I stand corrected. Hertz isnt and ouput but data from the MAF sensor. With the air/flow being the output function in grams per sec.
Very interesting write up.
Printable View
I definitely see your point. I believe GM abandoned the MAF-only system because of issues with response time during rapid throttle transitions. The older MAF-only systems (1986-89) had excellent drivability characteristics, but I'm sure the LS-style MAF/MAP systems offer improved fuel efficiency and lower emissions.
I tuned a VE Maloo HSV today with LS2 and e38, M6. Apart from spending 6 hours diagnosing why the motor ran of 4 cylinders.(A broken wire from the heavy handed exhaust fitter) I finally got it running on 8 and started to tune it mafless, we ran into some other minor tuning issues which were quickly sorted but once the sd tune was complete. The MAF was plugged back in. To my surprise the car now idled at 16.3:1 compared to the mafless tune of 14.7:1. Topend mixtures were spot on at 12.9:1 but I had to scale the bottom end of the MAF to get fuel mixtures spot on. And just to prove that this is the correct way of MAF tuning I tried to alter the VVE table(with the proper one saved) to alter afrs at idle and light loads, but It didnt help. Scaling the MAF sore immediate results. So when your MAF is out like this adjusting the VE tables won't help. It required significant adjustment. So I am all in favor for the mafless tune first then maf calibration last!:hihi: Not fudging figures to get required results:bad:
Joel, good job...:cheers:
I'm with Joel here as well. Over the past couple of weeks Joel and I have been talking about the 90mm MAF and how it works with the E38 ECM. I know there are a lot of guys out there that swear by the "SD Rules" philosophy, but things are changing with the E38 ( at least for the moment ). Here's a few points that I've come up with on this....
- The 90mm MAF is large enough to not be a significant restriction is most applications. Sure if you are chasing every last kW / HP for a full race car, it may be an issue, but for most street cars it's not too bad.
- The E38 ECM is much faster than the LS1 PCM. It seems to suffer much less from the old "MAF Delay" issues seen in the earlier cars. From what I've read and heard, the biggest advantage of SD is the responsiveness that comes with it. Both Joel and I have tested E38's and found that there is not much between a correctly set up SD tune and a correctly set up MAF tune.
- The new Transient Fueling tables for the E38 seem to rely heavily on MAF data. Turn off the MAF in the tune and you turn off Transient fueling as well.
- In std. form, the 90MM MAF is calibrated to just over 500g/S of air. Again, this should be OK for most street applications. If you need to measure more than 512g/S of air, then it may not work for you.
There is one big issue that the MAF suffers from. With a SD tune, you can place the IAT sensor in an optimal place to get accurate air temps while reducing the effects of heat soak. The 90mm MAF has the IAT sensor built into it and it can suffer from heat soak. You can cut the IAT sensor wires and splice in a remote mount unit, or you can modify the stock unit to reduce the effects of heat soak. I had a play on the weekend and came up with a mod that may help. Rather than clutter this thread, I'll post up some pics in a new thread.
Simon.
For what it's worth, I think the only significant advantage of SD over Mass Air is more consistent fuel control at idle on engines with long duration camshafts. Tuning LS1/E40/E38/E67 systems can be a bit confusing because the processor may be using speed/density calculations, MAF readings or some combination thereof to determine fuel flow. That being the case, the tuning sequence noted in previous posts is the only one that really makes sense. Since injector flow rate is a known constant, it's best to start by making sure IFR reflects actual injector flow. With that done, the next logical step is to disable the MAF and optimize the VE table. Following that, the MAF table can be optimized. When that's done, switching between Mass Air and speed/density should have a minimal effect on fuel trims-- until atmospheric conditions change.
I know you guys thought this thread was long gone...
Sorry to open this back up, but..........I've got an interesting question that I've always thought about and could never resolve a solid answer for myself.
Let's take a stock corvette and note the fuel trims after a drive.
