Summary

It’s increasingly common to convert existing cable throttle equipped engines to drive by wire. In this webinar we’ll take a look at what you’ll need to complete this work.

00:00 Hey team, Andre from High Performance Academy here, welcome to another webinar and this time we're going to be looking at drive by wire or electronic throttle conversions and these are becoming more and more commonplace as access to factory drive by wire throttle bodies and drive by wire pedals becomes more and more prevalent.
00:19 These days you're not going to find too many cars that roll off the showroom floor that don't have drive by wire throttles so obviously by definition we're getting access to more and more of these parts.
00:30 Now, the first thing before we actually jump into the process of converting to drive by wire, I guess a little background on why we may want to and how it works.
00:40 So, traditionally our throttle body was always actuated by a cable, a physical mechanical linkage between the driver's foot pedal and the throttle body itself.
00:49 And within reason that does work really well, it's been well and truly proven.
00:53 But as electronics advanced and we got the first OE style drive by wire throttle bodies or throttle systems, there are some advantages in doing so and particularly when we're getting a little bit more complicated with our electronics package and particularly when our engines are potentially making more torque than we could put to the road in most circumstances, being able to control the drive by wire or throttle opening more accurately using our ECU does give us some distinct advantages.
01:25 Now, every time we talk about drive by wire, inevitably there are a vocal minority that will point out that drive by wire is slow, lethargic, slushy, doesn't give you much feel and that a cable throttle is superior.
01:41 And of course everyone's entitled to their own opinion on this.
01:45 I think a lot of this comes from maybe the early days of factory drive by wire throttle systems where yes they could feel a little bit doughy or a little bit sluggish.
01:56 The reality was that, at least in my experience, that's very seldom to do with the actual electronics themselves and it's more to do with the programming in that OE application.
02:09 Often we can strip out the electronics package from one of these cars that felt doughy and unresponsive, put in a quality aftermarket ECU, set everything up with exactly the same inputs and actuators and find that the drive by wire actually feels just nice and sharp and crisp.
02:25 I would say that with modern technology as well, if we're looking at maybe an off the shelf Bosch Motorsport drive by wire throttle body, something that's been well set up, it'd be very difficult if not impossible for most drivers to tell the difference between cable and drive by wire.
02:42 So, that's the first, the lag or the perceived lag, really it's just not a thing, it's not something that's going to be holding you back.
02:52 The second one that is claimed is safety and yes there is some potential problems around safety with drive by wire but there are strategies involved which we're going to get into as we go through this lesson that basically eliminate that potential.
03:09 So, this is where we've got dual sensors on both our throttle pedal, both of these have to basically match what the ECU's expecting, otherwise it detects a fault and at the same time we have exactly the same in the throttle position sensor on the actual throttle body itself and again if those don't match what the computer's expecting, it knows there's a fault, it'll allow a little bit of error to creep in but as that accumulates, it will eventually decide that there is a problem with the drive by wire and it will shut that down.
03:41 There's a couple of ways it can do this, electronically it can force the throttle closed or if there is a physical problem with the throttle body that means that it can't be closed by the ECU, then it can instigate an engine rev limiter, maybe a 1800 or a 1500 RPM rev limiter, that's going to in most instances give us the safety strategy that we want.
04:04 Now, the people that argue that drive by wire is unsafe, probably also overlooking that, I don't have statistics on this but I could be pretty confident that more cars have crashed as a result of a mechanical cable jamming than have failed or crashed because of a drive by wire error as long as it's been properly installed and those safety strategies have been employed.
04:29 So, realistically I put a tick in the box for drive by wire over cable throttle any day of the week.
04:37 Alright, so this doesn't really answer the question of why though, basically we're looking at two technologies that essentially achieve the same thing in very different ways and there is no argument that drive by wire is certainly going to be both more expensive and more complicated than a cable throttle.
04:54 So, if it's more expensive and more complicated, where's the upside? Well I'm glad you asked because there are some significant ones.
05:03 And one of these is that we can get essentially indefinite or infinite control of the relationship between the driver's foot pedal position and the throttle body opening.
05:15 So, let's just jump into some software here, I've got a couple of examples here, basically all of the ECUs that control drive by wire give the same functionality so you're going to see something similar.
