Is there any downside/upside to running wideband sensors all the time? Will it confuse the PCM or will it provide opportunites for better all around performance?

 

OK how do you mean? You can't just hook up a wideband to the stock PCM and have it run all gravy. Your car is not setup to use wideband input, hence the use of narrowband O2s. If you want, hook up FAST or MS or all the other crazy stuff. That's wideband engine management. Otherwise, if you have a standalone unit, then run it as often as you like, it'll have nothing to do with your PCM.

 

That is not exactly true, Using the PLX wideband controllers, you can use the wideband to feed the stock PCM. The stock PCM will recieve a Narrow band signal but it is much more detailed in its results.

 

FJO has the same. Wideband O2 with narrowband waveform output to the PCM.

 

I will agree that it will be more accurate, but I was under the impression that the intial question was about running a wideband directly. Without the signal converter boxes or whatever PLX uses, there is no standalone wideband that will mate with the LS1 PCM as the OS is not setup for wideband input.

Besides, the narrowband signal is still only used at part throttle closed loop operation. Outside of that, you may have good data, but that data is ignored by the PCM as far as controlling fueling functions.

To answer the question though, there will be no confusion by the PCM since it does not truly accept wideband input. This narrowband conversion just allows for more accurate stoich detection but as far as the PCM is concerned there is no data acquisition at WOT. From a tuning standpoint though, this will help solidify the correlation b/w narrowband MV and true AFR. And if using any other standalone wideband, it operates completely independent of your EFI system.

 

The LM-1 and LC-1 have dual outputs for just this purpose. One defaults to "Narowband Simulation" mode and the other is for driving a Gage or ECU. You can actually change the switching point of the narrowband simulation to values other than 14.7:1 to fool the ECU.

 

Do you think a $35 "wideband" sensor is really more accurate at 14.64 than a $80 factory sensor, that is designed to just be accurate in that switching range?

Ryan

 

Where in the world are you buying wideband sensors for $35.00?

2nd if you are paying $80.00 for factory sensors you are buying in the wrong place.

Here is how it works.
A "Factory" narrow band sensor can read 10.1 to 20.1 A/F the problem is that it has a very narrow range of accuracy and when the PCM reads that voltage it is either on or off Or being 14.7 which is Stoichiometric pernounced sto-key-a-metric.

With a narrow band sensor the voltage range is 1 volt to 0 volts. as an example .900 may be 19:1 A/F and .100 may be 10:1. You can read it that way and you can actually see that with HP Tuners in the histogram when monitoring the narrow band field.

A Wideband is a 5 volt sensor allowing for a much more usable voltage range. That voltage range is then in turn converted to a ratio rich or lean.

It is the same principle as the difference between a 1 bar map sensor reading 5 volts and a 3 bar reading the same 5 volt range. You lose resolution. Most Wideband sensors range in price anywhere from $80.00 to $500. depending on how many wires, and if it is a heated or non heated unit.

Hope that helps the confusion.

 

As far as accuracy goes, the narrowband sensors are only accurate RIGHT AT stochiometric. That is their switching point. It is inherant in the chemistry of the sensor. The voltage range of the narrowband sensor is almost meaningless except that it switches polarity with respect to it's 1/2 scale output at exactly 14.7AFR. By how many millivlots it does this is dependant on AFR, EGT and exhaust backpressure. The "linear range" is just the effect of an instability. i.e. 900mV can mean almost any AFR depending on what the exhaust temperature and backpressure are at any given time. I know... I've tested it.

The "output" of a wideband sensor is the same as a narrowband sensor. It is the wideband controller that outputs 0-5v. The 0-5v that the controller outputs is proportional to the current that the contrroller has to apply to the sensor's pump cell to get the narrowband element inside the wideband sensor to switch on and off.

A couple months ago, i described how these things work internally. Since I am WAY too lazy to type all that again, here it is:

https://ls1tech.com/forums/showthrea...5&page=4&pp=20

need to scroll down a bit.


The cheapest wideband controller is probably the LC-1 $149 w/o the sensor or $199 w/ sensor.

 

the Bosch LSU4.2 sensors that most widebands seem to be using are only about 35... I picked up a few a while back for 40 or so each shipped

1stvwparts.com seems to be pretty popular for them


dr. mike, excellent info, Thanks!, dunno why I didn't see this the first time. I'm going to add your posts to the main sticky on this section as that is some very interesting information about wideband sensors and the LM-1

 

Can you tell me where to look for GM sensors for less than $75 then I was using prices I can get parts from dealers for. Sensors for my car I was told around $80, rear vette sensors for around $100 each

The VW sensor for my Lm1 i got a replacement for around 35.

I normally use sensors a lot more expensive for widebands, with a afm1000 being the norm.

I don't see closed loop working well with a wideband setup as a narrow band, as the pcm expects to see the voltage swing rich, lean, rich, lean. where a wideband will not have the same swinging, or even be fast enough to simulate a narrowband o2's output.

Ryan

 

Thaks horist Let me repost them with the spelling and grammar cleaned up 1st

 

The wideband controllers actually do simulate the 0-1v swings of the Narrowband sensor ( if they are set up properly ). You can even move the switching point of the simulation. i.e. you can have an LM-1 or LC-1 simulate a narrowband at 15.5:1 instead of 14.7:1. The ECU/PCM will see this and fuel the engine accordingly. May be good for a few MPG in part throttle/closed loop.



Most widebands are slow though. That's why it's best to use an LM-1 ( 100 samples per second ) or LC-1 ( 200 samples per second ). Thats faster than the ECU/PCM samples the narrowband input. Some other widebands, which shall remain nameless, can take 3 or 4 seconds to settle on an AFR value. The additional delay in the PCM's PID loop can cause overshoot/oscillation in the actual mixture.

 

On a semi-related topic, can you post links to current wideband O2 measurement vendors?

I know of this one: http://www.innovatemotorsports.com/

Others?