After adding a CAI and Headers, the fuel trims rise by about 8% or so clearly reflecting that we've added airflow and the compensation trend is for the pcm to need to add about 8% fuel to bring it back in line.
This seems to hold true for LS1 and E40 pcms using a conventional VE table and a MAF.
Back when the E38 first appeared in Vettes, I noticed that I'd add the same CAI and Headers and would take her for a drive and the fuel trims would be the opposite....The LTFTs I was tracking would go negative indicating that the pcm had subtracted fuel as if it was too rich...
But Why? In the LS1 example we've ADDED AIRFLOW and fuel trims went up indicating the need for more fuel.....
Why is it that in the E38/E67 controllers ADDING airflow caused the trims to dive negative? Doesn't adding airflow dictate adding more fuel?
While I've got plenty of these vehicles under my belt now and I've got it working to my liking....I often wonder if I've missed something simple here....
At times I've assumed that the new generation of MAFs gets overstimulated by an increase in airflow and/or over reacts at low airflow....or maybe that a vortex of air is responsible for the overstimulation?
Maybe it's just because the math and the lack of a conventional VE table has changed?
In principal, they should be the same....and it threw me for a loop for the first bunch I tuned.....But, then it just became the same old "just make em' happy" procedure....
Anyone ever thought about/notice this besides me?
I welcome your thoughts.
Chuck CoW
I've experienced the same thing with my 07 Z06. It had a CAI and headers plus a ported TB. Monitoring the trims showed -10% on average. I was like, WTF? It was completely opposite on my 02 just as one would hypothesize.
What does it mean? Hell if I know but, like you, would like to hear from the pros!
Chuck,
One thing to consider is that when making changes like a CAI to a "factory" calibrated air system, its not just the increase of air that effects the MAF's but also the change in airflow through the MAF has changed. Maybe more so with the new style MAF....
I think the other issue is the coefficients that are used for fuel computation in the E38/E67 process. More factors are used in the final calculation that we forget have an impact.
I agree with the "make it work" method sometimes!
HT
No doubt...I've considered that....But, I realize we've altered not just the MASS or VOLUME but, the flow characteristics....
Eventhough, it still should be as simple as we've added MORE air and with that the trims should reflect it by turning positive...
Another thought is that while we've obviously added more air (especially at WOT) but, maybe we've significantly added more WOT air, but with the larger more efficient flow potential, we've actually lost velocity in the idle areas resulting in slower air or that the meter just isn't efficient at the lowest air flow rates resulting in the negative trims???
Make sense to anyone?
Chuck CoW
I have been doing a maf only tune on my race car. I am bypassing the ve table to where anything over 400 rpm's is maf derived.
I went to the track and it was way lean with just maf tune in stock form. Injectors were scaled using marcin's spread sheet. I have 30 lb/hr redtop injectos .
When i first made a pass at the track ,(no dyno yet) my car was lean. WOT read around 14.2/1 which I only made a partial run to watch the log.
Instead of dicking with the table , I hooked up my old mass air flow translater and set it at 14% rich for wot and 10% pt. Went back and logged it, 10.2/1 air fuel at wot.
I pulled it back to 8% rich and the air fuel ratio went to 12.2- 13.2/1 in that range through out the rpm range.
the problem is that my maf was reading almost 11,000 htz and my timing went down from a commanded 26deg to a whopping 2.5-6 degrees at wot . It showed 1.2 deg of knock when I first started into the throttle but then the knock went away. as soon as I went to wot is when the timing went way down to the single digits.
Last night I took the Maft off and scaled the injectors -3% to see what it would do with the timing and launched the car in my driveway( neighbors love me) . ON the log, my timing bounced between 24 down to 16.5 and back up to 24 with 0 knock.
Air fuel went to 14.1 again , I was thinking that I need to drop the injectors down a little more to see if that helps.
