Practical Reflash Tuning: Step 5-C: Configuring Torque Tables
Watch This Course
$229 USD $114.50 USD
-OR-
Or
8 easy payments of only $14.31
Instant access. Easy checkout. No fees.
Step 5-C: Configuring Torque Tables
20.35
| 00:01 | - The last step of the tuning process that we're going to look at here under step five is what we now need to do with our torque tables. |
| 00:08 | We've already discussed the importance of these torque tables and they do need to be tuned accurately. |
| 00:14 | If we don't have these torque tables set accurately, we can end up with drivability problems and also the potential for the ECU to go into a limp home mode. |
| 00:23 | Now of course in our situation we are a little bit fortunate in that we are working from the base Roush calibration which is already calibrated for the supercharger installation. |
| 00:33 | This means that the torque tables are already relatively well developed. |
| 00:37 | And this doesn't leave us with any work to do. |
| 00:39 | In order to demonstrate the process, what I've done is purposefully modifed the torque tables so that we can see what we need to do and we can also see what we're going to be experiencing if these torque tables aren't quite right. |
| 00:52 | Now while we do need to spend our time and get these tables as accurate as we can, it is a little bit tricky developing these tables on a chassis dyno. |
| 01:02 | Fortunately there is a little bit of fudge room inside of these tables, they don't have to be pinpoint accurate. |
| 01:07 | But of course we wanna do our job to the best of our ability. |
| 01:11 | Before we have a look at the ECU tables, what we'll do is we'll just get our truck up and running and we'll have a look in our scanner to see what we can be looking for that will indicate a problem with the torque tables in the first place. |
| 01:25 | Alright we've got our engine up and running, we're in fourth gear, and we're around about 1600 to 1700 RPM. |
| 01:30 | Now the parameter that I am logging here in the scanner that we're interested in is IPC wheel torque error. |
| 01:37 | Now IPC stands for independent plausibility check. |
| 01:41 | And this is essentially the ECU's internal sanity check to make sure that the calculated torque base from the air flow does match what we've got in our torque tables. |
| 01:52 | So right now we can see that this number is sitting at zero. |
| 01:55 | And in a perfect world, this is what we want. |
| 01:57 | What I'll do now is apply a little bit more load and we'll come up and we start seeing that the IPC wheel torque error value starts to move around a little bit. |
| 02:07 | So this indicates, or is one indication that we've got some problems in our calibration there, and if we see this sort of situation, it does suggest that we've got some problems and some work to do in our torque tables. |
| 02:20 | Let's move on and we'll have a look at how we need to set up our ECU, in order to calibrate the torque tables. |
| 02:27 | Alright in our editor software here we are on our torque model tab and we're looking at our general tab, and under this we've got our torque calculation, these are our engine torque tables for each of our mapped points, along with the inverse table. |
| 02:42 | Now obviously we have one of these tables for each of these map points, let's open one up, and we can see that we've got a three dimensional table with our engine load versus our engine speed. |
| 02:53 | We've already discussed this previously but if we open up our inverse tab, we see that we have got exactly what the name implies, the inverse there, we're now, instead of load, we've got torque on our vertical axis, and this time the numbers inside of this table will show us what load value is required to achieve a certain amount of torque for a given RPM. |
| 03:16 | So the process here is that we are going to recalculate our torque tables, and then we're going to use the built in inverse calculator inside of the VCM editor software to populate the inverse tables. |
| 03:30 | It's really important that the torque table and the inverse torque table are correct, if these aren't correct, we're going to have big problems with our calibration. |
| 03:40 | Now of course as we move through the various mapped points, this adds another layer of complexity. |
| 03:44 | And while there are potentially a variety of ways we can go about calibrating these tables. |
| 03:50 | What we're going to do is start by setting our ECU to only use one fixed mapped point. |
| 03:56 | I've already gone ahead and done that but let's have a look at how I've done that. |
| 03:59 | We'll go to our air flow tab and our variable camshaft tab. |
| 04:02 | Let's come down to our mapped points configuration and if we click on that, we can see that I have only got map point zero enabled. |
| 04:11 | You'll remember that we were working between map point zero and map point four. |
| 04:15 | So this means that any time the engine is now running, the ECU is forced to use map point zero, there's no interpolation between the various mapped points, and this makes out work when we're calculating and calibrating our torque tables a lot simpler. |
| 04:31 | So what we're going to be doing here is working on the principle that if we have an error between the measured torque and the torque in our map point zero table, then it's likely that we're going to see an error of a similar magnitude throughout the other tables. |
