Tuning VE Questions??
I read a great paper from LS1Tech by .Silverhawk_02TA
They copied the LTFT Map to the VE table using HPTuner.
"Now, save the log, and similar to before, we are going to take our STFT histogram and “Paste Special – Add) it into the VE table. "
Can FlashScan do the same thing? I tried and could not get it to work.
Also how do you disable PE mode?
Thanks,
Bill
Below is the write up.
How to Tune Your VE Table Using Fuel Trims
Part I
Before we begin, it’s a useful thing to understand exactly how LTFT’s and STFT’s work, and their relationship to the MAF and VE table. You can skip this part if you so desire, and go straight to the step by step how-to. However, I urge you to read through this info, as understanding the how’s and why’s will help you better understand the PCM’s operation, and thus aid in you tuning.
First off, let’s define these terms:
STFT- Short Term Fuel Trim
LTFT- Long Term Fuel Trim
VE- Volumetric Efficiency
MAP- Manifold Absolute Pressure
MAF- Mass Air Flow Meter
SD- Speed Density
AFR- Air Fuel Ratio
PE- Performance Enrichment
Let’s start with VE, and the VE table. Volumetric efficiency for our application can be considered, in the simplest of terms, as the amount of air entering the engine. As far as it is used in the VE table, it is a calculation, rather than and actual measured value. This data is entered into the VE table, which is plotted on a MAP vs. RPM basis. VE values are used by the PCM to set the fueling of the engine, in conjunction with the MAF, or solely if we are in SD mode. These values are of course calculated by GM’s engineers to be correct using the stock components the come with our vehicles.
From the factory, our vehicles (for the most part) come equipped with a MAF, which is a device that measures the actual airmass entering the engine. So, if we have a MAF, why do we have a VE table at all? One thing the VE table does is provide a reference airmass calculation. The PCM uses this reference to compare the measured airmass supplied by the MAF to so it can ensure that the MAF is operating correctly. Also, while the MAF is accurate at measuring airmass that is entering the engine at a constant rate, it is not at all accurate when the airmass is changing, like during throttle transitions during normal driving. At these times, the PCM blends the MAF signal with the airmass calculations derived by the VE table to provide accurate fueling. So, above 4000 RPM, or during steady MAP conditions, fueling is set using the airmass measurement provided by the MAF alone. Under 4000 RPM, during unsteady MAP conditions, fueling is set using the airmass measurement provided by the MAF, modified by the airmass calculation derived from the VE table. Understanding the close interaction between the MAF and VE table, it is easy to see why it is important that they both are correct and working together. Unless of course we remove the MAF from the picture completely. More on this later.
During closed loop operation, the PCM attempts to maintain the stoichiometric AFR (usually set to 14.7:1, but may be changed in the PCM) using the airmass measurement/calculation and the readings provided by the front 2 O2 sensors. The PCM does this by utilizing STFT’s, which are moment by moment corrections applied to the fueling based on the signals provided by the O2’s, measured in +/- % of correction. So, an STFT of -7 would be a subtractive correction of 7%. Now, the stoichiometric AFR that the PCM strives to maintain is not steady, but more of an average. The O2’s will detect a rich condition, which will cause the STFT’s to go negative and subtract fuel. At this point, the O2’s will start detecting a lean condition, which in turn causes the STFT’s to go positive and add fuel. This is why a charted O2 signal looks like a sine wave. The average of these rich/lean swings in the stoichiometric AFR.
And now we come to LTFT’s. LTFT’s are the cumulative result of the average of STFT’s over time. They are in place to compensate for changes in environment, and also for variances in sensor signals, which change over time. LTFT’s start at zero. When the average STFT for a given cell (LTFT’s and STFT’s are plotted along the same MAP vs. RPM basis as the VE table) varies by more than 10% from the LTFT for more than 10 seconds, the LTFT is changed to compensate.
The problem that arises is that as we change the components in our engines (headers, camshafts, cylinder heads, air lids, etc.) we change the airflow characteristics of our engines. An engine is, after all, nothing more than a sophisticated air pump. As our airflow characteristics change, the VE for a given MAP vs. RPM value changes as well, rendering our factory calculated VE tables inaccurate to a greater or lesser extent. Luckily, our PCM’s have a built in means to compensate for these inaccuracies in STFT’s and LTFT’s.
