NefMoto

Hi

Hope someone can help, I have slowly been building up my xdf for my ecu 8E0909518P. My first couple of attempts at increasing boost are not giving me the expected results. My first attempt I basically increaed the entire KFMIRL table by 25% which put my max requested load at about 290+- I then increased LDRXN by the same scale ie 25% so the load values were about the same this increased my requested boost only slightly from 800mbar too around 1bar but I had part throttle surging and cold starts were very very lumpy and you could hear the turbo spooling. So my next attempt I only scaled the KFMIRL up from about the 30 load mark upwards again by 10% and did the same for LDRXN the same this has improved the cold start and part throttle surging but I am not not boosting above 1bar.

following is my KFMIRL map

0.00   0.00   0.00   0.00   0.00   0.00   0.00   0.00   0.00   0.00   0.00   0.00   0.00   0.00   0.00   0.00
11.70   11.16   10.95   10.73   10.43   10.27   10.10   10.01   10.01   10.01   9.82   9.56   9.23   8.93   8.81   8.55
23.37   22.34   21.94   21.28   20.63   20.60   20.06   20.37   20.04   20.02   19.27   18.89   18.19   17.91   17.58   16.97
40.03   36.75   35.41   34.34   33.49   33.07   32.32   32.27   32.11   31.99   31.15   30.73   30.21   30.21   29.49   29.46
52.92   48.40   46.20   44.91   44.02   43.57   42.75   42.24   42.14   41.98   41.23   40.88   40.78   40.99   40.06   40.41
66.31   60.10   56.79   55.22   54.24   54.19   53.56   52.36   52.13   51.89   51.66   51.33   51.35   50.86   50.30   50.39
79.55   71.98   67.83   65.70   64.57   64.34   63.59   62.77   62.37   61.97   61.90   61.50   61.41   60.56   60.24   60.07
92.98   83.98   78.12   76.13   74.84   74.37   73.50   73.17   72.70   72.14   72.07   71.44   71.18   70.15   69.89   69.38
107.02   96.47   87.99   86.46   84.89   84.33   83.56   82.90   82.64   82.01   81.92   81.21   80.70   79.83   79.34   79.13
134.75   123.05   110.86   107.11   105.73   105.17   104.30   102.73   102.26   102.10   102.12   101.58   100.85   99.68   99.05   99.66
159.33   146.75   133.41   129.24   128.09   127.41   126.38   124.22   123.28   123.26   123.05   122.56   122.16   121.48   121.10   122.49
178.39   165.89   154.95   154.10   151.01   148.48   147.52   145.92   145.15   144.57   144.35   144.40   144.40   144.19   145.43   145.32
187.25   175.83   166.32   164.72   162.80   158.91   157.48   155.18   154.48   154.41   154.83   155.98   156.66   156.07   156.66   156.33
196.20   186.19   176.46   174.36   172.69   169.36   168.52   165.94   164.18   163.74   166.18   166.32   166.48   167.86   168.96   167.65
202.64   194.14   186.21   184.81   183.38   179.25   179.21   175.79   174.71   174.19   175.64   177.38   178.55   180.75   180.12   180.17
219.12   216.40   212.82   212.72   213.05   210.10   207.97   206.77   202.36   204.45   204.29   207.59   207.38   210.80   211.95   211.97


and this is LDRXN

181.50   220.04   241.93   233.72   233.25   234.38   236.26   235.88   232.13   232.13   228.75   226.13   226.13   223.88   221.25   218.25


My k04 hybrid has a 50mm K04 compressor wheel and a k03 turbine wheel and I am pretty sure it can boost more than what I am doing currently. I would like too have about 1.2bar boost if any one has some pointers that would be great.

Is the hardware upto the job of 1.2bar?  I say this as I have a K03 hybrid on which wont boost over 1bar, even with pipe off n75 it will peak just under 1bar.  I added an external spring to actuator for testing and managed to gain a 1.6bar peak.  New actuator with uprated spring should cure my issue.

Nick

if it wasnt up to the job then the actual boost wouldnt meet requested but my requested boost never goes above 2000mbar, so my understanding is that the n75 is opening the WG and keeping actual boost at the requested boost level. I thought that boost is 10*(spec load)+300mbar so with my KFMIRL table having a max load level of 219 10*219+300mbar=2490mbar which I am not seeing in my requested boost. Also on my previous tuner map and my old k03 I was seeing requested/actual boost above 2bar

logs from a run I did now

Sunday   18   December   2011   12:45:05                                 
8E0 909 518 P      1.8L R4/5VT         0001                                       
                                             
   Group A:   002            Group B:   031            Group C:   115         
      RPM   Load   Inj. On Time   Mass Flow      Lambda Factor   Lambda Factor              RPM   Load   Absolute Pres.   Absolute Pres.
   TIME               TIME               TIME            
Marker   STAMP    /min   %    ms    g/s   STAMP               STAMP    /min   %    mbar    mbar