I am a little on the lean side through out the tune idle all the way up the rpm range , I command 14.63 at idle and it see's 16.5/1 up and down a little from there. I command 12.0/1 at wot (starting on the safe side) and am seeing 14/1.
I tried to do a auto ve tune first and no matter what I tried the car was so lean that it would not idle. I added the 15% to the ve table and it just wont stay running and when I touch the throttle , it bogs down and dies.
THis is why I started fooling with the maf tuneing. Less tables to fool with and changes with the weather.
I would not fool with injector flow rate on my Trans am that I drive, my race car is Idleing great and doesn not stall or surge when I come from high rpm back to idle with the maf only tune in it.
Am I completely wrong in thinking of playing with the inj. flow rate to get it in the air fuel that I want?
Aint Skeered,
Do what works for you...
there may be something going on that is not immediately apparent (maybe we need to look at your tune/logs again to figure it out).
Also do keep in mind that a racecar is different than a daily driver...
the racecar spends 95% of its time at idle or WOT, whereas the daily driver spends 95% of its time at idle or part throttle...
however, having all the tables correct has the following benefits (for both racecar and DD'r):
- part throttle driveability/repsonse of the the daily driver is better/improved (tip in, tip out, cruise, coast, ...),
- WOT AFR's are easier to set (what you edit in B3618/B3605/B3647 is the actual output AFR you see),
The tables all work together... so there are many combinations of values that will produce desired results in most cases...
...but if that is not working, then do what works... tweaking the IFR is not necessarily wrong, it is another method.
Cheers
Joe
:)
Aint Skeered-- and anyone else who's listening--
As joecar has noted, "Do what works for you". There are all kinds of theories about tuning-- some are right, some are wrong and some are both, depending on the situation. The philosophical problem I have with altering IFR is that it's the one variable that you can tie to an actual known value. On the other hand, I haven't found conversion factors to be all that accurate, and I'm pretty sure that some of the factory IFRs have a fudge factor built in. As an example, I have a set of LS2 injectors installed in a 5.3-liter truck engine. With the stock injectors, fuel trims were in the 0 to -4% range. Afer installing the new injectors and changing the IFR table, fuel trims were positive 10%. Fuel rail pressure is the same as test pressure, nothing else was changed, so the only thing that made sense was to change IFR.
Something else to consider-- on the old mass air systems, altering IFR was the only viable option because these systems didn't have a VE table and working through the mass air calibration tables was a painfully slow process. I've done a lot of chips for these cars using changes in IFR to dial in air/fuel ratios and gotten the results I was after.
In spite of all the theories and explanations, if wide band readings and commanded AFR are the same, it doesn't get any better than that. Someone can tell you that everything you're doing is wrong, but if the results are right, that's all that matters. Just be sure to keep accurate notes so that a month from now, you'll know what you did and why.
Thanks Guy's. I am dropping my ifr a total of 6% and I added 20% to my voltage correction table . I am going to try that Wensday night at the track to see if it puts me close. I also set my pe to command 1.22 eq to start. I am going to keep modding my Ifr untill I get air fuel to match commanded and then start reducing the fuel in pe.
I do have a little stumble on tip in , I guess that is going into pe mode. Any suggestions on getting that out? I only have the log of a short launch in my drive way with the tune that is in it now, Also can you see why on my log the timing at wot goes to 24 then down to 16.5 and back to 24?
Oh yeah, on the few cars that I do tuning on I save every tune with a discription. The files add up but untill I get one I am happy with , I save them before I change them.
The log is of the tune with -3% on fuel injectors.
The -6% is what I am going to load and try out.
Just so you know what I have for an engine and trans. It is a 347 with 11.63 /1 compression, cam is 239/243 at .050 with .612 lift int. .610 lift exh 112lsa 110 icl.
I have a truck intake with a 75 mm t/b that seems be not in need of porting. It is different then car t/b's and has a 3/16 th hole in the throttle body. It Idles really well as far as not surging but does seem to have a cleaner sound to it when it is closer to 14.7
When it is close to 18 /1 it still idles but it is not a clean lope.
