For Those of You With Dual Wideband Setups...
Having two widebands in a car is like having two watches, they can eash read a different time, but it doesn't mean that even one of them is exactly right. I have two in my car, and the passenger side originally showed leaner. Why would this be??? So I swapped the sensors (still connected to the same controller) and what do you know....now the drivers side is leaner. These are electronic devices reading current and changes in millivoltes and milliamps, Unless you have a way to "sync" the two, not all widebands are going to read dead equal...especially at light load. Under power, they both even up, no matter which bank they are in. It even does this on with the stock ecu in place with widebands.
The Pro48 can do dual wideband, however on the Z they are paired, so it isn't set up that way. After "synching" the two widebands so they read the same swapping bank to bank, and having my injectors flow tested so I know one isn't flowing more than the others. The variance bank to bank is neglegable.
The Pro48 can do dual wideband, however on the Z they are paired, so it isn't set up that way. After "synching" the two widebands so they read the same swapping bank to bank, and having my injectors flow tested so I know one isn't flowing more than the others. The variance bank to bank is neglegable.
Last edited by rcdash; Jul 30, 2009 at 09:40 AM.
Registered User
Joined: Nov 2007
Posts: 198
Likes: 0
From: AZ
I don't believe this conclusion is valid. I have measured plenum air flow on a test bed using a MAF sensor on each of the six runners and there is an overall variance at low flow velocities with the driver's side receiving more airflow (on the order of 5-8%). It's not a huge difference, but it is measureable and consistent. If you have widebands that you can free air calibrate (to "sync the two" as you say), then you can measure accurately even as the sensor ages.
When you did your flow testing of the intake on your bench, how did you account for reversion in the intake from the opposing cylinders? Bench tests don't mean much of anything at the end of the day, the engine as a package defines airflow...cam overlap, lift etc...
Did you free air calibrate during your one test?
It's possible you could account for the differences due to error in electronics/sampling, etc but your one data point is just that. I know what you're trying to say and while I also agree that a flow bench setup is an imperfect model, you must also admit that the conclusion you have come to is not really well supported. It's a nice thought though.
The flow inconsistency is there - empirical evidence from many sources is overwhelming in this regard.
I don't want to argue with you - it's no big deal, just another perspective. It is theoretically possible that all the observations to date have been unrelated to flow imbalance in the plenum, but at least there is some data to support that conclusion, both through wideband data published on the forums and through flow bench testing. I've logged countless hours of wideband data from Osiris and the Haltech, and using different plenums - stock, stock + spacer, and Crawford when testing internal plenum baffles. The plenums all generate different, but repeatable and statistically significant differences in bank to bank AFR (and only under heavy to moderate vacuum, not low load or boost).
It's possible you could account for the differences due to error in electronics/sampling, etc but your one data point is just that. I know what you're trying to say and while I also agree that a flow bench setup is an imperfect model, you must also admit that the conclusion you have come to is not really well supported. It's a nice thought though.
The flow inconsistency is there - empirical evidence from many sources is overwhelming in this regard.I don't want to argue with you - it's no big deal, just another perspective. It is theoretically possible that all the observations to date have been unrelated to flow imbalance in the plenum, but at least there is some data to support that conclusion, both through wideband data published on the forums and through flow bench testing. I've logged countless hours of wideband data from Osiris and the Haltech, and using different plenums - stock, stock + spacer, and Crawford when testing internal plenum baffles. The plenums all generate different, but repeatable and statistically significant differences in bank to bank AFR (and only under heavy to moderate vacuum, not low load or boost).
Last edited by rcdash; Jul 30, 2009 at 06:56 PM.
Registered User
Joined: Nov 2007
Posts: 198
Likes: 0
From: AZ
Did you free air calibrate during your one test?
It's possible you could account for the differences due to error in electronics/sampling, etc but your one data point is just that. I know what you're trying to say and while I also agree that a flow bench setup is an imperfect model, you must also admit that the conclusion you have come to is not really well supported. It's a nice thought though. The flow inconsistency is there - empirical evidence from many sources is overwhelming in this regard.
