How To: Power FC and FC Datalogit Tuning

Jesse IL · Nov 2, 2007

^^^

I'm getting there, Manny. I've just been very busy, so I haven't had a chance to cover anything but the Settings 1 page.

Jesse IL · Nov 2, 2007

Alright onto Settings 2.

The first thing I'd change here is the water temp correction coefficients. These are adding way too much fuel. Also, the left column is enrichment under no load and the right column is for enrichment under load. You should always have more enrichment under load. Here are the settings I usually use:

Left column: 1.000, 1.016, 1.031, 1.047, 1.063, 1.094
Right column: 1.000, 1.016, 1.047, 1.063, 1.094, 1.125

For Accelerate Injector, Cranking, Inj vs Accel, INJ vs. Air Temp and INJ vs Water Temp, I'd leave the default values alone.

For INJ vs Air Temp, I set these values all to 1.000, since technically the MAFS should be accounting for differences in air temp.

Now on INJ vs TPS, I would change the values here. This is for throttle position based enrichment. What you need to do is figure out the TPS voltage and idle, cruise and other throttle positions and convert them to percentages by dividing by 5. Set the lowest point to your percentage at idle and 1.000 value. Then set your next breakpoint as the most throttle you want with no enrichment and again a value of 1.000. Then between there and full throttle, increase the enrichment up to the AFR you want at WOT. This will require retuning and if explanation is difficult to understand , I apologize but this is a fairly difficult concept to understand. The benefit here is that it makes tuning for stoich while cruising much easier since enrichment is based on throttle position rather than on your map.

Jesse IL · Nov 2, 2007

The first thing I'll cover on Settings 3 is the airflow curve selection. The default airflow curve is what 99% of people should use. It works so there's no compelling reason to change it.

The airflow curve above will NOT work on a car with an Injen intake and even possibly other intakes. At idle, you will get MAFS voltages lower than the lowest point on the curve and the car will simply stall.

The reason the MAFS curve is modified is that the default curve has points that are 0.16v apart. There are typically a number of points above and below the range you need that can be truncated. This allows you to space the remaining points more closely, giving more resolution to the curve. Is this an improvement? Possibly, which is why I say most people should stick with the default curve.

If you wanted to change this curve, you would need to let your car idle funny warmed up and figure out the lowest recorded MAFS voltage. Then take the car to full throttle and measure the maximum MAFS voltage. Give yourself a little wiggle room above and below those numbers and you should be fine.

To set rpm breakpoints, I like to set my lowest rpm point right on my idle setpoint. I then like to add another breakpoint 100-200 rpm above that to account for idling while on A/C and the car's natural tendency to approach your idle setpoint slowly. Beyond that, I like to set one breapoint right before lift transition and if you have the opportunity, make the spacing between breakpoints more narrow in lift, where tuning is more critical. On the above map, assuming your 5600 cam change, I'd leave the point in lift alone, bump the rest of the points up by 50 rpm and change the bottom point to 1000.

Setting load breakpoints is more tricky. You need to figure out where the car "likes" to be depending on your intake. Every intake behaves differently and will register different MAFS voltages for a given amount of airlfow, which will translate into different calculated loads. An Injen intake will register significantly lower loads than a car with a BB intake mod. To determine this, you will need to do something like in the following thread:

http://www.newcelica.org/forums/showthread.php?t=233154

This will tell you where your loads fall while the car is idling. You should set your lowest load breakpoint to the lowest load seen at idle. In the above thread, I set the lowest load breakpoint to 1600 and set another one at probably 1800 to account for the fans or A/C running and fuel control when the car is returning to idle when coming to a stop.

You'll then want to take a drive where you'er doing cruising only, allowing for slight acceleration like what would happen on the highway. This will be your cruise load range. Then you'll want to do a full throttle run and see where the loads fall and what the maximum load is that's logged during the run.

When you have all this data, I like to set my load cell spacing fairly narrow in the cruise range to give myself more points for different conditions, then fairly wide in the part-throttle acceleration range where I spend limited time, then narrow again at full load, especially in the cells hit in lift, to give maximum control of fueling there. You should pik the top load reference about 10% higher than what you logged on your initial run to give yourself room to make power through tuning.

Throttle · Nov 3, 2007

The airflow curve above will NOT work on a car with an Injen intake and even possibly other intakes. At idle, you will get MAFS voltages lower than the lowest point on the curve and the car will simply stall.

Jesse, I'll adjust the airflow curve accordingly but I have been using this map for a while with my Injen CAI without any stalling. The car runs great with all these settings. I am creating a revised map with all your suggested changes that I will load.

Jesse IL · Nov 3, 2007

If your car is runing fine, there's nothing wrong with that airflow curve. All I'm trying to point out is that the lowest point on the airflow curve is 1.12v. If your MAFS hits 1.11v at any point, your car will immediately stall. I'm just pointing that out.

