Mitsubishi Lancer Evolution Forum banner

HOW TO: Adjust Injector Latency & Scaling for Larger Injectors

118K views 62 replies 32 participants last post by  GeoMitsu 
#1 · (Edited)
Code:
[B][COLOR=Red]DISCLAIMER: I have used the method outlined below successfully to tune my car. I am an amateur and not a professional. If you use this method then you accept full responsibility for whatever damage you do to your car. If you are concerned about voided warranty and damage to your car, you are better off leaving your car without any modifications to the ECU[/COLOR][/B].
Basic Info:

On average, your crusing injector pulsewidths will be double your idle injector pulsewidths. So, any changes that you make in the injector deadtime will affect idle twice as much as it will affect cruising. But, yes, it will affect both if they are both in the same Volt range that you are adjusting for.

Basically, as follows:

1. Choose a global setting (ECUFlash injector scaling number)
2. Log trims at idle and cruise
3. If trims are roughly the same, but positive, then you need to increase your global fueling, or in terms of ECUFlash, you need to lower your injector scaling number.
4. If both trims are roughly the same, but negative, then you need to descrease your global fueling, or in terms of ECUFlash, you need to raise your injector scaling number.
5. If idle trim is more positive than cruise trim, then you need to increase the dead time, or in terms of ECUFlash, you need to increase the latency value.
6. If the idle trim is less positive than the cruise trim, you need to decrease the deadtime, or in terms of ECUFlash, you need to decrease the latency value.

NOTE: After #5 or #6, you may need to readjust #1 accordingly.

NOTE 2: Basically, let's say that the IPW is 1ms at idle and 2ms at cruise. If you adjust the deadtime to add 100us (.1 ms), then you are affecting the idle fueling 10% and the cruise fueling 5%.


EXAMPLE:

Let me give a quick example using your numbers and show you how you would have to change the resulting scaling:

Idle trim: +3%
Cruise trim: -6%

Assuming that your idle pulsewidths are half of your cruising pulsewidths and idle is about 1ms (just using round numbers for the example...you would have to log to find out for sure), then let's see what adding 100us (.1 ms in ECUFlash) would do to the trims:

The 100us would add 10% fuel to idle, so it would subtract 10% from the idle trim, now making it -7%.

The 100us would add 5% fuel to cruise, so it would subtract 5% from the trims, now making it -11%.

So, now you have:
idle: -7%
cruise: -11%

So, now if you scaled your injectors to be 7% larger, then your trims would be:
idle: 0%
cruise: -4%

EDIT: Therefore it will look something like this for LTFT CRUISE/MID.

If you are using a scaling value of 731 (for example) and now you need to add 7% of scaling. Do the math: 731 x 0.07 = 51.17 = New scaling is 782.17.

So now, options in ECUFLASH are either to use 770 or 790 scaling. Using the 770 will make the injectors inject more fuel, but make the fuel slightly negative, since the computer has to take away from fuel. Now making it 790 will make the injectors inject less fuel, thus forcing the computer to compensate and add more fuel making the trim towards the positive. Now you have to work on getting the idle closer to 0.

I usually shoot for LTFT Cruise +/- 3% and LTFT Idle +/- 3-5%. These are averages. So if you are using EVOSCAN. Open up your log. Select the column with LTFT Cruise & Idle. Open the log in EXCEL or SPREADSHEET.

Select the correct column look at the first number you want to average. Lets say it is Column A ||| Row 02. The last log entry in the log for that column is Row 999. Therefore in that column select Row 1000. Then enter this equation "=AVERAGE(A02:A999)" this will average all your LTFT. Now you will able to adjust the scaling properly.


Of course, if you make any changes to your injector scaling, you will have to make sure that you make the necessary changes to your open-loop fuel maps, since the injector scaling change will change your AFR.

NOTE: Any changes to MAF Scaling will change your fuel trims. I suggest if your intake uses the STOCK OEM MAF tube as the housing. That you keep your stock MAF scaling and ONLY adjust the injector latency and scaling to get your fuel trims as close to 0 as possible and within +/- 5% of 0.
 
