S7: Use Secondary air system on Haltech Elite
Thread Starter
Full Member
Joined: Jun 2014
Posts: 75
Likes: 0
From: Germany
S7: Use Secondary air system on Haltech Elite
Hello,
I want to use the stock secondary air system on my S7 FD using a haltech elite 2000.
Everything is wired and Pump clutch, ACV relief solenoid and ACV Switching solenoid are working.
Because it's a S7, the ACV only has relief and switching. There is no Split Air bypass, Relief 2 and Port Air Bypass.
Mazda manual( I only have the manual for the "old" acv with the bypasses) says in idle, the pump is engaged and ACV is in mode 'Port air', so all air from the pump goes into the port. Which raises my lambda from 0.9 to 1.4. To correct this with a correction table i have to add about 45% more fuel to hit lambda 1 to pass emissions. With this much fuel the engine starts to stumble: This is how the ACV operated in this mode
So I tried to control the switching valve with a lower Duty cycle, the lowest possible DC was 70% on the switching valve, with a lower DC it isn't responding. This brings me down to a extra fuel amount of 35%. But when the port air is controlled by a DC on the switching valve, air is also going into the converter, which changes the AFR before the converter and the emissions-tester will see something lean-> it will fail
This is how the ACV operated in this version:
Next try: I tried to Control the port air amount by adding a duty cycle to the ACV relief valve. I was able to get the engine run with lambda 1 with a extra fuel amount of 10% and without passing air by into the converter. So the emission-tester will also see lambda 1.
But the result is a bit inconsistent and the o2 control hast to do a lot of work, because small fluctuations in RPM changes the flow of the pump a lot.
This is how the ACV operates in this version:
Any ides how it is easier to control? Or does anyone know how exactly the s7/8 ecu controls the pump?
regards
Bastian
I want to use the stock secondary air system on my S7 FD using a haltech elite 2000.
Everything is wired and Pump clutch, ACV relief solenoid and ACV Switching solenoid are working.
Because it's a S7, the ACV only has relief and switching. There is no Split Air bypass, Relief 2 and Port Air Bypass.
Mazda manual( I only have the manual for the "old" acv with the bypasses) says in idle, the pump is engaged and ACV is in mode 'Port air', so all air from the pump goes into the port. Which raises my lambda from 0.9 to 1.4. To correct this with a correction table i have to add about 45% more fuel to hit lambda 1 to pass emissions. With this much fuel the engine starts to stumble: This is how the ACV operated in this mode
So I tried to control the switching valve with a lower Duty cycle, the lowest possible DC was 70% on the switching valve, with a lower DC it isn't responding. This brings me down to a extra fuel amount of 35%. But when the port air is controlled by a DC on the switching valve, air is also going into the converter, which changes the AFR before the converter and the emissions-tester will see something lean-> it will fail
This is how the ACV operated in this version:
Next try: I tried to Control the port air amount by adding a duty cycle to the ACV relief valve. I was able to get the engine run with lambda 1 with a extra fuel amount of 10% and without passing air by into the converter. So the emission-tester will also see lambda 1.
But the result is a bit inconsistent and the o2 control hast to do a lot of work, because small fluctuations in RPM changes the flow of the pump a lot.
This is how the ACV operates in this version:
Any ides how it is easier to control? Or does anyone know how exactly the s7/8 ecu controls the pump?
regards
Bastian
Last edited by Slash200; Oct 10, 2018 at 09:49 AM.
Joined: Mar 2001
Posts: 31,835
Likes: 3,233
From: https://www2.mazda.com/en/100th/
the US cars run port air, they use the Split air at weird times, like transitions, and such.
if you put a wideband on an US car you would see idle readings around 18:1, its a false reading because of the air pump air, but it is also in open loop. its not in closed loop until 1250 or so RPM, and then it will be in closed loop until 1psi or 3200rpm whichever occurs first
if you put a wideband on an US car you would see idle readings around 18:1, its a false reading because of the air pump air, but it is also in open loop. its not in closed loop until 1250 or so RPM, and then it will be in closed loop until 1psi or 3200rpm whichever occurs first
Thread Starter
Full Member
Joined: Jun 2014
Posts: 75
Likes: 0
From: Germany
thanks j9fd3s for this information.
