Tuners Rejoice! Free Tuning For M4.4!


Recommended Posts

Well, in the end you tune up to the knock limit yeah. Though I haven't been able to download your spreadsheet, your description suggested you needed to search for knock first and go from there. I agree adding timing until you see some knock, and then backing off, is the proper approach. However, if you're suggesting you should insert history tables of knock events in your excel sheet, and the sheet then retards timing in the appropriate places, you could also try to do this live using the adaptive ignition retard feature:

  • Disable adaptive ignition retard 'recovery' (set "Number of ignitions per 1 increment advance" to something ridiculously high)
  • Disable knocking enrichment (only for your experimental runs)
  • Set "Retard per knock event" to something relatively small (i use 0.75)
  • Advance your timing map 
  • Do WOT runs until all knock has been removed by the ignition retard logic
  • Record resulting ignition timing

I guess that would do the same as a spreadsheet, though in-car and automatically.

 

Are you using closed-loop wideband control over the entire load-range? If not, such conversion doesn't hold ground. It only works that way if you're using the active WB control, since that mod interprets the values from that table as target AFR values. Without the mod you're going to be way off, as the VE map cannot be converted to an AFR map. That simply doesn't match, they're different things. There's no physical ground that would correlate those two by just a 1/14.7 division. Don't change your injector constant to compensate, just use the VE map and forget the idea that it should be read as an AFR map - it shouldn't in open-loop :P

It works in the WB-mod since it's a "guess in the right direction", followed by cleanup by the wideband control. In open-loop you'll just have to increase or decrease VE in the appropriate RPM and Load-range until you hit your desired AFR. Required VE can vary greatly with different turbo pressure and RPM.

Edited by Boxman
  • Upvote 1
Link to post
Share on other sites
  • Replies 7.5k
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

Attention: The first 30 or so pages of this thread are outdated. Please refer to the M4.4 Wikia article where all the relevant information is currently being collated. Before asking any questions p

Crush it.

After alot of testing and rewriting code, we finally got a useful new mod working. As we all know, some time ago my dad Piet found out how to convert to bigger maf housings with the maf factor. Conver

Posted Images

 @Boxman I was under the impression that the VE map was a multiplier for fuel injection time necessary for stoichiometric mix at 100%VE.
The highest value the adaptive/active knock "wait time" till it forgets one increment of knock retard is 255 ignitions. But i don't think the value in the map is the actual value though., on an interesting note, an engine cycle under fuel cut decel do not seem to count as "ignitions" in the eyes of active knock control.
Also setting the  "retard per knock event" to a low value might be dangerous, it might need to wait several knock cycles to retard to a safe level, meaning the KRCOUNT would go up.

As for people using the VE map as an AFR map in open-loop....

On 6/4/2014 at 10:42 AM, razorx said:

I was unclear. ;) I skipped a few steps in my explanation.

The VE map part load is in fact an inverted AFR map. And being used with all loads.

 

Change the calculation/conversion in the XDF to 14.7/(x*0.0078125) as I did a few months ago. Makes the table much easier to read.

Than his map looks like this:

2014060401.jpg

This was the original rev5b.

 

1. Lets say, you get an AFR of 13 at 3000RPM and 8.02load

You wanted an AFR of 11.8 at this load.

2. Simply multiply the value in that cell by 11.8/13 so 14.2545*11.8/13 = 12.94

To do all these cells will be a hard and time consuming task. What worked for me is this.

3. The original value was 14.2545, subtract 12.94 from it. The difference is 1.31

4. After this I subtracted 1.31 from the cells in the 3000RPM column with a load higher than 4

Repeat all four steps for each higher than 3000RPM column. Only use the highest load cell for the calculation.

This only works for the higher load cells in this map. In fact the cells where the map stops behaving as an AFR map and starts being a VE map..
You might even start at 2580.
For me this method worked fine in getting verry stable AFR values under varying loads

On 4/15/2014 at 12:44 PM, Piet said:

...

....
....

