O
Odin
Guest
Injoy.
Airflow Meter Bypass, Part 2
A massive airflow meter upgrade, step by step
By Julian Edgar
Advertisement
Advertisement
At a glance...
Double the flow capacity of any voltage-outputting airflow meter
Full control over air/fuel ratios
Cheap and easy
Email a friend Print article
Last week in Airflow Meter Bypass, Part 1 we introduced the idea of using an airflow meter bypass to lift the flow and measurement capabilities of a normal airflow meter. In short, by using a bypass around the airflow meter and the recalibrating the meter’s output with the Silicon Chip Digital Fuel Adjuster, we’re able to reduce the pressure drop through the intake system and also increase the voltage headroom before the meter maxes out.
This week we’ll take a look at the installation process on a guinea pig Nissan Maxima Turbo.
The Nissan
The 1988 Nissan Maxima runs the VG20ET turbo engine. In previous stories we've have covered the fitting of an intercooler, intercooler water spray, an electronic blow-off valve, a new exhaust and a new modified filter box. (For more on these stories, do a site search under ‘Maxima’.)
The airbox mod involved cutting away the lid so that air could reach the filter across its full area. A thick foam rubber strip then sealed this assembly to a bonnet opening so that only cold air could be inhaled. Making this simple modification dropped the total measured intake system peak load pressure drop from 30 inches of water to 20 inches of water. (Using a sensitive pressure gauge, the total system pressure drop was measured just prior to the turbo intake. With pressure drop measurements, the lower the numbers, the less restriction there is to flow.)
So why still the 20 inches of pressure drop? Following the airbox is a hotwire airflow meter that in turn connects to a large resonant chamber (top-left in this pic). From the chamber a long duct connects to the turbo intake. The pressure drop still being recorded comprises the flow losses through the airflow meter, through the resonant chamber, and along the intake pipe from the chamber to the turbo. (The pressure drop through the flat panel filter? Not measurable – the pressure drop with the filter in place or removed remained just the same, as is so often the case.)
Given that the resonant chamber is nicely built inside (the air has to do a U-turn but the exit duct is equipped with a bellmouth and the insides just look good), and that the pipe between this chamber and the turbo has no sharp bends, it seemed likely to me that plenty of the restriction was in the airflow meter.
At this stage I had never actually removed the airflow meter (it’s fairly hard to get to) but since the engine power in standard form is only about 110kW, I figured that it was probably a small one. I sourced a new airflow meter (from an Australian VL Commodore, a car equipped with a 3-litre Nissan six cylinder engine and up to 150kW of engine power) and then used a workshop manual to work out which pin on the new airflow meter did what. (As mentioned last week, even for an airflow meter from a known car, this often isn’t as simple as it first appears.)
I then pulled out the Maxima’s airflow meter – only to find that it was the same diameter as the one I’d bought to replace it! And furthermore, the replacement airflow meter looked as if it may well flow less, as it has many alloy heatsink fins projecting into the airstream.
Time for a rethink. Getting hold of a larger hotwire meter was out of the question – the budget would have been blown out of the water and plus, I couldn’t afford the time to go from wrecker to wrecker trying to find a larger meter for which I could get the pin-outs. That’s when the bypass approach was adopted.
The Layout
The most time-consuming part of the mechanical changes was working out exactly how to do it. With tight space to work with, and with a typical home do-it-yourselfer’s dislike of getting stuff welded (it always takes so long when you’ve gotta tape metal plumbing together and then take it somewhere to be welded – only to find that some of the parts moved in transit and so nothing now lines up!), it was a juggling act to get the right financial and flow outcomes.
My first approach was to use two off-the-shelf alloy boxes to form manifolds at each end of the airflow meter. As shown here, a new pod-type airflow meter would be bolted to one of the boxes, which would transfer air to both the airflow meter inlet and also the bypass inlet. On the other side of the airflow meter another alloy box would be used to join the flows so that this combined air could be sent on to the engine. The use of readily available flanges would allow easy connection of the pod filter to the box and the engine to the assembly.
However, the extra length that the two boxes gave to the assembly meant that it was difficult to fit it all under the bonnet. Time for a rethink. Looking at the standard underbonnet architecture I realised that I may be able to attach another flange to the metal resonant chamber into which the standard airflow meter directs air. I’d need two pod filters – one for the airflow meter and another for the bypass. The advantages of taking this approach was that no welding would be needed, two filters would flow better than one – and most importantly, it looked as if everything could be fitted in!