Now you may ask, what is the purpose of correcting the MAF and/or VE table if the PCM possesses such an adaptable method to account for these errors? Perhaps the single largest reason has to do with PE and WOT performance. At WOT, the PCM falls out of closed loop operation. When this happens, it relies on the PE table to set the AFR. The PE table is broken down by RPM, with the values being divisors of the stoichiometric AFR. Hence, if the stoichiometric AFR is the default value of 14.7, a divisor of 1.225 would equate to an AFR of 12:1. This is referred to as the “commanded” AFR. This AFR is not monitored and adjusted by the O2’s.
The modifications we perform on our engines are mostly designed to increase airflow. With a greater airmass entering the engine than is accounted for by the MAF and/or VE table, the AFR becomes lean, as the fueling is set for a lesser airmass. To compensate, the STFT’s, and eventually the LTFT’s, swing positive to a greater or lesser extent, depending on the modifications. Now, if the PCM has learned LTFT’s greater than +5, it will add fuel on top of what is commanded by the PE table, resulting in an AFR different from that commanded by the PE table. This is of course less than optimal, as later we will be using the PE table to set our WOT AFR exactly where we want it. We must also realize that the commanded AFR set by the PE table divisor is only accurate when the MAF/VE table is correct. If the MAF/VE table reflects a lesser airmass than that which is actually entering the engine, then the actual AFR at WOT will be leaner than the commanded AFR, possibly lean enough to cause major damage to the engine. This is very important to remember. It is possible to manipulate the commanded AFR divisors in the PE table until the desired actual AFR is achieved, but the “correct” way, the way which I will outline in this how-to, is to correct the MAF/VE table so that the commanded AFR equals the actual AFR at WOT.
Part II
So, where do we start in terms of correcting our MAF/VE table to reflect our mods? First, we will simulate a MAF failure. When the MAF fails, the PCM falls into SD mode, where instead of using the measured airmass provided by the MAF, it uses a calculated airmass value derived from the VE table. This enables us to tune our VE table without interference from the MAF. The PCM will also default to the Low Octane Spark Table as an added safety measure. To simulate this MAF failure, we will do the following:
1.) Disconnect the MAF from the engine’s wiring harness.
2.) Using HPTuner, copy the High Octane Spark Table to the Low Octane Spark Table.
3.) Flash the PCM.
4.) Start the vehicle, and using the scanner reset your LTFT’s.
5.) Viola! You are now running in SD mode!
Note: You can also set the MAF Sensor Failure Frequency to 0 in the PCM to enable SD mode. Whichever method you choose, you will throw at least one code and set the SES light. You can disable the SES light in the PCM for the corresponding codes, or just ignore it.
Tuning With LTFT’s + STFT’s
Now comes the tedious part. We get to drive around until the LTFT’s mostly learn themselves out. The general timeframe usually mentioned is about 1 hour or 100 miles. This is a guideline more than anything else, because the longer the engine falls into a particular MAP vs. RPM cell, the quicker the corresponding LTFT will be learned. In an idle or cruise cell, for example, it will be learned very quickly. The standard we will use to determine whether or not the LTFT’s are learned is the STFT’s. When your STFT’s are all fairly near 0, your LTFT’s will be learned. You can do this driving all at once, or over a few days or a week. Make an effort as you drive to hit various RPM’s and MAP values as you can, as often as you can. It’ll help if you have the scanner operating as you drive, so you can visually see which cells you are falling into on you LTFT histogram.
When you are ready to log after your LTFT’s have been learned, start by bringing the car up to operating temp. Begin logging, and again try to hit as many cells on your histogram as possible. A 20 minute or so logging session should be sufficient. Save your log file. (Note: If all the LTFT’s in your histogram are between +/- 10, skip ahead to “Tuning With STFT’s Alone”) Now, copy your LTFT histogram, and “Paste Special – Add” it into the VE table. Do the same with you STFT histogram.