What you should really do is do an idle test like I did above in that other thread and see what your MAFS signal really looks like under various conditions.

Jesse IL · Nov 3, 2007

For Settings 4, these look fine. I leave the default values in everything but IGN vs WaterT and IGN vs AirT. On water temp, I like to be a bit more agressive than the default settings. You should really not be beating on your car if the water temp gets up above 88 C, so I'd rather be safe than sorry. With the air temp, I found that the default air temps were higher than any air temps I've ever seen, even on the hottest day of the year in Chicago. I simply revised the temp breakpoints down from the default values here. You could go even lower if you wanted an even bigger factor of safety.

One thing I should point out is that if you have a car that is tuned on the very edge of knock under "ideal" conditions, you should be very conservative with these values. Because if your car is about to knock with cool air and with the engine at normal operating temps, you could hit a hot day with a heat soaked engine and blow the motor.

Jesse IL · Nov 3, 2007

Settings 5 is the most often overlooked part of the map. If you're running the stock injectors, you can leave the default values alone. If you're running larger (or smaller) injectors, you need to make changes here.

Larger injectors have heavier internal parts, so they take longer to open when they receive a signal. This means that they will have more lag time. If you have larger injectors, you need to look up their rated lag time. The MWR 630cc injectors are only rated at 14v, so what I did was take the MWR injector lag time and divide it by the stock injector lag time to get a percentage. Then I took that percentage and multiplied it through all the lag times at the other voltages. I'm not sure if this is 100% right, but the only real solution would be to have the injectors tested.

The other box can be used in one of two ways. Many people like to use it to make adjustments for switching to smaller or larger injectors. One thing you need to be aware of is that this setting affects ALL raw injection times, so besides the obvious, it affects two setings that most people overlook: Cranking and Accelerate Injector on Settings 2.

In a real engine, air distribution to individual cylinders is not 100% uniform. If you were able to monitor AFR individually in all four cylinders, you could adjust fuel trims up or down on individual cylinders to accuont for this. However, since you can't do that here, its best to just assume that everything is uniform.

Jesse IL · Nov 3, 2007

Now I'll focus on tweaking this particular map to suit his car. What we're looking at first here is the INJ map. Overall this looks very good. He has the car idling just richer than stoich, has it spiking rich off-idle where the clutch would be engaging, cruise is set at stoich and WOT is nice and rich.

Notice I said that it looks nice, but we don't know if this will work great on his particular car. In an earlier post I said that every intake reads differently, which will cause your logged load to be different, which will change where you want these commanded AFR's to fall. You need to figure out what cells the car is actually using diuring idle, cruise, taking off from a stop, etc and define these "zones" of the INJ map accordingly.

If we look at the log grid above, we see that at full throttle, the car starts out in the 7600 load row and runs up to the 12500 load row. Note: ignore the load references you see on the picture, count the cells and compare them to the INJ map. It should be pointed out that the actual logged load is lower than either of those two values. When you exceed a breakpoint, FC Edit will show you as being in the next highest load cell. We'll get to this more later when I show how the Chart function proves to be much more useful for tuning than looking at the grid.

Back to where our loads are falling, we see that this map probably works out fine on this car, as WOT is falling in the zone we've defined as being for WOT. If we wanted to look at areas where this map could be improved, we could change load breakpoints at idle as I've discussed earlier and we could also make the highest breakpoint lower as we're not taking advantage of the entire map. This would allow us to possibly make the spacing between breakpoints lower (its 700 now, could possibly lower it to 600). In order to do this, however, we'd need to look at this log either in Chart or open it up in Excel and see where what the actual loads were at certain points. Also, you can't arbitrarily change the load breakpoints without changing fueling to match. This requires recalculating fueling in an Excel spreadsheet, which I'll possibly go over later. For the time being, we'll leave this map alone, as its fine and is not causing problems or limiting power. I was only pointing out improvements that could be made.

Jesse IL · Nov 3, 2007

IGN map will follow the same rules as the INJ map, meaning that you need to make sure your "zones" are following the same behavior as the car. As we saw in the previous post, this car starts out in the 7600 load row when you go to full throttle. If you look at the IGN map above, you can see that we've dialed back ignition advance by the 6900 load row so we're fine there.

The first thing I'll point out is ignition advance at idle. The PFC for some reason wants to have 20 degrees ignition advance at idle, despite what you put in the map. I'm not sure if putting any lower values actually has any effect, but my feeling is that during the learning process, you should turn Idle-IG Cntrl off any lower these to a point where you achieve your target idle rpm. After restarting the car after learning, the PFC will most likely turn this setting back on and hopefully everything will work as expected. I've found this to be the most frustrating setting to work with on the car and if you find something that works, please post it here.