Trans is a th350 with a 8" converter.
My timing is copied from my Trans AM which is a 6 spd. Will the th350 want a different timing table to hit the converter harder?
+1.Quote:
Originally Posted by dfe1
Thanks Joe. I am sticking with this method for this car for a while to see how it responds at the track. I may try to fool with the ve tables if I ever get on a dyno but its to hard with out being able to drive the car on the street.
I have a friend with a 408 ls1 that he had tuned by a pro and that is what gave me the idea for maf tune. His car runs and drives perfect and its all maf. The tuner is use to doing fords and this is mostly how he has to do them from what he told me so he stuck with that method.
His car drives really good with no surging also.
I offer you another option (might get some flack for this - but oh well).
1-Disable VE by dropping MAF threshold from 4,000rpm to 200rpm.
2-Using WBO2, alter IFR table (down for more fuel - up for less) based on stock MAF curve (extend top end if necessary) and %AFR error. Rinse/repeat 2~3 times if necessary.
3-Unplug MAF and tune VE with newly developed IFR table.
4-Plug in MAF, restore 4,000rpm threshold and enjoy.
Now, timing won't shift grams/cylinder calculations from altered MAF tables. A4 trans pressure calcs too. Basically, anything that refers to MAF input for load calculation. :)
Or you can set your injectors, "Disable VE by dropping MAF threshold {B0120} from 4,000rpm to 200rpm" and use BEN against {B5001}.
this is easy with RTACS without RTACS i suggest increasing {B5001}
by X amount first so you don`t run overly lean.
after injectors are set and Auto VE, this is your best bet. IMO
"Auto MAF" :hihi:
Flack is not necessary... it will be interesting to see how close the IFR table gets. ;)
I am going to look at this again tonight and try them out. Not sure which way yet but if you can go more into detail on both methods as I dont understand what rinse and repeat meens or what you meen by setting AFR %error and extending the Maf curve.
Mr Prick, I have never tried messing with Rtacs, could you go a little further into what to do?
Thanks guy's., I am leaving work to go home. I will check back in a couple of hours . Thanks, Chad.
RTACS works with the RoadRunner LS1 Realtime Emulation PCM.
http://www.moates.net/images/box.jpg
I dont have the road runner. thanks anyways.
Rinse/repeat simply means, after you go out and log your data, upload the new tune with changes and log again. Make changes and upload that new tune. By the 3rd time around, it should be pretty close.
As for further detail....all your doing is logging the parameters of the table your changing along with the data you need to make the changes. So, in this case, you'd log MANVAC, WBO2 AFR, and Commanded AFR along with the normal stuff. AFR% Error for IFR table = Commanded AFR / WBO2 AFR. That way, if you see 15:1 when you're commanding 12:1, the result would be 12/15=0.8....which is saying reduce that cell by 20% to keep the injectors open 20% longer. If it was the other way around (seeing 12 when commanding 15), it'd be 15/12=1.25.....which is saying increase that cell by 25% so the injectors only stay open 75% of the time they were. Make sense?
Since there's only a handful of cells in the IFR table vs. the MANY cells in the MAF table, it shouldn't take too long. One trick is to use your filters so that you're focusing on the data in the middle of the cells. In other words, for the 35kPa cell, I'd filter out any data above 36kPa and any data below 34kPa. Filter out any transitional throttle (TP% is changing more than 3% over 500ms) along with any lower airflow scenarios (i.e. RPMs less than 1800 or so). You'll see the new IFR shape start to take place...
this is good advice for {B4001},Quote:
As for further detail....all your doing is logging the parameters of the table your changing along with the data you need to make the changes. So, in this case, you'd log MANVAC, WBO2 AFR, and Commanded AFR along with the normal stuff. AFR% Error for IFR table = Commanded AFR / WBO2 AFR. That way, if you see 15:1 when you're commanding 12:1, the result would be 12/15=0.8....which is saying reduce that cell by 20% to keep the injectors open 20% longer. If it was the other way around (seeing 12 when commanding 15), it'd be 15/12=1.25.....which is saying increase that cell by 25% so the injectors only stay open 75% of the time they were. Make sense?
a calc.pid comes in handy here commanded/actual ("reverse BEN factor").
to overcome a rich condition with {B4001} you must increase the values and,
vice versa for lean.
also you can change {B3701} with BEN, which may be easier for you.