05:27 This is in the Link G4 Plus software, it's a little older now but it's happened to have a calibration that works so let's have a look at how this works.
05:36 So, what we've got is on the vertical axis here, it's listed as AP main, stands for accelerator pedal, so that's the actual driver's foot pedal position, so where we are on the actual physical pedal.
05:51 So, obviously we're spanning there 0-100%.
05:54 Now, on top of that we also have the engine speed axis.
06:00 Now, that's not necessarily a requirement, we could just have this as a simple 2 dimensional table where our throttle position which I'm talking here about at the throttle plate is mapped against the accelerator pedal position.
06:15 By bringing in the engine RPM we do gain a little bit more control of the torque delivery so this can be helpful with turbocharged cars for instance where maybe they come on boost very sharply, we can manipulate the throttle opening around that area to smooth that out for example.
06:34 Another example of where we can use this 3 dimensional table, many years ago in my previous business I was the parity manager for a New Zealand V8 touring car series.
06:44 And in that series it was a controlled tube frame chassis with a variety of different engines and then a variety of different bodies.
06:52 And my task was to basically tune these engines, at the time we had a LS3 6.2 litre V8 from Chevrolet, we had a Ford Coyote V8 5 litre and we had a 5 litre Toyota or TRD Grand Am V8 engine.
07:12 So, very different engines with very different rev ranges and very different power and torque characteristics.
07:20 The aim was to match the power and torque delivery of those three engines within 1% throughout the rev range.
07:26 And we did exactly that by manipulating the relationship between the throttle plate opening and the driver's accelerator pedal position versus engine RPM.
07:36 So, basically if we had too much torque at let's say 3500 RPM and 100% throttle here, what we could do is basically take this particular zone here and we could bring that down to maybe 80% or whatever we needed to to match that.
07:51 And we could get really really granular with that and actually got exceptional results.
07:58 Now, of course when you've got a parity series like this, it's very difficult to keep all of the competitors happy anyway.
08:05 There were still claims of lack of engine parity but we could run them back to back on the dyno and essentially the cars are all made exactly the same power.
08:14 Alright, so that obviously is a fringe or very very niche use of drive by wire.
08:20 But the other thing we can do, often or with most ECUs that offer drive by wire functionality is that we can have multiple positions on a steering wheel switch or something like that.
08:33 So, what we might want to do here is for a wet race, we might want to remove a lot of the throttle position around maybe this mid range here where the driver is only at maybe part throttle, 40, 50, 60, 70% throttle, maybe 3 to 4500 RPM, and by reducing the actual throttle opening in that area, it's going to reduce the torque and it can make a car that otherwise could be quite peaky or hard to drive, it can give us a lot more control over the car.
09:06 And of course on a steering wheel switch you can have multiple positions like this so you could have a very reserved calibration, maybe even something like a valet mode where no matter what the driver does on the throttle pedal, the throttle body only opens to 30%, enough to drive the car around but not enough to go and actually thrash it.
09:28 So, really it's sort of a case of, it's up to your imagination how you go ahead and use this.
09:36 These days we're also seeing the use of drive by wire for torque management.
09:43 So, most late model factory cars now use a torque based programming or tuning model.
09:50 And that's all about basically the ECU calculating what percentage to open the throttle body in order to achieve the torque that the driver is requesting.
09:59 So, there's not necessarily a direct relationship between what the driver's doing on the foot pedal and the throttle plate opening all of the time.
10:10 Now, a good example of this, a little bit more complicated, but in Motec's GPRP Pro, which we run in our Toyota 86, this is torque based and I'll just show you a little bit of data from one of the last times we were at the track.
10:25 And we've got probably a little bit more information than we actually need here so I'll just get rid of a few aspects of the data, that'll do for the moment.
10:36 So, what we're looking at here which is the important part is our throttle pedal which is green, that is what the driver is doing on the foot pedal.
10:46 Then the yellow is our actual throttle position, so where the throttle body is at.
10:50 So, what we can see, and this is a classic example of the torque based model in action.
10:56 So, at this point here, the driver, me in this case, has gone to wide open throttle.
11:02 And we can see that the yellow throttle body opening also goes wide open, well 89%, pretty close.
11:08 But we can see also if I just zoom in a little bit, just how brief that is.