I don't want to argue with you - it's no big deal, just another perspective. It is theoretically possible that all the observations to date have been unrelated to flow imbalance in the plenum, but at least there is some data to support that conclusion, both through wideband data published on the forums and through flow bench testing. I've logged countless hours of wideband data from Osiris and the Haltech, and using different plenums - stock, stock + spacer, and Crawford when testing internal plenum baffles. The plenums all generate different, but repeatable and statistically significant differences in bank to bank AFR (and only under heavy to moderate vacuum, not low load or boost).
It's possible you could account for the differences due to error in electronics/sampling, etc but your one data point is just that. I know what you're trying to say and while I also agree that a flow bench setup is an imperfect model, you must also admit that the conclusion you have come to is not really well supported. It's a nice thought though.
The flow inconsistency is there - empirical evidence from many sources is overwhelming in this regard.I don't want to argue with you - it's no big deal, just another perspective. It is theoretically possible that all the observations to date have been unrelated to flow imbalance in the plenum, but at least there is some data to support that conclusion, both through wideband data published on the forums and through flow bench testing. I've logged countless hours of wideband data from Osiris and the Haltech, and using different plenums - stock, stock + spacer, and Crawford when testing internal plenum baffles. The plenums all generate different, but repeatable and statistically significant differences in bank to bank AFR (and only under heavy to moderate vacuum, not low load or boost).
....but it doesn't matter. If one is off, it would be consistantly off (I.E. a half a point no matter what bank it is installed in.) Has anyone logged FACTORY Longterm Trims for banks 1 and 2... This would give more consistant data with all things being COMPLETLY stock! I know when I logged the factory stuff my banks were pretty even, however the stock computer had been reset more than once so I can't say for sure how much drive time it had.
At WOT, under boost, the differences become insignificant on my widebands. A tenth of a point in AFR difference or dead even. At cruise/vacuum there is a very significant difference and that's where having dual widebands really helps to provide smooth engine response, adjusting bank to bank.
So you don't NEED to have it. One wideband is good enough - just keep in mind that the driver's side will be up to 5-10% leaner at low loads / cruise and adjust accordingly if you can. If you can't, you can still adjust low load target AFRs in the low 14 range instead of right at 14.7 and you won't be able to tell the difference in driveability.
So you don't NEED to have it. One wideband is good enough - just keep in mind that the driver's side will be up to 5-10% leaner at low loads / cruise and adjust accordingly if you can. If you can't, you can still adjust low load target AFRs in the low 14 range instead of right at 14.7 and you won't be able to tell the difference in driveability.
What he said.
I've seen even up to 1.5 pts difference between the two, with driver's side being the leaner of the two.
At full throttle, it evens out, most of the times it's identical.
I'm running a 3.12" Vortech with a Haltech Platinum.
My dual widebands helped me find an exhaust manifold leak. One side started reading leaner, and fuel correction on that side was maxing out. I went looking around to find that a heat shield bolt had backed out and was spraying hot gases on my coolant hard line on the driver's side. 5 min fix and back to normal. Could have been a disaster though...
accordfreak, some variance is normal but 2 points sounds excessive. It could be an exhaust leak like rcdash mentioned, or could be leaks on the intake side. Does it go away with throttle?
It's the asymmetrical neck of the plenum. The air comes into the chamber heading for passenger side runners (and consequently drivers side cylinders). No way around it without some kind of baffle intrinsic to the plenum design. Not sure why Cosworth didn't fix this - maybe hurt the #s up top?
I have dual wb and in my case, at WOT and full boost the rear bank is signifficantly leaner, as rpm goes up the front bank leans out and crosses afr curve becoming leaner. I do however have a FWD manifold which I am certain is the source for this difference(more than 1 afr). I corrected it with the emanage ultimate.
Registered User
Joined: Dec 2007
Posts: 154
Likes: 0
From: concord NC
im having this same problem..dual wb, i have the sts rear mount kit, so i have my wb in my test pipes. my drivers side wb usually reads more rich than the pass. side at idle and crusing. WOT bam no problems, diggin.. pretty sure no exhaust leaks.. tuning with uprev. any ideas
Does osiris have a way to trim only in the closed loop? It would be a bad deal to trim the entire bank all the time because he said it was equal under WOT