Performance wise, there are a couple areas for improvement I can see. The first is the total advance. Many people have made good, safe power with more ignition advance that this map has. This is intended to be a "base map", so ignition advance was purposely retarded. You should go through the procedure for calibrating your knock sensor, then see how far you can push advance safely here. The rule of thumb is that once you see knock, you back off two degrees from there.

The second place where some performance can be added is the high rpm ignition advance. Lets take a look at the dyno plot from this car.

Look at the lower blue curve, which is the stock ECU. Peak torque hits at about 6800 rpm, where you'd expect it to and them starts dropping off after about 7200 rpm. Once peak torque drops off, cylinder pressures start to fall. When this happens, you can hold power longer by increasing ignition timing. On this map, you could bump up ignition timing by one degree at 7300 rpm and more at the higher rpms. In other words, the engine will tolerate more ignition advance without knocking at redline versus at the toruqe peak.

Jesse IL · Nov 3, 2007

VVT is where I see the easiest place to make improvements on this car. The first place we'll look is the lift transition. Basically, the engine does not like huge swings in the VVT system. This car is obviously getting a stumble when it goes into lift, so we can try and fix this. To do this, we want to shrink the difference in VVT before and after the transition. Realize that lowering VVT before lift and raising it post-lift will reduce the power somewhat in both these areas, but unfortunately this is necessary to get the engine to behave well. What I'd do here is lower the VVT setting at 5500 to 41 and raise the VVT number at 5700 to 5. If this does not cure the problem, then you'll want to slowly start shrinking this. My next change would be: 5050, 5500 = 40 and 5700, 6100 = 6 and so forth. I wouldn't take 5050 past 36 or 6100 past 10, but you might keep playing with 5500 and 5700 to see if there's an improvement.

The most obvious thing going on here is that thecar with the PFC is hitting the torque peak sooner but is also not holding the torque as long and is dropping off faster. This is something that will need to be tuned on the dyno, but my suggestion would be to do a dyno run with the current settings, then increase the value in the 6500 & 6900 rpm columns by 5 and all the higher columns by 10, then dyno again. This will tell you if you're moving in the right direction with VVT or not. If you make more toruqe across the board, make the same changes again and do another pull. Make sure you give your car a chance to cool a bit between runs. What I like to do is sit with the car off for a couple minutes, then start it up on idle to circulate the coolant, then let it sit a couple more minutes. My suggestion with VVT on the dyno is make a fairly large change (like 10) and see what the effect is, then work towards smaller changes. This allows you to get a feel for the direction the engine wants to go without wasting your time and money doing incremental dyno runs.

As long as you're on the dyno, you can kill two birds with one stone and also mess around with the lift transition at the same time.

Jesse IL · Nov 3, 2007

Now I've covered basic setup of a map and general trends for tuning cars. What I'd like to do next is take a look at some actual street tuning and doing some of the things I've mentioned above. So what I'd need from Throttle would be the following, all in zip format. You can PM me for my email address and I can host all these files, so that someone reading this thread can work along with the example.

1. Copy of the map that was used to make the logs.
2. Log of car at full throttle with existing map.
3. Log of car idling from cold to fully warmed up.
4. Log of car cruising on the highway with no dramatic acceleration.
5. Six or more logs of the car at WOT in 3rd gear from a very low rpm to fuel cut, all with the ignition retarded 5 degrees from the 6900 load row up on the current map. You will need a straight, flat empty road to do this where you can start out at about 5-10 mph and go all the way to fuel cut in 3rd gear.

Jesse IL · Nov 5, 2007

If you have questions about the vertical axis in FC Edit, please read the following explanation. I receive numerous PM's about this so I felt it would be useful to review.

The breakpoints on the vertical axis in FC Edit represent engine load. When viewing a log, the logged value is PIM. On certain screens in FC Edit, this is also referred to as Boost.

Toyota engines run off of a mass airflow sensor (MAFS). This sensor does not measure pressure, but rather airflow directly. Cars such as Honda use a manifold absolute pressure (MAP) sensor. The type of fuel system uned on Hondas is called Speed/Density. In these cars, the ECU measures engine rpm, manifold pressure and air intake temperature to calculate the amount of air entering the engine. With a MAFS, this calculation is not necessary sine the sensor is measuring the air mass directly.

What this means is that the vertical axis in FC Edit for Toyota engines is different than in other versions of the software or on other ECU's such as the Hydra or the AEM. So rather than being a directly measured value (manifold pressure), it is a calculated value. The caluclated value represents load based on the engine. This engine load is caluclated from other directly measured values: engine rpm, throttle position and airflow. Load is directly proportional to throttle position and airflow and is inversely proportional to engine rpm.