11:11 So, basically at this point, and I don't have the requested and delivered torque unfortunately in this datalog, at this particular point we've requested X amount of torque, let's say 300 Nm, whatever that may have been.
11:25 Because the turbo is not spooled up, the throttle body goes to wide open throttle in order to achieve that torque target as quickly as possible.
11:33 And what we can see here is it achieves that and then it actually starts closing the throttle body down to maintain that torque.
11:39 The reason for this is if we stayed at wide open throttle, the engine continues to build boost and we would blow past our 300 Nm torque target.
11:47 So, it closes the throttle down, in this case all the way back down to 55%.
11:50 And then as we go through the rev range we can see it opens up a little bit more.
11:55 This actually is not quite tuned for our torque targets perfectly at higher RPM.
12:00 But the reason for this is another nice little trick that we can do with drive by wire throttle.
12:06 What we'll find, if we go onto the dyno with a turbocharged engine, and we go to 3500, maybe 4000 RPM, somewhere where the turbo is capable of making full boost, and we go to wide open throttle, and let's say we end up making 400 Nm of torque.
12:22 OK so we're at wide open throttle.
12:24 Now, if we close the throttle down to 80%, we will probably find that, give or take, we're still making almost exactly the same amount of torque.
12:33 Maybe we'll lose 10 or 15 Nm but not enough that you would actually notice.
12:38 And it's not probably until we get down to about 65, maybe 70% throttle opening that we actually start seeing the torque drop away significantly.
12:48 So, what that means is that if we're at wide open throttle and we start to lose traction, maybe through a fast corner or something of that nature, we actually have to back quite aggressively out of the throttle before we reduce the torque enough to reinstate traction and accelerate the car again.
13:05 Now, the idea with the torque based model that MoTeC is using is that while we're at wide open throttle, the throttle body will actually be closed partially, maybe around to 70, 75%, haven't quite got that here as you can see, we're asking still for a bit too much torque, we're at 88.6% throttle opening.
13:24 And the idea there is we're closer to that point where as we close the driver's foot pedal position, we'll see a more instantaneous and linear drop off in our torque being delivered.
13:35 And that means that it's much easier for the driver to modulate the throttle pedal and manage a car that has more torque than it can put to the ground in just about any circumstance.
13:46 So, it gives the driver a greater amount of control.
13:50 Also opens ourselves up for other strategies where maybe we want to incorporate throttle opening and closing as part of a traction control model or something of that nature, we're right on that point where if we close the throttle 5%, we're actually going to get a worthwhile reduction in our engine torque.
14:08 So, that's the idea behind it.
14:10 Now, one of the problems that comes along with this as well, if I can draw a little picture here, I'll try, is the relationship between airflow and throttle opening.
14:21 So, let's try and draw a straight line, that'll probably be helpful.
14:26 If we draw a graph here and we've got our, let's call this throttle opening, and this one here, let's call this airflow.
14:39 So, what we find is that the airflow through a throttle body is very very non linear.
14:44 And what I mean by this is if we've got the throttle body completely closed and we crack it just a little bit, let's just go to our overhead here and see if we can sort of demonstrate this.
14:54 So, our throttle body's closed at the moment.
14:56 We crack it just a little bit and we actually end up with quite a large increase in our airflow.
15:01 And as we go from zero through to 90°, as we open that we get a very non linear increase and then once we get closer to 90° to wide open throttle, let's say between about that point there, maybe 80° and 90°, we're actually going to see little to no increase in our airflow.
15:20 So, if we plot that, we'll come back to my laptop screen, we're going to get something that sort of looks a little bit like this.
15:28 And excuse my shoddy artwork but hopefully you get the picture.
15:32 So, what we can do then if we wanted to actually get a linear relationship between this, something that looks a bit more like that, excuse the fact that for some reason now we've got more airflow than our previous example, I think you probably get the picture.
15:46 We can now manipulate that throttle plate opening using our drive by wire throttle target tables in order to achieve a linear relationship between airflow and throttle opening.
15:58 Now, do we want to achieve that? Maybe not.
16:01 It really depends on the engine itself, how much torque it's got and our own personal preference.
16:07 If we've got a low torque engine then probably if we had a very linear relationship between our torque and our throttle plate, the engine's going to feel dull, it's going to feel dead at part throttle and it's just going to be horrible to drive.
