O2 Sensor
 

Is it reasonably possible to add an O2 sensor to a Merc MPI motor to ensure a more accurate fuel mix?

reasonably? basically no, you need to go to a dry exhaust for the O2 sensor to work properly

Hi ric232!
Jim at Arizona Speed and Marine has some info on O2 sensors in boats. The first problem you will have is the Mercruiser FI system has no provision for O2 so the only thing you can get is info on a monitor-not controll of your mixture. An O2 sensor that will do the job in a performance application adequately is a wide band unit that is very expensive, about $600-800. They do not tolerate leaded fuel very well. The oxygen sensor only works when hot no matter which type you use and the ceramic insulator will break with the first drop of water that touches it. Properly installed, proper computer, unleaded fuel, no water in the vicinity in the exhaust and this is the ultimate set up-VERY EXPENSIVE!!
You might call Jim at Arizona Speed-he is pretty good.

On top of the Thermo Shock that Blown1500 was talking about if you could get it last the Water in the Exhaust will add O2 to the exhaust and make the sensor lie to you.

If you had dry exhaust you could make a Wide band work as an Air/fuel ratio indicator. Bosch wide band ( 5 wire sensors) are a planar style and do have an integrated heater so getting them to operation temperature would not be a problem. I work here at Bosch and can forward anyone who wants info on theses sensors(Email me [email protected]). We do not make a stand alone system that would do this for you but I have heard of companies that do build them. There was a company I saw on the internet called LambdaBoy that made one. But like Blown1500 pointed out ...they are not cheap

MikeW

I don't see how the O2 sensor would work with the stock ECM unless it has some provision for it.
I've added a wide band (5 wire) sensor to the very top of the elbow of my CMIs. I also have "S" tailpipes so that the water is introduced well down stream of the sensor. The sensor is integrated to my F.A.S.T. ECM so that I can adjust it on the fly.
NOTE: I will only leave the sensor in for tuning purposes.

You can just buy a rich/lean indicator from PFM for what it's worth.

Later,
Dave

Mike,

A few other problems that come about besides the fact that the ECU can't recognize a wide band, ECU cannot self calibrate, sensors cannot take water, air velocity at certain speeds will give false readings, as well as misfires, if in collector, your only getting an average of 4cylinders, under some cases, sensor will not get hot enough to read in some cases (some cold starts, overly rich conditions), draws tremendous voltage from system as well as the fact that the 5 wire sensor is still not as accurate as the 6 wire UEGO sensor at the air fuel ratios that need to be run in high performance engine typically. I've seen variations of over 1 full air fuel at the 11's and 12's when compared to the Horiba units. With the tremendous duty cycle of a marine engine, if the air fuel is not dead on, then trouble could be in the future, especially if your ECU is basing it's calibration off of a 02 sensor and can richen or lean.

It's certainly possible, but more difficult than most think.

Thanks,
Dustin

So basically if the air density is high the A/F will be leaner then when the air density is low, like summer boating compared cooler spring and fall boating???? To me that kinda takes some of the benefit out of fuel injection? Gottahaveit posted some plug pictures and is concerned he is running to lean since he added CMI's and a Vortech flame arrestor. Is it likely he is just running leaner because of the higher densities that come with lower winter time temps? If that is the case, how does he fatten up his mixture?

Who is PFM?

I ran the FAST systems last year with Merc/Kinsler injection. FAST uses the 5 wire NTK L1H1 wide band sensor. I started off the year by letting the engines idle lower than I should have and pulled up some water...I guess. Both quit working and I had to replace them to the tune of $400 each. After that I kept rpm up over 1000 and life was good. I am running KE pipes and haave the O2 at the collector by the tail pipe flange, probably 30" away from the end where the water dumps. Supposedly a dry type tail pipe. I will say the system was much improved, from a driveability standpoint, over carbs. I also had a bung down by a couple ports for the O2. What was interesting was that cruising around 3000, the a/f ratio readout with the injection was basically within probably .3-.4 at the port. With a dominator on top it varied probably .6 to .8 cruising.

What was nice with the FAST system is that you can specifically dial in the air/fuel ratio desired at any rpm/intake mfld pressure. A world of dif compared to a carb. I am doing some blowers for next year and will probably use the systems with injection. If you do not like the way the O2 adjusts the a/f you can miniminze or eliminate the O2 and use just rpm, map, and air temp. However, the motor still needs to see a dyno upon setup just to establish the base line.

