL, MALT, both?).
    The Engine is not actually able to reach these "target" power levels due to physical limitations, so the PID control has this constant error (difference between target and actual).

    This is where the Integral is working against us. Since integral is looking in the past and sees the error it is trying to fix that, but can't for physical limitations.
    The driver lets off the WOT for what ever reason but the integral control is still fighting (due to it being based of past info) even though the proportional and derivative control are not trying.

    Is this an issue with the Integral settings or is this an issue where the DAL and MALT tables need to be adjusted to reflect the physical capabilities of the motor?

  5. #5
    The resultant torque from which the ecu pick from is more like knights of the round table only with more knights. While not all the tables are reference tables, they have have varying levels of input mostly none of which is independent of the others. Changing one table affects several others seen and unseen.

    While basic PID theory may help understand the functionality, the explanation above of how to manipulate the tables takes into consideration the limitations with which we have to work with. We won't be getting any more tables. Using the above method is a quick how-to around people having to fully understand PID controls.

    In a basic sense, your system goal is to ramp boost without oscillation. PID is for the boost aspect (assuming all other tables stay unchanged). Prop gain for ramp up, derivative for the overshoot, integral for steady-state and closing.

    Setting integral too high will attempt to force "closed" the boost causing steady-state oscillation at WOT. Too high a value will shut the throttle blade almost completely before the pedal even gets to 50% closed. I should look to see if I still have a log of this. Pretty cool how quick it reacts.

    Integral values which are too low will cause what the stock tuning does - throttle and boost hang after pedal lift. These are affects that drag racers don't notice. I had a "perfect" tune which I had to completely redo once I realized it was great for drag racing but not autox or road racing.

    What is too high or too low? There is not a single value as it depends on where all your other tables are set. If your tune is closer to where mine and MikeM is, I could give values which would get you close. But, based on what you have, it's more experimentation to see if you went too far. As an fyi I'm nowhere near stock. But, I think MikeM is. Different approaches to reach similar goals.

  6. #6
    Advanced Tuner projectlnf's Avatar
    Join Date
    Jun 2012
    Location
    warrenton va
    Posts
    452
    now how about the prop gain steady state table? leave it be? and the steady state table at the top of the turbo charger tables?
    2003 Ford excursion 6.0 PSD
    No limit intake
    Straight pipe
    Studded
    EGR delete
    KC billet single plane compressor wheel
    Self Tuned
    XDP Regulated return
    RDP Fuel sump
    AEM progressive water meth
    Custom suspension air bags
    8" lift
    35" tires

  7. #7
    I messed with the prop gain steady-state table, and it made no difference relative to the other testing I was doing. I may go back and play with it to see if I can tweak some minor things (situations where my repeated actions cause a minor stuttor). Otherwise, it's fine stock so far imo.

  8. #8
    Advanced Tuner projectlnf's Avatar
    Join Date
    Jun 2012
    Location
    warrenton va
    Posts
    452
    question for you guys im seeing when i go part throttle maybe like 5psi of boost just trying to get up to the speed limit in any gear. basically taking off from a light. ill see that sometimes it likes to over boost i cant control it my pedal hasent moved at all but it likes to slam to 15 and drop down to almost 0.

    and also if im able to keep the 5 psi when speeding up it will drop to 0 after about 3k.. again no pedal movement at all it just drops?

    if its confusing ill try to explain it again or send a log lol
    2003 Ford excursion 6.0 PSD
    No limit intake
    Straight pipe
    Studded
    EGR delete
    KC billet single plane compressor wheel
    Self Tuned
    XDP Regulated return
    RDP Fuel sump
    AEM progressive water meth
    Custom suspension air bags
    8" lift
    35" tires