16:20 A lot of this does though come down to the specifics of the engine and then personal preference on the part of the person driving it but the beauty of drive by wire throttle is we've got infinite control over all of that.
16:33 Alright, so I think we've probably done to death the advantages and probably fair to say that I am an advocate for drive by wire.
16:41 What we need to now understand is how we actually go about installing or changing to drive by wire.
16:48 And there's a few different ways that we can go about this.
16:51 The first and probably most common would be just to source the componentry we need from a wrecking yard off an existing drive by wire car.
17:02 Nice advantage here is that these parts don't necessarily need to be from the same manufacturer.
17:08 So, for example, just get this under our overhead again, this I'm pretty sure is a drive by wire throttle pedal assembly, yeah it is, off a Toyota GT86.
17:19 Maybe sort of in the region of $150 from a wrecker.
17:23 And we've got mounting points on this so we should be able to adapt this relatively easily to the firewall.
17:31 We've got our pedal assembly itself.
17:34 Probably in most instances we're going to find that if we're adapting this into another car, the pedal itself is probably not going to be in quite the right position, maybe not quite the right angle to match our existing pedals.
17:46 But of course we're really only interested in the actual electronics of this.
17:52 It's very easy to mechanically adapt the pedal itself to suit our application.
17:58 The other thing we will want to be careful of is ideally buying something where we can source replacement connectors to make our wiring nice and easy.
18:07 OK so that's the first part of that, the pedal.
18:10 Then of course is the drive by wire throttle body itself.
18:14 I've got this one here which is off my Toyota FJ40 which is running a Toyota 3UZ -FE V8, 4.3 litre V8.
18:26 All factory drive by wire throttle bodies are going to come in different shapes and sizes and they may look a little bit different.
18:32 In this case we've got two separate connectors.
18:36 Over here is for the drive by wire servo motor so that's just a 2 pin.
18:39 Basically, what we want to do is connect 12 volts to one side and ground the other, that'll drive the throttle body in one direction and if we reverse the polarity of those, it'll drive it in the other direction and I'll talk a little bit more about control strategy soon.
18:53 With this particular design we have the throttle position sensor on the opposite side.
18:59 Most of the time we'll find that those will be integrated together so we have the throttle position sensor and the drive by wire motor control all integrated into a single connector and quite often that'll be a 6 pin connector.
19:13 One of the easiest sources if you're after new componentry that's going to be a guaranteed result, it's not going to be potentially damaged from maybe 100,000 miles of use, I would be suggesting you look at the Bosch Motorsport range of drive by wire throttle bodies.
19:29 As the name insinuates they are designed for motorsport use, that's what we use on our Toyota GT86 and they're available in a range of different sizes to suit different engine applications.
19:40 One thing I would say about the Bosch drive by wire throttle bodies, the motorsport ones, is the connector on them I have found to be a little bit less than reliable.
19:54 This is something that I've seen a lot of other people report on as well so I don't think it's just us.
19:59 We ended up going to the trouble of removing the connector itself and actually potting and running the connector out to, or running a flying lead out to an autosport connector to get around that so something you do want to watch, particularly if you've got an application that is very high levels of vibration, like we're going to see in motorsport.
20:22 Now, the key to both of these components, the drive by wire throttle itself and the pedal, is that they both have two position sensor signals in them.
20:34 This is sort of for redundancy or the safety strategy.
20:38 Generally, what we'll find is, let me just bring up my little note again, similar sort of thing here, we're going to have our voltage in.
20:48 So, one of the sensors is going to go from maybe something like 0.5 to maybe 4.5 volts as we go from 0 to 100%, let's call this throttle opening and we'll call this voltage.
21:02 And the other one is going to do probably almost the polar opposite.
21:07 Now, there's a variety of different ways this all works out but basically we've got a two sensor system so that we've got the ability to basically fact check or sanity check the movement and make sure that no matter where we are, both of those are adding up and as I mentioned, if they don't, then the ECU knows something's wrong and it's likely to go into fault.
21:28 Exactly the same thing here, despite this looking like a conventional throttle position sensor, this is a two channel sensor so we've got that redundancy or sanity checking.
