I have set up my LTFT+STFT vs MAF table and have logged a drive from work. My question is can I simply just past special % over my MAF table to correct the error?
I have set up my LTFT+STFT vs MAF table and have logged a drive from work. My question is can I simply just past special % over my MAF table to correct the error?
You'll want to use data from after the engine is up to temp (coolant temp > ~175 degrees F or so), and then filter every thing so as to mask gear changes, coasting/DFCO, fuel enrichment, etc. Here is something I posted in another thread:
Here is the filter I used to tune the MAF and VVE; basically says:
ABS([50040.71]-[50040.71.shift(-50)]) < ([50040.71]/[50040.71.shift(-50)])*0.1 AND ABS([2311.71]-[2311.71.shift(-50)]) < ([2311.71]/[2311.71.shift(-50)])*0.1 AND [6310]<7 AND [2517.161.avg(2000)]=0 AND [2517.161.avg(-2000)]=0 AND [50118.238.avg(1000)]=1 AND [50118.238.avg(-1000)]=1
MAF airflow must change less than 0.1% over 50 ms;
VVE airflow must change less than 0.1% over 50 ms;
Torque Management = 0 (filter our gear shifts);
Fuel trim less than 7 (forget what the numbers mean offhand, but they change based on what the car is doing, like cruising, idling, coasting/DFCO is on, etc.);
Commanded EQ ratio = 1.00
My polling rate for the MAF and VVE is 50 ms but if you're polling quicker than you can adjust the filter. Other than this you just want to try to drive at steady state as much as possible. I used a long highway at night/weekends when no one was around and set cruise control from 100 km/h to 130 km/h in 5 km/h increments, in 4th, 5th and 6th gear to fill out VVE is much as I could, only took a few attempts to get the VVE and MAF within 1 - 2% using STFT + LTFT. Also keep your cams set to 0/0 (both cams at home position) because you can only adjust the base VVE table (both cams at 0); there is a table for every cam angle combo, but they all adjust when the base table is changed.
You could also add in the coolant temp parameter and filter for > 175 F. Also your parameter ID's and "Fuel Trim Cell" numbers might be different depending on your ECU. This is for an E39A. With steady driving/throttle and proper filtering, then yes, just apply the % change (or % half) to the existing MAF (and VVE) using LTFT + STFT. Also remember to reset the fuel trims via the Scanner after downloading a new tune and before recording the next log, so as to prevent old data from affecting new data.
I just add this filter math to the filter slot on my LTFT+STFT vs MAF table?
Do I need to set torque management to 0 under the engine or under the trans settings, or both?
What is the "POLLING RATE"?
Could you please tell me if I need to log all of these PIDS in the VCM scanner as well?
Also, and most importantly, I do not have a COMMANDED EQ RATIO PID, all I have available for my PCM is COMMANDED AIR FUEL RATIO (LO RES), Is there another option you could think of to use in place of this one?
Yes, copy and paste into the filter of your graph. You'll need to be logging/recording all of these parameters in your Channels list.
Don't need to touch Torque Management in your tune, you're just filtering out data when torque management doesn't = 0 (so when you shift, torque management kicks in to reduce torque to the transmission, and you don't want to be using data during these times).
You can right-click on each parameter in your Channels list and set the polling rate which is basically how often the parameter gets read/updated.
If you don't have EQ Ratio (Lambda) then you can use Air Fuel Ratio just as well, just set the filter such that it's only looking for whatever your AFR is during normal driving/cruising (I'm assuming this would be 14.7? Or something close to that - whatever your scanner shows while normal steady state driving).
Okay, I will set my tables up and filters up and try running a log this afternoon when I get off. I will get back to you and let you know how it goes.Thank you for your time and your help!
I am sorry to keep asking questions but, is there anything that I need to disable? I see a lot of people talking about failing the MAF or disabling SD/Dynamic airflow. Reason I ask is because I do not have an option to disable the dynamic airflow by setting the HIGH RPM DISABLE to 400 RPM, which ensures that I never go into dynamic airflow.
Don't know much about that, but I believe with Gen IV and V ECU's there is no need to disable the MAF to tune the VVE and vice versa. Don't know what ECU you're working with; the E39A is a Gen V (well, not sure if it's considered a Gen V since it's a V6? But it behaves the same as far as I can tell) and with it you don't need to disable anything to tune the MAF and VVE, and both can be adjusted at the same time.
I am really sorry my friend I should have specified in the original post, this is a 2000 Chevy Silverado w/ a 4.3 liter vortec engine. Again I'm so sorry to have possible wasted you time.Is there any way any of this is transferred over to the vortec platform? Again I'm sorry that is my mistake.