  9. #9
    Advanced Tuner projectlnf's Avatar
    Join Date
    Jun 2012
    Location
    warrenton va
    Posts
    452
    Quote Originally Posted by MikeM173 View Post
    Boost should only hopd for so long before it releases. Their is a lot of explanation to why but I'm on my phone so if need be I'll type it up later. If I'm understanding you correctly I don't see why you would want it to hold a certain boost level based on pedal position.
    its not necessarily by pedal position.. when im taking off from a light ill ill be lets say 25% throttle and ill get to about 3k.. all the way from take off until 3k im at 5psi then after 3k it drops to nothing.. i shift around 3500 in normal everyday driving. its just not smooth like id like it to be if that makes sense
    2003 Ford excursion 6.0 PSD
    No limit intake
    Straight pipe
    Studded
    EGR delete
    KC billet single plane compressor wheel
    Self Tuned
    XDP Regulated return
    RDP Fuel sump
    AEM progressive water meth
    Custom suspension air bags
    8" lift
    35" tires

  10. #10
    Potential Tuner
    Join Date
    Mar 2014
    Posts
    4
    Is there a PID to log pressure delta?

  11. #11
    projectlnf - would have to see the tune. Could be multiple tables wrong. Without knowing anything about your tune I'm guessing look up the threads on Optimum Spark tables.

    turbokid - not that I know of. Haven't found a need for one, personally.

  12. #12
    Advanced Tuner projectlnf's Avatar
    Join Date
    Jun 2012
    Location
    warrenton va
    Posts
    452
    i pm'd you codename bill
    2003 Ford excursion 6.0 PSD
    No limit intake
    Straight pipe
    Studded
    EGR delete
    KC billet single plane compressor wheel
    Self Tuned
    XDP Regulated return
    RDP Fuel sump
    AEM progressive water meth
    Custom suspension air bags
    8" lift
    35" tires

  13. #13
    Explanation of Prop Gain Steady-State added. Most people won't need to touch this (i.e. drag racers).

  14. #14
    Advanced Tuner projectlnf's Avatar
    Join Date
    Jun 2012
    Location
    warrenton va
    Posts
    452
    Quote Originally Posted by MikeM173 View Post
    If I had to take a guess the difference for pressure delta would most likely be desired vs actual. I'll have tom look at logs because I don't normally concern myself with anything below boost but I would assume this would apply there as well. I can't say for sure without looking at logs but it might be a difference in kpa or psi. So if your asking for 255kpa WOT and your just starting to boost and are at 100kpa then it might be the equivalence of 155 pressure delta difference. Their would be no need to have the pid table read out that far so they stop at 50 for pressure delta. You could simply raise that as well if need be to help test it.

    which table is the difference table that would do this? and im not moving the pedal at all etc is staying the same so i dont see why it would think im trying to go wot?

    ill try to get a log this weekend of it doing it right now its in my shop getting the compressor housing rotated a bit. i think i might be the onl cobalt ss turbo with the zzp front strut tower brace, now i see why they tell you it wont fit! the charge pipe runs right into it lol
    2003 Ford excursion 6.0 PSD
    No limit intake
    Straight pipe
    Studded
    EGR delete
    KC billet single plane compressor wheel
    Self Tuned
    XDP Regulated return
    RDP Fuel sump
    AEM progressive water meth
    Custom suspension air bags
    8" lift
    35" tires

  15. #15
    Tuner
    Join Date
    Aug 2010
    Location
    Detroit
    Posts
    64
    Well, maybe this is a good place to post this, but I have some questions about my tune after I did some modifications / drag racing last night.

    First, I'm trying to eliminate the small amount of boost drop in between shifts. Its not bad now, but it is definitely noticeable on the track. Can this be done by modifying the PID tables?
    Second, I know we are not trying to control throttle blade here, but I find it excessive that it takes over 1 sec for my TB to return to 100% open after I command it. Can I at least reduce this using the PID tables?

    I have already started to change them, but I've enclosed my tune without the changes I made. I'm curious to see where people here would suggest I go versus where I went. Also, any general comments on my tune are always appreciated.

    the 1/4mi run is at 2:27 in the file.

