Oxigen Sensor question

[Buzo]

I couldn't find the stupid projects forum so I'll throw it here. And I'm sorry if it is too stupid.

A friend of mine gave a narrow band oxygen sensor. It has the heater wires and it's brand new. And I got it for free.

But it is half of the size of the standard O2's because it is used in motorcycles. I did read the technical information and it says nothing about limits in exhaust gas flow or anything that prevent its use in our engine.

I may give it a try, just need to find the proper thread reduction.
The advantage would be the heater, and that its 24 yrs newer than my stock O2.
Disadvantages, too small for the size of the exhaust tube?

I do have a WB O2 sensor on hand, but I have not ordered the controller yet. So this would be an interim install just to see how degraded my stock sensor is.

I could easily tell the difference because I have hundreds of traces of the behavior of my stock sensor. So I can directly compare any change in performance.

[Buzo]

I just realized all my projects start with a picture in this little table hahah

Here is a comparison of the size of a WB sensor vs the bike's NB sensor.

I'm doing this just for the heater, the spec says the O2 signal is usable after 30 seconds. Right now I am waiting until the coolant temp is above 50 C to start the closed loop control in my MS, which is 1/3rd of my daily trip.

Hope it sounds less stupid now

http://img580.imageshack.us/img580/5168/48607374.jpg (http://imageshack.us/photo/my-images/580/48607374.jpg/)

[Oldblueaccord]

SO are you asking a question or telling us?


I think if you can find the smaller bung you could weld it in or find an adapter. I would imagine there both metric threaded. I think the larger one is 14mm. Moroso makes weld in bungs.


wp

[Buzo]

When I started the thread I was going to ask.
But then I said, I have everything to answer (with data) my own question.
People usually would look for the correct replacement part and don't experiment with this sensitive function of the car.
And People usually don't have a data acquisition system permanently tied to the car.

I just want to know what's different between a 24 yrs old sensor and a new one. Some says the sensor gets lazy with time, so I hope to get a faster response besides my earlier closed loop control engagement.

This is my car at cruising speed. The blue line is the narrow band O2's closed loop response. Will see if that changes with the new sensor.

http://img11.imageshack.us/img11/8646/currento2reading.png (http://imageshack.us/photo/my-images/11/currento2reading.png/)

[ecogabriel]

You can get a new O2 sensor for as little as $10 + shipping at rockauto.com
But as you said, if you want to experiment it is understandable trying that bike sensor... especially because the OE sensor is unheated. Just for the record, 4G Accord oxygen sensors are heated too.
Anyway, it may be an interesting experiment.

[import racer]

Why do some have one wire and some have four wires.

[ecogabriel]

Why do some have one wire and some have four wires.

two wires operate the heater element inside the sensor; the other two are sensor signal and ground.

[import racer]

So could we run heated ones on ours,and where would the heated wires go.

[Buzo]

So could we run heated ones on ours,and where would the heated wires go.

One heater wire go to the +12 V ignition signal and the other to the ground, but they recommend to limit the voltage getting to the heater to no more than 13.5 V.

All the ECU's need to wait for some time until the oxygen sensor gets the working temperature (600 C aprox) inside the exhaust, so there is a period of time after startup where the amount of fuel is only calculated and the car runs open loop (no feedback of the air/fuel mixture).

Adding the heater to the sensors was designed to make that waiting time short.

I don't know, however, what is the algorithm in our cars (carbs or FI) before they check for the O2 sensor. So don't know if there will be any benefit for a stock ECM.

[Buzo]

Got the sensor installed today and started to run all the the wiring.
If this location works OK for this sensor, then I'm going to use the other side of the tube for my WB sensor when the time comes.

http://img718.imageshack.us/img718/7060/o2installed.jpg (http://imageshack.us/photo/my-images/718/o2installed.jpg/)

[ecogabriel]

I suspect the ECU in regular EFI cars wait for the engine to reach certain temperature before reading the O2 sensor; that gives the sensor enough time to heat up. Since you are running megasquirt I believe it operates heated sensors right?

[Oldblueaccord]

I think its preferred before the Catalytic converter. I cant tell from your picture.



wp

[obdriver6]

I suspect the ECU in regular EFI cars wait for the engine to reach certain temperature before reading the O2 sensor; that gives the sensor enough time to heat up. Since you are running megasquirt I believe it operates heated sensors right?

