l2r99gst

iTrader: (2)

Hopefully the logs will show the issue and we can work on getting the VE tables straightened out.

bigric09

iTrader: (37)

if they dont, i got bigger fish to fry then i guess

I was thinking about flashing the settings Ryu has on the thread you linked earlier and seeing if that gets us to a closer place to start. The only difference may be im on a 3582 for now until my TS setup gets here, so i dont think i will need to be at 100 on the rpm scale until later on in the range.

l2r99gst

iTrader: (2)

Just keep what you have now. I can tell you what to change to based on the logs and go from there.

ziad

iTrader: (1)

lol i agree do what eric says.. he helped me understand SD and get my trims better. and i can get SD reasonable then so can u...... saying that no i never was able to get logworks to behave properly.... in the end gave up and just wrote some code to do it for me

i do have a bad stutter at 3700-4x00 at cruising load, playing around with VE made it better but then i didnt like the effect on a full throttle run at those rpms.

hence my RPM VE is 100 after 3500 or so. and 1:1 after 100load. seems ok enough

bigric09

iTrader: (37)

Eric... can you PM your email address. My cruise log compressed is a hair over 100kb. I will post my idle log though once my damn computer is done copying from my external.

l2r99gst

iTrader: (2)

OK, I have the logs. After I make the charts and look at your VE table settings, I will make some recommendations. I will just post here on what I am doing so anyone else that is interested can read and have it as reference as well.

bigric09

iTrader: (37)

sounds good... how hard would it be to explain the actual making of the graph portion. I have LW3 installed on my laptop, i just need to know as well as it would be helpful for everyone what it is you are doing so we can get things in line. Again, thanks for your help.

knochgoon24

iTrader: (2)

Thanks for the help.

Is MAT the temperature from the MAS sensor or from the IAT sensor in my UICP? I don't think I have my IAT sensor getting the correct readings at the moment with the ROM I'm using. (IAT in through fuel temp sensor pin #77)

Now here's the next bit of fun. I'm running a 2g DSM MAS in my car. I managed to tune the MAS stuff to get it to work quite well. If I tune the SD using this method, I'll have to copy all the MAS table values over, correct?

I have some free time tonight, so I may try to get some logging done at the very least.

I was just thinking that I'd make it a calculated value in EvoScan or I could calculate it afterward. It might be nice to see a live VE value. Either way, I could then load the data into either LogWorks or Excel and make it work. I'm better with Excel, but I haven't played around with LogWorks too much.

EDIT:
So after about an hour of searching through the SD threads, just jumping link to link, I came across this https://www.evolutionm.net/forums/ec...irst-test.html and I think this may be the best option for my situation. I'll still get some logs with the MAF so I have something to fall back on.

If I just ditch messing with the MAF stuff and used the Evo maps and apply the changes outline in that thread, it will probably put me near where I need to be. My MAF tables are a bit screwy. The "easy" setup just makes more sense to me right now. My VE is probably a bit less than an Evo, so if anything, I'll run a bit rich until I get stuff straightened out.

EDIT 2: I also saw that MAT is the sensor in the UICP or manifold, and IAT is the MAF temp.

l2r99gst

iTrader: (2)

OK, so I received the logs from bigric09 and put them into LogWorks to analyze the fueling using:

LTFT Lo + STFT for idle
LTFT Mid + STFT for cruise

The charts below show the results for the idle and cruise logs:

Idle - average


Idle - number of points


Cruise - average


Cruise - number of points


As you can see from his idle charts, he idles in the 50-60 map range at around 1000 RPM. The rest of the points for this chart are just extraneous data. We care about the cells where the most data points fall. So, looking at the chart, it looks like he needs about a -11% correction in fueling at idle.

The same goes for cruise, so we will concentrate on the cells that have a larger number of points, not the outlying cells that were hit a few times. We want the corrections numbers from the cells with the good sample rates. This also helps out in averaging transitions in throttle, etc. So, looking at the chart and the meat of the data, rough corrections for a few areas are as follows:
50-70 map/2500RPM needs -14%...50-70/3000 needs -12%, 50-70/2500 needs -13%
80 map - 2500/-4, 3000/-9, 3500/-12
90 map - 2500/-7, 3000/-8, 3500/-8

continued below....

l2r99gst

iTrader: (2)

OK, so now we have to see what the 3D VE table looks like based on the current VE tables. Bigric09 posted his setting on the page before in post #377.