   285.04   1720   39.1   3.06   12.83   285.27   0.984   1           285.51   1760   37.6   980   1000
   285.74   1800   37.6   3.06   13   285.98   0.984   1           286.22   1840   38.3   980   1000
   286.44   1840   39.8   3.4   14.78   286.69   0.969   1           286.92   1880   39.8   1030   1000
   287.15   1920   76.7   7.82   32.25   287.39   0.766   0.953           287.61   2040   88.7   1960   1150
   287.85   2080   94   9.86   40.81   288.09   0.727   0.922           288.31   2200   104.5   1940   1350
   288.56   2280   110.5   9.18   51.44   288.79   0.906   0.914           289.02   2400   127.1   1950   1590
   289.26   2480   140.6   12.58   74.64   289.5   0.891   0.906           289.72   2680   171.4   1960   2030
   289.96   2800   181.2   16.66   98.75   290.2   0.914   0.899           290.43   2960   182   1970   2190
   290.67   3120   175.2   16.32   100.36   290.91   0.899   0.899           291.13   3280   168.4   1970   2060
   291.37   3360   167.7   15.64   108.92   291.59   0.906   0.899           291.83   3560   162.4   1960   2040
   292.07   3640   160.2   14.96   111.25   292.3   0.899   0.899           292.54   3800   160.2   1960   1990
   292.78   3880   160.9   15.3   118.89   293   0.875   0.867           293.24   4040   160.2   1950   1970
   293.47   4160   159.4   15.64   126.17   293.7   0.836   0.836           293.94   4280   158.6   1960   1940
   294.17   4400   160.2   16.32   134.72   294.41   0.821   0.821           294.65   4560   160.9   1960   1920
   294.87   4640   162.4   17.34   142.61   295.11   0.797   0.805           295.35   4800   163.2   1960   1930
   295.58   4880   162.4   17   150.44   295.82   0.782   0.789           296.06   5040   164.7   1960   1920
   296.3   5120   163.2   18.36   162.03   296.52   0.743   0.782           296.76   5280   162.4   1930   1900
   297   5360   162.4   19.04   167.53   297.22   0.735   0.766           297.46   5520   159.4   1880   1860
   297.71   5600   159.4   19.38   170.72   297.93   0.735   0.75           298.17   5720   158.6   1840   1840
   298.41   5800   156.4   19.38   174   298.67   0.735   0.735           298.9   5960   151.9   1810   1790
   299.15   6040   150.4   19.04   173.72   299.38   0.735   0.727           299.61   6160   148.1   1760   1780
   299.85   6200   145.1   18.7   171.56   300.07   0.735   0.711           300.31   6320   143.6   1710   1770
   300.55   6360   140.6   18.36   173.25   300.78   0.735   0.704           301.02   6480   135.3   1700   1740
   301.26   6560   135.3   18.36   166.83   301.48   0.735   0.696           301.72   6680   133.1   1660   1690

maybe you miss KFLDHBN ...
your LDRXN values are way to hight !
when I was doing some log on a S3 8L 210 HP AMK, I found that requested boost is about 13* requested load .

Thanks for the feedback, yes I have not modified KFLDHBN could you tell me what the relation is too boost?

ok my understanding was that KFLDHBN was for altitude adjustment, I did a quick search and saw that it is mentioned when people modify KFMIRL/LDRXN but I cant seem too find any info on how it relates too those maps?

my current KFLDHBN table

KFLDHBN set the max alowed boost presure ratio vs RPM and ambiante temperatur or intake temperatur selcted by CWRLMX

Set KFLDHBN to FF in the entire table (to disable this boost limiter totally) and you will see the boost increase you are  looking for but before that, decrease your LDRXN to sane levels, i.e. something like 190-195 at the most midrange for your desired 1.2 bars of boost...

I looked a little harder at KFLDHBN and I think I understand whats its doing.... I am running a dual side mount IC setup on my car and IAT's never go above 50degrees. I multiplied the whole KFLDHBN table by 10% and lowered my LDRXN slightly. Car is running really strong now logs attached I havent touched fueling or timing yet but while logging I was getting a fuel trim rich multi DTC not sure why fueling would go over the 25% adaption limit if I havent touched anything fueling related? Thanks for the help.

If you did it like this, you are using KFLDHBN as limiter instead of LDRXN which is kind of backwards. KFLDHBN is for protection, not a limiter per say. My recommendation would be to temporarily FF KFLDHBN totally and focus on getting LDRXN straight in accordance with your desired boost curve, and once that is done, adjust KFLDHBN properly to act as protection, not as a hard limiter.

This.

thank you for all the help. OK too clarify I should take my stock file and FF everything in KFLDHBN then modify LDRXN. Do I modify KFMIRL? or just LDRXN to start? It looks like my K04 hybrid is handling the boost fine, any tips on what you guys are running boost pressure wise, 1.2bar fine or if my turbo/IAT are good can I ramp it up too the MAP sensor limit of 1.5bar? I have heard of guys running k03's with unichips that are boosting 1.5bar.

Standard K03-52 here, hits 1.3bar and tails down to 0.9 (using catless pipe).

my car is running like an absolute monster  ;D  , I think my fueling issue (rich) multi is because of the S3 injectors I have just used lemmiwinks too remove 16% from primary fueling adaption channel while testing, this has brought my fuel trims down too -3% on both idle/multi so fueling seems good.

Now onto my next issue which I wasnt really expecting because none of the pro tuner files ever threw this code on my car but I dont think they had things as aggressive as I do right now but anyway... the car had a "16487 - Mass Air Flow Sensor (G70): Signal too High" fault. I wasnt logging at the time so I can only assume that my MAF hit its limit of 229 g/s which caused the fault. Now I understand that if I lower boost etc this will bring things down and probably within the limits of the current MAF but what other options do I have? Do I need too scale the MAF table or I see lots of people changing MAF housings?

Keep in mind that you need to request enough load in KFMIRL but by the looks
of things you can intitially keep it as is.

Secondly, FF the entire KFLDHBN to disable this limiter totally. The reason you want
to do this is that need to avoid interferance whilst calibrating LDRXN.
I would strongly suggest that you keep LDRXN at a maximum of 190 intially in the midrange
and decrease it towards 150-160 at high revs during evaluation. If you need a complete
LDRXN curve to start from, I'd be happy to suggest one.

Also, there is another limiter that might be necessary to adjust and that is for maximum
allowed flow through the air mass meter. You will see a fault code for this when you push
the boost pressure but this limiter is of course adjustable as well if you need it.

Finally, there is an ME7 function for boost deviation handling thart might need to
be disabled during boost calibration. If you get a code called 17963 your actual boost
deviates too much from your requested boost. To prevent this from happening, FF the entire
KFDLULS function during boost calibration to get your boost control in order, then recalibrate
the KFDLULS properly. Having more than 0.2 bars of deviation indicates that you need to work
som more on your regulation.
If you have difficulties reaching the higher boost levels requested, it's not uncommon that you
need to increase KFLDIMX for higher boost pressure levels. This has been the case for me on
quite a few cars.

During calibration I always use analog knock detection, readings of oil temperature, air after
compressor temperature, air after intercooler temperature, exhaust gas temperature and back pressure
levels to give me an indication that I'm within sane operating conditions and I suggest that you do
the same when cranking the boost severely.

any help would be great, if you could post your suggested LDRXN curve that would be greatly appreciated! I am having an issue trying to figure out how too FF the table, I have the demo winols which wont let me save the bin, I cant figure out how too find the hex address in tunerpro hex editor either and my other linux based hex editor hexer cant seem to find the starting hex address of  KFLDHBN either... suggestions on what you guys use?


Thanks I posted just before you and it looks like I have run into that exact issue, which table do I need to adjust for the MAF?

