Correct Tuning Procedures
Hi, I'm only a newbie, and there are so many parameters to change but in what order????
I recently came across the following tuning procedure which looks pretty good (based on some of the info on this forum), so i just wanted to get some opinions before i used. it It came from: http://ls1edit.slowcar.net/
I also found an awesome log file analyser, so much better than excel, get it at: http://www.slowcar.net/analyzer/
Tuning Guide
This document will attempt to layout a step by step process to guide you in the tuning of your ls1 powered vehicle
Requirements
In order to properly tune your car there are a few items that you need to have available
LS1 Edit Software + Cable
Laptop computer (pentium 133 or greater - p233+ suggested)
Logging software - EASE, EFILive, Autotap, etc.
Outline
There are many different ways one can tune a car, but below I have outlined a methodical process which should work well
1: Set absolute parameters
2: Start car, verify
3: Tune idle
4: Tune part throttle
5: Tune WOT
1: Absolute parameters
Okay, at this point you have connected your laptop to your car, downloaded your base program with LS1 Edit (you *always* should start with your program), and are ready to start. The number one thing to remember when editing your pcm is to make only one or a few small changes at a time.
First we are going to start with what I call "absolute" parameters. This are values that can be known and set with total objectivity. These generally include gear ratio, fan turn on temps, rev limiter, etc. You should also choose a desired idle rpm and enter it initially. The main idea here is to keep the program as stock as possible - we only want to change values that have a concrete value - leave shift points, fuel, spark, and other subjective (from our point of view) values alone.
The only tricky parameter to set here will be injector constant. This is because the LS1 does not have a manifold vacuum regulated fuel pressure, so as manifold vacuum changes (and rail pressure stays constant) the pressure drop across the injector change, so the injector flow/constant itself changes. A spreadsheet to simplify the calculation is made available:
Excel spreadsheet to calculate injector constant
If you are using a vacuum referenced regulator (some supercharger setups, etc.) then you will simply fill in a constant value across the range.
2: Start car, verify
Now we will start up the car and verify that everything is working properly. Before we actually key on we should have our scanning/logging software hooked up and ready to go. I would suggest logging at a minimum the following parameters
RPM
MAF flow (g/sec or lb/min)
MAF Frequency
MAP
Spark Advance
Knock Retard
Injector Pulsewdith (left and right banks)
O2 sensors - B1S1 and B2S1
IAC Position
L-trim (left and right)
S-trim (left and right)
in addition to any parameters you have personal interest in. You should be logging all data to disk.
Now we are ready to start the car. Start it and begin logging. You should first verify everything seems mechanically sound (oil pressure, etc.). Next look at the MAF flow and/or frequency. As you blip the throttle this value should increase/change. You need to let the car run for a while to heat up and go through it's DTC tests. You want to watch out for any kind of SES light. If you get one you need to determine if it is a mechanical/electrical problem, or is simply caused by your new combination (camshaft, etc.). Once the vehicle is up to temp you should first verify that your O2 sensors are responding to changes in the throttle. If the vehicle seems to be running decently (no pinging or potentially problematic situations) you should drive it around for awhile. Monitor your s-trims, l-trims, and O2 values and ensure you do not have any great splits, exhaust leaks, etc. If your O2 sensors are old or just aren't switching fast enough (anti-freeze and RTV can both kill them easily) you should replace them before you go any further.
3: Idle
Now that we have verified that everything is in proper working order we can begin tuning. Idle is the best place to start. You previously set your desired idle rpm - you should now subjectively decide if you are happy with this rpm. If not change the value with ls1 edit and re-evaluate. Once you have found a desired idle rpm you can begin tweaking it for stability. Take note of your IAC counts. In a no load situation (neutral, no ac) you want them no lower than 30, and no higher than 50. You may need to drill a hole in your throttle body/enlarge the one that is already there to bring down the IAC values. Do this until they are acceptable. Now put a load on your car (D if automatic, and put the AC on). The values here should be no higher than 120 or so. If they are enlarge the hole.
Repeat the above process until your IAC values fall inline. If you are still not happy with the idle you can try adjusting the timing. Be carefull of adding to much timing - it can give you a great no load idle, but with any kind of load will become erratic. A "hunting" idle is a sure sign of too much timing. To adjust the timing at idle the base spark tables are the easiest place. The tables are scaled vs. rpm and g/cyl of airflow. The rpm part is evident. To calculate the g/cyl use the following formula
g/cyl = 15 * MAF(g/sec) / RPM
This formula takes massflow per unit time and converts it to massflow per cylinder. The 15 is a constant which corresponds to the characteristics or a V8 running a 4 cycle combustion cycle. The easiest thing to do is to load your log file into excel (export it as a CSV), then create a formula in excel which applies the formula above - a short mockup is provided
example of excel spreadsheet to convert from g/sec to g/cyl
4: Part Throttle
Not that we have our idle down and have verified that the car is in proper working order we can begin part throttle tuning. The first step in this is to record a long log file of driving. I would suggest a minimum of 20 minutes, and the longer the better.