IMO
it is easier to alter {B0101} to a set {B4001} than the reverse.
Good points... B4001 is in "reverse"... increase B4001 to go leaner.
Again - keep in mind 100kPa on the MAP sensor is probably 0 or 1kPa in MANVAC. MAKE SURE YOU'RE USING MANVAC AS A REFERENCE WHEN ADJUSTING IFR!!! :) So, assuming you're commanding 12.0:1 at WOT and you're seeing 13.67:1, you would adjust the 0kPa cell of IFR down by 12.2%.
MAVNAC 0kPa = WOT
MANVAC 75kPa = idle
I am not using the long term fuel trim pids in my black box as my nova that I am tuning does not have narrow bands. Could I take them out of my options ini and replace them with manvac?
Another question, can you log wideband through serial cable in pass through mode yet?
ok I think I got it right in my options. I replaced the eops with cylair , Longterm 1 wiht dynair and also longterm 2 with manvac.
How do I go about manually validateing serial pids ? Just select pids and plug into car to validate?
Connect to car,
select pids,
save pid selection to file (for future reference),
File->Connect,
Info->Validate PIDs.
I will do that tomorrow Joe, Thanks. You guy's have been lots of help, I have learned a lot the last few days.
I will post results of how the nova runs tomorrow night if i can get the tune done in a few passes. Hope to be close enough to run the bracket race tomorrow night.
Races were cancelled after I made the 90 mile trip. The freeze busted the tracks water mains. I will try again later.
:bawl:
Went to the track yesterday. car was pulling timing again but I figured that out. Tq management in the transmission even though I dont have a 4l60 in it, (th350) was pulling timing. I also dont have a vss so the car did not know it was moving . I turned everything that I could find off that could pull timing and that fixxed it. with tq mng. I ran 12.06 @ 110 mph.
After fixxing the timing issue It went 11.24 @ 119.5mph with a 1.62 60 ft. Not spinning, the car is leaving like it has a 2 step. Intake air temp was at 115+ leaving the line and it was weak off the line. once I go wot it leaves soft and Then breaths a good gulp of air in and its like nitrous hits it. air fuel on that pass was around 12.0/1
next pass was a 11.12 @ 119.8 with a 1.59 60 ft leaned it out to 12.3/1
After that pass I pulled it back to about 12.6 but have not read the data logg on these yet. It reacted to nice with a 11.05 @ 120.2 and a 1.54 60 ft. (still have that 2step going on) and I was so wanting that 10 anything that I went back around without changeing anything. Car had a long break to cool off since a friend of mine smashed his oil pan ,the track crew spent an hour cleaning up.
After the cool down(not sure yet on iat yet) my car left with a 1.57 60 ft(still hesitateing) and ran 10.96 @ 120.69 mph.
I am moving my Iat sensor and my filter all the way out by the grill to get the hot header air away from going into the engine. That should fix that 2 step feeling and ram some air in at the same time.
Good job... :cheers: ...post some logs.
Good work mate. Great to see it is all coming together.
Joe they are on my laptop and I need a new charger for it. I broke the wires to my charger last night.
I have to reroute my cold air intake to the grill. Its breathing some hot air in but i do believe there is something else going on with it. That bog off the line is on every pass and then it comes alive.
Here is a video a friend took though of the 11.05 pass
http://s273.photobucket.com/albums/j...2009020525.flv