Based on the year I'm guessing this is a Gen III ECU, or at least functions as such. If that's the case you could still use the same filter as above except for maybe the VE Airflow wouldn't be applicable (and the fuel trim cells might be different, but easy enough to log it and see what it's doing as you drive around), but I really don't know - my E39A ECU is the only thing I've tuned. If you need to disable the MAF to tune the VE and vice versa, I wonder if you can ask in the Gen III forums what tables (ID's) are needed exactly, and request HP Tuners to add them for you. Chopperdoc181 on YouTube has a bunch of videos too (which I watched when I first started learning how to tune); you could maybe use his videos to find the tables you need.
Minor correction, but 0.1 = 10%Originally Posted by KillboyPowerhead
0.1% would be 0.001 (and would probably filter out all the data)
If your MAF read 100 g/s at some point (let's say right now) and read 95 g/s 50 ms ago (a 5% difference), then the formula says 100 - 95 < 100/95*0.1, or 5 < 0.105, which is false and the data would be filtered out. If your MAF reads 100 g/s now and 99.95 g/s 50 ms ago (0.05% difference), then you'd get 0.05 < 0.10005, which is true and the data would be accepted.
Or am I incorrect? I noticed a lot of data gets filtered out this way - even to confirm I would look at a minute or two long section of a log, totally steady state driving (cruise control set on a pretty flat/level road) and see very few samples taken, despite the MAF changing less than 10%.
When setting my base AFR, should i change this to like 13.1 or leave it at 14.7 cause of 93 pump gas.
You'd want to set to 14.7 during normal running conditions (normal driving, engine up to temp). The factory O2 sensors are designed to read very close to stoichiometric AFR (14.7). The LTFT and STFT error is derived from the O2 sensors and is a % difference from stoic, and if you're trying to get to as close to 0% error as you can you're effectively trying to get to as close to stoic as you can. During power enrichment (PE) is when you'd stray from stoic. For AFR during PE, every engine is different; my engine (LFX) seems to like 12.5 from what I've read. Could try changing yours in 0.2 increments or something, do a pass, see if there's any change, etc.
Ah, right you are; I misread the filter on first go-around. It's not clear to me why the ratio of the current and prior readings comes into play.
This simpler version probably works just as well: ABS([50040.71]-[50040.71.shift(-50)]) < 0.001*[50040.71.shift(-50)]
Or, in words: MAF delta must be less than 0.1% of MAF prior. So if the MAF read 100g/s 50 ms ago, then the current reading will pass the filter if it's between 99.9 and 100.1.
If it's filtering out too much data, you can always experiment with incrementally increasing the multiplier to 0.002, or 0.003, and so on.
Killboy, whenever you logged your car how high in the ROM band did you go in each gear?
Once my log is done do I change both the MAF table as-well-as the VE table at the same time?
Like I said my truck is a 4 speed auto so when I went out last night I set Cruze to 4 in 3rd and went up 500RPM at a time until 70 mph then shifted to 4th and continued going up 500 RPM until I was going to fast and stopped recording data cause at that point it was only hanging around 2500 to 3000 RPM.
Also, did you tune the PE the same way or did you use another filter along with a wide band? I have a wide band installed and set up on HP TUNERS, just not sure how to implement it. Once again, THANK YOU for taking the time to help me. And thanks to all who have shared your thoughts.
I set cruise control to 100 km/h (~60 mph) and drove for ~5 minutes or so in 6th gear, then same thing in 5th, then 4th, then set cruise to 105 km/h and drove for ~5 minutes in 6th, then 5th, etc. all the way to 130 km/h (~80 mph). I noticed once I was around 3500 RPM or so for a bit the COT (cat over-temp) would come on (can see EQ ratio/AFR drop) so I didn't drive at the higher speeds in 4th gear much, so as to not toast the cats. So all together covered 1600 RPM to 4000 RPM of the VVE, and 2500 - 5500 Hz on the MAF. Going up and down hills also helps to hit the upper and lower parts of the VVE table. Can also turn on the AC which adds load on the engine, and so you'll hit higher rows on the VVE compared to not having the AC on. Anything above 5500 Hz and 4000 RPM I did via wide open throttle pulls, so in those cases made the filters much less restrictive to get more data in those smaller time frames.
Again, don't know anything about your ECU, but for mine I adjusted MAF and VVE at the same time.
For PE you'll need to get your wideband into the channels, and also disable whatever you need to to run without a factory O2 (assuming you removed a factory sensor to use the wideband which is what I did). And then see what the wideband says compared to your tune. I prefer to go by lambda/EQ Ratio rather than AFR. Whether you're tuning PE or normal driving, you just need to get lambda to equal 1.00 (plus or minus a couple %) and then the ECU does the rest: whether you're requesting 14.7 AFR or 12.5, so long as you got everything close to 1.00 lambda then the ECU knows how much fuel to add at any requested AFR.
Reply With Quote