21:39 How this works is going to be a little dependent on the particular manufacturer, that Toyota throttle body that I've just pulled up from memory, I set this up quite a while ago from memory, one of the channels will go maybe 0.3 to 4.2 volts and it's nice and linear, the other one I think from memory starts at maybe 4.5 volts and drops down but then it flat lines at about 50% throttle.
22:03 So, you've got to have to make sure that the ECU that you're using can work with the throttle body that you've got or you're intending to use, otherwise you can get yourself into a situation where there's a compatibility problem.
22:16 But again both of those need those two channels and that will give you your sanity check for safety's sake.
22:24 Now, I will point out that it is possible to kind of cheat and basically take one of those sensor outputs and run it into both of your accelerator pedal position inputs on the ECU or likewise both of your throttle position sensor inputs on the ECU and I do not recommend that because you're just stripping away that safety strategy that the OE manufacturer as well as the ECU manufacturer has strived and worked really hard to get in there.
22:55 So, don't do that, make sure everything is working properly.
22:59 Now, there are a few other options as well.
23:02 If we come across to my laptop screen here for a moment, if you've got a pedal box which is pretty common in race cars, this is the pedal box in our Toyota GT86.
23:14 And most of these traditionally have been set up for throttle cables.
23:20 In recent times though they have incorporated or given easy incorporation of a drive by wire system so that's what we've got right here.
23:29 We can see that we've got a mount for a drive by wire or throttle position sensor, again dual channel, a little bell crank on that and that simply attaches through that little rod there to our throttle pedal.
23:44 What we want to watch with these as well is often these pedal boxes will incorporate some kind of non linear bell crank anyway.
23:53 So, basically as we move the throttle pedal through its travel, we might not see a linear jump in our throttle position, or our accelerator pedal position I should say, versus throttle movement.
24:05 So, what I'm getting at is 5° of throttle pedal travel might not always add up to 0 .2 of a volt or something like that, might be non linear but again we can basically get around that with our throttle body, our throttle position, drive by wire throttle position mapping.
24:21 Right so that's a throttle, sorry a pedal box installation.
24:28 There are other options as well, just let me jump back to my notes here for a second.
24:32 Because if you've got an existing pedal box that doesn't really lend itself easily to adapting a throttle position sensor, jump across to my laptop screen, we can use linear potentiometers and I tried really hard to find the exact one but I couldn't.
24:51 This is a penny and jiles sensor, this is probably more likely to be used for shock position, damper position or something like that.
24:58 But you can get these in a range of different lengths and a range of different travels.
25:03 And importantly you can also get them in single channel or dual channel.
25:07 So, with these it's not that difficult to adapt these onto a pedal box and then you've got those dual channels you can just run into your ECU.
25:15 Alright, we'll head back across to our photos here.
25:17 The other drive by wire installation we've done in the not too distant recent future, past is what I'm looking for, we'll get there in the end, is our FD RX7.
25:28 And there are kits available to convert to take a GM drive by wire throttle body right onto the manifold but these run quite an unusual sort of triangular throttle body with three butterflies in it, primary and a secondary.
25:42 And I wanted to keep that so what we did here was build a bell crank assembly.
25:47 And in the next slide here, a little bit hard to see because it's really well hidden, what we used is a BMW E46 M3 S54 drive by wire throttle body actuator.
25:59 Bit of a mouthful but this is a remote actuator that's designed for the S54 six individual throttle bodies and it just actuates those via a rod.
26:11 So, we just adapted that, mounted it down below the power steer reservoir and then we can see we've got the rod here that comes up and mounts up to our little bell crank.
26:21 And that's actually worked out quite nicely.
26:24 One thing I would mention with this sort of remote application is we want to be mindful of where we mount the throttle position sensor.
26:34 That S54 actuator actually has a single channel throttle position sensor on it.
26:39 So, originally we used that for one channel and then we had the factory throttle position sensor on the Mazda throttle body which I think was mounted over the other side here.
26:52 And as you can understand, probably it all became pretty apparent once we had it up and running, you could start to see a small discrepancy, I mean only maybe 0.2% discrepancy between the two because there is always going to be potential for a tiny amount of slop in that assembly.
27:09 And when it was trying to control off the drive by wire actuator down the bottom and then using the backup, the one on the actual throttle plate as a reference, it just didn't work that smoothly so we ended up changing to a two channel Penny and Giles rotary sensor that mounted in place of the factory throttle position sensor and everything worked really nicely.