I created an Excel spreadsheet to multiply out the 2D tables to create a view of the 3D table based on his settings. Then, below that, I have the same thing, but with new proposed settings, where we can change the 2D VE tables to arrive at the corrections that we found in the post above:



What I like to do is scale the RPM VE so that the highest value is 100%, which should be somewhere around the midrange, then taper off from there to either extreme.

So, from there, I started to adjust the map VE values to accomodate the corrections that were needed from the logs above. For example, his idle needed about a -11% correction. Remember, his idle was around the 60kpa/1000RPM cell. So, looking at his original VE table, that cell showed a VE of 67%. To make a -11% correction to that, we want to have a value of .89*67=60%. The values I came up with for now to achieve that was a map VE at 60kpa of 50 and an RPM VE value at 1000 of 73.

I then used that same method to go through the entire chart, trying to get as close to the corrections needed as possible. Again, it's not easy with just 2D tables, as changing a value in the map table changes an entire row in the 3D table and changing a value in the RPM table changes and entire column in the 3D table. So, when you change one map value, you will most likely have to readjust many RPM values and vice versa...change one RPM values may lead to changing several map values. This is where you may go nuts. This the reason why I advocate for a 3D VE table. But anyway, just use that method and then notice the trend in the table and try to blend the transitions...ie towards 100-120 map the VE starts to rise from the relatively flat spot in the cruising range. Also, as RPM increases towards the mid-range, the VE increases, and then falls over either side. So, for now, bigric will try out these new settings and we will keep tweaking from there.

One important note about the new settings and this car. The values that I chose for the VE tables will cause the WOT fueling to be about 10% leaner in this case. In talking to bigric09 over email, he said that it was too rich anway, so it may turn out good. But, anytime you are tuning VE tables, take caution on the first time you are going WOT, just in case it's leaner than you expect. If so, you can either use the RPM VE to tune it back to normal, or use the fuel maps. In this case, I would recommend using the fuel maps, and use the RPM VE to adjust for further mods in the future.

Hopefully this helps a few people out.


Eric

Slo_crx1

iTrader: (9)

To some degree that's pretty much how I adjust things for my SD setup. I never got to play with the 3 bar sensor so perhaps it is a more 1:1 scaling, but I know the Omni 4 bar does not like a 1:1 setting under 100% load. With that said, I've never seen my RPM VE ever jump above the 100% mark and honestly I'd probably adjust values in that range if it ever did jump above it. Lately when I have time, I've been trying to fine tune the RPM VE's vs LTFT Low and Mid by looking at the variances of o2 feedback and actually adjusting the MAF compensation tables to see how well the combination works for more of a "3D" effect.

In the long right for me right now it's pointless since by end of march I'll have a completely different setup on and running, so all this little tuning bits and pieces is just to try and make life easier in the next few months to fine tune as much of the VE as possible. A turbo that big on a stock motor is going to be interesting to tune, so I need to keep variances to a minimum.

bigric09

iTrader: (37)

im about to flash this and go cruise and idle again...

what my question is... what was the determining factor that in needed to lose 11% ? Im assuming you were looking at the fuel trims and seeing i was neg 11 to 12, and those are the cells i cruise or idle in therefore we are taking 11 percent.

bigric09

iTrader: (37)

im on a 3582 now with a built bottom end... im not sure what you are running to, but i may be a good candidate for a starting point.

bigric09

iTrader: (37)

also... how did you import the log into a format logworks will use ?

and... you are only showing 7 values on the map ve table and there are 8.

l2r99gst

iTrader: (2)

That's right. Your fuel trims tell you the percent of fuel that is being added or subtracted to maintain a stoich mixture.

I will zip up your logs in the format that LogWorks can read so you can take a look at what they look like. Just open it up in Excel and you will see.