| 04:48 | So this just speeds up our process. |
| 04:50 | Rather than needing to individually calibrate each of the map point tables for our torque model, we're going to simply do all of our work on our map point zero torque table and then we're going to apply the error that we find on that table to all of our other tables that are being utilised. |
| 05:08 | It's really important obviously to remember that once we've gone through this process and we've got our torque tables dialled in, that we also re enable the other mapped points that we are using when the engine is normally running. |
| 05:21 | Alright now we've got our ECU set up to only use our mapped point zero torque table, we're going to have a look at our excel spreadsheet that we've developed and see how we can use this with the torque table tuning. |
| 05:33 | Now the excel spreadsheet on face value does look a little bit complicated, but the way we're actually going to use the premise of its operation is really straightforward so let's go through that now. |
| 05:43 | We've got a variety of tables here and the top table you can see we've got labelled as dyno torque values. |
| 05:50 | So this table, as we can see here, has break points for both load, as well as RPM, that match exactly with the break points on our torque table that we've got over here on the left hand side. |
| 06:04 | So that's really important. |
| 06:06 | The idea here is we're going to simply populate this table with the numbers that we read off our dyno. |
| 06:12 | Next we've got our dyno correction factor. |
| 06:15 | So this is quite important, our dyno is obviously going to be reading the torque being produced at the rear wheels or at the rollers or at the hubs depending on your type of dyno. |
| 06:25 | In our case we are using a parameter from the Mainline called derived torque which essentially takes the gear ratio out of the calculation. |
| 06:33 | Regardless of this though, we're not directly measuring torque at the flywheel and there's going to be some level of drive train loss here. |
| 06:42 | So we need a correction factor. |
| 06:43 | So you'll see that I've got this set to 1.35 and this is one of the complexities of tuning these torque tables. |
| 06:50 | What we need to do is have a relatively accurate compensation factor or correction factor here to work with. |
| 06:57 | And a really good way of developing this is if you have a completely stock standard late model Ford that works on this torque based model, what you can do is compare your dyno values to the factory torque tables and adjust your compensation or correction factor until the two correlate really closely. |
| 07:16 | So that's what we've done here on our Mainline dyno. |
| 07:18 | Moving down, our next table is our flywheel corrected torque, so this is really straightforward, there's just a simple calculation that is going to take the numbers that we enter in our dyno torque values table, and it's gonna multiply it through by our correction factor. |
| 07:34 | So in other words it's just going to multiply our dyno values by a factor of 1.35 The next table down if we scroll down a little further, I've labelled our original torque table. |
| 07:45 | So what we're going to do here is head back across to our editor, we're going to highlight our entire torque table that we're working with, in this case mapped point zero, control and C, we'll copy those values, let's head back across to our excel spreadsheet, and we're simply going to then paste these values into our original torque table, we can do this with control and V. |
| 08:09 | So now we've got the same numbers in our excel spreadsheet as we've got from our mapped point zero torque table. |
| 08:16 | Now the rest of this is going to require us to gather some data on the dyno, but the last table we see here which is labelled percentage change, what this does is it shows us the difference between the original values in the torque table, and the values that we've just calculated in our flywheel corrected torque table. |
| 08:36 | So essentially this will show us what error there is between what the dyno's showing us and what the original torque table is, we can then apply that error to the original torque table in our ECU, that'll correct that error and our torque values should line up. |
| 08:51 | Alright so let's go through the process now, we'll get up and running on the dyno and we'll see how we can do this. |
| 08:58 | Alright so the numbers that we wanna take notice of here, we've got our load being shown on our scanner, so that correlates with the load values on the vertical axis of our dyno torque values table. |
| 09:11 | We've also got obviously our RPM which correlates with our RPM values across the horizontal axis. |
| 09:18 | And then if we jump across to our dyno, we've got the value that we're going to be populating that table with, which is our derived torque value. |
| 09:26 | It is also really important to make sure that you are working in the same units. |
| 09:32 | We can display torque in both our HP Tuners VCM editor software, as well as on our dyno, in either newton metres or pound feet of torque. |
| 09:42 | We obviously need to make sure that we are populating our tables with the same units, otherwise we're going to end up with some very big inaccuracies there and the system isn't going to work. |