There is seldom a real need for a full version of WinOLS as you can change what you want to change in your file in the "demo version" and then just "copy and paste" the changes (found in the "text window" of the WinOLS working space) into any freeware HEX-editor where you have the stock file (overwriting the stock values for any given area) and thereby acheive basically the "same thing" as working with a full version of WinOLS. It just takes an extra minute to copy the changes into the right positions in your file in the HEX-editor. No big deal.

I'll get back to you regarding the MAF-limiter.

The scaling for the MAF is called MLHFM and in this map you will see when the MAF saturates. Re-scaling the MAF-curve won't help you one bit if the MAF is really saturated. If it is saturated it is saturated and you need a bigger one. On the other hand, if it's "only" the MAF-limiter and not the MAF-hardware per say (hard limit) then you just need to adjust the limiter (sof limit).

The MAF-limiter is called KFMLDMX and here you can set an absolute maximum rating for the MAF vs engine speed and %DK as a safety feature. It doesn't have to have anything to do at all with when the MAF actually saturates, but is rather used as a warning when higher MAF-flows than expected are recorded (as a safety function for a broken hose to the wastegate actuator among other things).

Increasing this map to suit your new driving conditions (where you have issues with MAF-warning) will most likely solve your issues. As for instance LDRXN works as an overall limiter for WOT, the KFMLDMX handles other driving scenarios than WOT.

once again thank you for the assistance fredrik_a I appreciate it!

Ok so here is my new KFLDHBN which has been FF'ed

Ok, next step would now be to recalibrate LDRXN and possibly adjust KFMLDMX as well.
Do you still need a suggestion for your LDRXN setting? Also, running at a desired 1.2 bars
of boost, I strongly suggest that you use a wideband AFR measuring unit to log fuel and
adjust it accordingly.

OK I am busy flashing the new file, I adjusted LDRXN/KFMIRL right down as well. I havent touched KFMLDMX yet

 

logs from a quick run I did now (not sure why I logged coolant temp instead of iat) car is much smooother but doesnt feel as strong as it did last night, actual boost is usually a little higher than requested so I assume I can increase requested slightly. I did see my EGT's spike too a little over 950degrees whats a safe limit for EGT's?

There are few things in your LDRXN I am a bit concerned about.
for instance, why do you request 194.82@1.750 rpm? Is it reasonable
to beleive that you will be able to run the equivalent of about
1.2 bars of boost at that engine speed? There is no need to request
something the engine can never give you.

looking at a graph of requested/actual boost I see actual boost gets a little lumpy up in the RPM. Would this be because KFMIRL and LDRXN are not lining up at that point in the RPM axis?

Could you post your suggested LDRXN please?

new ldrxn comments?

Looks better low-end, will produce a shitload of boost midrange and will probably throw a MIL for MAF-flow. ;-)

please could you post your suggested LDRXN, i dont really like the way the car feels with it like this. if i put my foot down WOT then it sort of waits and then goes feels weird.

This is probably because you run KFLDRL stock which in many stock files suggest 95% dc at high loads which is usually problematic. Try decreasing it to 85% midrange and 80% top end.

As for the LDRXN, after having tuned quite a few cars, I tend to have it more powerful in midrange as backpressure is horrific top end, so I'd suggest something along the lines of the following...

100   120   130   140   160   170   185   190   195   190   186   177   170   167   161   155

This will give a healthy curve for you to start from.

Also, make sure that LDRXNZK is changed accordingly. ZK is used during knock but as pulling boost is not commonly used as knock prevention (sompared to ignition retard and fuel enrichment) you can usually use LDRXNZK as a copy of LDRXN, perhaps if prefered you can decrease it somewhat.

Regarding your EGT question... If you are logging calculated EGT (not using a separate EGT sensor) then you need to keep in mind that this ME7 calculation is quite coarse. 950°C calculated can easily be well over 1.000°C actual so proceed with caution if you have no true EGT measurement values. Changing the hardware of the car (turbo, exhaust etc.) will affect the EGT model to make it more and more inaccurate and nothing beats true measurements.

thanks again.

I am battling to find KFMLDMX in my file, the template I was using too compare maps doesnt have KFMLDMX in it so I am not really sure what the map looks like.

http://nefariousmotorsports.com/forum/index.php?action=dlattach;topic=1184.0;attach=1481

Look at 13C42 adress. 9x8 map.

Thanks but I am not sure it doesnt look the same as another file I am using as a template. template screenshot is the first attachment.

I'll have a look tonight...

appreciated.

You must to view the maps in 16 bit and not in 8 bit.

Thanks, you wouldnt happen too know what addresses I could use for a rpm and dk axises in relation too this map?

13C38, 8 bit, factor 40 for rpm and 13C28,16 bit, factor 0,024414, for throttle position.

Thank you for all the help so far, does this look right? I am a little unsure why the map starts at 300kg/h throughout 438 RPM? I did try and move the start address but nothing else seemed right.

It looks quite funky, I need to double check that you have the right map.
A standard 180hp engine usually looks something like below, with engine speed from about idle to 6.5krpm's. %dk-axis seems correct though.

50,000   59,000   90,400   110,400   135,200   155,200   165,200   175,200
55,200   65,300   97,900   117,900   142,300   164,300   174,300   188,300
60,000   69,800   104,600   128,000   152,200   172,700   186,400   200,600
66,900   77,200   119,100   148,900   174,200   193,500   204,600   219,500
68,700   83,400   137,300   185,600   229,300   258,700   291,800   342,800
69,500   86,200   144,100   209,100   262,500   311,200   388,300   524,400
70,100   88,300   145,700   231,700   289,800   344,600   461,100   661,800
85,000   103,200   149,000   274,700   339,500   399,200   533,900   746,100

It also looks quite strange to have the last cell at over 9krpm's.

It may be like that because the xdf def for that map has 9 columns instead of 8, if I make it 8 then the rpm axis stop at 6500rpm but I still have the 400rpm column all 300s?

So, you must to use 8 bit for RPM axis and not 16 bit like in your screenshot.

the xdf settings

The fourth and fifth screenshot must be modified. Adress (HEX) is 013C38 and not 013C37, Data Size is 8 bit and not 16 bit (in screenshot number 4) and conversion factor is 40 and not 0.25 (in screenshot number 5).

again thanks for your help! Made the changes as you suggested. Still have the first column starting in the 300kgh mark? RPM axis seems fixed.

