Page 2 of 3 FirstFirst 123 LastLast
Results 21 to 40 of 41

Thread: LSJ WB VE Tuning

  1. #21
    Senior Tuner cobaltssoverbooster's Avatar
    Join Date
    Dec 2008
    Location
    Nevada
    Posts
    4,452
    your still reading air fuel ratio commanded instead of eq ratio commanded.

    your wideband is reporting lambda and your ecu is reporting afr.
    in order for the fuel error math to work these must be the same. now its up to you to choose but i would run lambda error if you do anything that is involving ethanol.
    so leave your wideband in lambda but now delete the commanded afr channel and use commanded eq ratio (which is lambda) as your ecu reporting pid.
    2000 Ford Mustang - Top Sportsman

  2. #22
    Tuner in Training
    Join Date
    Apr 2018
    Posts
    20
    Ok, you'll need to help clarify things for a newbie like me. I was messing with layouts and so I did have some erroneous junk in there. I have added an external sensor WB. I have the AEM 30-4110 and am running standard not pro for HPT. When I add the sensor I add it as reading Lamda and tell it that it is a AEM 30-4110. As far as voltage offset, when I have the scanner read the WB it reads 1.264, which is the lamda in open air from the manual, I don't seem to need an offset.

    As far as having the channels read commanded EQ, it only lists the sensor itself already added. If I have it read Commanded AFR it reads 10.1422, my commanded AFR for 80% ethanol.

    In the charts I have it display EQ and EQ error by pulling formulas. For EQ commanded its [50121] / 10.1422, where 50121 is the sensor in AFR and 10.1422 is my commanded AFR, and since they are the same it should be 1. For EQ Error its 100* ( [31000.238] - ( [50121] / 10.1422 ) ) / ( [50121] / 10.1422 ), where 31000.238 is the AEM EQ ratio. It seems to all work.

    I'll attach my most recent layout to see if that's right.Layout 06.03.19.Layout.xml
    Last edited by blackred; 06-03-2019 at 10:46 PM.

  3. #23
    Senior Tuner cobaltssoverbooster's Avatar
    Join Date
    Dec 2008
    Location
    Nevada
    Posts
    4,452
    i never have done conversions like that. i guess as long as you verify your ethanol ratio in afr by testing through lambda error then you could do that.
    whatever i set my gauge up to read on the face of the gauge, thats what i also read as a commanded value. afr - afr and lambda-eq
    i dont have vehicles lying around so i cant connect and recommended the correct channel for direct lambda eq commanded.
    2000 Ford Mustang - Top Sportsman

  4. #24
    Tuner in Training
    Join Date
    Apr 2018
    Posts
    20
    Well, assuming I set everything up right, and I think I did, do you see anything glaringly wrong with the MAF tune in process as is?

  5. #25
    Senior Tuner cobaltssoverbooster's Avatar
    Join Date
    Dec 2008
    Location
    Nevada
    Posts
    4,452
    when you make your custom user paramters save those in the creation screen and then upload them. luckily you posted your formula here so i just made my own new one.
    from what i can see you are doing good on the maf tuning side.
    i do notice that you are having some fueling trends that will not follow the maf curve very well. i suggest you set the maf to about 1500 or 2000 rpm and tune the ve in below whatever rpm you select to run ve. the multi axis correction that the ve table provides should be able to cover the issue i see. remember to keep the tables smooth. if that means you need to have +/- 5% error in a few cells then thats ok. when the error gets to around 8% that becomes a critical point and i would apply a small correction to drop it closer to a 5% error.
    2000 Ford Mustang - Top Sportsman

  6. #26
    Tuner in Training
    Join Date
    Apr 2018
    Posts
    20
    Sounds like a plan.

    I know most people on the lsj platform tend to MAF only. Definitely wouldn't hurt to do VE at idle or low RPM. Did you happen to look at the VE table? Are the spikes seen typical of the lsj P12 ECM? Should I smooth them? I already ran several runs for VE tuning.

  7. #27
    Senior Tuner cobaltssoverbooster's Avatar
    Join Date
    Dec 2008
    Location
    Nevada
    Posts
    4,452
    Everything should be reasonably smooth. I'd have to look at the ve table again to comment on yours. I've been in and out of some CNC machine classes so I'm sporadic at best on replies. I only get a few minutes here and there

  8. #28
    Tuner in Training
    Join Date
    Apr 2018
    Posts
    20
    No worries. Im working on MAF and PE tuning for a bit. I have to go out of town for another week, then I can hopefully get things dialed in. Let me know what you think when you have a second to yourself

  9. #29
    Senior Tuner cobaltssoverbooster's Avatar
    Join Date
    Dec 2008
    Location
    Nevada
    Posts
    4,452
    id say your ve needs some help. getting close to the correct injector offsets and base data will help the table quality out a lot.
    2000 Ford Mustang - Top Sportsman

  10. #30
    Tuner in Training
    Join Date
    Apr 2018
    Posts
    20
    Are you referring to the fuel injector control section in the tune?