| 09:53 | Now obviously we've got a lot of data points to fill in here and it's also going to be impossible to access absolutely every load site in this table. |
| 10:03 | That's OK though, we don't necessarily need to access every point. |
| 10:07 | What we're looking for is a general trend that we can apply into our torque tables. |
| 10:12 | So for this demonstration what we're going to do is simply look at the values at 2000 RPM and we're going to fill in as many of those values as we can. |
| 10:21 | So we'll get up and running here and we'll just bring our RPM up on our dyno to 2000. |
| 10:31 | And once we've got ourselves up to 2000 RPM we're going to increase the throttle pedal position until we're at a load of 0.30 And it is obviously important while we're doing this to make sure that we're as accurate with those values as we can be. |
| 10:47 | We wanna be as close to the centre of each of these break points as we can get. |
| 10:51 | So right now we're sitting at around about 0.2 load and we can see that our derived torque's sitting at around about 20 newton metres, it's moving around a little bit. |
| 11:01 | Alright what we can see here is on the dyno, about the lowest load point I'm going to be able to get to is 0.3, can't quite get down to 0.2 If I do that I simply don't have enough torque to keep the rollers turning and the engine RPM just decreases. |
| 11:18 | So we'll start from 0.3 load here, 2000 RPM. |
| 11:21 | I'll just try and get as accurate as we can into the centre of that cell. |
| 11:26 | And we'll have a look at our torque values on our dyno. |
| 11:29 | And we can see our torque is moving around a little bit but it's pretty close there to 140 newton metres so I'll enter that value there. |
| 11:37 | Now we can increase our load a little bit here up to our next point which is 0.5, again making sure that we stay on top of our engine RPM, it stays nice and accurate at 2000. |
| 11:48 | Come up to 0.5, just allow everything to stabilise there, and again our torque on our dyno is moving around a little bit. |
| 11:56 | I'm gonna average that and call it 245 newton metres. |
| 12:00 | Come down to our next point which is a load of 0.6 so I'll just apply a little bit more throttle to get us to that point. |
| 12:09 | Again just watching our engine RPM doesn't creep up there as well. |
| 12:13 | And with our load at 0.6, just averaging our numbers on the dyno, I'm gonna call that 295 newton metres, so let's enter that value in there. |
| 12:26 | Come down to 0.8 And 0.8, 2000 RPM, looks like we're around about 390 newton metres, so we'll enter that value in. |
| 12:45 | And the last point that we're gonna be able to get here at 2000 RPM is a load of 1.0 and we are about 460 newton metres there so let's enter that value. |
| 13:00 | OK we'll bring the car back to idle here and we'll shut it down and have a look at our results. |
| 13:06 | Alright so obviously that's a fairly small slice out of our table there. |
| 13:11 | But let's just have a look at what's happened with that data. |
| 13:13 | So when we move down, first of all we can see that our corrected flywheel values have automatically populated, remembering those are just the numbers that we've entered from our derived torque, multiplied out by our dyno correction factor of 1.35 We'll slide down a little bit further and now the excel spreadsheet is directly comparing these values that we've just calculated, to the values for the same cells in our original torque table. |
| 13:42 | And what we're looking for here is the error. |
| 13:45 | So if we move down, we can see that we've got actually quite a nice trend forming there. |
| 13:51 | We've got an error between about 4% at the low end and around about 6% at the high end. |
| 13:57 | So this shows us how inaccurate or how much error there is between the torque tables inside of the ECU and the torque that we're actually registering for our engine in real time using our Mainline dyno. |
| 14:10 | So the process now is to apply that correction factor to the the original torque tables. |
| 14:16 | Now of course we have only looked at that thin slice and there is a variety of ways we can deal with this excel spreadsheet. |
| 14:23 | Particularly if we've got a lightly modified engine. |
| 14:26 | Then there's no real problem with just looking at the torque table in one or two places. |
| 14:32 | Perhaps we might choose 2000 RPM and then another point at 3000 or 4000 RPM and gather some data. |
| 14:39 | Assuming that we are seeing a relatively consistent trend in our error, then we can go and apply that error to the entire table. |
| 14:47 | Of course things are going to be a little bit more complex if we're making engine modifications that are going to have a dramatic effect on our engine torque. |
| 14:54 | And in those instances it might make a lot more sense to be a lot more accurate with our data that we're gathering and look at a lot more data points to really get a good feel for what's going on and what our error is. |
| 15:08 | I'd always be very suspicious if you're seeing really dramatic errors that are changing as we move through this table. |
| 15:15 | This is a really fiddly process to go through and we do need to be as accurate as we can gathering this data. |
