• Got the Contributing Memberships stuff finally worked out and made up a thread as a sort of "How-To" to help people figure out how to participate. So if you need help figuring it out, here's the thread you need to take a look at -> http://www.corvetteflorida.com/forums/showthread.php?t=3581 Thank you, everyone! Rich Z.

Carchip data log for 07/30/06

Rich Z

Internet Sanitation Engineer
Staff member
OK, I didn't do much but press on the go pedal a little bit, but I'm concerned about the amount of advance I am seeing in the 2,000 to 3,000 rpm range. Anyone got any comments about this log:

carchip002.gif
 
Somewhere you said you changed your inj.slope on your Predator.I am new to tunning (Just bought a Predator) and would like to know more about the inj.slope.I have changed fan sittings and shift firmness so far.Thanks for any help.
 
From what I could gather, you need to monitor the real time data on the Predator looking at your long term fuel trims. The target is to be as close to 0.00 as possible across the rpm range. In my case, it appeared I was running around +4.00 in a lot of instances, so what I did was to set the injector slope to -4.00 on both banks. That should have brought the averages closer to what is considered optimum.

Of course, I am a newbie at this as well, so take this with a grain of salt, subject to verification........... :D
 
I will check it out and see where I am at.I put a Vararam on and have a surge problem at slow speeds I am trying to get ironed out now.
Thanks
 
A little while back, I was cruising through a bunch of sites looking for notes on programming the predator. As I found things of interest, I just cut and pasted the text into a text file for review later. I don't have the original sources, and hope I don't step on anyone's toes, but I will post that entire file here in case it will be of any help to anyone else.

I spent the last two hours tuning. I started with the factory HP tune and modified. My LTFTs were about +12% to +14%. I dialed those in so that they were in the range -2 to 0. I then started advancing timing until I got knock retard (as indicated via the real time data) and I backed it off a bit from there. I ended up using 0 for the low RPM range, +2 for the mid range, and +5 for the high range. The only other thing I did was go +10% on the fans and made the WOT fuel mixture just a hint leaner (I think -2%) based on the fact that my WOT O2 values were .91-.92 to start which in my experience is just a hint rich. After my -2% adjustment to fuel mixture, they were down to about .89 and .90 for sensor 1 & 2 which in past experience is what my C5 liked. My other mods are Vortex Rammer intake (possibly what caused the LTFTs to be off in the first place) and Corsa Touring exhaust. Now, after the tuning, the car absolutely stands up on its rear wheels on takeoff. I don't detect any lag when shifting and it's difficult to keep the tires planted now when taking off in 1st. Big difference! I made some dyno runs before the tune so I'll make some after driving the car a week or so in similar weather to make sure everything settled down and it's a good comparison.
----------------
My trims started out at +12 to +14 so I dialed in -12 on the injector slope and that was all I needed: that one iteration pulled them down to near zero. I then noticed that just holding a steady speed (like 40-55 MPH) at very little throttle tends to show trims that are about 2 points off from other trims where the gas is pressed down a little more. As you press down on the gas, you enter new cells based on both throttle and RPM. I found the low RPM low throttle cells to be a little skewed from the rest of the cells so when you do your trims, be sure to go up a hill where you can hold the throttle at say 1/4 throttle for a little while and see what those trims are. I set mine to favor the trims with a little more throttle since they seemed more consistent. Again, I don't think it makes too much difference since they were only off by about 2-3 points.

As to your final question, I'm seeing anywhere between 1 and 2 points difference between the left and right banks. For examlpe, I might see about 0 in one bank and -1.5 in the other. Pretty typical from what I've seen.
---------------
Been messing with the factory HP tune again tonight. We really need a synopsis of exactly what each of the "tunable" parameters are, what they do, and what to look for in the real time data when you change them. The fan, well, that just acts strange no matter what you do. Then there's the question of WOT fuel mixture trimming. I had been setting mine so that the o2 sensors read .89 to .90 which is right where my C5 had the right A/F ratio. That'll do until I can get to the dyno up the road and look at the real A/F ratio. I found an interesting parameter in the full parameter list: "Air fuel ratio". It was at 14.75 at less than WOT so it looked good at first. Then when I went to WOT it went UP to about 16.7 to 17.0! If you make the fuel mixture leaner, the number goes down which doesn't make sense. Can anyone explain what the "Air fuel ratio" parameter is and why it goes down with a leaner mixture and up with a richer mixture?
---------

Most all modern day vehicles are equipped with a MAF (Mass Air Flow) system. The MAF is what tells the pcm the incoming aircharge, and allows the pcm to calculate load values (not weight, but as in "aircharge") for fueling and spark calculations, amongst other areas. The MAF allows for more precise tuning, and also allows for changes in climate and elevation conditions. This is how you're able to get tuned say at sea level, then drive to the high heavens of Colorado and your tune still be dead on accurate as the day it left the dyno.