As Razorx already stated, when I convert the VE-table into AFR values with his formula 14.7/(x*0.0078125), I'am getting values very close to the measured AFR's in my logs

I suspect that if the maf table, injectorconstant and injector deadtimes are really, really spot on, the logged values for AFR could be very close to the values in the VE/AFR table.
I'am planning to build an injector flow meter this weekend to determine exactly the injector constant and dead times of my injectors to see if I can get really close with my logged AFRs to the values in the table that way.

 

Link to post
Share on other sites

Yes.. that was back in 2014.. forget about that post !

It appears that the original (open loop) VE table is a last correction on the calculated injector time taking the Volumetric Efficiency into account.

Edited by Piet
Link to post
Share on other sites
9 minutes ago, Piet said:

Yes.. that was back in 2014.. forget about that post !

It appears that the original (open loop) VE table is a last correction on the calculated injector time taking the Volumetric Efficiency into account.

So i can use this map as an (open loop) AFR map?
It has worked pretty well for me so far especially near peak torque as that's when the VE is closest to 100%
As I move further from the high load, peak torque values it drifts a bit but not bad, only about a .4 AFR deviation.

Also, @Piet what is the final word on the 960 MAF conversion? Is the best way to just multiply by 1.34?
Or is it to use one of the two list of values you and @venderbroeck came up with by directly comparing airflow with the MAFs hooked in series?
It might be skewing my measurements.
Thanks!
-Calvin

Edited by Calvin Sonnik
Link to post
Share on other sites

Not really,  the values in the original table  are not some sort of a request AFR but only a compensation for volumetric efficiency.

 

Only with the wideband mod the table has turned into a real AFR request table.

  • Upvote 1
Link to post
Share on other sites

@Calvin Sonnik The map acts on fueling, but isn't numerically convertable to AFR. You have to keep check of the AFR you're hitting in each cell, and adjust the corresponding VE cell accordingly (either up- or downwards) until you hit the desired AFR. The actual conversion formula between VE-value and resulting AFR value will be different for every cell, which is why you can't substitute the entire table for a separate AFR map with one catch-all conversion formula.

Since they scale somewhat in the right direction with the simple catch-all 1/14.7 division, it may *seem* to do it's job, but really that's a pitfall as the actual conversion formula for our VE to AFR is unknown and differs per cell. As you say, you're still a few points of AFR off when converting the table. With wideband control you can get away with it, since the difference between the underlying VE value will be altered via STFT until the AFR setpoint is reached.

As for this:

Quote

The VE map part load is in fact an inverted AFR map. And being used with all loads.

Is just not true. With the knowledge back then it may have seemed like it scaled that way, but the conclusion as drawn there was premature.

Edited by Boxman
  • Upvote 1
Link to post
Share on other sites
49 minutes ago, Calvin Sonnik said:

 The highest value the adaptive/active knock "wait time" till it forgets one increment of knock retard is 255 ignitions. But i don't think the value in the map is the actual value though., on an interesting note, an engine cycle under fuel cut decel do not seem to count as "ignitions" in the eyes of active knock control.
Also setting the  "retard per knock event" to a low value might be dangerous, it might need to wait several knock cycles to retard to a safe level, meaning the KRCOUNT would go up.

Oh and also regarding this, I mean to only do it that way when you're going to allow knock for research purposes anyway. When done, obviously set them back to default values for proper failsafe behavior. I have not used this method myself, but just figured this could be a useful approach if you want to quickly tune ignition and utilize the features the ECU provides to do so.

If knock-detect is as accurate as @Piet says (on-par or even superior to listening in with a headset), it should be fine for the most part.

Edited by Boxman
  • Upvote 1
Link to post
Share on other sites

Okay, i'll take it all into account and i will improve. I also noticed if i take a completely stock T5 and use the stock tune and datalog, i get tons of timing pull and fuel enrichment. 
2 of the 3 cars i've done ECU work on were in good stage 0 condition. while the stock ECu is not getting much past a load of 8MS many people run a MBC for years, i'm guessing that it has timing pull through most of its life.

