O2 location

[webebob]

It appears several people are using FI and data logging.
Just wondering were is the best place for O2 sensor to be located.
The pipes collect so far from the engine. Is that too far? IE... Too cool?
Closer to the engine would read a single pipe. Is that too "jumpy"?
I've tried both with reservations on both.
Any suggestions?

[dalola]

webebob,
I use O2 sensors for carb tuning as well, and install them in the collector, both sides. This is the first place to tap into to get all 4 cylinders.
Dave

[CCX33911]

My car is not totally complete, but I have run the engine a bit, etc. I have them mounted in the sidepipe at the collector. I have HEATED sensors. I think at this distance that you need heated sensors.

[Toivo]

I have unheated sensors in the collectors.

Only use one for the ECM.

I believe in symetry (sp?) though, so there's one for each bank.

Seems to work fine and I always have a spare.

Reading down the posts, I should also say I'm using Tri-Y headers which are relatively short, but it's still a ways down to the collector, just not quite as far down as with long-tube headers. And it does seem to work fine there, in fact I have fairly low correction factors (and only allow +-10% max anyway) of 1,2 or 3% +-.

You asked for actual experiences

[Wade Chamberlain]

I run a blown and injected 302 in my FFR. I had the 02's in the collectors at first, but had problems when the computer tried to transition from open to closed loop control. The 02's were not hot enough, even though they are heated, and would indicate a rich condition. The computer tried to compensate by leaning up the mixture, which would cause a lean misfire, stumbling, etc. Once the car was up to full operating temperature and the 02's were hot, it seemed to run OK.

I relocated the 02's to a single pipe, about 13-14" out from the head and the problem is gone. However, I am concerned about only sampling one cylinder.

[Double Venom]

Just a repeat:

The Viper engine, engineer, (yup-the engineer for USA) told us not to run the 02's in a single cylinder as it would definately confuse the computer. But then again, we have 5 tubes

Ist DV we made a 1" solid plate, drilled the 5 holes then on end drilled and tapped for the 02 to read all five pipes prior to reaching the collector. A sandwhich plate if you will.

All subsequent DV's including the DV II are using short headers, allowing us to use a long one piece elbow right into the sidepipes. At the very end of the headers is now where we are installing the 02's. So far this seems to work out best.

Bottom line, keep them close, better yet keep them close and reading all 4 tubes.
DV

[Leo]

Mr. DV,

Please elaborate on your initial solution. I'm envisioning a plate that connects all 5 header tubes together via a small hole and the O2 sensor in the center of this 'manifold'. Did it work? I'm afraid long tubes are a must for me. Thanks!!!

[RICK LAKE]

Webebob HI I have speed-pro F.I. system in my car. I have O2 sensor in both side pipes. The right side is for the computor, the left side is for the lean,rich meter so I can do quick checks when driving. I used a infared pyrometer for the hottest point at the collector in the exhaust and mounted the O2 sensor in those places. I have my header tubes rapped in thermo shield tubes so the heat stays in the exhaust. the system is real close between the meter and the laptop when checking the mixture. I am going to put a EGT meter in my car when I have the blower running as to not lean it out and damage the motor. Hope this helps. my sensor are wide band for supercharging but are not heated. My Alum. motor gets warm in about 2 minutes I have a 180 thermostat in it. It will run about 195 in the heat under hard driving and hold. good luck with it. Rick Lake

[leadfoot]

I have installed the O2 sensor just where the 4 pipes meet the collector outside of the cars front quarter panel. This is almost 30 inches from the port.

I am running a Holley Commander 950, and the sensor reads 0.02 (VERY lean). The manual says that installation too far away will show this reading.

My question is with heated vs. non-heated. How do I know what I have? Will a heated sensor allow this distance?

My only other option is to run it out of one pipe, or custom plumb the 4 into 1 (a-la DV style) into 4 and then into the collector, to retain the look on the outside of the car.

Is an O2 sensor a sensor is a sensor?

Thanks...

[CCX33911]

heated usually has four wires (sometimes three). Two are for the heater and two for the signal. Usually they read ~0.5 V for stociometric (sp?). They are very non-linear though.

The heater is turned on with relay from 12V source and ground.

[dalola]

A heated O2 sensor is generally suited for measurements in the lean range, and come into operation in the "warm-up" phase, before the stoichiometric ratio is reached.(the term refering to the ideal A/F ratio of 14.7:1)

In MOST of our Cobra applications, the use of heated sensors would not be of much benefit.

Hope this helps.

Dave

[CCX33911]

I'm not a O2 sensor expert, but my understanding is that they produce accurate results when the materials are at a particular optimal operating range (something like 600 F). The thought with the Cobra is that the O2 sensors are mounted some 30" down stream and may not be operating at the optimal heat range to produce accurate results. The heater would suppliment the heat produced by the exhaust to keep the sensor hot enough. Again I have not measured the temperatures involved nor have I seen specific numbers which would indicate the required operating temperatures, etc.

[dalola]

Mr. Gibbs,
You are correct, in that the sensor must achieve a minimum operating temp of around 400F to function properly. It is at these temps(and higher), the electrolyte becomes conductive and generates a characteristic galvanic charge, and this voltage is an index of the gas oxygen content. Heated sensors reach this temp much quicker, and will maintain optimum temps as well. If you need continuous imput from your O2(computor controlled EFI engines), then heated may be the way to go. On carburated engines, when O2's are used for tuning purposes, you would only need heated sensors if the exhaust gas temp was not sufficient once at normal engine operating temps. If unsure, I suppose it would be wise to error on the heated side.

Dave

[leadfoot]

I believe that I have a non-heated O2 sensor (3 wires), and because of the 30-inch distance, the Holley Commander show VERY lean no matter what I do. Can I swap in any old heated O2 sensor?

[TButtrick]

I had put my O2 sensors in a single tube about 9" from the head on a small block and I think I may have fried the sensors being this close. The sensors I used were the new Bosch large barrel types. I swapped these out for the older small stem sensors that were Ford OEM before they switched to the newer Bosch types. I speculate that the older style O2 sensors are more dependant on heat to operate (600F) but are less susceptible to damage by extreme heat. In contrast, the newer large barrel types seem less dependant on heat to operate but can be damaged by extreme conditions. I believe that both the four and three wire sensors are heated. The three wire uses two for the heat element and one for the O2 signal using the case for return. The four wire is the same but includes a forth wire for signal return rather than the case. There is more information I've gathered at the end of the below thread that shows pictures of the two types of sensors I refer to...

http://www.cobraforum.com/cgi-bin/ul...1&t=020955&p=2

[Ken Oikawa]

leadfoot

3 wire O2 sensor is the heated version. Hot, ground and signal.

I use a single wire O2 sensor for my AF gauge and it takes more than 15 minutes of idling before the needle on the gauge moves.
I have the O2 sensor mounted at the collector.

[leadfoot]

Ken...

Thanks... Your 15 minute comment helped! I have been trying to tune the car, and since I am not getting a proper reading on the sensor, I have never let the car idle long enough to get everything up to temp....

Now with our snow... it might just be done for the year.

[Dave Samson]

I have a single wire Bosch sensor in my Holden (Chev) engine and have made up a 0 - 1000 mV meter plus a Summit air/fuel guage to back it up. Both show wild fluctuations between 100 - 900 mV after only 2 - 3 miniutes.
At cruising speeds, no acceleration or deceleration I see around 400 - 500 mV.
The sensor is positioned just after the weld where the 4 header pipes join the collector. I believe the sensor is working ok.