In Closed Loop the system is always trying to obtain a stoich ratio, which is 14.7 a/f. It uses the adaptive strategy system to calculate how far off from stoich the fueling is, and this is why you see the Long Term fuel trims correcting to keep the Short Term Fuel Trims as close to 0% as possible. In Open Loop there isn't any adaptive strategy, therefore the system ignores the O2 sensor input, and runs off of the VE and PE calculations. Because of the MAF system, both Closed Loop and Open Loop should always stay consistent.

However, if Long Term Fuel Trims are positive during Closed Loop, and you go WOT, they generally will stay positive during Open Loop. This can affect PE by allowing additional fuel on top of the PE calculations. Now, when you adjust the Injector Slopes you are either leaning or enrichening the ENTIRE fuel curve, in both Closed Loop and Open Loop. This is why you must lean out the PE table in order compensate for such.

--------------

Now, you are saying for my +12 LTFTs, I should go -12 on the injector slope (fine here, that zeros LTFTs nicely) and go +12 (rich) on the power enrichment. RWTD is saying the opposite: still go -12 on the injector slope but go -12 (lean) on the power enrichment, which is counterintuitive to me.

To me, logic dictates that if you have +12 LTFTs, the car is adding 12% more fuel to richen the mixture and get 14.7:1 A/F ratio at part throttle. No problem there. But if the LTFTs are +12, then the WOT mixture (power enrichment) will be +12%. Now you change your injector slope so that your LTFTs are between -2 and 0. Now no fuel is being added at WOT.

Here's the $64,000 question and what no one as yet has answered... and what MUST be answered to get this right: does the default WOT enrichment BEFORE any LTFT values are added work off the injector slope? The answer to this question is critical in understanding how to set the power enrichment adjustment! There are two possibilities:

(1) Injector slope and power enrichment are mutually exclusive. The injector slope only applies at part throttle (closed loop) and injector slope and any changes to injector slope are ignored at WOT. In this case, if you go -12 on injector slope to lean things out and zero LTFTs, you'd have to go +12 (richer) on the power enrichment to compensate for the fact that WOT doesn't know about the change you made to injector slope.

(2) At WOT, the computer starts with the current injector slope and bases the WOT fuel mixture on a point on that curve (the rightmost WOT point). In this case, WOT will be taking the modified (-12) slope and calculating injector pulsewidth based on that curve/line. In this case nothing should be done to the power enrichment: it should be left at zero because WOT will already be calculated from the -12 injector slope.

I can't logically see any case where the RWTD/Diablo recommendation works. If you have to go -12 to zero out +12 LTFTs, it doesn't make sense that you'd have to go -12 on the power enrichment table. In theory (if case 1 above is true), if anything, you should have to go +12 (richer) on the power enrichment because in zeroing out your positive LTFTs, WOT will end up leaner (by 12%).

The relationship (1 or 2 above) needs to be clarified else we all risk making the wrong settings.

Edit: I still think the best method of adjusting WOT mixture is what I've always done in the past. In the absence of a wideband O2 which is ideal, it's probably best to get your LTFTs at or just under zero and then adjust the WOT (power enrichment) so that both pre-cat O2's average about .90 volts at WOT. That is, once you've zeroed your LTFT's and you've driven around a bit, make some WOT runs and watch O2 bank 1 sensor 1 and O2 bank 2 sensor 1 and adjust the power enrichment so that those two O2 sensor values hover around .90 for both sensors at WOT. Also while at WOT, confirm that your LTFTs show zero at WOT. That's what I do in the absence of a dyno and wideband and I think I'll stick to that... but I'd still like to know how the innards work/relate.

