Practical Reflash Tuning: Injector Scaling
Watch This Course
$229 USD $114.50 USD
-OR-
Or
8 easy payments of only $14.31
Instant access. Easy checkout. No fees.
Injector Scaling
08.09
| 00:00 | - Just like the MAF scaling we've just covered, scaling the ECU calibration to suit a different set of injectors is also vital to properly tuning a factory ECU. |
| 00:12 | If you're tuning on the factory fitted injectors then obviously there's nothing to do here. |
| 00:17 | But frequently the modifications to the engine will exceed the flow capabilities of the stock injectors and larger injectors will be a necessity. |
| 00:27 | The reason the injectors scaling is so critical is because of the way an OE ECU works. |
| 00:34 | Instead of having a fuel table that directly requests a certain injector pulse width, which is common in aftermarket ECUs. |
| 00:43 | The OE ECU is commanding a specific air/fuel ratio. |
| 00:49 | Remember that the ECU is measuring mass air flow and if it knows how much air is entering the engine and the desired air/fuel ratio, it's quite easy to calculate how long to open the injectors in order to get the required mass of fuel. |
| 01:06 | This all hinges on the ECU knowing what size the injectors are though or in other words how much fuel they'll supply for a given pulse width. |
| 01:16 | There are two dramatically different techniques used by OE manufacturers to define the injector characteristics. |
| 01:23 | In many ECUs the injectors are dealt with in a simplified manner giving just two parameters that need to be adjusted. |
| 01:31 | Injector flow scaling or reference flow, and injector latency or dead time. |
| 01:37 | The more sophisticated ECUs however from the likes of Ford and GM, the injectors are dealt with more thoroughly and complete characterization data is required by the ECU. |
| 01:50 | What I mean by this is that rather than a simple reference flow and dead time table these ECUs will much more thoroughly characterised the injector flow based on pulse width, fuel pressure, and battery voltage. |
| 02:05 | This data deemed necessary by the manufacturers in order for the ECU to be able to do the best job possible of insuring exactly the required mass of fuel is delivered and hence accurately controlling the air/fuel ratio. |
| 02:22 | If you're dealing with an ECU that includes full characterization data for the fuel injectors then you need to make sure that you can replicate the required data for any new injectors you fit. |
| 02:36 | Since it's not possible for us as tuners to generate this data, you need to make sure that any injectors you purchase come with the required data. |
| 02:46 | Shown here is the GM characterization data that Injector Dynamics make available for their ID1000 injectors. |
| 02:55 | Swapping to a set of ID1000 injectors is as simple as copying the individual tables from the supplied Excel file and pasting them into the matching table in your ECU Calibration. |
| 03:09 | It might seem like a lot of trouble to go to but the reason for all of the data becomes obvious when you can turn the key after flashing the new injector data into your ECU and the engine starts and runs exactly like it did on the old injectors. |
| 03:25 | If you have an unknown set of injectors with no data then you literally have no chance guessing this data and it will be impossible to achieve good results. |
| 03:36 | If your ECU uses the simpler injector data of scaling and latency, then it is possible for your to rescale the calibration to suit a new set of injectors, but there are still some precautions. |
| 03:50 | Since the parameters are injector flow and latency, on the face of it, copying the relevant flow and dead time data from the injector manufacturers data sheet would seem like an obvious choice. |
| 04:04 | Unfortunately it's not quite that simple though and these values might get you close but definitely won't be correct. |
| 04:12 | We'll typically find that the injector flow scaling number ends up around 10 to 20 percent lower than what an injector manufacturer would quote. |
| 04:22 | So for instance an injector that's rated at 1000cc’s at three bar would likely end up with a flow scaling value around 800 to 900cc’s per minute. |
| 04:34 | While the injector latency data should follow the same curve as the latency values provided by the injector manufacturer they may need to be increased or decreased to suit. |
| 04:47 | The process of dialling in the scaling and latency for a different set of injectors in this type of ECU involves logging the fuel trims at idle and at cruise, and then making adjustments based on the results. |
| 05:02 | What we're trying to do is adjust the scaling and latency values in order to get the fuel trims as close to zero as possible under all closed loop operation. |
| 05:14 | In reality a trim of zero is pretty hard to achieve and I'm generally happy if my trims are within plus or minus 5 percent, although obviously the closer we can get the better. |
| 05:26 | The process involves logging the fuel trims at idle and then again at cruise. |
| 05:32 | Since the latency value has more impact at the low injector pulse widths we see at idle, the latency values have more effect on the fuel trims at idle and less at cruise. |
| 05:45 | I start by checking the fuel trims at cruise and making changes to the scaling here. |
| 05:51 | If for example the trims at cruise end up to be positive 10 percent, then this means that the ECU believes the injectors are bigger than they are and is supplying less fuel. |
| 06:04 | In this case I've reduced the injector scaling value by about the same amount as the trim value, in this case 10 percent, and then I'd try it again. |
| 06:15 | Once the trims at cruise are close you can test at idle. |
| 06:21 | This time we want to make changes to the latency values. |
| 06:25 | If the trims are idle at positive, this means the injectors need to be open longer, and this can be achieved by increasing the latency values. |
| 06:35 | Likewise if the trims are negative, we would reduce the latency numbers. |
| 06:40 | Of course changing the latency values will have a smaller impact on the cruise trims too, so it becomes an iterative process of adjusting, flashing, and logging. |
| 06:52 | An issue that I've struck a number of times is the situation where a car is being fitted with both a larger MAF, or a modified intake system, and a larger set of injectors at the same time, and is then being delivered ready for tuning. |
| 07:08 | If the ECU uses full injector characterization tables like GM and Ford, then this isn't necessarily a problem. |
| 07:17 | But particularly in the case of Subaru ECUs, makes the tuning impossible. |
| 07:23 | The reason is that we need to recalibrate the ECU to suit both larger injectors and a different MAF calibration however we rely essentially on the same data for both tasks. |
| 07:36 | What I mean is that if air/fuel ratio is rich we could create this by modifying the MAF calibration or we could get a similar result by altering the injector scaling and latency. |
| 07:49 | There's no way for the tuner to know where to make these adjustments and the result will usually end up as a disaster. |
| 07:57 | The best solution in this situation is return the MAF or the injector to stock and deal with one of the adjustments at a time. |