    Car details: 2007 solstice gxp, custom exhaust, IC, piping, catless, ported intake manifold, ported TB, ported head on the intake side.

    13.1.hpl
    09-03.2-14 93oct with opti4.hpt
    '07 Solstice GXP. Mods - Tune, IC, CP, catless DP, ported intake manifold, ported TB, ported Head on the intake side, catback, K&N CAI, LV front brace, WERKS backbone, HPS pads up front, and HP+ pads in the read.

  16. #16
    Advanced Tuner |V3nom|'s Avatar
    Join Date
    Apr 2009
    Location
    Denham Springs, LA
    Posts
    418
    Quote Originally Posted by hockeymandan2828 View Post
    Second, I know we are not trying to control throttle blade here, but I find it excessive that it takes over 1 sec for my TB to return to 100% open after I command it. Can I at least reduce this using the PID tables?
    im sure these guys can help with that but after reading this thread my understanding is that if the throttle blade goes wide open too fast you wont be able to control the boost ramp up (or control it as well) and itll turn into a on/off switch feeling. now that might not sound bad for drag racing but my guess is that could be too hard on the motor and drivetrain (think clutch) and shorten the life span of said items.
    2008 Sky RL

  17. #17
    TB opening is unimportant. Boost ramp, overshoot, and boost hang after pedal lift are.

    boost ramp - prop gain tables & prop max
    overshoot - derivative tables (notice zeroed columns stock)
    tb and boost close - integral

    Integral tables will affect boost after lift. integral tables can make engine hold boost or cut it short on throttle lift. OS tables are urgency tables affecting all variables. TB opening taking 1 second is irrelevant. Stock k04 maxes out at .485 sec boost ramp from 0-20psi.

  18. #18
    Tuner
    Join Date
    Aug 2010
    Location
    Detroit
    Posts
    64
    Interesting Bill, thanks for the insight there. I ran a track day yesterday, and it was extremely apparent that my mid throttle pedal modulation really screws with my tune as it sits. There were several times when I would start to exit a corner and would try to slowly add throttle (pedal at 30% or 40%) and the ECU would command full boost and full power. Then, there were several times when I released my foot from the pedal, and the engine stayed at 100% throttle and full boost for 0.5+ sec. Almost spun me a couple times, not too fun.

    So, if I have this right, increasing the values in all of the tables increases the "effectiveness" of that table. Maybe a better way of saying that would be: By increasing the values in the prop gain / prop max the boost will come on quicker. By increasing the values in the integral tables the boost and TB will close faster. Is that right?
    '07 Solstice GXP. Mods - Tune, IC, CP, catless DP, ported intake manifold, ported TB, ported Head on the intake side, catback, K&N CAI, LV front brace, WERKS backbone, HPS pads up front, and HP+ pads in the read.

  19. #19
    Senior Tuner cobaltssoverbooster's Avatar
    Join Date
    Dec 2008
    Location
    Nevada
    Posts
    4,452
    Changing your midrange dals will he'll with your bog in conjunction with pids and o's tables. Problem with dals is once you track orient them the car doesn't feel real dd friendly. Your pedal will be more linear like a cable throttle rather than the smooth exponential pattern similar to stock.

    Bill has changed settings to get the feel I'm talking about
    Last edited by cobaltssoverbooster; 09-05-2014 at 10:12 AM.

  20. #20
    Tuner
    Join Date
    Aug 2010
    Location
    Detroit
    Posts
    64
    Here is a good screenshot (ignoring the KR...) of what I am talking about. Look at desired boost and actual throttle % in respect to my foot. scary.

    MKE Mile 2.JPG
    '07 Solstice GXP. Mods - Tune, IC, CP, catless DP, ported intake manifold, ported TB, ported Head on the intake side, catback, K&N CAI, LV front brace, WERKS backbone, HPS pads up front, and HP+ pads in the read.