| 15:22 | The better the data we gather on the dyno, the more accurate our results are going to be and the better our finished results So by looking at this data again, what I'm going to do is say that on average there, we're probably about 5% out and the torque table in our ECU is about 5% lower than what we're actually registering on the dyno. |
| 15:44 | So that's the important aspect. |
| 15:46 | Our torque table is set too low. |
| 15:48 | So what we can do now is head back across to our editor and we'll see how we can correct this table. |
| 15:54 | Now while of course you could correct that values in this table directly inside of the table here, we actually don't want to do this. |
| 16:01 | We want to use the inverse torque calculator function, which we can get to by going to the edit menu, we'll go down to torque inverse calculator and here we can select originally or to start with our torque table group. |
| 16:16 | So remembering that we're on map point zero so this is the table that we want to be modifying. |
| 16:22 | Of course we have all of our other tables available that we can select. |
| 16:25 | So what I'm going to do is select that entire table and we are going to multiply this out by 1.05 Once we do this we see that the torque to inverse calculate option becomes available, we can now click this button and the inverse calculator will automatically populate our inverse table map point zero with the correct values. |
| 16:48 | Now of course as we've discussed here, we're going to make the assumption that if our engine torque zero table, map point zero table was inaccurate by 5%, we're going to go through and make the same adjustments to the other torque table. |
| 17:04 | So we'll be making that exact same change to engine torque tables one through to four and of course calculating the inverse for those tables as well. |
| 17:13 | Let's jump ahead, we will make those changes and flash them into the ECU, and we'll have a look at the results on our scanner once we've made those changes. |
| 17:23 | Alright we're back up and running at 2000 RPM, we're looking at the same point that we just gathered data for our excel spreadsheet. |
| 17:30 | We want to have another look at our IPC wheel torque error parameter. |
| 17:33 | So right now that's sitting at zero newton metres, and you'll remember in our first test, as we applied load, this became quite large, the error value became quite large. |
| 17:43 | Ultimately we're always aiming for this to be zero or as close to zero as we can get it. |
| 17:49 | That's what we should see with a well calibrated torque table. |
| 17:52 | I'm just gonna apply some load here, and we'll see what happens with that wheel torque error value. |
| 17:58 | So we see that it does move around slightly but it's always very very close to zero, certainly much closer than what we saw with our first test. |
| 18:08 | And this is an indication that we've got our torque table well calculated or well calibrated. |
| 18:16 | Now I'll also mention that the IPC wheel torque error is an error that's cumulative. |
| 18:21 | So if we've got an error that's occurring for too long, this is likely to trip a limp home fault code in our ECU. |
| 18:29 | Now this is why one of the steps we went through during our configuration process was to raise, back in our editor, our wheel torque error max value, we've set that at 500000. |
| 18:42 | So we're setting that at a value that's going to allow some room for that IPC wheel torque error value if it's not sitting exactly at zero all of the time. |
| 18:53 | Alright so at this point you should have a process that you can apply to modify and calibrate your torque tables. |
| 19:00 | Much like the other aspects of the calibration that we've looked at so far, this is also an iterative process. |
| 19:07 | You may not get this done in one adjustment. |
| 19:10 | So if you go through and you've still got significant IPC wheel torque error values then that would indicate that there's still some work to do. |
| 19:18 | The last point I'll make here is that once we've got our torque tables calibrated accurately and we really understand what the engine torque is doing, this is another opportunity for us to go back and re address our driver demand table, remember that we also looked at that during our configuration process, and we want to make sure that our driver demand table is realistic, and that we are actually targeting torque values that are at or slightly above the torque the engine is producing. |
| 19:49 | Remembering if our driver demand table is targeting torque values that are less than the engine is capable of producing, then the ECU can actually reduce torque or close our throttle body, and this will affect the power and torque the engine actually delivers on the dyno. |
| 20:04 | So at this point in our tuning process, we've completed all of the work necessary on the dyno. |
| 20:10 | Everything's calibrated correctly, we've got a nice increase in performance over our original tune, albeit with a safer state of tune with less knock occurring while the engine's running under high load. |
| 20:22 | In the next step, we would take the truck off the dyno out into the real world and gather some data to make sure everything we see out in the real world confirms what we saw here on the dyno. |