I bought two oiled cotton pod filters (AUD$25 each on special) and two ~75mm plastic flanges (AUD$20 each). I then used a holesaw to cut a hole in a relatively flat wall of the box, then doing a little panel beating to give a surface that the flange could bolt up against.
To make sure that the seal was good, I cut a gasket out of sheet rubber and sandwiched this between the flange and the box (yellow arrow). The bolts holding the flange in place were inserted from within the box and then doubled nuts (tightened against one) another were used to lower the likelihood of a bolt coming loose and then floating through the turbo compressor. These nuts are shown by the red arrow.
With the box back in place, the approach can be more clearly seen. The top opening is the new one created for the bypass while the lower opening is the original that connects to the airflow meter.
Some short lengths of thick-walled plastic pipe were then used to join the flanges to the pod filters.
Tuning
As you would expect with such a large bypass, the car wouldn’t start or run without the Digital Fuel Adjuster in place. The DFA was wired-in at the ECU, with the wire coming from the airflow meter connecting to the ‘in’ terminal and the wire from the ECU connecting to the ‘out’ terminal. With power and earth supplied, the electrical installation of the DFA was then finished! At this stage the oxygen sensor was also disconnected so the tuning wasn’t confused by ECU’s learning behaviour. The DFA was also set so that it always intercepts, ie the car doesn’t have to first start before it starts work.
Cranking of the engine showed on the DFA hand controller the load sites being outputted by the airflow meter. I increased the correction at these and all surrounding load sites - but I initially increased these outputs by far too much. I had thought that a massive correction would be needed at all loads but as I subsequently found out, only very small corrections were needed near idle. In fact, with the DFA set to its normal coarse mode of operation, a +6 correction was all that was needed to have the car idling happily. (I’d started at +60!)
(For those who are interested in more numbers, measurement showed that the ECU needed a 2.8V signal from the ECU to run properly at idle, and the actual bypassed airflow meter output was 2.4V. The +6 correction brought the airflow meter voltage up from 2.4 to 2.8V.)
With the car idling happily, I plugged in the same +6 correction at higher load sites – that is, I worked ahead as much as possible, putting in figures based on the corrections being used at the lower value load sites. At this stage I also attached a MoTeC air/fuel ratio meter to the exhaust so that I could see exactly what I was doing.
(As we have covered in another story - Real World Air/Fuel Ratio Tuning - much tuning can be carried out using just a multimeter and a normal narrow-band probe – and certainly getting the car running at idle and light loads would be no problem at all using just this type of sensor. However, for full-load tuning, an accurate air/fuel ratio meter must be used. Tuning the DFA is so easy that even 20 minutes on a dyno will be enough to finish the tuning if the light loads have already been set.)
At this stage the DFA map looked something like this – the car idled at load site #30 but I put in the correction for Load Sites 25-29 for better starting and idling behaviour.
Load Site
Adjustment
25
5
26
5
27
5
28
5
29
6
30
6
31
6
32
7
With this much tuning done we decided to hit the road. My driveway is very steep and it took us three goes to get up it – the air/fuel ratio was varying between 10:1 and 18:1! However, some more load site adjustment of sites between 33 and 38 (shown below) gave good light load driveability. At these loads I was aiming at an air/fuel ratio of around the mid-fourteens (eg 14.3 – 14.9:1), the near-stoichiometric ratio used for best emissions performance.
Load Site
Adjustment
33
7
34
7
35
7
36
8
37
9
38
9
The next loads involved going lightly into boost, where I set the DFA to give air/fuel ratios in the mid-thirteens. Note how the amount of correction needed to the airflow meter output is increasing with load as more and more intake air takes the bypass route (table below).
Load Site
Adjustment
39
9
40
9
41
10
42
10
43
10
44
11
Above Load Site 45 (see below) the engine was on substantial boost, with the peak load site being 58. (Always put in numbers above the max load site in case the engine is in a situation where it develops more power than when being tuned – eg on a very cold day.) At these loads I set the DFA to provide an air/fuel ratio going into the mid-twelves and then progressing into the high-elevens at absolute peak load.