    I'm running ID1000 injectors and have all the tables in spreadsheet provided by them specific for our cars. I'm running a DW300 fuel pump with a fuel labs BRFPS. I am using the vacuum port on the external fuel pressure regulator set to 42psi (at 0 psi). I have set up the injector tables accordingly. My flow rate vs. KPA is static across the range, as well as my offset in the offsets vs. volts vs. PAC tables.

    Let me know if that's what you are referring to. Not only is this data from the injector manufacturer, but in the car, setting it up this way, it idles and and runs well.

  11. #31
    Senior Tuner cobaltssoverbooster's Avatar
    Join Date
    Dec 2008
    Location
    Nevada
    Posts
    4,452
    Use the filter options in your scanner to only analyze data that is in steady state. your table quality could be due to constantly changing data. you need to analyze the data as it stays for a period of time in each cell. you can do that by using the filter system to remove any data where tps is not steady and/or rpm/map/ gcyl are not steady. you would normally monitor the derivative to do this. The program has this built into the filter generation menu under the option "slope".

    lower slope values indicate more constant signal sources. you will have to play with the sensitivity to get your desired results. This feature basically turns your scanner log into a steady state dyno recording once you get it fine tuned. also make sure to filter out any rpm changes that have negative slope. i do this to remove any chance of disabled cylinder fueling for coast down. some weird stuff happens in coast, its easier for me to let the ecu do its thing and i just remove that section from my log file.
    2000 Ford Mustang - Top Sportsman

  12. #32
    Tuner in Training
    Join Date
    Apr 2018
    Posts
    20
    Would you be willing to share what filtering formulas that you would use for VE, MAF and PE tuning?

    I get you on the coastdown and fueling. I don't know if it's an issue, or one that can be remedied, but when driving around (not tuning) when I take my foot off the accelerator it very noticeably cuts fuel to the point it suddenly slows down (WB shows it off the charts lean), but not always. It happens at different RPMS. I don't know if this is a DFCO area or something that can be adjusted? I think that when this happens during tuning it can throw cells off. So filtering this out would be the best.

  13. #33
    Senior Tuner cobaltssoverbooster's Avatar
    Join Date
    Dec 2008
    Location
    Nevada
    Posts
    4,452
    usually changes based on vehicle but you could say
    rpm(slope) >0 and rpm (slope) < 200
    tps slope > 0 and tps slope < 8

    its based on whatever pid you use and how constant you want the result to be. more specifically its based on the pid units so rpm is rpm and tps is %, both are allowed change
    you also have to set a ms time in which your change is allowed, so if i set it to 10 ms then i would be saying i want to see tps changes from 0 to 8 % positive change only over a 10 ms window. this 10 ms window keeps getting analyzed to check for valid data reporting. my explanation is not doing this feature any justice. its one of the topics i struggle to explain.
    2000 Ford Mustang - Top Sportsman

  14. #34
    Tuner in Training
    Join Date
    Apr 2018
    Posts
    20
    Ok, I'll give it a shot. Unfortunately heading out of town for another week, then I hope to fine tune this.

    I'm assuming I can string along the RPM and TPS filters with AND in between. I'm not quite sure how to limit the time frame in ms. I'll mess around with it unless you have more pointers.

    I'm also assuming that the prior logs have all the data I would need, and if I go back and put in the filter, it would smooth out the results and I wouldn't have to start all over with new runs.

  15. #35
    Senior Tuner cobaltssoverbooster's Avatar
    Join Date
    Dec 2008
    Location
    Nevada
    Posts
    4,452
    so if your a computer code kind of guy its called boolean operations.
    filters are true/false labels (boolean constants). when outside the filter bounds the results is false. when inside its true.

    when you say OR, if a value lies out side of two filters separated with an OR then its false. if any of the two filters are within bounds(one or both filters) then its true.
    when you say AND, values lying outside of the two filters is false. if one filter is within bounds and the other isn't then its also false. BOTH filters have to be within bounds for an AND statement to produce a result of true.
    in this boolean test any result of TRUE will be scan results that you will see in the histogram plot. if the test is FALSE then those results get omitted from your view.

    if you string 4 filters with AND, then all 4 need to be within the bounds of the filter to be TRUE and allowed passage to the histogram view.
    i try my best to use AND statements to specify specific windows of operation they seem easier and all adding a new filter does is basically narrow the range of filtered data. in the example provided above that is how i would recommend you try using the filters the first time around.
    2000 Ford Mustang - Top Sportsman

  16. #36
    Tuner in Training
    Join Date
    Apr 2018
    Posts
    20
    This is my formula: [50090.156.slope(10)]>0 AND [50090.156.slope(10)] <8 AND [50070.56.slope(10)] >0 AND [50070.56.slope(10)] <200

    Loading VE logs this definitely cuts down on aberrant data. If I adjust the milliseconds up it narrows the band even further. When I get back I'll try some runs with the scanner with the new filters and see if this smooths things out.