------------

Exactly wrong when you said:To me, logic dictates that if you have +12 LTFTs, the car is adding 12% more fuel to richen the mixture and get 14.7:1 A/F ratio at part throttle.

Untill you get this pooint you wont get the rest:
The LTFT being +12 are doing nothing to part throttle fueling. I keep saying it and you don't get it.

Closed loop = part throttle is set at 14.7 no matter what your LTFT's (unless they are above =+26%). I will say it again differently: The car runs 14.7 and is not putting in +12% fuel in at part throttle. Again differently: The part throttle a/f ratio is not adjusted by LTFT's. the car is not adding 12% fuel at part throttle. It is at 14.7 no matter what because the car uses the O2 sensors to measure the exhaust and adds or leans the fuel to be 14.7 no matter what you do to LTFT's.

What are the LTFT's then? They only measure how much fuel needs to be added when you go WOT.
SO:
If you take away the +12 on the LTFT's then at WOT there will be 12% less fuel dumped in the a/f so you go add that back in from power enrichemnt---PE.

THERE IS NO SUCH THING AS PART THROTTLE FUEL ADJUSTMENT. IT IS PART THROTTLE TUNING.

+20% LTFT's =14.7 a/f ratio
-11% LTFT's =14.7 a/f ratio
0% LTFT's =14.7 a/f ratio

However in each of the above examples the WOT a/f ratio changes.

Now to answer:

Here's the $64,000 question and what no one as yet has answered... and what MUST be answered to get this right: does the default WOT enrichment BEFORE any LTFT values are added work off the injector slope?

Yes but they do not alter the part throttle fueling they only lessen the fuel dumped at WOT. Changing the Injector values will do nothing to part throttle since the car will read a deprature from 14.7 and adjust.

.....so..... without the added fuel from the LTFT's since you zero them out you need to richen the mixture by the adder you deleted from the closed loop (LTFT) adder that you defined as zero. At least you now recognize that there are 2 things affecting WOT

Power enrichment and the fuel added fromt he positive LTFT's....so if you add nothing by zeroing out the LTFT's what must you do to the fuel mixture since it missing that adder? You richen it back up. You go to PE (power enrichment and add it).

In this paragraph you again disregarded everything i said about your terminology:

(1) Injector slope and power enrichment are mutually exclusive. The injector slope only applies at part throttle (closed loop) and injector slope and any changes to injector slope are ignored at WOT. In this case, if you go -12 on injector slope to lean things out and zero LTFTs, you'd have to go +12 (richer) on the power enrichment to compensate for the fact that WOT doesn't know about the change you made to injector slope.

You said that if you go -12 on injector slope to "lean things out". You aren't leaning out LTFT's you are adjusting them.

You also have lean and rich reversed with LTFT's. Here, this will confuse the crap out of you.....if you go above +26% LTFT's you will get a lean condition for closed loop because you exceeded the computer's ability to compensate for stoic. Positive would be lean if it didnt compensate. When it goes WOT the O2's are disconnected and the car needs more fuel. Positive LTFT's are lean but the car corrects at closed loop on its own but at WOT you have to add fuel so it does by locking at positive for LTFT's and dumps that amount of fuel. If they lock zero it dumps no fuel at WOT. If it weren't for the O2's in closed loop +LTFT woulld be lean or above 14.7 and -LTFT's would be rich in closed loop. The car adjusts automatically using the O2's for measurement with in the range of -20 to +26%. outside this range you get codes which is why some aircleaners trigger a check engine light.

It appears that your confusion comes from the following:

You think adjusting LTFT's adjust the closed loop A/F ratio-not true

You think +LTFT's are a rich condition in closed loop...it is lean but tsince the O2's measure that it is still stoic...14.7.

You think -LTFT's are lean but they are rich but in closed loop the PCM corrects and it is still 14.7....however in this case since you are negative there is no fuel added to the WOT mixture. This is an adder only and doesnt pull fuel if you go negative.

Since the polarity of the LTFT's is at issue with you, it is easy to understand why you think fuel needs are in reverse. The car reads lean at closed loop and self adjusts...it measures the LTFT's as lean (+20 or so) and says to the PCM to command +20% more at WOT since it cant do it....open throttle means no O2's.
--------------

I know you know what you are doing here, but I too have a lot of experience with this stuff and have been fooling with this stuff since 1983. Now, maybe my observations and conclusions over the last 20+ years is wrong, but I respectfully disagree with the above!