For RPM axis the offset is 0 and not -48. In the rest, all is ok.

like this?

sorry wrong screenshot

you still have factor 0.25 insead of 40 .

Here are some s-shots... 1.8TT 225 to 280+

OK changed from "0.25 * X" to "40 * X" the only issue is that I still have all 300kgh in the 320RPM column is that normal?

I have not worked with your particular ECU HW/SW revision earlier but I've never come across (as far as I can remember) this fuction as anything else than 8x8, so in your case the 9x8 is not that common at least.

Regardless, MAF restrictions at sub idle speed is not relevant anyway so...

OK so my attempt then to modify all other values EXCEPT that column was a good idea then.

Probably. If I were you I'd log the actual flow reported (using VCDS) to make sure that the current settings are actually not sufficient and that this is the reason you get a MIL.

After modifying that table no more MAF mil, I was logging VCDS block 002 values higher than was in that table. So now boost is good, cold starts are very lumpy but I still feel like there is more potential for my car so now I am looking at KFMIOP to see if that is influencing the power delivery. According too everything I have read here its the inverse of KFMIRL but now to find it in my file......... I dunno I guess my ecu is one of those difficult ones because not one of the maps seems too fit the 16x11 map I am using as a template  :P

never mind found it

16A7C

scrap that 16AC0 looks correct, the load axis is the next issue?

KFMIOP?

start address 16AA8 11x16

load axis address 16C08

rpm axis address 1276C

how do you figure out the factors for a map and its axis?

I mean, I don't know if there is a technical way to do it, but I open up a closely related damos in WinOLS and copy the offsets over. I just make sure when I'm done it looks correct; text and 2d.

thanks Rob, having a hard time with the RPM axis for this map and I am not 100% sure about the load axis either.

You have loaded text and 3d, from looking at it, your load looks ok, but rpm looks "f"ed. I'll have to look at it in a hex editor to see what I come up with. Will do after work- 5:00ish (sure would be nice to have WinOLS on the smart phone, lol). Maybe someone sitting with a pc can help you faster.

Here you go; later then expected, but... ;)  It looks like both of your axis were "f"ed, I think your load you started at 16C08, instead 16COA. You were very close! For Y though, which is much harder to find, you were way off. No big deal, I'm happy to help. Believe me, I'm not great at it, but getting better.

Thanks Rob, I am slowly getting a hang of this too thanks too the help of you guys! Hopefully I will be able too contribute back to this forum soon.

Perfect. Good luck with the torque map. It might take a few testruns, but I'm sure that you will get the car running properly soon.

OK KFMIOP, thanks Rob

looking a little closer at KFMIOP I dont understand why the load axis doesnt correspond too my stock KFMIRL which had a max load of 191, ie KFMIOP load starts at 10 and ends at 145 is that correct?

I'm sure it's correct. A load value of 191 in stock KFMIOP isn't really necessary if your stock LDRXN limits the output way earlier than that.
Your KFMIRL might have load values that are never used in real life as they are limited by LDRXN (for instance) and I'm sure that your stock LDRXN isn't no way near the 145 you are referring to in your stock KFMIOP? It happens that 150hp engines and 180hp engines share the same KFMIRL and the difference in output power is being taken care of by different LDRXN (among other things) although KFMIRL can actually tbe he same.

If you need to rescale KFMIOP when running higher loads, just do so to avoid having the same settings for "anything" above 145. The axis is editable as well and it might come in handy now that you are turning up the boost.
 
Keep in mind that all the maps in your stock file are more or less made for "stock purposes" and I'm sure that the VAG tech's didn't concider someone wanting to runt 1.3 bars of boost instead of the 0.45 bars you have stock.  ;) That's why some maps needs adjusting once you start to tune, and not only the map values but the axis ranges as well.  ;D

thanks again fredrik, now too figure out how too edit the axis in tunerpro.

To edit the axis, I defined a new table in tunerpro called KFMIOP - load axis, I used the same starting address for the data and the columns. Is that right? Now too scale the axis for this correctly would I need to ensure that the highest load level in that axis should be the highest load level I have in IRL?

this is the modified axis, comments?

loaded the file up with the modified IOP axis and car feels worse. Forgot too test part throttle surging etc but WOT in 2nd the TM light started flashing on the dash (no fault in vcds).

I've never used Tunerpro to be honest, I do everything in WinOLS and a HEX-editor actually so I can't help you with this. In WinOLS, it's quite straight forward to change the axis values as they are 1D and you can basically change the data as you please. If you have mapped the axis data correctly to your corresponding map, you will see the changes in the map right away.

yes the changes I made too the axis reflected in IOP immediately so I was confident that it was at least one way of doing it  :D, now too figure out how too tune IOP correctly because just changing the axis has made the car feel worse.

Well... As far as I can tell, in KFMIOP you have about 69 for 4krpm and 213, and your KFMIRL says about 99 for 4krpm and 219. You migh want to align a bite more...

so the figures should be closer? This is what confused me because of the post regarding IOP being the inverse of IRL?

better?

I would avoid having more than 99.9%, but other than that... Give it a testrun :-)

much better, will drive it like this for awhile and do some more logging. the only thing is the car now has a funny idle... not a miss or anything but it idles up then idles down, so for 2 seconds it idling at 720 then for 2 seconds it idles at 800 if that makes sense?

They are related,  inverse or not... the IRL have % in the axis and LOAD in the main values, and the IOP it's the inverse...

For me it's more like  double crossing.... I am very far from a person that understands the ME7.5 logic, it's just an idea...

[EDIT]:
When you set the MIOP at 100%, you are requesting the max load in IRL, it's that what happens?

Good to hear. As a rule of thumb, I always align these two maps as well as possible and as a result, I've never experienced any problems with part throttle behaviour or anything else actually. 

any idea about the idle issue? Also I never catered for the s3 injectors I had fitted and whats happening now is that I get a fuel trim rich DTC and all I am really doing currently is using lemmiwinks too pull 16% from the primary fuel adapation channel which seems too pull fuel trims more inline but I guess my next step would be too work on my fueling... whats the best approuch for to tuning fueling for bigger injectors?

krkte and tvub

thanks, suggestions on tuning those maps?

also examples of what the maps look like in your file please.

krkte?