27:31 Now, to give you a sort of more real world example of this, first of all I've got here, if we go to our overhead, this is an aftermarket actuator for drive by wire from Gen-V who manufacture a lot of throttle body systems from the UK.
27:46 So, we've got our little bell crank actuator here.
27:49 Do have a position here to mount up a throttle position sensor.
27:54 And I'll just slide these into view.
27:57 This is what this kind of looks like as a full application.
28:01 This is a set of Gen-V throttle bodies for a Honda K20 or our Honda K20.
28:06 And of course if you are using the Gen-V actuator with the Gen-V throttle bodies, it is all just plug and play, all bolts up really nicely.
28:14 So, there's a variety of ways that we can actually go about adding drive by wire to an existing system depending on whether we are a single throttle body or individual throttle bodies and there's no necessarily right or wrong way of doing that.
28:31 Now, we do need to also understand what we actually need from our ECU.
28:36 And this kind of can get a little bit complicated.
28:39 We obviously are going to need an ECU that's capable of controlling drive by wire which these days you'd probably be hard pressed to find one that isn't.
28:49 But let's just jump across to my laptop screen and I just chose Haltech as an example here and this is their R5 VCU.
28:58 Pretty advanced ECU so clearly we're more than capable of handling up to dual drive by wires here but what we want to look for first of all is what's often called, we're usually called an H bridge output.
29:13 So, we need two H bridge outputs for each drive by wire throttle that we want to use.
29:19 So, in this case we've got four so this means we could run dual drive by wire throttles, pretty common on a lot of applications these days to have dual drive by wire.
29:29 So, the H bridge basically allows us to pulse width modulate the operation of the drive by wire throttle body or servo motor.
29:38 We can reverse polarity, we can run those H bridges either to 12 volts or to ground.
29:43 So, by running one to 12 volts and one to ground, that will drive the motor as I mentioned earlier in one direction, we reverse that and it'll drive it in the other direction.
29:51 So, that's the fundamental aspect that we need in order to be able to drive a drive by wire throttle.
29:57 The part that's easy to overlook though is that we also need four additional analogue voltage inputs.
30:04 So, that's for our accelerator pedal position sensors, remember we've got two at the accelerator pedal and then two also at the drive by wire throttle body as well.
30:16 Whereas traditionally we'd only need one.
30:18 So, basically that means we need three additional analogue voltage inputs compared to what we'd need for a cable throttle.
30:23 And if you're running an ECU that you've kind of already got close to maxed out with maybe sensor inputs, that's where you can get yourself into a position where you don't have enough sensors and you can't do that which kind of leads onto that situation I talked about earlier where people sort of cheat the system or don't incorporate those safety strategies which again, don't be doing that.
30:50 Alright, so that's what you need, basically an ECU that has enough inputs and outputs available and at least two H bridge outputs for each drive by wire throttle body that you want to actuate.
31:01 And of course it does need a drive by wire control strategy built into it but again generally anything late model is now going to be set up for drive by wire applications.
31:12 We will go into questions and answers shortly.
31:17 So, this is a good point to sort of remind you if you do have any questions on anything related to the topic, please ask those in the chat and we'll get into those in a moment.
31:26 The next bit that we need to talk about is how we actually calibrate a drive by wire throttle system.
31:31 So, let's go back into our Link G4 Plus software and if we open our menu here, we can see we've got a section that is dedicated to our drive by wire or electronic throttle as Link refer to it as and what we want to do is go into eThrottle1 and we can go to eThrottle setup.
31:49 So, in the Link software, and everyone does this a little bit differently, we've got our eThrottle mode and if we click on that we'll get our drop down menu.
31:58 So, we can have it on, off or in this case we're in setup mode.
32:02 Setup mode does what it says on the label pretty much, it is used while we're setting this up but in setup mode, and it'll give you plenty of warnings about this, it does disable the safety strategies so obviously once we've got everything set up, really important to change that to the on mode.
32:19 Coming through this we've got the number of active tables.
32:22 So, as I mentioned we can have multiple tables driven off switches or rotary knobs or whatever, in this case for simplicity we have one single table.
32:31 This is the bit where we need to be a little bit careful.