Load Site
Adjustment
45
12
46
12
47
12
48
12
49
13
50
15
51
15
52
15
53
16
54
16
55
16
56
16
57
16
58
16
59
16
60
16
61
16
62
16
Note that during this tuning process the DFA was set in its coarse mode of adjustment, that is, each up/down increment moves the output voltage much further than if fine mode is selected. This was done to allow speedy tuning, and in fact the accuracy with which the air/fuel ratio is maintained – and so the excellent driveability – means that the DFA can be left set in this mode.
Testing
The bypass and DFA were fitted in order to reduce intake flow restriction – so how well did it all work? Testing needed to take into account that there have been two major changes here – the replacement of the stock airfilter box with the two pod filters, and the use of the bypass. To (mostly) separate the effects of these two changes, two tests were undertaken.
Firstly, the bypass was completely blocked and the DFA tune returned to normal. (Just removing power to the DFA does this – it’s then bypassed.) This test was designed to show the gains made by the removal of the airbox and the fitting of the pod filter to the end of the airflow meter. However, full load testing showed that these changes had made no difference at all to the intake system flow restriction! It still remained at a measured 20 inches of water.
The bypass was then re-opened, the DFA switched back on and the testing done again. This time, the total intake system restriction had dropped to just 10 inches of water – the use of the airflow meter bypass and extra pod filter had halved the total intake restriction! Furthermore, the maximum output voltage of the airflow meter is now well down from its ceiling voltage – this airflow meter is now probably a 350hp design...
Remember that the total measured restriction includes that caused by the U-turn the air needs to take through the resonant box and the flow down the resonant box to turbo feed pipe. It’s likely that the restriction of just the combined filters, bypass and airflow meter is now only 2-3 inches of water.
Conclusion
The use of a bypass in conjunction with the Silicon Chip Digital Fuel Adjuster is a cheap and very effective way of upgrading the airflow meter capability and at the same time giving control over air/fuel ratios. In this case I chose to use a very large bypass but a smaller one could instead be used, which would allow the airflow meter to work across a broader proportion of its original range.
Either way, we can’t think of any reason why you’d now want to swap an airflow meter for a large unit - not when it’s this easy to improve the power capability of the one you’ve already got...
rob
Airflow Meter Bypass, Part 2
A massive airflow meter upgrade, step by step
By Julian Edgar
Advertisement
Advertisement
At a glance...
Double the flow capacity of any voltage-outputting airflow meter
Full control over air/fuel ratios
Cheap and easy
Email a friend Print article
Last week in Airflow Meter Bypass, Part 1 we introduced the idea of using an airflow meter bypass to lift the flow and measurement capabilities of a normal airflow meter. In short, by using a bypass around the airflow meter and the recalibrating the meter’s output with the Silicon Chip Digital Fuel Adjuster, we’re able to reduce the pressure drop through the intake system and also increase the voltage headroom before the meter maxes out.
This week we’ll take a look at the installation process on a guinea pig Nissan Maxima Turbo.
The Nissan
The 1988 Nissan Maxima runs the VG20ET turbo engine. In previous stories we've have covered the fitting of an intercooler, intercooler water spray, an electronic blow-off valve, a new exhaust and a new modified filter box. (For more on these stories, do a site search under ‘Maxima’.)
The airbox mod involved cutting away the lid so that air could reach the filter across its full area. A thick foam rubber strip then sealed this assembly to a bonnet opening so that only cold air could be inhaled. Making this simple modification dropped the total measured intake system peak load pressure drop from 30 inches of water to 20 inches of water. (Using a sensitive pressure gauge, the total system pressure drop was measured just prior to the turbo intake. With pressure drop measurements, the lower the numbers, the less restriction there is to flow.)
So why still the 20 inches of pressure drop? Following the airbox is a hotwire airflow meter that in turn connects to a large resonant chamber (top-left in this pic). From the chamber a long duct connects to the turbo intake. The pressure drop still being recorded comprises the flow losses through the airflow meter, through the resonant chamber, and along the intake pipe from the chamber to the turbo. (The pressure drop through the flat panel filter? Not measurable – the pressure drop with the filter in place or removed remained just the same, as is so often the case.)
Given that the resonant chamber is nicely built inside (the air has to do a U-turn but the exit duct is equipped with a bellmouth and the insides just look good), and that the pipe between this chamber and the turbo has no sharp bends, it seemed likely to me that plenty of the restriction was in the airflow meter.