 

Link to post
Share on other sites

ugh. i gotta get back in touch with Aaron... i STILL cant seem to connect and log and i had a few flashing CEL's this weekend with knock. i do need to stage 0 and or go to his COP system asap too. Not his fault - all mine lol... i am an absent parent to my R lately :(

Link to post
Share on other sites
9 hours ago, gmsgltr said:

ugh. i gotta get back in touch with Aaron... i STILL cant seem to connect and log and i had a few flashing CEL's this weekend with knock. i do need to stage 0 and or go to his COP system asap too. Not his fault - all mine lol... i am an absent parent to my R lately :(

Hey Greg, could you e-mail about these problems? No need to clutter this thread. Thanks

Link to post
Share on other sites
On 10/16/2017 at 2:20 AM, Calvin Sonnik said:

Alright guys, this week its happening (unless I break something or I burn through the money I have been given to make this car fast), S70 T5M M4.4 
 

  • Green Injectors From ME7 S60R
  • Upgraded MAF
  • TD04-19T
  • 3" Angle Flange Downpipe
  • AEM X-Series 3000 AFR Gauge With Wideband Lambda Mod
  • IAT Post Intercooler
  • MAP Sensor for helping PID for TCV and LRD
  • LOTS of time tuning and gently creeping up on a goal of 300TQ  300HP to start, the ultimate goal being 300WTQ, 300WHP.

But I have a few questions,
What is my RPM limit? Stock N-Block, 150K motor, stock valve springs, camshafts. I think valve float will be my primary issue, rod strength and head flow will be secondary issues? 6750RPM? or more like 7500RPM?
What should I target my AFRs to? As per usual, I will be targeting low 12s at first, I'm wondering if I should richen it a bit at 3600-4800RPM or so to prevent knock? I already need to pull a bunch of timing there too before adding a bunch above 5000RPM or so.
Is the stock catalytic converter and muffler severely restrictive? if so, I will not be spending a bunch of time to get tuning right if I'm only going to need to change it due to lower EGT and EBP when the Cat and muffler are changed.
Where should I target peak Load/Torque at? Now that I will have larger injectors I will not have to worry about maxing injectors at high RPM high Load.


On an unrelated note:
I can't seem to flash an ECU when it's hot (Bank 0 failed to flash). The ECU gets really warm as it pulls air from the fan shroud post cooler sandwich.
If I pull the ECU and wave it in the air like a madman it flashes just fine.
It happens with all 3 of the ECUs I've used, and I'm getting steady 15V to the bootloader pin.
I wonder if anyone has a solution related to cooling the ECU better or possibly another hardware solution.

Thanks!!!!!!!!!!!!11!!!!!!!!!!!11111!!!!!!!!!!!!111!1
Calvin Sonniksen

Was re-reading old posts in this thread and noticed no one replied to your post above so I can give my two cents worth... 

The td04 hot side is going to be your biggest restriction generally when you start modding the N.  Go with the 7cm angle flange if you can; it helps a bit.  Don't forget either the R or Japanfold exhaust manifold too.  And yes, the N is not going to rev (esp w/stock valve springs) like the RNs with the lighter valve train, so don't plan on going much higher than stock redline.  But your pwr goals are not too unreasonable for your planned mods except you'll likely need a few more.  Like a bigger intercooler to start.  The do88 is a nice drop-in or you can go the fmic route.  And of course, rip!

What size maf are planning on using?  Probably 3.25" is about right.  I'm using the S90 maf housing right now but I should probably go up a bit.

You might be able to get away with stock exhaust with the 3" down pipe, but turbo engines hate back pressure, so why?  For those pwr goals 2.5" back from the downpipe would be fine coupled with good flowing cat and muffler.  I like my setup (link in my sig) but it does have more drone than stock for sure.  Either I've gone deaf in the last few years since I've installed it, or I've gotten used to it.  It sure sounds nice at WOT though..   ;^)

Assuming you're running pump gas, you are going to need wmi, for sure.  Again, because of that restrictive hot side, you have do something to cool the charge, or the knock will just constrain your pwr severely.  You can add fuel (run rich) but that robs pwr and wastes gas.  I just got my wmi installed and working, and though I"m still dialing in some details, it's awesome!  Wish I'd done it a long time ago.

Greens will probably be fine for you, but you may get close to maxing them out, so keep an eye on your injector duty cycle.