32:34 So, we've got proportional, integral and derivative gains.
32:38 And this is really key to the accuracy of control of our drive by wire throttle.
32:44 Like a lot of closed loop control, or basically like all closed loop control strategies, boost control, idle speed control, continuously variable cam control, we're using this PID algorithm to decide what to do, what to send out to the drive by wire throttle in order to move it to match the new target that the accelerator pedal has set without undershooting or overshooting, while maintaining as fast a response as we possibly can.
33:12 The problem is that with PID control, if we get this wrong, we can end up in a situation where we get very poor control of the throttle body.
33:22 And in a mission critical safety aspect like the throttle actuation, we don't want to be risking that.
33:30 For this reason what we'll find is that most ECU manufacturers, when you're setting up your drive by wire throttle body, will have a dropdown menu and they will list the part number for the particular throttle body.
33:43 So, it might be a Bosch drive by wire throttle body and then the Bosch part number.
33:47 So, what we want to do is basically find the part number on our throttle body and make sure that we've selected the correct one.
33:54 And the reason for this is that the ECU manufacturer will have actually bench tested each of those throttle body part numbers and basically set up PID gains that suit that throttle body which means we're going to get good accurate fast control without overshooting or undershooting.
34:13 Really really important, unless you are very very confident with PID control tuning, this is something that I would recommend that you leave to the ECU manufacturer.
34:24 If they don't have your particular throttle body part number, then quite often they will offer a service where for free or for a modest fee, you can send the throttle body into them, they will bench test it and send it back with the correct PID gains that you can just enter and then you'll know it works.
34:43 Takes all of the risk and anxiety away from setting something up and you're going to know it will work.
34:48 The other aspect with this, it's not just about safety because what you'll find is that if your PID control gains are not accurate but aren't resulting in severe overshoot or undershoot which could be a safety element, what you might still find or most likely will still find is there will be some error between the throttle body requested target and what you're actually receiving.
35:12 And this is another part of the safety strategy where if there's a significant error and it's there for too long, again the ECU will go hey something's not right here, I'm not comfortable with this and it will go into that safety shutdown.
35:25 Which can be a little bit frustrating if it's a road car, can be pretty devastating if it's a race car so you really want to make sure that you've got this right.
35:34 We'll come back into our software again here.
35:37 We've got our dead band so basically what this means is that if we're targeting say 10 % throttle, with a dead band of 0.1%, the ECU will not try and make a change if we are between 9.9% and 10.1%.
35:53 Basically, it's just a little dead band that's going to be quite tight to our actual target where the ECU is comfortable that we're close enough to our target, we don't have to worry about that.
36:03 We've even got our minimum and maximum clamps, so this is the duty cycle being sent to the drive by wire servo motor.
36:09 So, typically we would normally leave this at plus or minus 100% but you can use this or manipulate this to try and reduce overshoots if you're struggling with that.
36:21 I find this is a bit of a band aid and the other aspect is if we artificially limit the maximum and minimum duty cycle, what this can do is therefore make the control a little bit more lethargic, it can't drive the throttle open and closed as quickly as you'd like.
36:38 Then we've got when stalled, so in this case quiet throttle, if we open this up, basically a lot of strategies because the drive by wire, if you actually hear it with the engine not running, it emits quite an audible high pitched humming noise.
36:54 And that can be a little bit off putting so quite often when the engine is stalled, after two or three seconds the ECU will shut down control of the drive by wire motor to reduce that noise, hence quiet throttle.
37:06 Whether you do that or not really is just personal preference.
37:10 Now, we've also got some control strategies here so if we've got, sorry some setup strategies, we've got throttle position sensor so looks quite confusing.
37:19 We've got our TPS main and sub open and closed so that just incorporates these voltages here so we can calibrate this.
37:28 But there is a little calibrate icon here which will go through and do this for you so you don't have to manually set these up.
37:35 And then we've also got our TPS sub 100%.
37:38 So, this is for applications like that Toyota 3UZ throttle body that I mentioned earlier where they will actually max out at a certain point.
37:48 Instead, of being a full 0-100% travel, you can set that up through the Link ECU.
37:54 And again every ECU has its own different ways of doing this but most of it's going to look pretty similar.