You have some info about krkte here:

http://www.ecuconnections.com/forum/viewtopic.php?f=2&t=1742&start=40

^^^This location is wrong;

KRKTE 1D744 16bit lo/hi (1by1)

You have to find datasheet for injectors to figure;

TVUB 13FD4 16bit lo/hi (5by1) (x axis is 8bit) ;)

your the man Rob! Could you post a screen shot of what your 3d map/text for tvub looks like please?

Sure, right after work.

Is this TVUB??

TVUB - Factor 0.003200 precison 4 - units (ms)
X-axis unit (V) with factor 0,070400 precison 4

KRKTE - factor I think it's 0,002133 using precision 4

For the TT225 the stock value is 0.0832 (ms/%)

I think KRKTE is .000111? I will check when I het home, that's just from memory? Not 100% on that.

I've seen different factor in other ECU's:

0.000167 -> 4z7907551R (A6 2.7T)
0.000111 -> 8d0907550G (S4)
0.000111 -> 8d0907551F (S4)
0.002133 -> 8N0906018F (1.8T 225cv)

The TT has an 512kb ECU so the krkte it's 8bit, and the 1024kb ecu uses 16bit krkte... the difference could be here!

^^^Thats some interesting info. I've never really looked at 512k files (really didn't want to confuse myself to bad). I'm still pretty confident factor is .000111, Here's some screen shots. I made them quick, but you'll see. I have my maps from my bin up in the left for reference. Fwiw, even if .000111 is off, just watch your fuel trims and adjust accordingly. You might have to do that anyway unless you can find the stock bin to go along with your injectors.

As far as I know, TVUB (battery voltage compensation) should be found from injector data online. Google your part number off the injectors.

I'm kind of surprised you didn't attack your fueling first ??? Any reason? From everything I've read, KRKTE and associated maps are one of the most important elements in ME7.

Forgot to answer this... Read your Funktionsrahmen and you will see how it works.

Hi Rob, fueling on my car has been rich since I put the s3 injectors on nearly a year ago already and I ALWAYS watch my fuel trims so I wanted to focus on what I thought were the more complicated maps first ie boost etc then focus on fueling (kinda backwards I know)

Hmmm searched google for the datasheet for my injectors not finding a single hit, the injectors which were from an audi 8L S3 part number 0208156063 I assume they are Bosch?

nevermind part number was wrong 0280156063 datasheet attached

looking at the formula posted here

http://nefariousmotorsports.com/forum/index.php?topic=1189.0

I cant figure out how I would calculate this "260 @ 3Bar (43.5psi) = 260cc" if my car is the returnless fuel system?

What fuel pressure are you getting in your system? There must be a regulation somewhere.

Regardless,

0280156063 = 393cc injectors @ 3Bar = 269 measured at n-heptane

22.3793336/269 = Calculated KRKTE value @ 0.08319

Get an injector calibration sheet and edit TVUB while you're at it.

No idea hence my question, I dont have an analog fuel pressure tester.


what is the formula here too calculate the 269?


injector calibration sheet?

Thanks for the assistance.

If you are using S3 injectors they are same of TT225, so use the krkte that I posted above... If your ECU it's a 512kb one, than use the factor that I've posted. It's 8 bit

(The calculated value from krkte it's almost the same that I've posted and seen in one Damos file)

393*0.684 = 269

Check your fuel pressure or if stock find out what the stock fuel pressure regulator is for your engine/year/model.

As you can see from the image below the injector voltages are given on the spec sheet from the manufacturer (FNPW_OFFSET). Take these values and plug them into TVUB. The values below are for EV14 550cc injectors however.

(http://img.photobucket.com/albums/v677/elizondon/0280158117cs.jpg)

been searching cant find anything re a spec sheet for tvub voltages for these bosch injectors.

http://www.bosch-motorsport.de/content/language2/html/2659.htm

thought I would just copy the values of TVUB from an s3 file, looks like the files are pretty close. My defined TVUB is top

 airtite your file is 512kb?

I have a Damos from a TT25 (512kb file) that I use to define the stuff for my S3 they are very similar if you want I can post the Damos file

Nice work. Copy over temin as well if you trying to be spot on.  :)

nope my file is a 1 mb file, the s3 file I was using is from

http://nefariousmotorsports.com/forum/index.php?action=dlattach;topic=1137.0;attach=1439

The issue is that I am unsure about TVUB in the s3 file it doesnt look the same as mine in text or 2d?

Pic?

one page back

http://nefariousmotorsports.com/forum/index.php?action=dlattach;topic=1274.0;attach=1837

and here is the hex

I did a quick search to find out fuel pressure. Turns out the wonderful internet has different numbers all over the place. I know its 3.5 or 4 bar?

What's your engine code, AMB?

AVJ

ok looks like TVUB is at 13FD4

I have copied across the values from an s3 file for TVUB and TEMIN and set KRKTE as suggested above.

ok file loaded went for a quick 10 minute, drive pushing hard in 1st - 3rd couldnt really get into 4th BUT looking at block 32 after the drive, I now have the car adding fuel ie I have positive values in both block 32-01 (1.2%) and block 32-02 (9.4%) so it looks like its running lean under load.

Increase KRKTE?

also noticing a bit of knock under load, so I guess I have too pull some timing. According too the s4wiki they reference KFZW/2 but I see there is also KFZWOP/2?

increased KRKTE to 0.10304 and block 32-01/02 are both 0.0% so looking better. If requested lambda doesnt stick too specified then I will see this in block 32 right or should I be logging block 31 and monitoring it there?

 I have found this. I have not tried it however.

http://nefariousmotorsports.com/forum/index.php?topic=884.0

I logged block 032 to check mine.

Here we go, this is what I was trying to find earlier.


http://www.nefariousmotorsports.com/forum/index.php?topic=651.0

do you reset your adaptation values to 0 with lemmiwinks ?
your KRKTE is for stock 317cc injectors .

what I have seen is that as soon as I reflash the adaption values are reset.

Any more news? :)
Did you get the car to a dyno or something that can measure your results?

My car is getting a k04/e05 this week and I want to make some improvements by myself, like you did. Nothing but just the k04/e05 and a 4bar FPR that I'm planning to use.