After adding these values in, take a look at the 3-D graph of the VE table. You may have large depressions or spikes corresponding to cells that you didn’t have logged LTFT data for. Hand smooth these using the 3-D graph. Just click on the peak or valley and move it up or down so that it conforms with the rest of the graph. When this is done, save the .bin file and flash the PCM with it. Here comes the tedious part again. Repeat the LTFT learning process again, and then make another log. What we are looking for now are LTFT’s between +/- 10. If the preceeding steps were followed correctly, they all should be. However, it may be necessary to repeat the (LTFT + STFT) + VE procedure again before your LTFT’s fall in line. When they do, rejoice! Because now we can move on to…
Tuning With STFT’s Alone
Now things get quite a bit easier, because we no longer have to wait for LTFT’s to be learned. First off, open the editor and go to General>Fuel Control>Open & Closed Loop, and set LTFT enable to “OFF”. Then go to General>Fuel Control>Power Enrich. Look for the MAP field under the “PE Enable” heading, and set it to 640 (This will disable PE mode. We don’t want it interfering with our next step). Save these parameters and load it onto your PCM. Now we get to do some more logging.
Method here becomes very important. We want to start out by getting the vehicle up to operating temp, like always. Try to find a long stretch of road with no stops. A high speed limit here also helps. Start the STFT histogram as you’ll be referencing it quite a lot while you’re driving, and start logging. Get into third gear (for an M6, if you have an A4 you may have to experiment to find a gear that works well for the following procedure), and level the RPM’s off at about 1200. Now, we’re going to “pulse” the gas pedal, constantly going from 0 – 100% TPS (this is why we disable PE mode) at a steady rate. This can be a bit tricky, but the object is to hit every MAP cell for the 1200 RPM column from off-idle to WOT. You should be able to cycle through the column a couple times before the engine falls into the 1600 RPM column. Continue doing this for all subsequent RPM columns. If you plan on using your MAF again after tuning your VE table, you can stop if you want at the 4000 RPM column. If you plan on going straight SD, keep on truckin! The higher the RPM’s get, the tougher it’s going to be to get all the cells in the column. It’s okay if you miss a few (even a bunch) because we can always hand smooth later. Also, the most important cells to hit in these higher RPM’s are going to be the WOT ones. The more correct we can get them, the easier it is going to be to tune the PE table (if going SD), so try to hit them all if you can.
If you are good, you can hit all these cells often enough to get a good average STFT. You can tell that you’ve found the average STFT for a cell (or close to it) when you “pulse” through it and the STFT value doesn’t change anymore. Just keep at it until you get a good average of as many cells as you can. Now, save the log, and similar to before, we are going to take our STFT histogram and “Paste Special – Add) it into the VE table. Save the table, flash the PCM, and do another logging run using the same procedures as in the previous paragraph. Now your STFT’s should be very close to zero. This time, we’re going to “Paste Special – Multiply by % - Half” the STFT histogram into the VE table. Hand smooth any large spikes or dips. Save the table and flash the PCM.
Now we can log again just to verify where we’re at. If you are happy with your STFT’s (they should all be within a couple points of 0 now) down low, you can skip spinning up to the high RPM’s if you want. You could also keep logging and applying your STFT’s into the VE table as many times as you want, but you probably won’t be able to improve your VE table very much more (assuming your STFT’s are very close to 0. If not, wash, rinse and repeat.). Now we get to spend some quality time with the 3-D graph of the VE table. At this point, there should not be any big peaks or valleys. Hand smooth only for cells in which you didn’t log any data, so that they correspond with the logged cells around them. Personally, I don’t use polynomial smoothing at all. It modifies the VE values enough so that it throws my STFT’s off significantly. That and my VE table is pretty smooth already between the application of my STFT’s and hand smoothing. Generally, it’s okay if the VE table is a little choppy, and long as it doesn’t look like the Rockies and you aren’t experiencing any burst knock.
Now that your VE table is so pretty and correct, go back and re-enable PE mode by setting the PE Enable MAP value back to 15. Also switch LTFT Enable back to “ON”. Flash your PCM. Drive around for a few days or a week, and re-log to see where your LTFT’s ended up. They should all now be within the +/- 5 standard. Congratulations!
Reply With Quote
Originally Posted by SSbaby