38:01 Generally, you'll find there will be a setup wizard for your drive by wire where you run that wizard and it will tell you to go to wide open throttle, press a button, go to closed throttle and press a button and then it will generally do this automatically at the throttle body as well.
38:18 Interestingly doing that automatically can actually end up, let's see if I can show you this, quite often these throttle bodies, if you go and drive it all the way home, probably a little hard to see but at its maximum travel, this is actually a little bit beyond 90%.
38:38 So, if the ECU is doing an automated, sets it to maximum travel in one direction and decides that's 100% and then goes the other way and drives it closed, that's 0%, you're probably going to find that you'll be a little bit beyond 100% throttle.
38:54 Does that matter? Probably unlikely that it's going to show you a difference in power but for example with the Motec M1 when we go through this, we disable the drive by wire throttle actuator and then manually hold the throttle plate in the closed position and then we manually hold it at that 90° position so it's as accurate as it can be.
39:16 Alright, we'll jump into questions now, if you've got any more, please keep them coming.
39:27 David has asked, "Is there a way to program the throttle to open when on overrun with the injectors off to reduce pumping losses?".
39:36 Yes you can.
39:40 It's going to depend on what the ECU is using as an input for the overrun strategy.
39:48 So, whether it's going into overrun based on the throttle position or the accelerator pedal position.
39:54 I should have also mentioned this, naming strategies become quite important here.
39:58 So, generally for drive by wire systems we're talking about TPS being throttle body position and we're looking at APP being accelerator pedal position.
40:08 So, whereas a conventional cable throttle we obviously only talk about throttle position, now we have to distinguish between the two.
40:15 So, as long as we can make the closed loop system go into closed loop overrun or fuel shut off based on our throttle pedal position, then yes that wouldn't be a problem.
40:26 We actually do this with a closed loop control strategy for gearbox control.
40:35 So, if we've got a paddle shifter transmission, particularly for the downshifts, what we actually find is that by artificially increasing the minimum throttle body opening, we end up basically reducing the vacuum in the intake manifold and then when there's a blip to match revs for a downshift, the engine can actually respond quicker and match revs faster than if the throttle body was completely closed.
41:00 I haven't looked too far into what you're talking about there in terms of reducing pumping losses but short answer to that is yeah it's probably very possible but the way you're going to go about it will be dependent on the particular ECU you're dealing with.
41:15 F Fernandez has asked, "Can you use the throttle as a sort of traction control, I would say coming off a corner, but even if you're 75-90% pedal open, the throttle will slowly open?".
41:27 Probably not in the way that you're thinking there.
41:31 I would definitely not be an advocate for artificially bringing in a lethargic or purposely slowed down opening or closing of the throttle, because as a driver you're going to detect that, you're going to sense it and it will make you feel very, very uncomfortable.
41:46 You want a very very quick response to any input you put into the pedal.
41:53 You can use this as part of a traction control strategy but what we do need to understand is that even drive by wire throttle, there is always going to be some level of latency to its operation compared to maybe bringing in an ignition retard or an ignition cut event or fuel cut for that matter which can be absolutely instantaneous.
42:14 So, generally with traction control strategies that employ throttle control, it's normally in addition to some of those other strategies so it might instantaneously use an ignition retard and then as the drive by wire closes down, the ignition can be phased back in so it becomes somewhat seamless but yeah you wouldn't want to slow down the opening of your throttle plate for a traction control strategy, I don't think that would feel very nice at all.
42:44 GW2's asked, "What about implementation on a car with an automatic transition and a kick down cable?".
42:51 Good question, actually something I haven't really considered too much there.
42:55 That might be a little bit tricky in terms of you don't have a bell crank that you can actuate the kick down cable off.
43:03 It might still be easy enough to incorporate that with an external bell crank but in that instance I'm not sure, maybe it would be easier to actuate an existing throttle body that's designed to have that kick down cable attached to it with an external actuator like that Gen V1 that I showed you earlier.
43:22 Anything can be done, it just comes down to how hard that's going to be, how complicated that's going to be and then by definition how much it's going to cost you.
43:31 Alright, looks like that's got us to the end of the question so we'll leave it there.
43:34 As usual if our members are watching this at a later point in the archive and have got further questions, please feel free to ask those in the forum and I'll be happy to answer them there.
43:43 Thanks for watching and we'll see you all again next time.