Nice thread by the way. Thanks for all contributors.

havent had a chance to put it on the dyno but the car has been flying! easily beats modded 2.0T love the surge of power and the noise from the DV.

artite: when you first swapped the turbo and installed the k04, your car runs fine with k03 software?

Just picked up my car today and it's seems a bit weird. After 3000+ rpm the car seems to lose boost... acting like if I released the accelerator and I can hear the blowoff valve sending boost to intake again... but I'm WOT, which seems a lot weird. I read something about this can be "limp mode" because k04 is overboosting.

This can be fixed with ECU mapping or I'll need to trick my MAP with some kind of diode? Anybody have a clue on this?

if you are running stock software then you shouldnt be hitting limp because requested boost would still be low, and yes I did run stock software when I first changed too k04 with no issues

You need to log to find out what's going on. It would be a good idea to find the root of the problem before changing anything, don't add a diode. That would just be putting a band aid on it, if it did in fact fix anything.

Hi guys, thanks for replying me.

Today I get the car and a vag cable in hands... nicely dumped my flash to start analyzing it. First of all, let me explain: my car is a 1.8T GTI 2003. It was broken and forgoten in my house for almost 5 years... the turbine has blowed up. (Yeah, I'm kind of weird guy having a gti stopped for 5 years... hehe). Just leaved the gti and bought a bmw :)

The story: The turbine that blowed up was a k03s modified. My software is remapped too and the guy told me I'm making like 260hp with everything else stock (only K&N filter on original car box). The software was remaped by a well known brazilian chip tunning company who work only with german cars. They say that their chip is "powerfull but with safety". So I guess is very conservative.

Another thing to known: here our gasoline have at least 15% of sugar cane alcohool. Even the best gasoline (podium from Petrobras who says that the RON is 101 have 15% of alcohol). So, our octanes here is a little higher.

The car is really fast, but not so fast. It's faster than a k03s stock remaped for sure. I beated a lot of cars with it. Now, the car has a modified k04 (e05) and everything else stock (even the cat). I'm planning to remap it by myself as high I can and then buy some parts, like new injectors, maf, etc. Resume: it's a toy/hobby. I'm a computer scientist, so asm programming are simple to me, and these stuff are of my interest.

The car came back from the workshop and today I found a new situation: broken alternator... hehe... I bought a new battery to get my car back to the greasemonkey's... :) Found no other DTC's, just low voltage only, and ABS get weird when the battery died. Maybe this is the problem... the car isn't getting enough power to supply enough fuel pumping or other things... will fix this in the next week.

I'll upload my binary here to everyone who wants to explore it. Please make comments if you do so.

For now I'm looking LDRXN and I see:

90,0019   108,0023   111,5180   117,0025   154,5033   192,0041   192,0041   192,0041   192,0041   184,0821   176,1834   167,8395   158,6284   148,3391   138,1670   128,0184

These values are for sure really low compared to what artite are doing with their k04. Will start looking another things today, and I'll paste the results here for whom interested.

If anyone have any comments, please feel free to share :)

After some hours of examination, here's the things I spoted:

Km/h limit
- VMAXNB increased to 300km/h; (+50)

RPM limit (I'll raise it up to 7000 in my tests, someone can tell me how safe is it?)
- NMAX increased to 6900; (+100)
- NMAXDVG increased all table to 6900 (+300)
- NMAXGA increased all table to 6900 (+100)
- NMAXDV increased to 6900 (+300)
- NMAXOG increased to 6900 (+100)

boost and their limits
- LDRXN map table is the same that a original 180PS until 1720RPM, then it's increased, but seems to be a lot different/conservative than suggested here by artite/
fredrik_a; max value = 192, which I can guess 1.2bar
- LDRXN and LDRXNZK are using the same values; (ignoring knock issues?!? or dealing with it on another maps)
- KFMIRL are identical maps from 180PS original (untouched); which I can guess is limiting boost to 1.2bar
- KFLDHBN is entirely FF'd; (unfinished work I guess... the tuning guy wont did their job well)

As far I can see, and understand (and I'm a completely newbie here), my car was never "tuned", just raised some boost and limit's and that's it...
Paid a real good amount of cash, and now I'm seeing these guys with different eyes... sadly.

This is why so many "non-pros" are doing their own tuning.

I was told go to pro tuner over and over so i did posted the map on here and told it was total shit lol

Just to everybody knows: got my car back. Sucessfully loaded original k03 1.8T 180PS file on ECU. Came from a ride right now and guess what: the boost still cuting at 3000rpm.
 
Will spend my next hours trying to figure it out. I have no idea how to log things yet... if someone can point me, I'll be thankfull.

Nevermind, just figured out how to log on vag-com. Did a log like artite one, here the results. Can someone please give me some advice that what is going on?

Group A:   '002            Group B:   '031            Group C:   '115         
   Idle speed   Engine load   Inj. period   Air mass in      Lambda Factor   Lambda Factor   Bin. Bits   Bin. Bits      Engine Speed   Engine Load   Spec. Boost   Actual Boost
TIME   700-820 rpm   15-25%   2.0 - 4.0 ms   2.0-4.5 g/s   TIME               TIME   700-6800 RPM   15-150%   990-1800 mbar   <=1800 mbar
STAMP    /min   %    ms    g/s   STAMP               STAMP    /min   %    mbar    mbar


365.07   680   21.1   2.72   3.42   365.39   1.016   1.000                         364.78   720   19.5   920.0   940.0
365.99   1440   44.4   5.44   12.53   366.29   1.000   1.000                         365.68   720   21.8   920.0   940.0
366.90   1480   72.9   8.84   21.89   367.20   0.969   1.000                         366.59   1360   70.7   1370.0   950.0
367.80   1720   83.5   9.52   29.22   368.10   0.945   0.953                         367.50   1600   78.2   1480.0   1020.0
368.71   2040   98.5   11.56   40.58   369.01   0.938   0.945                         368.40   1920   91.7   1690.0   1120.0
369.61   2400   119.5   13.26   57.33   369.91   0.938   0.945                         369.31   2280   112.0   1700.0   1290.0
370.52   2840   143.6   16.32   82.06   370.82   0.899   0.789                         370.21   2680   133.1   1690.0   1560.0
371.42   3400   162.4   26.52   100.61   371.72   0.750   0.891                         371.12   3200   168.4   1650.0   1990.0
372.33   3880   151.9   24.48   123.03   372.63   0.789   0.766                         372.03   3720   125.6   1620.0   1470.0
373.23   4400   119.5   17.68   111.83   373.54   0.789   0.727                         372.94   4240   118.0   1650.0   1630.0
374.14   4880   115.8   22.10   90.56   374.44   0.750   0.836                         373.84   4760   160.9   1660.0   2300.0
375.04   5360   162.4   34.00   169.56   375.35   1.000   0.883                         374.74   5200   148.1   1680.0   1900.0

best is to do a 3rd gear run starting at 2000 rpm till redline BUT from your log above at 3200rpm your boost spikes too 1900 where as requested boost is 1600 this is more than 300mbar and would cause limp, either your wastegate is not adjusted correctly or your n75 has a vacuum leak or isnt working properly

My wastegate isn't touched by nobody. Just bought the k04 and the shop installed it as it is. Do you have any clue how can I adjust it?

