Boost Control: Open Loop Boost Control Tuning
Watch This Course
$79 USD $39.50 USD
-OR-
Or
8 easy payments of only $4.94
Instant access. Easy checkout. No fees.
Open Loop Boost Control Tuning
13.49
| 00:00 | - Right, now that we've looked at how the boost control system and the M1 is configured and set up, we've got a system ready to go, and we're going to look at how to start tuning that on the dyno. |
| 00:11 | We're going to look at this in two ways, we're going to start with an open loop system, and as you already know, this is going to be an important first step to our closed loop boost control anyway. |
| 00:22 | So let's just have a quick look again at some of the parameters that are important to this. |
| 00:27 | We've got our boost activate, which I've set at 10 kPa. |
| 00:30 | We've got our boost margin at 20 kPa, so there is the margin above and below our boost aim, and we've got our boost maximum, so this is our overboost cut. |
| 00:41 | Now, again, before we start tuning any boost control system, it's important to set up an overboost cut-out, just to protect the engine in case we've got any of our setup parameters wrong, or something goes wrong with the wastegate and we end up with more boost than we want. |
| 00:55 | When we're on the dyno, we've got a lot of things going on, and it can be easy just to have the engine briefly overboost, and we don't want that to do any damage, so at the moment I've got that set up at 90 kPa. Now, the example we're going to be looking at on our Toyota 86, it's a turbocharged naturally aspirated engine, which is quite high in compression. |
| 01:17 | So we're not going to be targeting a lot of boost pressure, and we're not going to be moving the boost aim very far above our wastegate spring pressure. |
| 01:27 | But we'll see how that progresses anyway through this worked example. |
| 01:31 | So at the moment we have got our boost aim table, our boost aim main set to zero. |
| 01:37 | Now, remembering at this point, we're looking at open loop control, so this table doesn't really come into play, because the ECU isn't using any closed loop control to try and track our target. |
| 01:51 | So, we've got our feed forward table here, which is the duty cycle that the ECU will be applying, and we've got our proportional, our integral, and our derivative gains. |
| 02:02 | So, because we are looking at open loop control here, we want to start by setting the P, I, and D parameters all to zero, which means that we've got no feedback. |
| 02:12 | So I've done that already, we've got those all set to zero, and our feed forward table is set currently at single value of zero. |
| 02:20 | If we press the A key, we can set this up as a 3D, 2D, or 4D table, depending on what we're trying to achieve. |
| 02:30 | I'm going to enable the engine speed axis, and we're going to enter starting value of 2,000, and finishing value of 7,000, and 500 RPM increments. |
| 02:44 | Now, when you're setting up the boost aim table, or the feed forward table, we don't really need to worry about what's happening until we get to a point where we can actually start getting reasonable boost, generally an area slightly below where we're first going to achieve peak boost pressure. |
| 03:02 | That's why I've selected 2,000 RPM as the first point in that table. |
| 03:06 | Doesn't really matter what's happening below that, because I know I can't get anywhere near my aim boost value. |
| 03:13 | Okay, so now that we've got that all set up, again at the moment, we're going to be in open loop mode, and the ECU is not requesting any duty cycle at all, as you can see from the zeroes in the feed forward table. |
| 03:26 | So we're doing our first control runs setting up, just to see how well the engine is capable of controlling boost on its own. |
| 03:36 | Alright, so we'll perform our first run, and then we'll have a look at the time graph, the data logging, and see what we've got. |
| 03:58 | Okay, so we can press the T key, and we can pause the time graph, and I'm just going to press F6, and expand it out so we can see it a little bit more clearly. |
| 04:09 | So what we're really looking at is what our boost was doing during that run, remembering we've got absolutely no control being applied by the ECU at this point. |
| 04:17 | What we're really looking for is reasonably stable and consistent boost, and you can see that we've got our inlet manifold pressure here. |
| 04:24 | This is inlet manifold pressure absolute, and down the bottom here, on this graph here, group here, we've got our boost pressure kPa gauge. |
| 04:33 | So this is our gauge pressure, again this is just a reiteration that the M1 is looking at gauge pressure for any of our boost aims. |
| 04:42 | So, you can see that we do have reasonably stable control, which is exactly what we would like to achieve. |
| 04:49 | Boost pressure reaches a peak of 51 kPa gauge, and it drops very slightly down to about 44, 45 kPa at 7,000 RPM, which was the top of that run. |
| 05:02 | So, now that I've established that we've got two fundamental aspects that we know of, first of all, our engine is capable of providing stable boost, and secondly, that we know what our base boost pressure is on our wastegate spring before we apply any electronic control whatsoever. |
| 05:21 | Okay, so let's now look at what we can do to achieve a aim value of 60 kPa gauge. |
| 05:29 | So again, this is just a defence to the fact that this is a naturally aspirated engine, and we don't want to be putting a lot of boost pressure into it. |
| 05:38 | Now despite the fact that we are not using closed loop control, even so, I would still like to see my boost aim main value, to 60 kPa. |
| 05:51 | Now, it's not going to have any effect on our control whatsoever, but what it will do is it will display our boost aim on our time graph, and that's just going to make our job a little bit easier when it comes to deciding where we are in relation to our aim. |
| 06:06 | Now the next thing we can do is go to our feed forward table. |
| 06:11 | So again, what we want to do is start by requesting a change in duty cycle, so that we can assess how much additional boost that gives us. |
