Would this be true of the TR O2 sensor?

[2QUIK6]

While trying to figure out some problems with my TPI over on the 3rdgen board, I learned that the heated O2 sensor used on 90-92 f-bodies becomes very inaccurate and unreliable the further from a "perfect" a/f ratio you get. So if your BLM's are at 127-129, then the O2 sensors readings will be almost perfect, but the further your BLMs deviate from 128, the more unreliable the O2 mv readings become for the computer to use, thus making the BLM correction amounts less reliable to keep you motor running "clean".
Therefore, you cannot say that the "Computer will compensate for changes in air temp, fuel pressure, etc by the adjustment of the BLM" , in fact the further off from the optimal 128 you are, chances are the a/f ratio will not get corrected reliably, maybe only close, therefore allowing you to still fail emissions testing.
*
Would this be true for the stock non-heated O2 sensors on our cars?

 

[OKTurbo]

I'm no expert, so take what I say with a grain of salt.....

You are correct in saying that the stock O2 (narrow band/switching style) is very inaccurate away from stoichiometric (14.7:1). This is why you shouldn't take WOT O2 readings as an actual A/F ratio. A good wideband O2 is MUCH more accurate at reading A/F ratio.

If your BLM's are within a controllable range your A/F mixture will be correct.....even if the BLM/INT are at 100. This is what the BLM is doing....it's adjusting the injector pulse widths to maintain the proper A/F ratio. Unless the BLM is bottomed out/topped out the computer is still in control of the mixture.

WOT is another story. The computer is making no corrections at WOT anyway....the car goes into open loop. ANY attempt at reading the stock O2 to actually get an A/F ratio is futile. You can use it as a comparative value, but not as an actual ratio. The computer is learning a certain BLM value (can't remember which cell it is) and this is the BLM used for WOT fueling. If this BLM is off of 128 by very much....Yes you'll have some tuning to do. This is the main reason many race chips have a WOT BLM lock at 128....this way you can adjust your fuel pressure at idle to adjust your WOT fueling without it going haywire.

If you raise your FP to get a richer WOT and then drive your car up to the starting line the BLM's have just been adjusted for the added FP....this will screw up your WOT fueling if you don't lock the WOT BLM at 128.

Any of this would be true for a single wire O2 or a heated O2 sensor.

That's my story...and I'm sticking to it.

John

 

[gofstbuick]

I'm no expert either, but I think it doesn't matter whether the O2 sesnsor is heated or not. The only thing the sesnsor does is sense O2. It must respond from full rich to full lean or visa versa in less than 100milliseconds. When the engine is cold, going to closed loop too soon is too lean and at WOT, you need all the fuel the engine demands, so why let the O2 try to make compensations at WOT.
IMHO, some emissions problems are because our engines are inherently lean (probably in an effort to meet federal emissions requirements).

 

[TurboDave]

Originally posted by OKTurbo
I'm no expert, so take what I say with a grain of salt.....

You are correct in saying that the stock O2 (narrow band/switching style) is very inaccurate away from stoichiometric (14.7:1). This is why you shouldn't take WOT O2 readings as an actual A/F ratio. A good wideband O2 is MUCH more accurate at reading A/F ratio.

If your BLM's are within a controllable range your A/F mixture will be correct.....even if the BLM/INT are at 100. This is what the BLM is doing....it's adjusting the injector pulse widths to maintain the proper A/F ratio. Unless the BLM is bottomed out/topped out the computer is still in control of the mixture.

WOT is another story. The computer is making no corrections at WOT anyway....the car goes into open loop. ANY attempt at reading the stock O2 to actually get an A/F ratio is futile. You can use it as a comparative value, but not as an actual ratio. The computer is learning a certain BLM value (can't remember which cell it is) and this is the BLM used for WOT fueling. If this BLM is off of 128 by very much....Yes you'll have some tuning to do. This is the main reason many race chips have a WOT BLM lock at 128....this way you can adjust your fuel pressure at idle to adjust your WOT fueling without it going haywire.

If you raise your FP to get a richer WOT and then drive your car up to the starting line the BLM's have just been adjusted for the added FP....this will screw up your WOT fueling if you don't lock the WOT BLM at 128.

Any of this would be true for a single wire O2 or a heated O2 sensor.

That's my story...and I'm sticking to it.

John

And you should stick to it. You pretty much nailed it.