My N75 is stock, and last week I just disconnected it and it retains some boost ( I guess 4psi ) without the car going to limp mode. Any other way to verify if N75 is working good?

Thanks again for the response, and sorry for asking a lot of newbie questions.
 

Update: I guess I'll need to remap N75 to work with my actuator. I will deal with this later.

Just mapped to request 29psi at all just to test, and voilá. 22psi starting at 3,4k rpm, max of 24psi at 3,8k rpm. Never had a chance to pass 4k rpm in 3th (too fast/powerfull for a ordinary day/road drive) :) I read a little over the inet and saw that e05 turbo's are capable of 28/29psi, but it blows fast. I guess the manufacturer of this one that I bought have configured the wastegate for 24psi which people says that is the max for reability. So, the wastegate is good, the question is about N75 mapings to do the right job.

Cold hard facts: my K04 isn't a K03/hybrid, it's a K04-001 with E05 spool (probably a k04-020 internals). The turbocharger is way bigger than k03s and you only start to feel it at 3k rpm.

MAF limit DTC showed up. Will limit it higher and see what happens. Now it's time to improve my logs and log N75.

Another strange thing is that my car is a VW golf, but the ECU/software is for an Audi 1.8T ECU. Sincerely I'm no sure if my car is running 3bar or 4bar FPR. Need to verify this thing too.

But for sure, an upgrade to e05 definitely can't run with stock software. Limp mode will happen at 2500rpm and the car feels like shit.

the n75 should still control the boost and hold the wastegate open even with your e05 on stock software, I have had a few guys with turbos where the wastegate has been either tightened right up so it cant open or the wastegate was actually stuck....... my suggestion would be too load your stock map first and log things like n75 duty cycle, boost, throttle position

Today logs:


Thursday   17   May   2012   10:13:58                                 
06A 906 032 HP      1.8L R4/5VT     G   0001                                       
                                             
   Group A:   '003            Group B:   '115            Group C:   '114         
      Idle speed   Air mass in   Throt Angle   Ignit Angle      Engine Speed   Engine Load   Spec. Boost   Actual Boost      Eng Load Call   Corr. Load   Engine Load   Bypas Dut Cyc/0-100%
   TIME   700-820 rpm   2.0-4.5 g/s   0.2-4.0%   0-12 BTDC   TIME   700-6800 RPM   15-150%   990-1800 mbar   <=1800 mbar   TIME   <=150%   <=150%   15-150%   
Marker   STAMP    /min    g/s   %    ｰBTDC   STAMP    /min   %    mbar    mbar   STAMP   %   %   %   %

   454.10   2680   9.25   4.7   0.0   453.49   2720   12.0   930.0   1020.0   453.79   183.5   174.6   12.8   0.0
   455.00   2680   21.89   13.7   21.8   454.40   2680   17.3   930.0   970.0   454.70   182.7   173.1   28.6   0.0
   455.91   2800   59.06   99.6   11.3   455.30   2720   63.9   1500.0   990.0   455.61   184.2   174.6   86.5   95.3
   456.81   2960   87.53   100.0   6.0   456.21   2840   117.3   2270.0   1370.0   456.51   187.2   180.6   131.6   95.3
   457.72   3200   113.83   100.0   3.8   457.11   3000   160.2   2370.0   1870.0   457.42   191.7   189.6   169.9   95.3
   458.62   3480   135.94   100.0   4.5   458.02   3320   189.5   2550.0   2260.0   458.32   191.7   198.5   191.7   95.3
   459.53   3800   145.19   99.6   6.8   458.92   3600   191.7   2550.0   2420.0   459.23   191.7   198.5   191.7   95.3
   460.43   4080   151.08   100.0   6.8   459.83   4000   191.7   2550.0   2500.0   460.13   191.7   197.0   191.7   94.5
   461.34   4400   157.36   100.0   7.5   460.73   4200   191.0   2550.0   2500.0   461.04   191.7   200.0   188.7   95.3
   462.25   4640   146.06   30.6   7.5   461.64   4480   178.2   2090.0   2540.0   461.95   191.7   203.0   160.9   60.8
   463.15   4680   142.11   32.5   7.5   462.54   4520   154.9   1840.0   1560.0   462.85   191.7   203.0   160.9   74.1
   464.06   4800   149.69   39.6   7.5   463.46   4720   112.8   1550.0   1770.0   463.76   191.7   204.5   137.6   72.2
   464.97   4880   131.31   91.4   11.3   464.36   4840   126.3   1600.0   2470.0   464.67   191.7   204.5   145.1   5.5
   465.86   5040   75.03   25.5   12.8   465.26   5000   155.6   1900.0   2080.0   465.56   191.7   206.0   133.8   54.5
   466.77   5160   158.83   52.5   6.8   466.16   5040   120.3   1580.0   1230.0   466.48   191.7   206.0   115.0   66.3
   467.67   5200   154.83   96.1   9.0   467.08   5080   97.0   1350.0   1790.0   467.37   191.7   204.5   104.5   70.2
   468.59   5320   127.64   94.5   9.8   467.97   5240   142.9   1380.0   2440.0   468.29   191.7   203.0   85.0   0.0
   469.48   5360   75.14   27.5   12.8   468.89   5360   148.1   1760.0   1960.0   469.19   191.7   203.0   112.8   33.3
   470.40   5480   61.14   23.5   16.5   469.79   5480   115.0   1680.0   1460.0   470.10   191.7   203.0   178.2   61.2
   471.30   5600   65.56   24.3   13.5   470.69   5600   117.3   1410.0   1230.0   470.99   191.7   203.0   172.9   56.5
   472.21   5680   84.39   24.7   15.8   471.60   5600   105.3   1590.0   1290.0   471.91   191.7   203.0   177.4   69.0
   473.10   5720   44.72   22.0   18.0   472.51   5640   119.5   1490.0   1250.0   472.81   191.7   201.5   162.4   62.0
   474.02   5840   51.69   22.4   13.5   473.42   5880   95.5   1450.0   1180.0   473.71   191.7   201.5   149.6   67.5
   474.93   5800   50.17   22.0   18.8   474.32   5800   115.0   1460.0   1200.0   474.61   191.7   201.5   157.9   62.7
   475.83   5880   52.22   22.4   18.8   475.23   5840   97.7   1460.0   1190.0   475.52   191.7   201.5   149.6   69.0
   476.74   5960   49.19   22.4   18.0   476.13   5920   97.0   1500.0   1180.0   476.42   191.7   201.5   164.7   71.4