Let's start from a clean slate where the computer is cleared and first started up. There are tables in the PCM that relate things like MAF (airflow) readings to injector pulsewidth (mixture) so that for a given amount of air entering the engine, the injectors are pulsed on/off at a frequency that will produce the proper mixture for that amount of airflow. These are the default or "base" tables.

The computer starts by looking up the default amount of fuel and spitting that amount of fuel into the engine. It then monitors the O2 readings to see how well the default worked. If the O2 voltage looks too rich, the short term fuel trims (STFT) are immediately changed to go negative. If the O2 voltage looks too lean, the STFTs go positive. On the next cycle, the computer looks at the STFTs and modifies the default table based on the last STFTs. In a rich condition where STFTs show +5 (more fuel needed), the base table is used and then +5% fuel is added to that. This cycle repeats at a relatively fast pace and the feedback ensures that you are always getting stoich mixture.

Now, the long term fuel trims (LTFT) are nothing more than a long term average of STFTs. STFTs tend to bounce around as the computer adjusts A/F ratio and while the STFTs bounce around, the computer is averaging them to determine the average amount of fuel to be added/subtracted. This average goes into the LTFTs. Your STFTs may show +4, +5, +2, +6, +8 and the computer averages those and sets your LTFTs to +5. The LTFT +5 is put into the cell corresponding to the current RPM and throttle position.

Now, it's been my experience that the next time the computer hits that cell, it starts with the default table plus the LTFT. It then spits that amount of fuel and the STFTs then show the mixture change needed from base+LTFT. This is why when you first start the car after a "reboot" your STFTs may be very high one way or the other. As your LTFTs are modified and used along with the base table, your STFT start to come down and hover around zero +/-.

You seem to be under the assumption that STFTs are all that are used at part throttle but I believe that to be incorrect. The LTFTs are not just used at WOT. If they were, why have LFTFs in every throttle/RPM cell? You'd only need one LTFT for WOT. Your mixture would be too unstable if it depended on STFTs alone.

Now, I'm ready to be proven wrong as a lot of my experience has been in "reverse engineering" PCM data, but I don't think I am. In my experience, your fuel mixture at part throttle is determined by base+LTFT+STFT. At WOT, it is base+LTFT.
-----------------

Tuning with the Predator: synopsis of my methods

I feel like I've refined my methods for using the Predator so I though I'd post my own personal workflow here:

First, as I found out, communication or other errors can cause partial loads of a tune. Your car can be rendered inoperable due to a load failing partway through, so I'd recommend doing the tuning at home. One of my tunes stopped at 28% load and the car wouldn't start. At this time, it takes a call to DiabloSport for them to give you a password that you enter into the Predator so that you can force-load the original backup back into the car. Since a password is needed to recover in the event of a load failure, I now only tune at home (not in parking lots on my test routes) and only during DiabloSport work hours of 9:00am to 5:30pm eastern time. Not that a load failure will occur to you, but you should plan for it just in case.

TUNING: PHASE I (Finding LTFTs)

I prefer to start with the Factory HP Tune just because I want to start at baseline and try to get the best tune from there. I load the Factory HP Tune and then go to "Real Time Data" and choose the option to "Reset Live Tables". Then I make sure to warm the car up fully and after it is warmed up completely, I go for a 10 minute drive while monitoring the following real time parameters from the "Diagnostics" menu, "Real Time Data", "Gas General Short", and "Select from list":

Long Term Fuel Trim Bank A
Long Term Fuel Trim Bank B
Oxygen Sensor 1 Bank 1
Oxygen Sensor 1 Bank 2
Total Spark Retard

Go for a drive and make sure to use part throttle a lot. Do things like go up hills (if possible) while holding your foot at say 20% throttle. As you hold the same throttle position and the same RPM, the LTFT (long term fuel trims) will update over time. Sometimes when I can't find a hill I've even held my foot lightly on the brake with my foot on the gas just to hold the same RPM and the same throttle, making sure not to hold it too long (maybe 20 seconds) so the brakes don't overheat. It's not good for the brakes but I use it to "cheat" sometimes in order to determine the LTFTs. Anyway, with your foot still at the 20% throttle position, note the LTFTs once it looks like they've stabilized. The LTFT from the left and right (A and B) banks may be different but usually not by more than a few percent. Let's say one LTFT shows about +10% on average and the other shows about +12% on average. I then make a note that the LTFTs are about +11% meaning that the computer is having to add 11% more fuel than expected.