At this stage I had never actually removed the airflow meter (it’s fairly hard to get to) but since the engine power in standard form is only about 110kW, I figured that it was probably a small one. I sourced a new airflow meter (from an Australian VL Commodore, a car equipped with a 3-litre Nissan six cylinder engine and up to 150kW of engine power) and then used a workshop manual to work out which pin on the new airflow meter did what. (As mentioned last week, even for an airflow meter from a known car, this often isn’t as simple as it first appears.)
I then pulled out the Maxima’s airflow meter – only to find that it was the same diameter as the one I’d bought to replace it! And furthermore, the replacement airflow meter looked as if it may well flow less, as it has many alloy heatsink fins projecting into the airstream.
Time for a rethink. Getting hold of a larger hotwire meter was out of the question – the budget would have been blown out of the water and plus, I couldn’t afford the time to go from wrecker to wrecker trying to find a larger meter for which I could get the pin-outs. That’s when the bypass approach was adopted.
The Layout
The most time-consuming part of the mechanical changes was working out exactly how to do it. With tight space to work with, and with a typical home do-it-yourselfer’s dislike of getting stuff welded (it always takes so long when you’ve gotta tape metal plumbing together and then take it somewhere to be welded – only to find that some of the parts moved in transit and so nothing now lines up!), it was a juggling act to get the right financial and flow outcomes.
My first approach was to use two off-the-shelf alloy boxes to form manifolds at each end of the airflow meter. As shown here, a new pod-type airflow meter would be bolted to one of the boxes, which would transfer air to both the airflow meter inlet and also the bypass inlet. On the other side of the airflow meter another alloy box would be used to join the flows so that this combined air could be sent on to the engine. The use of readily available flanges would allow easy connection of the pod filter to the box and the engine to the assembly.
However, the extra length that the two boxes gave to the assembly meant that it was difficult to fit it all under the bonnet. Time for a rethink. Looking at the standard underbonnet architecture I realised that I may be able to attach another flange to the metal resonant chamber into which the standard airflow meter directs air. I’d need two pod filters – one for the airflow meter and another for the bypass. The advantages of taking this approach was that no welding would be needed, two filters would flow better than one – and most importantly, it looked as if everything could be fitted in!
I bought two oiled cotton pod filters (AUD$25 each on special) and two ~75mm plastic flanges (AUD$20 each). I then used a holesaw to cut a hole in a relatively flat wall of the box, then doing a little panel beating to give a surface that the flange could bolt up against.
To make sure that the seal was good, I cut a gasket out of sheet rubber and sandwiched this between the flange and the box (yellow arrow). The bolts holding the flange in place were inserted from within the box and then doubled nuts (tightened against one) another were used to lower the likelihood of a bolt coming loose and then floating through the turbo compressor. These nuts are shown by the red arrow.
With the box back in place, the approach can be more clearly seen. The top opening is the new one created for the bypass while the lower opening is the original that connects to the airflow meter.
Some short lengths of thick-walled plastic pipe were then used to join the flanges to the pod filters.
Tuning
As you would expect with such a large bypass, the car wouldn’t start or run without the Digital Fuel Adjuster in place. The DFA was wired-in at the ECU, with the wire coming from the airflow meter connecting to the ‘in’ terminal and the wire from the ECU connecting to the ‘out’ terminal. With power and earth supplied, the electrical installation of the DFA was then finished! At this stage the oxygen sensor was also disconnected so the tuning wasn’t confused by ECU’s learning behaviour. The DFA was also set so that it always intercepts, ie the car doesn’t have to first start before it starts work.
Cranking of the engine showed on the DFA hand controller the load sites being outputted by the airflow meter. I increased the correction at these and all surrounding load sites - but I initially increased these outputs by far too much. I had thought that a massive correction would be needed at all loads but as I subsequently found out, only very small corrections were needed near idle. In fact, with the DFA set to its normal coarse mode of operation, a +6 correction was all that was needed to have the car idling happily. (I’d started at +60!)
(For those who are interested in more numbers, measurement showed that the ECU needed a 2.8V signal from the ECU to run properly at idle, and the actual bypassed airflow meter output was 2.4V. The +6 correction brought the airflow meter voltage up from 2.4 to 2.8V.)