 

[bruce]

Originally posted by OKTurbo
I'm no expert, so take what I say with a grain of salt.....
Any of this would be true for a single wire O2 or a heated O2 sensor.
That's my story...and I'm sticking to it.

Well yes, and everything you've said is true.

But, a heated O2 gives a more consistant ie repeatable output since it's min operating temp isn't a function of EGT. So for around town and running to the grocery store a heated one can improve drivibility.

For day in and day out best drivibility, I've gone to a heated one, and relocated to where the converter was. This does two things, it allows the heater in the O2 to maintain a more constant temp. (and hense output), and the sensor is in a area of less backpressure, both of which skew it's output.

 

[OKTurbo]

I agree bruce, but 2QUIK6's original question about whether the computer is compensating for changing conditions with a BLM off of 128 is not true. As long as the BLM's are between...say 90 and 150...the computer is still maintaining the "proper" mixture.
((except at WOT))

While we're discussing O2's, bruce, have you been doing anymore tuning with the DIY_WB? I rec'd the parts kits from Steve C. a couple of weeks ago but haven't had time to solder the board up with T-ball still going on. I'd like to try one of Bruce Roe's LED displays, but I still need to contact him for a board.

What set-up are you using? Data logger on PC / DVM / LED display. How has the DIY_WB been holding up with use?

I'm excited to give it a try. There seemed to be a lot of interest at first, but things have quieted down with respect to the DIY_WB. Any other TurboBuick guys out there using the DIY_WB?

John

 

[boostcreep]

So let me ask this...let's say we live in a perfect enviroment, the temp never changes, the baro never changes the humidity is constant, and any other piece of the enviorment never changes..would we still have a need for BLM's? And let's say the load on the motor never changes either. So, do we need 'em?

Nick

 

[2QUIK6]

You would still need BLM corrections if all of those things did not change because spark advance and rpm would be changing even if load was not, unless you are able to program that tables perfectly in the chip for a perfect mixture at all rpm/spark/air flow combinations
*
Yes I know what the BLM cells are used for, but my original question was whether or not the O2 sensor reads more inaccurately the further off from a perfect mixture you get?? For example, if it reads inaccurately at say a current BLM reading of 105, how can the ECM possibly cut fuel correctly to give you a perfect A/F mixture thats desired?
(Hence still being able to fail an emmissions test even though the ECM is compensating.)
Whereas having a BLM reading of 120, your more likely to pass because the O2 sensor is providing more accurate data for correction.

 

[OKTurbo]

It's really not a matter of accuracy since the O2 sensor (both single wire and heated) are switching type sensors. So to answer your question...yes the O2 sensor is less accurate the further you get from 14.7:1 A/F ratio.

But the BLM doesn't really affect this accuracy. If your BLM was at 105 the computer's ability to adjust the mixture would be the same if the BLM was exactly 128. You also have to consider what the INT #'s are doing also. I'll often see (esp with my RA93 chip and 009's) BLM's of 100 but INT's of 135. One is saying rich...the other is saying lean. This has to do with the relationship between the BLM's and INT's. BLM's are like the base (long term) corrections and the INT's are the shorter term. If the INT's stay lean for long enough the computer will bump up the BLM number to maintain the mixture.

The O2 sensor accuracy isn't really a function of where the BLM's are at in their range. Along the same lines....the accuracy of the computer to adjust the mixture is no greater if the BLM's are at 100 or if they are at 140.

John

 

[2QUIK6]

Originally posted by OKTurbo
The O2 sensor accuracy isn't really a function of where the BLM's are at in their range. Along the same lines....the accuracy of the computer to adjust the mixture is no greater if the BLM's are at 100 or if they are at 140.

I guess its how the computer arrives at a given BLM that I'm having trouble with. Correct me if I'm wrong, but the only way the computer knows if it is running rich or lean is by the O2 voltage thats being fed to it right, so if the O2 sensor is more inaccuarte as you get away from the A/F of 14.7, how can the BLM be a near perfect correction if the O2 mV was off by an unknown amount? Maybe the O2s were enough to make the BLM 110 one time, but the the very same A/F at some other time is enough to make the BLM 112 because the O2 is inacurrate.
At least that's what I'm trying to get at anyway.
I guess to sum it up, is it best to try to program the chips fuel tables for a near 128 BLM / INT at all rpm/air mass/load combinations for best correction results?