| 06:20 | So, in this case, I'm going to start by requesting a duty cycle of 15%. |
| 06:27 | Again, remember this M1 ECU will remove the deadband at the bottom of the solenoids control range, so when we enter a value of five or 10 or 15 in here, that is actually what we're going to be getting. |
| 06:41 | So we're gonna start with a small change there of 15, and we can start the time graph running again, and we'll do another run with our dyno, and we'll see what that gives us as a result. |
| 07:06 | Okay, so we've performed our second run there, and we'll pause the time graph again and we'll expand it out and see what we ended up with, and we actually see that we've really had no change in our boost pressure at all. |
| 07:20 | In fact, our peak boost was actually very marginally lower. |
| 07:23 | It's one aspect that I did need to, I should address right now as well, is when we are doing these open loop tuning runs on the dyno, it's important to always start from as close to the same parameters as we can. |
| 07:36 | So what that means is, we want to start with the same amount of engine temperature, we want to start with relatively similar intake air temperature, and it's always best if we run the car for a little while and get a little bit of load into the engines, some heat into the exhaust, and the turbo charger before we do a run. |
| 07:55 | So, we've seen very little effect from that 15% change, and that's not uncommon with some boost control systems. |
| 08:03 | As I said, there's no, in the main course, there's no one set of number we can apply to all situations. |
| 08:11 | So what I would do when I've seen no change from duty cycle target there of 15%, I'd make another change. |
| 08:19 | So in this case I'm going to increase that to 30%, and we'll do another run and see how that works out. |
| 08:40 | Okay, so we've performed our third run. |
| 08:42 | We'll again pause our time graph, expand it and have a look at it. |
| 08:46 | Now this time we have seen a small increase to our boost. |
| 08:50 | We've now gone up to 52.6 kPa. |
| 08:55 | You can see we've still got a general trend with this particular turbo charger, and this engine where the boost will drop off a little bit in the higher revs. |
| 09:04 | And we're going to use the duty cycle feed for a table to correct that. |
| 09:10 | So, we've seen an increase, we're still below our aim, and we've only seen a recently small increase thanks to that additional 15% duty cycle. |
| 09:20 | So I would simply make the same change again, and this time, we're going to target 45% duty cycle. |
| 09:27 | Before we do that, let's try and speed up our process a little bit, though. |
| 09:30 | You can see that between right about 4,000 and 5,500 RPM we're sitting close to our peak of 52 kPa gauge, and it's... |
| 09:43 | Often this higher rev range, from about 6,000 RPM and up, that we drop off, so what we're going to do, in this case, I want a flat boost curve, and we're going to achieve that by commanding a little bit of additional duty cycle from 6,000 RPM and above. |
| 10:03 | And you know, I'm just going to select between 5 and 6,000 RPM, and I'm going to press the I key to interpolate those values, so we get nice, smooth control. |
| 10:13 | So this will help pull our duty cycle up, at higher RPM enhance also increase our boost. |
| 10:20 | Okay, let's do another run and see what that gives us. |
| 10:40 | Okay, let's have a look at that on the time graph there, I've paused there and again we'll enlarge it. |
| 10:44 | Now you can see we've got control pretty close to our target. |
| 10:49 | We're coming up to 60 kPa, we actually overshoot very slightly, so we're at 62 kPa from about 5,000 RPM, so I'm looking at the RPM here. |
| 10:59 | 5,000 RPM through to about 6,250 RPM. |
| 11:04 | And then in the higher revs, we're right back on our target. |
| 11:09 | So, when we're doing these runs on the dyno, I should also point out, because we're not 100% sure exactly what boost we're going to end up with, we do want to keep an eye all of the time on the boost pressure the engine's running, either on the dyno or on our ECU software, and be ready to abort the run if, for any particular reason, we end up with more boost than we wanted. |
| 11:32 | In this case, I'm only two kPa over my target, so I'm quite happy with that. |
| 11:36 | So what I'm going to do is, from 5,000 to 6,000 RPM, we will pull a little bit of duty cycle back out of that. |
| 11:46 | So let's highlight those, and I will take 3% out. |
| 11:51 | Now, the change that I'm making there, I'm really just basing my decisions on what I've seen from the increase and boost relative to a 15% increase in our duty cycle. |
| 12:05 | So it's really a case of starting to feel out the system, and understanding what sort of duty cycle change results in what sort of boost pressure change. |
| 12:15 | And it is an iterative sort of test-and-see process, we can't sort of get it exactly right in one run. |
| 12:22 | However, we are getting pretty close to our target now, so we'll do one more run and see how that works out. |
| 12:47 | Right, we'll pause that time graph and have a look at it. |
| 12:50 | And you can see now boost pressure is very very stable, and very very close to our aim right through that run. |
| 12:59 | Anything right in the top end, we're down about two kPa. |
| 13:03 | And that's gonna come down to how fussy you want to be with those targets, remembering it is an open loop system, so we can still expect some fluctuation day to day, and depending on the amount of load we've got on the car. |
| 13:17 | So, in essence, we've done though, we've got a boost control system set up in open loop that will give us reasonably accurate control within a couple of kPa of our boost target. |
| 13:31 | And that's about as much as we can expect to achieve with open loop. |
| 13:35 | In the next module we'll look at how we can build on this, and we can now add close loop control to improve the accuracy as those conditions change that the engine's subjected to. |