What really piqued my interest about your post was that you're running a post intercooler IAT.  Curious about what you're planning there?  What sensor are you using, how are you hooking it up, etc?  I just got my IAT hooked up and installed in the post-intercooler pipe (with programming help from venderbroeck).  Once I get things dialed in, I think it will help a great deal, in the summer time especially.

On the flashing; I am still only bench flashing but I've found I need to run the pwr supply up to 16v or I get flashing errs; give that a try.  BTW, update your sig w/your mods so we don't have to remember what you've got.  Good luck!

Edited by gdog
Link to post
Share on other sites

@gdog Thanks for the input!
The car i'm modding is a friends so he decided he wanted to spend X dollar amount and I wanted a M4.4 car to tune so i did the work for free if i got to tune it and use it as a showcase car for what i can do for not a lot of money. I'm going to get as much performance out of the car as i can with the money allocated to me.  I hope to use this car to get more performance jobs with whiteblock/redblock cars and less oil/brake changes as a underemployed mobile mechanic. (I have Three M4.4 Cars under my care currently)  I'll add the Mods to my signature once they are on MY car, V70XC that has a M66C and 1999RN in the trunk along with a bunch of small electronic goodies.
I am perpetually broke so i don't have money to get my M4.4 car running yet.
As of the tune now:

  •  Green injectors are maxed safe at 90%DC.
  • 960 MAF getting close to maxing with spikes of 1150 KG/HR
  • Hard Limiter set to 6810 RPM 
  • Soft Limiter set one HEX digit below
  • Stock 19T with TCV maxed after 5000RPM
  • 3" catless to stock exhaust after where the cat would be.
  • Flowing a steady max of 1100 KG/Hr of air at peak flow (MAF calibration might be slightly off, but only 1-2%)
  • Open Loop VE map
  • AEM X-Series wired for data logging using tank pressure ADC
  • Highly reduced LDR P-Part and I-Part to prevent oscillation and overshoot. (Was not easy, 19T spools fast AF)
  • Lots of other small changes to change drivability, economy, and research unknown (to me) parameters.

I have not reached my goal yet, but above 3000 RPM in second gear, i can spin at least 1 wheel if not both if i put my foot down.
Tires are decent but not great, full tread depth though.

IAT is a GM IAT from a Junkyard car. not calibrated yet, just used as reference to get an idea of what is "hot" and "cool"
I have a later model ME7 IAT/BPS. Its a 4 pin Boost Pressure Sensor and IAT in one. I'm going to check its calibration and then see if i can change the calibration in the .bin. (also talked to @venderbroeck about this) then install in a stock intercooler pipe like i have the GM IAT fitted.
Hot side is restrictive for sure, going to change cam timing to try and gain 0.75 to 1.5 deg of timing at peak torque.

I'm trying to keep AFRs in the very low 12s and high 11s to stay safe. (Doing a good job of that)
Changed accelerator enrichment to prevent terribly rich AFR on tip-in.
For flashing im using a step-up board to put 15V into the B8 Pin. I am going to put a filter capacitor in the output and check it with my oscilloscope to see if i have too much ripple on the output, another one of my ECUs flashes at battery voltage (12.8V) just fine. I think some require wayyy more voltage than others. Getting sick of flashing and am going to get an Ostrich 2.0 for tuning then I will flash the tune onto an ECU with a step-up board and switch on the ECU itself when i'm done.

Thanks for your interest, It means a lot!

-Calvin Sonniksen

Edited by Calvin Sonnik
Link to post
Share on other sites

So after some tinkering today and being pretty happy with how it pulls, I figured I should probably drop a youtube vid to show off a bit. Just before the run I thought to myself "ah you'll see, this is the pull where I blow the thing to pieces". But, no guts no glory:

 

Not quite the 100-200 km/h pull I had in mind. Always double - no, triplecheck your turbo hose clamps :'D

On the bright side, the entire run was knock free, peak load of 15.5 ms. When this all is tightened down properly again (clamp was probably just skewed ever so slightly), time to see where I can pull some timing and how much. I'm still running very conservative here. All in 4th gear btw, wasn't doing it for the time but wanted some good data too.

Edited by Boxman
Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.