Once we have our log file we can begin the data reduction. The first element we will tune will be the fuel delivery. At part throttle the computer uses the MAF meter to find out the amount of air entering the engine. It then calculates the amount of fuel required to maintain a 14.7:1 A/F ratio. It injects this fuel by controlling the injector pulsewidth. The O2 sensors, which are very accurate at 14.7:1 (o2 response curve information can be seen here) provide feedback to the computer and let it know how close it is to the goal. The computer uses this feedback to tweak the fueling of the motor to achieve a proper 14.7:1 a/f ratio. This "tweaking" is exhibited to us through the s-trim and l-trim parameters. These values indicate how the computer is correcting. Since injector flow and pulsewdith are known with great precision, and we have no control over the internal algorithms we will assume that any inaccuracy (which is exhibited by nonzero trim percentages) is a result of an incorrect MAF transfer function.
In tuning part throttle we will tweak the MAF transfer function according to the l-trim values we logged. There are 2 ways of doing this, the simplest is to simply view your l-trim values/average them, and scale the entire MAF table by a percentage which will give you l-trims of -2 to -8. L-trims are in units of percent so this is easy. If our average l-trims are around +5 and we want to shoot for say, -4, then we would just multiply the entire table by 109% (or an increase of (+5 - (-4))=9 percent). Likewise if we were at -10 and wanted to shoot for negative 4 we would decrease the entire table by 6 percent, or multiply by 94%.
Once you have done this you need to repeat the logging process above an check your new l-trim values. We want to avoid positive l-trim values since they will be applied at WOT and will lead to inconsistent fueling. Negative values are okay, though we shouldn't go too far out of whack.
The second option is a little more complicated. It uses the same premise above, but instead of taking the average value it applies a localized l-trim correction to each point of the MAF tranfer function and derives a new curve. An example of this is made available:
Excel spreadsheet of above tuning method
This method is not for everyone, but in certain instances is very usefull.
After repeating the above method until L-trims fall in line your fueling should be done. Now we can address spark. Spark advance is a rather difficult item to tune directly yourself, but I have developed the following method to help. This method relies on a properly functioning knock sensor without any desensitization.
Assuming you have a NA car which does not ping with the stock timing advance: Take the entire timing table and increase it by 5 degrees. Now start driving the car while you are logging. You want to try and emulate every possible driving condition you can think of. If you hear pinging at any point of course back out, and if you hear pinging constantly reduce timing across the board.
When you are done logging you need to export your data to a CSV and open in it excel. Here we will make another pivot table. You need to create a column with g/cyl (as shown how above), and you need a column containing spark retard and rpm also. You will use these three items to make a pivot table. Scale the table with g/cyl on the x axis and rpm on the y axis. Put spark retard in the middle and set it's mode to average. You should group the axis along the same lines as they are grouped in the PCM.
We now have a table of the average spark retard taken out at each timing point. Now go to your table in the pcm and subtract 75% of this value from the actual spark advance at each point where spark retard occured.
You should then re-logg the car. Repeat the procedure until you do not get any spark retard. Your timing curve should now be tuned.
Finally, if you have an automatic this is when you will start tuning shift pressure, points, and TCC. I will put in hints on this at a later time
5: WOT tuning
This is what you have been waiting for. The first thing to do is make a quick WOT pass in a low gear (so you have a low load) and check both your o2's and your knock retard. O2's are *not* accurate nor precise at this a/f ratio, but can still be used for a ballpark estimate. If they aren't between 850-950 I would adjust your PE vs RPM table according. This table is where you will do all fueling changes at WOT. If you have knock retard we need to localize it to a point in the timing table, so using the method we used above for part throttle tuning we will do the same thing for WOT tuning. Just be sure to let out of it if you pick up large amounts of knock retard or hear pinging.
Once you have changed either spark or fuel you should go back and check the other. There really isn't much you can do with fuel without a dyno + wideband setup. If you do have wideband feedback available you will generally end up running the richest at your torque peak and leaning out as you go towards your hp peak
Once you have fuel and spark set you can begin playing with your shift points and tranny paramters if you have an automatic. If you have a six speed you are good to go!
Follow-Up
After a week or so you need to re-verify all your logged values and ensure they haven't drifted. If they have repeat the processes neccecary to bring them back in line.