I am not familiar with other systems but FAST has been trouble free. There are nicer systems out there, but I think the prices aare a little bit up there. For what it's worth!

Dustin
I never heard of a 6 wire sensor. A Bosch Wide (broad) band is 5 wire (2 for the heater circuit one is ground and 2 signal wires). We don't make one. All I was stating is it would be neat to use the wide band in a stand alone system only. It would be nice to use it as feed back to control but that would require an ECU that could read a Wide band. Bosch wide bands would work very well with a Hi Po systems. They can read richer and leaner Air/ fuel ratios than a tube type or planar sensor. As I agree with you water will give you a false reading but air velocity (I am assuming you mean Exhaust velocity because intake velocity will have no effect at all on a O2 sensor) will not affect a Bosch Wide band. As a wide band works under the same principle (Nernst principle) the ECU reads Pump cell current rather than voltage and this sensor gives an unmistakable, continuous signal.
MikeW

Mike, no question, they are mucho importante, you really shouldn't tune without, but I was stating that it's much more difficult for a wide variety of applications, you can surely make them work, but you also have to know when your getting false signals. Typically speaking, many tuners use the Bosch 5 wire sensors with their Motec systems, which are far cheaper than the Horibas, but the Horibas are very, very accurated when calibrated properly. As for the 6 wire sensors, they have an extra heating circuit, UEGO sensors is what there called. Because of this extra heating circuit, you get much more accurate air fuels when your in the peak power range such as 12.5:1 and below. The extra heating circuit helps burn the excess gas off the sensor for a more accurate reading. Other than that, there very similar in design. This is what the Horiba uses.

Another problem you have is the life of the sensor and with some air fuel systems, it does a self calibrated offset. We have to test our sensors in a special chamber to check its accuracy for the automotive side because of emissions, it's so critical. Boats, its more about just making power and making it live, so no matter what meter you have, if you've proven that combo to work whether it says 13:1 @ WOT, your making peak power, it's not detonating

As for velocity, I did mean exhaust air velocity. So when I said that air flow could cause a problem, with our data logging system in 8 cylinders, you can actually see when the cylinder was fired because the scope is that fine. But when the exhaust is trying to scavenge at slow speeds, there is tremendous turbulence in the marine exhaust and I can guarantee with many, many data files that show how imporant sensor location is at slow air speeds in the exhaust. But when you get to the collector and take the average, it's typically no where near the same air fuel. If you don't run 8, then your assumming that all 8 are within a certain % of distribution, you could have one 14:1 and one 10:1, your average would be skewed. Most of the problems come from cold start, as the motor will misfire a bit if air fuels are not dead on, if it misfires from bad ignition, improper spark, overly rich, etc. then you can get the cylinder to flame out and pump air, this will give the user a reading that typically would not make sense. Its also possible to be so rich, your lean. This happens when the fuel simply cannot be fired and runs right out the exhaust, but what was combusted could be near what the proper air fuel is. So you add more fuel, you can just make it worse, it gets leaner (combusted). These are just some of the things I've run into so thats why I worry some about the "stand-alone" systems, theres just so many things that can vary.

To my knowledge, PFM uses a 4 wire sensor.

FAST and Accel both can use a 5 wire sensor which are much more accurate than the 4 wire. It's certainly better than nothing and typically everything works out, so didn't mean to be negative, just wanted to let people know it's not always that simple.

Thanks,
Dustin

Dustin
AAHH.... 6 wires for multiple heater circuits. That is interesting. Bosch pioneered the "Planar" type sensor. We then incorporated this technology into the our Broad Band. I am not sure anyone else uses this technology. First a planar is built in layer and the sensor element is only 1.5mm thick. So keeping it heated to the 1292° - 1472° F it needs to work in is not a problem. Also it's operating range is: ë = 0.7 to 2.2 (atmosphere) (10.29:1 to 32.34:1). The Broad band Planar uses a "Pump Circuit" to draw negative Oxygen Ions into the sensor so movement in the Exhaust is not that critical for this sensor to operate- And your right, dose not mean be accurate. I understand better what you where referring to and I do agree that sensor placement is important.
MikeW

Oh... as for the bosch being cheaper..lol . Let me just say there was a little pricing problem. If you know anyone who use them and can still get them for the lower price. better stock up now. We kinda drop the ball on that one.
MikeW