The ecu doesn't wait, it stay in open loop until the O2 heats up enough to produce a signal. Thats when it goes into close loop.

[Buzo]

The ecu doesn't wait, it stay in open loop until the O2 heats up enough to produce a signal. Thats when it goes into close loop.

If that's the case then a heated O2 sensor would be useful for any 3geez.

I finished the installation of my new sensor and fired the car up and waited until it went to closed loop. There were no problems but I'll wait for a longer trip to log and post the results. So far it looks like -in this specific case- the size doesn't matter :)

[obdriver6]

Yes they would but I think that the heater can't be ON all the time so you would have to find a way to control it. The benefits of having an HO2 would be that it goes to close loop faster (from a couple of minutes to a couple of seconds) which equals less pollution and less gas wasted. Thats why some OBD1 cars and all OBD2 have them.

[Buzo]

Yes they would but I think that the heater can't be ON all the time so you would have to find a way to control it. The benefits of having an HO2 would be that it goes to close loop faster (from a couple of minutes to a couple of seconds) which equals less pollution and less gas wasted. Thats why some OBD1 cars and all OBD2 have them.

There is a formula to control the duty cycle of the heater:

Duty Cycle = (13.5)^2 / (actual battery voltage)^2

So you need something to keep continuously reading the battery voltage & calculating the equation, like a microprocessor. But if your voltage regulator is working OK, you can connect it directly with very few risk of damage.

In my case I implemented the control in one of the microprocessors I already have installed in my car.

[Buzo]

This is my car at cruising speed. The blue line is the narrow band O2's closed loop response. Will see if that changes with the new sensor.

http://img11.imageshack.us/img11/8646/currento2reading.png (http://imageshack.us/photo/my-images/11/currento2reading.png/)

And this is my car after I put the new tiny O2 sensor.

http://img20.imageshack.us/img20/7783/newo2.png (http://imageshack.us/photo/my-images/20/newo2.png/)

The scale in the X axis is the same, both with a length of 55 seconds.
Its evident the new sensor's response is lot faster.

During start up this morning it was at 10 C, the IAC valve almost took more time to go to the start up position than the new sensor to be ready. So now I can go back and re-tune the after start and warm up enrichments in my MS.

[2drSE-i]

That's pretty awesome! Good work!

[Buzo]

Before I move anything in my tune, I'll check if there is a fuel economy improvement just by removing the lag due to the age of the old sensor.

In the mean time, it is only a nice before-after chart :)

[Buzo]

I tested only with 5 gallons, but I got 25 MPG compared to the 22 I was getting before!

The only change to the tune was to start closed loop at 20 C instead of 50 C (I think it was in 70 C during February).

I don't know if the benefit was due to the early closed loop, or to the faster response of the sensor, or both. But 3 extra MPG sounds excellent for me!

[Buzo]

Confirming 250 miles / 10 gallons = 25 MPG. City drive.

If we take into account that my wife's 2012 Direct Injection Focus is at 29 MPG, I think I have done a good job tuning my Accord!

[MessyHonda]

good stuff man. i need to learn how to tune my car instead of paying people to do a half ass job

[Buzo]

Today I made another change to my tune that must give me another improvement in fuel economy.

I have coil on plugs in my car and I was firing them in wasted spark mode, it means that cylinders 1 & 4 and 2 & 3 were being fired in pairs at the same time. One day during a hard acceleration I blew the coil's fuse out. It was 15 Amps.

So I was "wasting" more than 7.5 Amps in firing the coil that was in the expulsion cycle. In order to do sequential ignition in megasquirt I had to put a cam sensor so I modified my dizzy and added a hall sensor and wired individual outputs to my coils.

Don't know how much load is 7.5 Amps for our alternator, but hope to get "something" extra. We'll see...

[Buzo]

good stuff man. i need to learn how to tune my car instead of paying people to do a half ass job

I have learned a lot by switching from Carb to TBI. From understanding the fuel equation, to learning the effect of the temperature in the mix of fuel and air. I thank Internet and the forums like this one where we get answers to almost any question. So go ahead and put hands on.

[Buzo]

I captured the Battery voltage before and after the modification.
Any coil will store current when energized and will return it back to the battery when disconnected.

The before pic has a lot of voltage spikes as product of that current getting returned to the battery by the two coils, and also product of the higher current demand.

The after picture has a flatter voltage. The continuous line is the 13.5 V. I don't know if this is going to mean something in fuel economy, but the voltage regulator system is working more stable now.

