Practical Reflash Tuning: MAF Rescaling
Watch This Course
$229 USD
-OR-
Or
8 easy payments of only $28.63
Instant access. Easy checkout. No fees.
Learn more
MAF Rescaling
06.54
| 00:00 | Since the ECU's calculations for mass air flow are all based off the MAF sensor's calibration, understandably it's vital that this is correct. |
| 00:11 | If the MAF calibration is wrong, the entire tune will be affected, and you're likely to waste a lot of time trying to correct poor drivability issues, stalling, hesitations, and poor control over both the fuel and ignition timing. |
| 00:27 | Remember that the MAF sensor scaling is used by the ECU to calculate engine load, and engine load is used as an input for jut about everything the ECU does, including fuel delivery, ignition timing, and boost control. |
| 00:43 | This is why it's so vital that the MAF sensor scaling is correct. |
| 00:49 | A MAF sensor will either output a voltage or a frequency, depending on its design, that varies in relation to the mass of air passing through it. |
| 00:59 | In a standard application, the MAF sensor, as well as the air box, filter, and the remainder of the intake tract will be run on a flow bench to generate an accurate calibration for that particular sensor. |
| 01:14 | What this means is that anytime we modify an aspect of the intake system, it's likely to affect the accuracy of the MAF calibration, and in turn the accuracy of the tune. |
| 01:27 | This might be obvious enough if you've made a dramatic change, such as fitting a completely different MAF sensor or possibly fitting the MAF sensor into a different diameter housing, but even more subtle changes, like a cold air intake or different intake plumbing can also impact on the MAF calibration. |
| 01:50 | In order to rescale the MAF sensor, what we need to do is calculate the error between the air-fuel ratio the ECU is commanding and the air-fuel ratio that is being measured in the exhaust. |
| 02:04 | Any time there's an error, this means the ECU thinks that either more or less air is entering the engine that what there actually is. |
| 02:13 | The specific technique for rescaling the MAF will depend on exactly what ECU you're tuning, and we'll look at some examples a little late on to show you how it can be done. |
| 02:25 | Essentially though what we're going to be doing is to log a lot of data across as much of the MAF sensor's measurement range as possible. |
| 02:34 | The key piece of data we're looking at is the error between the target air-fuel ratio and the measured air-fuel ratio. |
| 02:42 | If we find, for example, that at a specific point the measured air-fuel ratio is 5% leaner than target, this could be corrected by increasing the MAF sensor calibration at this voltage by 5%. |
| 02:57 | Proper rescaling of the MAF sensor involves a lot of data logging, and there are some considerations to keep in mind while you're making a log for this purpose. |
| 03:06 | We want to make sure that the operating conditions of the engine are as stable and consistent as possible across the entire log. |
| 03:15 | So performing your logging around town where the car will be regularly stopping and starting and suffering from heat soak isn't going to give you the best data. |
| 03:26 | Likewise, you want to ensure you're making smooth changes to the throttle position as abrupt changes will result in sharp changes in air-flow through the MAF, and the ECU will also be adding acceleration enrichment that can affect the measured air-fuel ratio and the accuracy of the air-fuel ratio. |
| 03:47 | It's also useful to perform large logs, so that the results can be averaged before being applied to the MAF calibration curve. |
| 03:57 | This helps reduce the effect of outliers in your log file, and will give you a more accurate result. |
| 04:05 | If you look at a MAF scaling table graphically, it will generally have a smooth logarithmic shape like the one shown. |
| 04:14 | When it comes to rescaling the MAF calibration, we want to ensure that the same general shape and smoothness is retained. |
| 04:22 | If you find that your calibration is becoming irregular and bumpy, this is an indicator that you may have another issue. |
| 04:31 | One of the more common reasons for rescaling a MAF sensor would be due to fitting the sensing element into a larger diameter housing. |
| 04:40 | This is often done to extend the usable measurement range of a factory MAF sensor, which may be essential when making large increases in air flow on an engine. |
| 04:51 | A good way to estimate the required change to your MAF calibration is to measure the cross sectional area of the old housing and the new housing, and then express the difference as a percentage. |
| 05:06 | Let's say the original housing we had had an inside diameter of 65 millimetres and we're going to be swapping to a housing with an inside diameter of 80 millimetres. |
| 05:18 | We can calculate the cross-sectional area by using the equation area equals pi multiplied by radius squared. |
| 05:28 | The radius is simply half of the diameter, so putting the numbers from our two housings into the equation, we end up with a cross-sectional area of 3,318 millimetres squared for the original housing, and for our larger 80 millimetre housing, we're looking at a cross-sectional area of 5,027 millimetres squared. |
| 05:53 | To work out the percentage difference, we can simply divide the cross-sectional area of the new housing by that of the old housing. |
| 06:00 | In this case, that would give us an an answer of 1.52, which means that the new housing is 52% larger than the original. |
| 06:11 | You can see that when it comes to the diameter of a MAF housing, it doesn't require a large increase in diameter to have a significant affect on the cross-sectional area. |
| 06:24 | Now we've calculated this difference in area, we can highlight the entire MAF calibration table and multiply it by 1.52. |
| 06:34 | This isn't going to be the perfect answer, and we'll still need to rescale the MAF from here, but this will give you a starting point that should let you get the engine running so you can then begin logging some data and fine-tune the calibration further. |