Note that LTFTs when at idle and very little throttle can be misleading. Use LTFTs at part throttle, say at least 10% throttle up to 20% throttle.

TUNING: PHASE II (Zeroing LTFTs)

Now that we've found the LTFTs, +11% in this example, we need to modify the injector slope to get the LTFTs as close to zero (or slightly negative) as possible. We do this by going to the "Performance Tune" parameter, selecting "Factory HP Tune" (assuming that's the tune you are using), "Modify Parameters". Just scroll down to injector slope, enter the opposite of your LTFTs (my LTFTs were +11% so I entered -11% in the injector slope). Press the left arrow until it asks you "Install tune now". Answer Yes.

After the tune loads, go back out and repeat phase I and II until you notice that your LTFTs are hovering between 0% and -2% at the preferred 10% to 20% throttle position. Once you've done that, you've effectively zeroed your LTFTs. The purpose of this procedure is to get the "baseline" of your computer as close to correct as possible so that it isn't always "hunting" for the right fuel mixture, especially after a reset, battery replacement, etc.

TUNING: PHASE III (WOT fuel mixture)

Zeroing LTFTs has an effect on your fuel mixture at WOT (wide open throttle). Once you've zeroed your LTFTs, add the same five parameters above from the real time data and go for another drive. Just make sure you've driven around a bit (10 minutes or so) after your last tune-load so that things have stabilized reasonably well. Make some WOT runs. It's better to have a passenger so you don't have to make WOT runs while reading the parameters but the Predator does do data logging so that you can make WOT runs and then just play them back later. What we are looking for are O2 readings in the .90 volt range at WOT. Make sure your foot is to the floor and note the O2 voltages while RPM increases through say 4,000 RPM. Do it in something like second or third gear so the engine isn't blazing up through the RPM's fast. It takes a second or two at WOT for the O2's to stabilize. As you approach 3,000 RPM, the O2 readings should stabilize. If you see .86 to .87, you need to ADD fuel as you are running lean. If you see .93 to .94 you are running rich and you need to subtract fuel.

TUNING: PHASE IV (Modifying WOT fuel mixture)

Use the "power enrichment" variable in the custom tune to adjust your fuel mixture until your WOT O2 readings are about .90 volts on average between the two sensors (again they can differ a bit). I found it pretty easy to get them right at .90 at WOT. Some people will argue that you must have a wideband O2 sensor to get accurate enough readings to do this but the .90 should get you close and is certainly better than nothing (or guessing). For me, I found that going -11% on the injector slope forced me to go the opposite direction on WOT (power enrichment) mixture to +11% (richer). This may be a good rule of thumb but I still prefer to read the WOT O2 sensor voltage. Of course, if your car is going the opposite way (compared to mine) and you showed -5% LTFT's, you'd go +5% on the injector slope and -5% on the power enrichment: you just negate the signs on my example above.

TUNING: PHASE V (Timing)

The last parameter to tune once we have fuel mixture close to correct is timing. For me, I found that I had to keep the low RPM timing setting at zero. Anything over 0 produced knock at low RPM for me. For the mid and high range, I went +4 advance and +6 advance respectively and still had no knock (per the spark knock retard reading). I set these values and then went out and monitored the total spark knock at WOT to make sure no spark knock was being detected. Theoretically you could keep advancing timing until you start seeing non zero values pop up in the total spark retard real time data, but I wanted to be a bit conservative. I might have been able to go further than +4 and +6 on the timing because I got zero knock doing that but I didn't want to push it. Call me chicken... for now. I'm not 100% confident that the knock sensors (or my ears) are good enough to pick up light knock at higher RPM so I chose to be a bit conservative.

TUNING: PHASE VI (Other stuff)

The other things I did were to change the fan power % to +10% and turn CAGS off (just to eliminate the 1-4 shift light). I can't say I'm really convinced that the fan parameter even does anything as I've never really seen a difference in coolant temperature. I think DiabloSport needs to work on the fan parameter to give us something more predictable. YMMV.