    I think my MAF tune is getting closer as well. Any recommendations for filters for tuning MAF and PE?

    Thanks for your help

  17. #37
    Senior Tuner cobaltssoverbooster's Avatar
    Join Date
    Dec 2008
    Location
    Nevada
    Posts
    4,452
    Not really any special filter other than what has been discussed.

    You can use the smooth feature in the editor to smooth the tables out. If your logs after smoothing generate some minor error don't worry about it.

  18. #38
    Tuner in Training
    Join Date
    Apr 2018
    Posts
    20
    So it's been a while and finally back home and have been tuning in the mornings on way to work. I tuned VE tables first with the new filtering formulas in. It's definitely smoother, but I also don't think you can tune much of the table in our particular cars. I also just installed Fuel It bluetooth ethanol scanner and works great! Last fill up showing 78% ethanol (commanded stoich is for 80%) So some questions:

    1. How does the VE table look, any recommendations?
    2. How does the MAF table look?
    3. Loaded the PE tune as well for WOT. Any recommendations? I have kept the PE ratio at 1.25 across the board. Again, this is what I'v read on some forums, but should it be ramped up as well?
    4. Now that the base tune is completed, I think, how do I fine tune it by adjusting the timing. I showed zero knock retard in my scan. My histograms were set up from a forum recommendation. Do you use the timing advanced histogram, copy and paste by a percent, or by adding 1 or 2 to the scan? Only in certain areas? If so, where does this table exist? In the high or low octane tables?
    5. My car seems to respond better when tuning MAF, which means no dynamic is on. Can I just put dynamic to 400-500 or even zero so MAF ONLY?
    6. My car also does better with all DFCO off. What I mean is that when I turn everything back on (DFCO, STFT, etc), when I coast down or just take my foot off the accelerator, it cuts all fuel and almost lurches back and when you reapply gas it lurches forward. Not smooth at all. And when coasting right before a downshift to turn a corner, or to a full stop, right before low RPM of say maybe 1500 or so, it hesitates right before pushing the clutch in. But when in MAF tuning mode it never does this and seems to drive better. Any recommendations?

    Any and all help is appreciated for this noob who is willing to learn!

  19. #39
    Tuner in Training
    Join Date
    Apr 2018
    Posts
    20
    Bump for some advise

  20. #40
    Advanced Tuner
    Join Date
    Apr 2011
    Location
    Chicago, IL
    Posts
    763
    Im going to be mobile for the next week so i cant look at any files so i will rely on others for that. Hopefully cobaltoverbooster will chime in as well to take a peak. 1.25 is fine for afr target/pe ratio. Just make sure the wideband is reading what you are commanding for ve and or maf. I put widebands in lambda because to me its just easier to deal with on pump and e85 because lambda is always 1 no matter the stoich. Shoot for a .8 lambda which would be a 11.8 target afr for pump gas and then move to a dyno to finish it off to find what the vehicle likes best. You need to make sure the base fueling is correct before you start messing around with ramp in and ramp out pe setups imo.

    Timing is done by adding 1 degree here and there but you will only find the best timing for your setup on a dyno so you can see the graph for torque and horsepower. Same same for wot fueling. Your vehicle might want a richer afr or leaner but you wont know unless you see it. Copy paste % is for airflow model tuning. Yes ho and lo is where its at but also the ecm also uses subcharts that add timing like for ect, iat and afr. I dont run those and use solely my ho and lo spark to control timing but everyones different. I leave the negative for safety measure - jic. I also adjust the iat negative portion for e85 or race fuel users since the better knock protection so you can push more timing in the hotter iat temps

    As i said before. i found maf to be a smoother operation then ve especially for me since my lsj is cammed. Ve just made things to touchy and dont worry about the ve cells in the lower map/high rpm or higher map/lower rpm area since you dont hit them. Just make sure it looks nice and smooth. I just use 200(disable)/100(re-enable) for dynamic airflow so its in maf only mode.

    You can turn off dfco but your mpg will suffer a little bit if you use dfco mode - meaning in gear for coastdown since the injectors will remain active. The lurch is normal operation of dfco going on and off. You can adjust the settings to try to make it a little smoother to your liking if you wish. The hesitation i suppose could have something to do with having ve enabled but i cant a test to that but you said maf only it doesnt do it. maf is more of a general airflow model vs ve thats more cell specific so it will be more touchy/sensitive than maf so that could possibly being playing a role in the hesisiation but im not sure on that sorry
    Last edited by TCSS07; 07-08-2019 at 08:58 AM.