See less See more
#2 ·
nj1266 said:
EDITED BY: CLIPSE3GT

Injector Scaling

The stock injectors on the Evo are rated at 600. Generally speaking, the stock injectors will give you adequate fuel flow with a TBE, an intake, and 19-23 lbs of boost by redline. Once you get cams, then it is advisable to get bigger injectors. If you get a bigger turbo, then definitely get bigger injectors.

If you install bigger injectors, then you will have to scale them properly and use the correct injector voltage latency, otherwise your car will idle poorly and stall on occasion.

So how do you go about scaling your new injectors?

Scaling injectors is a PITA. It involves a lot of trial an error. It is not enough to simply put numbers in the rom tables and simply declare the injectors scaled. You must test and make sure that the scaling is accurate.

Here are the steps I follow when scaling injectors

1. Open Ecuflash and open your rom. Under fuel locate the Injector Scaling table and the Injector Battery Voltage Latency table. On a stock Evo they look like this:



The table on the left refers to the injector value that the ECU is using when making its fuel supply calculations. The number in the table is always smaller than the actual size of the injectors on the engine. For example, the stock injector size is ~600, but the number in the table is 532. As a general rule of thumb, enter a number in the table that is 15-20% less than the size of the injectors installed on your Evo.

EVO X: I found that the difference is about 11% in the OEM values.

Starting Point for DW Injectors:

800cc: ~713 Scale Starting Point

1000cc: ~890 Scale Starting Point


For example, let us say that you are using 680 size injectors. Then you enter in the injector scaling table a number between 578 and 544. Please note that this is only the starting point and not the end point of injector scaling. The final number will be determined through multiple sessions of logging and testing your fuel trims. More on this later.

The table to the right is the injector battery voltage latency table, aka, dead time table. The numbers in the right column are in milliseconds. The numbers refer to the amount of time that the injectors take to open completely and produce maximum flow. The numbers in the left column are in volts. As battery voltage decreases, the time between the injector receiving the signal to open and when it actually opens increases. Therefore, you must send the signal sooner to have the injector open at the appropriate time. The larger the injectors, the more time is needed for them to open. So you will have to increase the numbers in the millisecond column to compensate for larger injectors. The trick is to find the accurate numbers.

ou need to make sure that the numbers that you have entered are working properly. To do that you MUST log your fuel trims for an extended period of time. The trims to log are:

LTFT Idle = Long Term Fuel Trim Idle
LTFT Cruise = Long Term Fuel Trim Cruise

Both of these trims fluctuate between +/- 12.5%. The LTFT Idle is for idle and the LTFT Cruise is for cruising. Your aim is to keep both trim to +/- 5% or less. If the fuel trims are too positive, then the ECU will add fuel and this will royally make your AFR too rich. If your fuel trims are too negative, then the ECU will remove fuel and this will make your AFR too lean.

How do I log my fuel trims?

Logging fuel trims takes a lot of time and you will waste a lot of gas to get your fuel trims as close to 0 as possible.

Let us start with the LTFT Cruise. You must drive the car at a steady speed for at least 16 minutes. Why? The fuel trims cycle approximately every 4 minutes. You will need to have them cycle multiple times until they settle on a number in your log. 16 minutes will alow your trims to cycle 4 times. That will give them ample time to settle.

Let us say we had the scaling at 552 for the 680 injectors. And we cruised at a steady speed of 60 mph for 16 minutes. We found out that the trims went way negative and hit -10%. So we pulled over and incremented the injector scaling twice. The scaling went up from 552 to 572. We flashed the new numbers into the ECU. Then we went logging again for another 16 minutes.

We found out that the trims are still going negative but not as much as before. This time our trims hit -8%. We now know that we are on the right track. But we are nowhere near the +/-5% that we would like to hit. So we pulled over on the side of the freeway again. Incremented the injector scaling in the rom twice. This moved it from 572 to 597. We flashed the new scaling into the ECU and went for another 16 minute log.

The numbers in the log were very close to 0. The LTFT cruise registered around -3. We are almost there. So we pulled over on the side of the freeway and incremented the injector scaling one last time. The new scaling was 609. We flashed the scaling into the ECU and went for yet another 16 minute log. The logs showed a final fuel trim of -1.86. We have dialed the LTFT cruise to as close to zero as possible.