Finally, let me say that the above is what worked for me and my car with the Factory HP Tune. The values will most likely be different when starting from the Diablo Tune but the methodology is the same. My car is the base car with MN6, Vortex Rammer intake and Corsa Touring Exhaust.

Hope this helps someone and I reserve the right to refine this post as needed or as questions are asked... or as I'm corrected for any mistakes.

-------------------
Ok, I am really confused. I initially set my injector slope by reading the LTFT's in 5th gear (said 3rd yesterday...error) going 55mph - this was RWTD's instructions. I was +9 so I backed off the Injector slope by -9.

Now when I do a WOT run, my LTFT's go to a rock solid -6.25% and when I hold a steady rpm at partial throttle, they stay between 0 and -2.34% with -1.56% seeming to be it's most common. At WOT the O2 sensors are about 830mV and I get 0 knock counts. From what I hear you guys saying, I should still have the 0 to -2 LTFT at WOT, but I don't. I did initially change my PE table by LEANING it out 9 per the RWTD instructions (this is opposite of what you guys are doing) to coincide with the injector slope change.

The above data is with the car warmed up and 6 separate runs, all giving the same results...low LTFT (-6.25%) but 0 knock at any rpm. BTW the car feels like it is absolutely running like a beast.

--------------
Adjusting Tunes

LTFT's (Long Term Fuel Trims):

You should always adjust the tune to get the LTFT's (both Bank 1 & Bank 2) within ideal range. Ideal range is as close to 0% as possible, with a slight tilt into the negative territory (0% to -2%). This ensures that the LTFT's will go to 0% at WOT (Wide Open Throttle) and not affect your PE fueling (Power Enrichment, aka WOT fueling). This can easily be done by modifying the Injector Slope parameter within your Predator's end-user options, and then the WOT PE fueling to correct for the new global change in overall fueling resulting from the change in the Injector Slope.

Take the vehicle for a drive and put the car in OD (4th for the A4, 5th for the M6) and test the LTFT's at a steady state 55mph speed (RWTD suggests not to use the cruise control, but to instead hold the throttle as steady as possible). Get an average of the LTFT's and then take and change your Injector Slope by the % of the fuel trims.

Once you get the average long term fuel trim value:

Positive LTFT's - For a positive value you subtract that value in percent in the Predator Injector Slope end-user option. It is also recommended to lean both fuel tables by the same percentage.
Negative LTFT's - For a negative value you add that value in percent in the Predator Injector Slope end-user option. It is also recommended to enrichen both fuel tables by that same percentage.

i.e. if you have i.e. a +6 on the LTFT's as average, then subtract in % the same amount from the Injector Slope. This will also globally add fueling everywhere, even in PE, so you generally will want to lean out the PE fueling by the same % you just added in the overall fueling.

Note: Keep in mind that when you make end-user adjustments the Predator is only uploading the calibration changes back into the pcm, therefore you will need to reset your adaptive strategy after a change to the fuel trims. You will need to do this by removing your negative battery cable, turning on your headlights, hitting the brake (doing these last two things helps to remove any residual power in the system), then hooking the negative cable back up, and then going out and retesting.

----------------
 
Z06 Rocket said:
Whew, that was a mouthful Rich!:hehehe:

Yeah, and the Predator is considered as kid's stuff for tuning. The real heavy duty stuff is with programs like HPtuners and EFI Live. I just don't want to blow up my Z while I am climbing the learning curve, however....

Man in the old days, tuning your car meant having the plugs, points, condensor and air filter replaced. If things had gotten bad since your last tuneup, maybe a new distributor cap and ignition wires. The WORST that could happen is it might run worse then when you took it in if the tech (mechanic in those days) goofed something up or the points weren't set correctly. NOW, a tune means hooking up a computer and you hope the keyboard jockey knows what he is doing. Worst that could happen? Well melted or broken pistons, for one......... :ack2:

One thing I can't figure out, however. If a person can look at logged data and make a determination of an OPTIMUM result based on values presented in the data, why can't the PCM program itself be smart enough to be self adjusting to attain optimum values? It's only a program. If it is programmed with target values, then is SHOULD be able to adjust the same parameters a human being can via the keyboard in order to reach those optimum results. Why DOESN'T the PCM ALWAYS have the optimum air/fuel ratio?

Can you tell I used to be a programmer by trade? :D
 
Back
Top