Sale ends todayGet 30% off any course (excluding packages)

Ends in --- --- ---

Understanding AFR: Effect of AFR on Power

Watch This Course

$89 USD

Or 8 weekly payments of only $11.13 Instant access. Easy checkout. No fees. Learn more
Course Access for Life
60 day money back guarantee

Effect of AFR on Power


00:00 - While the amount of power an engine produces is related to the chosen air fuel ratio, it's not as critical as many people think.
00:09 The amount of power the engine can produce is actually defined by the amount of air that the engine can consume during the intake stroke.
00:17 What we need to do is supply sufficient fuel to make sure that we properly combust all of the available oxygen.
00:25 To do this under full load, we'll be running the engine richer than stoichiometric, and there are a couple of reasons for this.
00:33 Firstly, we want to add additional fuel to make sure all of the oxygen is mixed with fuel.
00:39 This ensures we are releasing all of the available power from the engine.
00:44 Secondly, we are using the extra fuel to quench, or cool, the combustion charge, and hence control the combustion temperature.
00:53 This is critical to the safety and reliability of any engine, and should always take precedence over outright power.
01:03 There's a widespread belief that to make the absolute maximum power, any engine must be run at the absolute lean limit, which consequently can be a very dangerous way to churn.
01:16 There is some truth to this belief, but the reality is that engine power actually varies a very small amount over quite a wide range of air fuel ratio.
01:28 As an experiment, we ran our Nissan 350Z to show how the power was affected by different air fuel ratios.
01:36 For this practical demonstration, we're going to have a look at how the lambda or AFR target effects torque using the torque optimization function on our mainline shessy dynode.
01:47 Now for this demonstration we're going to be using our Nissan 350Z fitted with a Haltech elite 2500 ECU.
01:55 Of course, it's the concept that's important here and this is applicable to any engine and any engine management system.
02:03 Let's start by having a quick look at the Haltech laptop software so you can understand what I'm doing.
02:10 So here in front of us we have our lambda target table.
02:14 And these are the lambda targets we want the engine to be running.
02:18 On the horizontal X axis we have our engine load, measured in KPA, and on the vertical axis we have our engine RPM.
02:27 Now for this test we're going to be operating at 2000 RPM in wide open throttle.
02:33 So we're going to be operating in these zones out here on the right hand side of our graph.
02:39 Now on the right hand side of the ECU software you can see we have our lambda target displayed along with the actual measured air fuel ratio from a wide-band sensor in the exhaust.
02:52 You'll be able to monitor those while we're running the engine.
02:55 And we'll jump across to our Dyno software and we'll have a look at what we've got here.
03:01 So this is our torque optimization screen, and on the horizontal axis we have our measured lambda which is coming from the Haltech ECU.
03:11 And on the vertical axis we've got our engine torque which is being measured, obviously, by our Dyno.
03:19 And what's going to happen is while we're running the engine the Dyno will plot the torque in real time so we'll be able to see a curve, we'll be able to see that torque curve of how it's affected by our lambda target.
03:32 The top right hand corner of the screen, we have some live data, we've got our lambda value coming from our ECU, we have our torque value our engine RPM, and our throttle position.
03:44 Okay so what we're going to do is start by going into our laptop software and we're going to highlight the entire wide-open throttle area that the engine's going to be running in, and we're gonna start by targeting a very lean air fuel ratio.
03:58 We're going to enter a target of 1.15 lambda.
04:02 Now this is much leaner than we'd ever expect to run a naturally aspirated engine at wide open throttle.
04:08 And what I'm going to do is we'll start at 1.15 lambda and then I'll gradually richen the air fuel ratio all the way through to 0.75 lambda.
04:18 And we'll see how those lambda values affect our torque.
04:22 So let's get started now.
04:23 We'll go to 2000 RPM. We're in fourth gear.
04:26 And I'll go to wide open throttle.
04:29 And once the engine's stable, I'll begin our test.
04:33 Now the Dyno's going to plot the torque and what I'm going to do is just drop our lambda target by about 0.01 lambda every second, and you can see what's happening now is that the Dyno is plotting the torque on the screen.
04:53 And we're down to 0.97 lambda now.
04:57 And what you can see is as we've moved from our very lean starting point, 1.15 lambda, you can see that our torque has increased.
05:05 It's hit a plateau though.
05:07 Now we're getting down to 0.88 lambda, and you can see that our torque again is starting to fall away.
05:14 We're down to 0.84 now, and you can see that the torque is continuing to drop off.
05:20 And we're going to go all the way down to 0.075, which again, for a naturally aspirated engine running on pumped fuel like this, that's much richer than we'd likely run.
05:32 Alright, our test is finished now, so I'll just back off the throttle and we can talk about those results.
05:37 So on our Dyno screen you can see our red line which is our torque plot, and you can see there's a cross here showing the point where we made peak torque.
05:49 It's also being shown up here.
05:52 So what it's saying is that at your ratio of 0.92, we made our peak torque which was 342 newton meters.
06:01 So for this particular operating point, the peak power or peak torque was being produced at 0.92 lambda.
06:08 The important point, though, to take away from this is that over what is a relatively wide range of lambda values, probably everything from around about 0.86, 0.87 through to about lambda 1.0, you can see that we've seen almost no change in torque across those lambda targets.