It's new for me that the factory ecu isn't running in closed loop until 1250 and that the factory ecu isn't trying to reach lambda 1 in idle. Maybe it isn't required from old emission tests to reach lambda 1 in idle.
When it is true that the factory ecu uses closed loop not in idle, why stumbles the factory ecu when the Pump is removed?
It's new for me that the factory ecu isn't running in closed loop until 1250 and that the factory ecu isn't trying to reach lambda 1 in idle. Maybe it isn't required from old emission tests to reach lambda 1 in idle.
When it is true that the factory ecu uses closed loop not in idle, why stumbles the factory ecu when the Pump is removed?
Last edited by Slash200; Oct 10, 2018 at 03:10 PM.
Joined: Mar 2001
Posts: 31,835
Likes: 3,233
From: https://www2.mazda.com/en/100th/
thanks j9fd3s for this information.
It's new for me that the factory ecu isn't running in closed loop until 1250 and that the factory ecu isn't trying to reach lambda 1 in idle. Maybe it isn't required from old emission tests to reach lambda 1 in idle.
When it is true that the factory ecu uses closed loop not in idle, why stumbles the factory ecu when the Pump is removed?
It's new for me that the factory ecu isn't running in closed loop until 1250 and that the factory ecu isn't trying to reach lambda 1 in idle. Maybe it isn't required from old emission tests to reach lambda 1 in idle.
When it is true that the factory ecu uses closed loop not in idle, why stumbles the factory ecu when the Pump is removed?
some portion of the Port Air is going into the engine, and it gets cycled around on the next intake stroke, think of it as the reverse of an EGR valve. when you take that away, idle speed drops, and its richer...
you might investigate using a different air pump pulley, in the US there were a couple different ones
Read this thread: https://www.rx7club.com/3rd-generati...ntrols-841963/
and read this more recent thread where I helped someone pass emissions in California: https://www.rx7club.com/3rd-generati...d-out-1119012/
Do you have a link to the test cycle you have to meet? There should a graph of time vs vehicle speed. I am assuming it's an actual driving test on a dyno (similar to second thread) rather than just an idle + steady speed test. If it's a steady speed test, well that's pretty easy comparatively speaking.
The key to remember is that the port air is for oxidation reactions to reduce HC and CO at idle and low load, NOT to run lambda 1, because there is no NOx concern in those areas. The split air is for reduction reactions to reduce NOx at heavier non boosted loads. Then during heavy acceleration you have to stay lambda 1, because CO emission increases when you get richer than lambda 1. If you go richer during heavy acceleration you will fail CO regardless of the air pump. So you have to look at the test cycle and then adjust your secondary air operation window and enrichment strategy according to that.
and read this more recent thread where I helped someone pass emissions in California: https://www.rx7club.com/3rd-generati...d-out-1119012/
Do you have a link to the test cycle you have to meet? There should a graph of time vs vehicle speed. I am assuming it's an actual driving test on a dyno (similar to second thread) rather than just an idle + steady speed test. If it's a steady speed test, well that's pretty easy comparatively speaking.
The key to remember is that the port air is for oxidation reactions to reduce HC and CO at idle and low load, NOT to run lambda 1, because there is no NOx concern in those areas. The split air is for reduction reactions to reduce NOx at heavier non boosted loads. Then during heavy acceleration you have to stay lambda 1, because CO emission increases when you get richer than lambda 1. If you go richer during heavy acceleration you will fail CO regardless of the air pump. So you have to look at the test cycle and then adjust your secondary air operation window and enrichment strategy according to that.
Thread
Thread Starter
Forum
Replies
Last Post
Captain_Panic
The Bad & Fugly Businesses
8
Dec 10, 2020 02:56 PM
red_force1
2nd Generation Specific (1986-1992)
1
Jun 4, 2018 12:08 AM