With the car idling happily, I plugged in the same +6 correction at higher load sites – that is, I worked ahead as much as possible, putting in figures based on the corrections being used at the lower value load sites. At this stage I also attached a MoTeC air/fuel ratio meter to the exhaust so that I could see exactly what I was doing.
(As we have covered in another story - Real World Air/Fuel Ratio Tuning - much tuning can be carried out using just a multimeter and a normal narrow-band probe – and certainly getting the car running at idle and light loads would be no problem at all using just this type of sensor. However, for full-load tuning, an accurate air/fuel ratio meter must be used. Tuning the DFA is so easy that even 20 minutes on a dyno will be enough to finish the tuning if the light loads have already been set.)
At this stage the DFA map looked something like this – the car idled at load site #30 but I put in the correction for Load Sites 25-29 for better starting and idling behaviour.
Load Site
Adjustment
25
5
26
5
27
5
28
5
29
6
30
6
31
6
32
7
With this much tuning done we decided to hit the road. My driveway is very steep and it took us three goes to get up it – the air/fuel ratio was varying between 10:1 and 18:1! However, some more load site adjustment of sites between 33 and 38 (shown below) gave good light load driveability. At these loads I was aiming at an air/fuel ratio of around the mid-fourteens (eg 14.3 – 14.9:1), the near-stoichiometric ratio used for best emissions performance.
Load Site
Adjustment
33
7
34
7
35
7
36
8
37
9
38
9
The next loads involved going lightly into boost, where I set the DFA to give air/fuel ratios in the mid-thirteens. Note how the amount of correction needed to the airflow meter output is increasing with load as more and more intake air takes the bypass route (table below).
Load Site
Adjustment
39
9
40
9
41
10
42
10
43
10
44
11
Above Load Site 45 (see below) the engine was on substantial boost, with the peak load site being 58. (Always put in numbers above the max load site in case the engine is in a situation where it develops more power than when being tuned – eg on a very cold day.) At these loads I set the DFA to provide an air/fuel ratio going into the mid-twelves and then progressing into the high-elevens at absolute peak load.
Load Site
Adjustment
45
12
46
12
47
12
48
12
49
13
50
15
51
15
52
15
53
16
54
16
55
16
56
16
57
16
58
16
59
16
60
16
61
16
62
16
Note that during this tuning process the DFA was set in its coarse mode of adjustment, that is, each up/down increment moves the output voltage much further than if fine mode is selected. This was done to allow speedy tuning, and in fact the accuracy with which the air/fuel ratio is maintained – and so the excellent driveability – means that the DFA can be left set in this mode.
Testing
The bypass and DFA were fitted in order to reduce intake flow restriction – so how well did it all work? Testing needed to take into account that there have been two major changes here – the replacement of the stock airfilter box with the two pod filters, and the use of the bypass. To (mostly) separate the effects of these two changes, two tests were undertaken.
Firstly, the bypass was completely blocked and the DFA tune returned to normal. (Just removing power to the DFA does this – it’s then bypassed.) This test was designed to show the gains made by the removal of the airbox and the fitting of the pod filter to the end of the airflow meter. However, full load testing showed that these changes had made no difference at all to the intake system flow restriction! It still remained at a measured 20 inches of water.
The bypass was then re-opened, the DFA switched back on and the testing done again. This time, the total intake system restriction had dropped to just 10 inches of water – the use of the airflow meter bypass and extra pod filter had halved the total intake restriction! Furthermore, the maximum output voltage of the airflow meter is now well down from its ceiling voltage – this airflow meter is now probably a 350hp design...
Remember that the total measured restriction includes that caused by the U-turn the air needs to take through the resonant box and the flow down the resonant box to turbo feed pipe. It’s likely that the restriction of just the combined filters, bypass and airflow meter is now only 2-3 inches of water.
Conclusion
The use of a bypass in conjunction with the Silicon Chip Digital Fuel Adjuster is a cheap and very effective way of upgrading the airflow meter capability and at the same time giving control over air/fuel ratios. In this case I chose to use a very large bypass but a smaller one could instead be used, which would allow the airflow meter to work across a broader proportion of its original range.
Either way, we can’t think of any reason why you’d now want to swap an airflow meter for a large unit - not when it’s this easy to improve the power capability of the one you’ve already got...
rob