06:31 We've reached a plateau.
06:34 However, if we look at the very rich side and also, conversely, at the very lean side of our graph, we can see that at both sides of that spectrum, the torque drops off quite sharply.
06:48 And if we continue to go richer, or continue to go leaner, our torque would have dropped off further.
06:53 The point of the demonstration, though, is that over a relatively wide range of lambda values, in this case around about 0.87 through to about lambda 1.00, 0.98, we've seen almost no change in our engine power.
07:09 So we can actually choose a safely rich air fuel ratio in this case, without sacrificing torque or power.
07:16 For the second part of our demonstration, we're going to look at the effect of air fuel ratio on power during a ramp run.
07:24 Now, we shouldn't really be expecting anything different than what we saw during the torque optimization test, but this is just another way of visualizing exactly what's going on.
07:33 So we're using the same car, this time we've swapped to a link G4+ ECU, but again the actual ECU and the engine don't really matter, it's the concept that I want you to understand.
07:44 Let's have a look at the laptop software and we'll see exactly what we've got in here.
07:48 So the table that we're going to be interested in here is our AFR lambda target tables is really the only place I'm going to be making changes.
07:57 Now for our first run, we're going to be targeting lambda 0.90, and you can see that I've filled out the entire wide open throttle running area of that table with this target.
08:09 So in this case, we've got our load manifold gauge pressure, manifold pressure on the vertical axis.
08:16 And again, we've got RPM on the horizontal axis and this lower area of the table is where we're going to be running at wide open throttle.
08:26 So let's do our first ramp run now on the Dyno, and we'll see what result we get.
08:32 (engine revving) Ok so there's our first ramp run complete there.
08:51 And the engine's made 149.7 kilowatts, or 200.7 horsepower at the rear wheels.
08:59 Now you can see there at the bottom of the screen we have our power graph in kilowatts, and the red line above is our lambda target.
09:08 And you can see our red line is running pretty close to our target of 0.90.
09:13 There's always going to be a little bit of a discrepancy as we go through that run.
09:18 So when I talk about a target of 0.90, I'm generally meaning 0.90, plus or minus around about 0.01 lambda.
09:28 That's what I would consider to be on target.
09:32 Okay so we've got our first run done there.
09:34 What I'm going to do is save that and we will call that lambda 0.90, just so we can reference that a little bit later on when we want to overlap with our other runs.
09:47 And for the sake of simplicity, we'll show that live on the Dyno while we're doing our next run.
09:53 Okay let's jump back into our laptop software and what I'm going to do now is highlight again that entire wide open throttle running area.
10:01 And this time we're going to target 0.86.
10:04 Now remember from our torque optimization test we saw that our power was reasonable, or our torque was reasonably consistent down to about 0.86 lambda.
10:15 So obviously I'm not expecting a huge difference in the power during a ramp run.
10:21 So let's do our 2nd run now and we'll see what result we get.
10:24 (engine revving) So that's our 2nd run complete there.
10:45 And we've got almost exactly the same result.
10:48 We're down to around about a kilowatt there.
10:50 But you can see because we overlaid those runs during that test, the green run, which is our current run overlays almost exactly with the previous run we did.
11:01 If we look at our lambda at the top here you can see referenced against our previous run you can that our red line now is a little bit richer as we'd expect.
11:10 And you can see that for the most part of that run we're sitting at our target or very close to 0.86 lambda.
11:19 So what we'll do is save that one now.
11:22 We'll call that 0.86 so we can reference it again.
11:26 And this time we'll drop back into our laptop software and this time I'm going to target 0.95 lambda.
11:32 Again from our torque optimization test we saw up to at least 0.95, perhaps even 0.97 lambda that our power wasn't massively influenced, or our torque wasn't massively influenced.
11:45 So again, I'm not expecting a huge difference from our ramp run power tests.
11:51 Let's do that ramp run now and we'll see how that works out.
11:55 (engine revving) So this is our 3rd ramp run done.
12:14 And you can see that our results are almost identical to our first run.
12:19 Less than half percent difference between those runs.
12:23 So we've got 149.9 kilowatts or 201 horse power at the rear wheels.
12:30 Again, because we've overlaid that with our last graph, you can see the green run, our power is almost identical throughout that entire run compared to our last run at 0.86 lambda.
12:43 And again our air fuel ratio, our lambda numbers at the top of the graph there, you can see that we were sitting on our target at 0.95, plus or minus about 0.01 lambda.
12:58 So the purpose of this test is simply to show you that the power is not actually influenced that dramatically by our air fuel ratio target.
13:08 We don't need to be churning for the leanest possible air fuel ratio in order to make good power.
13:15 And churning with a richer air fuel ratio also means we've got extra fuel there to help control and cool that combustion chamber temperature.
13:25 To highlight these differences we can also analyze the three runs overlaid with each other which we can see now on our Dyno screen.
13:33 And again, throughout those runs you can see that the variation of power that we've got is negligible.
13:40 We've got almost no difference.
13:41 At the bottom of the screen here you can see our three lambda plots, our purple run, which was our first run with our target of 0.90, the red run which was our 2nd run with the 0.86 lambda target, and then finally our light blue run which was targeting 0.95.