What this means is that if you increase the throttle position, load goes up. If the measuerd airflow goes up, load goes up. However, if airflow and throttle position remain constant but engine rpm increases, load will go down. In this way, load is representative of volumetric efficiency of the engine. If you look at a log from a car at full throttle, you'll notice that PIM peaks at exactly the same place the torque peaks on the dyno curve. Peak torque occurs at the point of peak volumetric efficiency.

Load/PIM is scaled from 0-20,000. If the MAFS voltage reaches 5v, you will register PIM = 20,000. In general, the higher load you see, the more power your car is making. The shape of the PIM curve will closely follow the torque curve of the engine and the magnitude of PIM will correlate to how much power is being made.

shadytunerGTS · Nov 12, 2007

I don't know if you like my crazy theories or not, but here it goes. I'm thinking that since my car seems to idle better in the evening compared to the daytime that the air is cooler at night. So, during the daytime I would need to adjust ignition timing to get the car to idle like it should. Would adjusting the IGN vs AirT setting cure this ongoing problem I can not seem to find an answer for? Oh, I just scanned through above and found you suggested changing the IGN vs AirT setting. I think this is going to do it.

How does this sound:
IGN vs WaterT: First column 100, 88.
IGN vs WaterT: second column 6, 3.

IGN vs AirT: First column 36, 26, 16.
IGN vs AirT: second column 6, 3, 0.

many · Nov 14, 2007

Jesse, I'm trying to look at my logged and saved maps but I can't seem to make them appear on the grid.
I can see the log on the chart but for some unknown reason it only appears on the grid on just the upper corner(P01/N01)but only that one has numbers in it.
I'm on the newest version of the FC-edit v2.123
It works fine if I just logged a map and is still unsaved ....I can go true it with no problem.
Any clue ?

Jesse IL · Nov 14, 2007

Rather then continuing to speculate on "crazy" theories, what I'd suggest you do is come up with some hard evidence of what the car is doing. Take a log of the car idling when its warm and one when its cold. Once you have those you can post some basic screen prints and the problem should become pretty clear.

I've seen that happen before and I'm not entirely sure why it happens? It happened to me once over the summer and I'm trying to remember why it was happening... Maybe I didn't have loggging of MapRef enabled? I don't honestly remember. Have you tried switching the source back and forth between MapRef and Advanced? Honestly, I do very little tuning via the table any longer. I find it to be far to rough and it tends to be misleading, especially when you get very close to your target AFR.

many · Nov 14, 2007

Yes I tried it back and forth, clicked on about everything for about 1 hour(&$#@)....the only thing that works is by opening the log file in text format ...and modify the file .. ex: [2zz-ge v2.71A] to---> [2zzge] and then using an older version(v1.100) to open in map watch.

on another note..I just did an excel chart for the air mv on low and high cam for vvt settings....wanna take a look at it Jesse?
I can't host it (I tried but it doesn't accept excel files) I could send it to you(I still have your email)

Jesse IL · Nov 15, 2007

I took a look at the Excel charts and emailed you back some comments. The one thing I'll say here that I think needs to be made clear at this point in my explanation is that any time you're plotting data from an FC Edit log, the chart type you want to use is XY Scatter. The x-axis will always be your rpm, and its never uniform. On the plots many sent me to look at, he had tried to force everything to uniform data points. Scatter plots don't care if the rpm points are all uniform, which is why you want to use them.

many · Nov 15, 2007

Thanks.

On the chart it's asking me what to put on the value of X and Y axis?

Jesse IL · Nov 16, 2007

1. Highlight all the mV values you want to plot and click the toolbar button to make a chart.

2. Choose XY Scatter

3. When the dialog box pops up, cliick over to the Series tab and click the little box next to the x-axis values. this will pop you back to your data. Select all the rpm values that correspond to the mV you already chose.

Example:

A1:A99 (rpm values)
C1:C99 (mV values)

zero2toy · Nov 28, 2007

When logging in different gears, ie 3rd or 4th, wont you get different logged load cells, ie load cells logged will be higher do to more load in 4th vs 3rd? I was thinking about this the other day, and currently do not have the means to test this. Just me overthinking things to the point of tuning lower rpm points.

Sorry, ^ dumb question, load is not equated the way i was thinking earlier today, disregard unless there is something i missed here.

How To: Power FC and FC Datalogit Tuning

Jesse IL

Jesse IL

Jesse IL

Throttle

Jesse IL

Jesse IL

Jesse IL

Jesse IL

Jesse IL

Jesse IL

Jesse IL

Jesse IL

shadytunerGTS

many

Jesse IL

many

Jesse IL

many

Jesse IL

zero2toy

Top Contributors this Month

Recommended Communities