The next step was to log the LTFT idle. This is done by logging the car for 16 minutes at idle. We did that several times and found out that the LTFT idle stayed at the -1.66 level. This is a similar number to the one we achieved with the LTFT cruise trims.

We have finally dialed our fuel trims. Are we done done now? Not yet.

Now you MUST go back and adjust your High Octane AFR fuel map to fit the new injector scaling. You are basically going to have to re-tune your fuel map to fit the new injectors.

The whole process is very time and gas consuming and it can get very frustrating. But with time and patience, it can be done.
THANKS to: l2r99gst & nj1266
 
#3 ·
UPDATED with more info!!!
 
#18 ·
yip - ^ :+1:
 
#7 ·
Of course, if you make any changes to your injector scaling, you will have to make sure that you make the necessary changes to your open-loop fuel maps, since the injector scaling change will change your AFR.

&

Now you MUST go back and adjust your High Octane AFR fuel map to fit the new injector scaling. You are basically going to have to re-tune your fuel map to fit the new injectors.
Clipse3GT - this may seem trivial, but can you briefly go over how this is done? Thanks.
 
#13 ·
In my testing... I know that my method works, with the steps posted in the first post. I tuned several cars with the same method. Also got close to target.
 
#15 ·
well it's been a couple of days so i'll answer my own questions. When you increase the deadtime you actually richen the system. which means steps 5 and 6 are backwards. so if your cruise trim is more positive than idle it means the ecu is adding that amount. so you need to richen the system, by Increasing the deadtime.
 
#16 ·
Did you just test this on idle?
 
#19 ·
Has anyone tried this with the AP? We don't have LTFT Idle. We just have 1 LTFT which is the cruise one. I know this is the ECUFlash section, but the process is / should be the same.

Here is what Cobb's documents say:

Injector Scaler
This table contains a singular value used to represent the fuel injector size or flow rate. Any changes to this value will affect ALL tables within the ECU related to fuel delivery and load calculations. When using stock injectors with Petrol fuel, this value DOES NOT need to be altered. When adjusting this value, a lower number represents an SMALLER injector, whereas a larger scale value will represent a LARGER injector.

Tuning Tips – To calculate a starting value for a different injector size than stock, use the following formula:

New Scale Value = [(New Injector Size * Original Scale Value) / Original Injector Size].

For example, let’s say you are replacing your factory EVO X injectors (~550cc) for aftermarket 750cc injectors. The formula would look like:
New Scale Value = [(750cc * 532) / 550]
New Scale Value = 725

Input the calculated value as a starting Fuel Injector Scaler value. To fine tune the injector scale value, we suggest you install the stock intake system and run with stock level fuel pressure levels. You will want to display the Short-term Fuel Trim and Long-term Fuel Trim values with the Dashboard. With the engine idling at full temperature (coolant temperature between 180-195 F and intake air temperature +/- 15 degrees F of ambient temperature), you can make adjustments to the scale value until the A/F Trim Mimed. + A/F Learned are as close to zero as possible, +/- 5% is generally acceptable. We have seen that you will also need to fine tune the Intake Calibration tables in order to get the calibration closer to optimal. The closer you can get to 0% is ideal. DO NOT attempt to tune for an aftermarket Intake and aftermarket injectors at the same time. An aftermarket intake will affect your A/F Trim Mimed. and LTFT values at idle and part throttle, making it nearly impossible to find an accurate Injector Scale Value. If you have an aftermarket intake please use the above equation to establish your initial Fuel Injector Scale value then proceed to the Intake Calibration section if necessary. If you plan to use an aftermarket intake, it can be installed and the necessary tuning can be performed AFTER you have found the optimal Fuel Injector Scale value.

Injector Latency
This table contains latency values used to tell the ECU how much latency is needed to properly control the fuel injectors; the breakpoints are in battery voltage. All fuel injectors require a certain amount of time to fully open which is referred to as Injector Latency. The amount of latency an injector requires is dependent on several factors such as the size of fuel injectors, viscosity of fuel, manifold pressure, and fuel pressure. Lower battery voltage requires increased injector latency (dead time). Likewise, higher fuel pressure may also increase the injector's latency. The data in this table is represented in milliseconds. A higher value will open the fuel injector sooner, thus the total IPW will be greater; a lower value will open the fuel injector later, thus the total IPW will be less.