My n75 is working, we can see duty doing the job and the boost decreasing. I guess.
The weird thing: this log was made in 3th gear WOT like you told me to do. This entire time I'm pressing the gas to the floor and you can see the throtle angle closing down... The boost is cutting because of throttle cut.

The car is running good until 4krpm, and then it "chokes" and it's really hard to get to 6k+ rpm. Any clue?

is this a stock map? Your requested boost looks fairly high for a stock map.

also please attach the csv next time makes it easier to read

Nope, still my "tuned" file that I'm working on it.
I'm fighting with me7logger... it's blowing up an error randomly... can't log things right yet.

Did a little run and logged the pieces when the car "chokes". CondLambdaForPartProtection (B_lambts) and CondEGTForPartProtection (B_tabgbts) are both triggered (1). But I guess these 2 things only affects my fueling.

My car has no misfires and I can't see any ignition retard due knocking. Both values show as 0's. Knock voltages are below 10V. These can be wrong in higher RPM's, but for now still can't pass 4000rpm without the car "choke".

The problem stills the Throttle that cut's while my gas pedal still's 100%. Maybe the problem is MAP? I'm for sure hitting boosts near MAP limit (2550)...

So have you read the s4wiki article on throttle cut?

Rpm 4480

Actual boost exceeds requested by more than 300mbar! Throttle cut triggered as a result.

my suggestion is load your stock map and make sure you have no issues ie no throttle cut etc that way you will know whether its hardware or not

Thanks for the right path nyet. Now I see that my boost are getting over 24psi (MAP limit) and I really need to be carefull (uncontrolled boost)...

Thanks artite that point me to load original file and make things right without raising boost. You told me this twice. This is what need to be done in first place at all. Now I see this clearly.

My turbo is a k04/e05, and my original car PID is trying to deal with k03s wastegate. This is the problem. K04/e05 is based on k04-001 with k04-020 internals (TT225 one). So guess what I did: copied TT225 PID maps and did a little bit of math (because the maps axis values isn't the same), and tchadám: the car PID is now working fine with "original boost maps". The N75 cycle is doing fine and the ECU can cantrol boost fine now. The car is smooth until 6000rpm now.

But another things happened :) The ECU is throwing DTC: 17963 Charge Pressure: Maximum Limit Exceeded and the logs still showing that past 4000rpm the car is lowering the boost request and the N75 is working to decrease boost (without throttle cut), even LDRXN and KFMIRL remaped asking 22PSI. Now it's time to read more info...

Hey guys, attached is a 4th gear WOT log I did right now... can't go far 4800rpm because the road ended... For now it's what I can do.
You can see that in 3600RPM my requested boost start to decrease, and this is the problem. I really doesn't know why it's requesting less...

Attached is my LDRXN and KFMIRL. I'm trying to request 2500mbar ( ~ 22psi) past 4000rpm, so I really doesn't understand what I'm doing wrong now... The throttle is cutting again because I'm delivering too much more boost (I known, that's the idea now... I don't want to tweak PID again, because this loweringin request boot isn't what I desire).

LDRXNZK is exact copy of LDRXN. I have FF'd KFDLULS. KFLDHBN is FF's too. 

that is NOT a csv, it is a cut and paste from excel.

Please attach an unedited csv with full headers.

Also, don't bother with VCDS, you're not going to be able to do any real further tuning w/o a proper logger

nyet: nefmoto logger neither me7logger are running fine here... I guess my obd2 bus is high loaded... the log starts and then both softwares stops at random... lost comunications I guess... :/ I'll try ECUx and see what happens.

Attached .xls now...

Not an xls either. the original csv please.

ecuxplot can only read real csvs, since there is no telling what people do to their xls' before they save them.

In any case, make getting a real logger working a priority.

Also, 3rd gear is fine, although 4th might be easier if you are only trying to log ramp.

Your LDRXN is ridiculously high at low rpms.

Please read up on positive deviation.

Found the problem with loggers.... my notebook is a IBM/Lenovo with an accelerometer who stops the HD when the latptop shakes. Disabled this feature and the logger runs fine.
Attached the first log I did to try, so no long WOT's... just a test.

Found 2 problems in my car: EGT sensor probably broken, because always displaying 1035C and coolant temps getting so high that the corrected load decreases (causing my requested boosts to drop). The car is cutting throttle when temp gets high and then when it runs from 100 throttle to 30 (cutting throttle to decrease too much boost) the KFMIRL map get's in place and try to get my load even down (max load at 30% is very low) causing the choke's.

Will fix these 2 issues before trying any more mods on the maps/ECU.

btw, loaded this file on ECUx plot and your tool is really great!

Please if you have comments, feel free :)

Another lesson learned: you will never play nice with stock injectors :) No matter what turbocharger you`re talking about, fueling is the first thing to increase before increase boost :)

The temperature problem was because my car is running extremely lean with high boosts (Thanks to ECUx plot I can see this clearly). Injector duty cycles extremely high. Did a run with 1.2bar max and I can go far to 6000rpm without chokes, but again, the car start to starving out at 4.5k and throttle cut at 6k. Sincerely, now I can see this things and I`m wondering again about the tuners I paid to get my car chipped and they never told me this stuff... what a pirates. Owwwww. Increase boost and let the sponge bob play!