 

[TurboTR]

Well let's keep one thing straight here, warmed up properly and all the switching O2 sensor is ~ dead nuts accurate for doing the job it was intended for. That job is to tell the control loop either rich of stoich or lean of stoich. That's it. The control loop doesn't look at how far away from stoich directly, it just looks at rich of stoich or lean of stoich. Maybe that's the part you're missing; sounds like you're assuming it's like a linear control scheme based around O2 volts. It's not. It's more of a Hi/Low scheme. Say I'm currently lean of stoich, ok ramp up fuel until the sensor switches over to rich. Now I'm rich of stoich, ok ramp down fuel until the sensor switches to lean of stoich again. Repeat as quickly as possible to try and average an overall stoich a/f ratio, with minimal sized swings into rich and lean territory

TurboTR

 

[TurboDave]

Originally posted by 2QUIK6

I guess its how the computer arrives at a given BLM that I'm having trouble with. Correct me if I'm wrong, but the only way the computer knows if it is running rich or lean is by the O2 voltage thats being fed to it right, so if the O2 sensor is more inaccuarte as you get away from the A/F of 14.7, how can the BLM be a near perfect correction if the O2 mV was off by an unknown amount? Maybe the O2s were enough to make the BLM 110 one time, but the the very same A/F at some other time is enough to make the BLM 112 because the O2 is inacurrate.
At least that's what I'm trying to get at anyway.
I guess to sum it up, is it best to try to program the chips fuel tables for a near 128 BLM / INT at all rpm/air mass/load combinations for best correction results?

You may be misunderstanding the purpose of BLM and how the O2 work together.
Yes, your assumptions of the innacuracy of the O2 sensor as it moves away from 14.7 are correct.
However, the purpose of the BLM/INT is to correct the fuel delivery to the point that the O2 is seeing 14.7 again.
A lot of people incorrectly assume because they see a BLM of 105 then their car is running rich. Well, no. since the BLM is 105, it may have at some short point in time been running rich to cause the BLM to go down to 105, BUT since it's at 105 (not blocked at one end or the other) it IS NO LONGER running rich. It is running stoiceometric 14.7:1. The BLM value brings the fuel delivery to a point that the O2 sensor is again reading in it's correct range.

And no, just because you're running a blm of 105 isn't going to cause a failed emmisions test. Something else is going on. At 105 the car should still be running 14.7:1 and there's still room for more adjustment, since the lower adjustment limit in most chips is 100 (some 90).

 

[bruce]

Originally posted by OKTurbo
I agree bruce, but 2QUIK6's original question about whether the computer is compensating for changing conditions with a BLM off of 128 is not true. As long as the BLM's are between...say 90 and 150...the computer is still maintaining the "proper" mixture.
((except at WOT))

While we're discussing O2's, bruce, have you been doing anymore tuning with the DIY_WB? I rec'd the parts kits from Steve C. a couple of weeks ago but haven't had time to solder the board up with T-ball still going on. I'd like to try one of Bruce Roe's LED displays, but I still need to contact him for a board.

What set-up are you using? Data logger on PC / DVM / LED display. How has the DIY_WB been holding up with use?

The DIY-WBs are approaching McDs burger sold. Well past a thousand. And being used in shops as well of little clusters of guys using them. I've put over 10K miles on one sensor and managed to cripple one by leaving it on for 3 days.

I'm using one of Bruce Roes display boards and it's at 9K without a problem.

 

[bruce]

What seems to be being ignored here is the strategy of how a O2 Sensor works.

It's output swings from high to low, and the fueling algarhythm (?) is set to allow that to happen depending on far the O2 overshots the PID (control devise) reverses the correction to swing the AFR in the other direction, The O2 and it's strategy is not desgned to maintain a 14.7:1 AFR, but only AVERAGE IT. So the system is rather sloopy. As you run converters hoter and hotter you have to add fuel to them to keep them active, so the topend fueling is too rich for max performance, using a converter.

 

[2QUIK6]

Thanks Dave and bruce, that was the missing piece of information that I didn't understand.
I pulled the chip out of my TPI and pulled the bin off of it (just got my pocket programmer yesterday) and the minimum allowed BLM setting was 108, so I was bottomed out and couldn't go any lower for correction. When it failed emissions it was at 108 because I recorded the test with my scan tool. I found if I set the idle speed up a bit then the BLMs would go to 112-115, I did that and it passed emissions, so I guess its because the BLMs were bottomed out and could not adj anymore.
Thanks guys!