Tuning Tips – Most fuel injector manufacturers will be able to provide you with this latency data and the voltage they are referenced at. Although, the drivers used to develop these latencies may be different than the injector drivers in the stock ECU. You can use the published values as a starting point and modify from there. Don't be afraid if your final values differ from what the manufacturer provided. To tune this table, we suggest that you first establish a good Injector Scale value.

One way to find the correct latency (or at least the latency that works best with the injector drivers in the ECU and your particular injectors) is to have your fuel system running stock fuel pressure and have the stock intake system installed then;

1st - set the proper scale value for the injectors you are using based of the scaler calculation.
2nd - start the engine and let the car warm up to temperature (coolant temperature between 180-195 F and intake air temperature +/- 15 degrees F of ambient temperature) then re-set the ECU so your fuel trims start at zero.
3rd - start the vehicle again and watch the SUM of your fuel trims, Short-term Fuel Trim + Long-term Fuel Trim.

If you see that the SUM of your fuel trims (A/F Trim Mimed. + LTFT) is positive then add injector latency until you see the SUM of your fuel trims come closer to zero. You will have to test this throughout the operating range of the engine...the entire MAF curve. Try to avoid sudden throttle movements during this process, you want to avoid seeing any corrections based on the Enrichment table settings.

If you see that the SUM of your fuel trims is negative then reduce injector latency until you see the SUM of your fuel trims comes closer to zero. You will have to test this throughout the operating range of the engine...the entire MAF curve. Try to avoid sudden throttle movements during this process, you want to avoid seeing any corrections based on the Tip-in Enrichment table settings.

This is part of a calibration process that should be able to get you close to the ideal settings necessary to properly control your fuel injectors. Please take into account that you will most likely have to fine tune the intake calibration table as the final step. This will be necessary to match the characteristics of these new fuel injectors.
 
#20 ·
For cobb, log MAV. 1.88 and below = Idle trim. 1.89-2.62 is cruise. Depending on how your open loop load and throttle tables are done, the upper voltage open loop switch range can varry.
 
#23 ·
^ I have LTFT disabled for open loop anyway, but it is good advice.

I'm going to log my MAF Hz vs MAF Volts to confirm the crossover when I have a chance. Another thing to look into would be to log the "current" LTFT, which should switch from Idle to Cruise when it does, but will only need one parameter logged.
 
#26 ·
btt
 
#29 ·
Should the scaling and latency numbers be the same if you switch to a different brand injectors with the same size?
 
#36 ·

NOTE 2: Basically, let's say that the IPW is 1ms at idle and 2ms at cruise. If you adjust the deadtime to add 100us (.1 ms), then you are affecting the idle fueling 10% and the cruise fueling 5%.

I'm not positive, but I think this might be incorrect. Let's look at the case of a theoretical injector that has a dead time at the given voltage & pressure of 1ms, and your total IPW as logged is 2ms.

Now, consider the case where my latency value is incorrect. Say it is .9ms instead of 1ms. My LTFT would be 10% (adding 10% fuel to make up for the missing .1ms of latency).

If I used the math above, then I would arrive at the conclusion that I wanted to add .2ms to latency (10% of 2ms), but if I do then my latency is now 1.1 and my LTFT is now -10%. Clearly what I really want to do is add .1 and bring my latency up to the proper value of 1ms. I really want just a 5% increase here vs total IPW.

Now lets look at another injector, maybe this one has the same latency but flows half as much, requiring a total of 3ms IPW at the same load. Again, my latency is .9ms, off by .1ms in this case LTFT would be 5% (.1 ms is 5% of 2ms). But I don't want to take 5% of 3ms, because that would be .15ms and adding that to my .9ms latency would yield 1.05ms which is wrong again.

Instead, what I want to do is this:

Start with my total IPW and subtract out my latency, yielding in this case:

3 - .9 = 2.1ms

Now divide by the amount of LTFT (in this case adding 5% would mean 105%) so:

2.1 / 1.05 = 2ms

Now divide that by the original IPW (2ms/3ms or 2/3rds) and then multiply back the % of fuel we are missing:

5% of 3ms is .15 ms. take 2/3 of .15 and we end up with .1, which is what we actually want to add to our latency value!

Does this make sense or have I completely missed something?
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top