Before (wasted spark -firing two coils at the time))
http://img406.imageshack.us/img406/7735/battbefore.png (http://imageshack.us/photo/my-images/406/battbefore.png/)

After (sequential ignition - firing one coil at the time)
http://img4.imageshack.us/img4/8229/battafter.png (http://imageshack.us/photo/my-images/4/battafter.png/)

[Oldblueaccord]

Before I move anything in my tune, I'll check if there is a fuel economy improvement just by removing the lag due to the age of the old sensor.

In the mean time, it is only a nice before-after chart :)

I should let you test mine sometime there original with 315,000 miles on them. I pulled a 32 mpg with 87 octane and a few ounces of acetone.

What fuel octane do you use or do you have where you live?

I also wonder how much the stock Honda ECU samples the O2 sensors and how much they effect the fuel mix. I have thought about unhooking them and doing a mpg test to see.


wp

[Buzo]

I should let you test mine sometime there original with 315,000 miles on them. I pulled a 32 mpg with 87 octane and a few ounces of acetone.

What fuel octane do you use or do you have where you live?

I also wonder how much the stock Honda ECU samples the O2 sensors and how much they effect the fuel mix. I have thought about unhooking them and doing a mpg test to see.


wp

WOW 32 MPG! I would not move anything! Not sure about the octanes in our fuels, all I know is that we have only two options, one expensive and another less expensive, so I use the less expensive.

What is that acetone thing? Do you pour it in the tank with each refill?

[Oldblueaccord]

Well my car is stock fuel injection so to me 28-32 mpg is about average maybe a little high. The only two things that might help it is I have a K+N filter cone directly on the throttle body and newer RCeng. 250cc injectors. I have some wideband O2 checks and for the most part the cars runs very well with it being rich 12.5 at higher rpms. its idles at 14.7 when warmed up.

Acetone is something I have been using on/off for about 5 years. you just need a small amount like 2 ounces per 10 gallons of fuel. it works very well in carburator cars and helps a small amount on most,but not all,fuel injected cars. I think it smoothes out the burn rate of gasoline but thats just a theory. theres is alot about it on the internet if you google search it.

The last 2 years I have been running about 30% Ethanol with 87 octane fuel. I have been following your post because I am wanting to goto 100% with a minimal amount of modifications to my injection system. if it goes ok I would then rebuild my old motor and up the compression to take full advantage of Ethanol fuel. its is rated about 105 octane.

wp

[Buzo]

Running our cars with ethanol would be the most incredible project thought!

When switching to 100% ethanol, you must upgrade the components inside the fuel tank. Ethanol is going to corrode any non-plated and copper materials. I'm sure you knew it and are planning for it too.

I know a good source of ethanol-ready fuel components in town, so let me know when you are ready to start and I'll see what I can get for you.

[Buzo]

Question for my friend Cygnus...

I did switch my setup to sequential ignition for a week or so, just enough to measure the new fuel economy and I got 28-29 MPG.

However, I had to switch back to wasted spark because I couldn't get rid of a sporadic loss of synchrony. I'll say 2 loss of sync in a 15 minutes trip.

A loss of sync in Megasquirt means your engine shuts off for a fraction of a second and then turns on again, so you see your rpm gauge going to zero and back. And you fell the engine how it suddenly shakes to one side.

OK, the question. Could you please tell me how is your cam sensor installed in your car? I am doing something wrong, I know it, but after I applied all my knowledge of signal filtering I just couldn't fix that noise that causes the loss of sync. Thanks.

[Oldblueaccord]

It this cam signal wire shielded or with a drain wire possible? it might need to be.


wp

[ecogabriel]

The sensor wires in the EFI cars are shielded; I took the wiring maze apart to repair a broken connector near the distributor and saw the two shielded wires going to the distributor (one CYL, the other CRANK).
From what I have read about EDIS and seeing the sensors and triggers' positions inside the distributor, they should be pretty close and on a solid (vibration-free) surface.... EDIT: I have read your thread and your hall sensor inside the distributor
TIP: you can use the cam sensor from a 86-87 USDM Acura Integra in place of your old distributor. It has a crank and cyl sensor inside. The only modification would be finding a way to screw it to the a20's cylinder head, and possibly keying it differently; otherwise it fits perfect (I tried one in the junkyard). Those sensors are magnetic (is that the proper word?) using only two wires.

[Buzo]

Thanks for your advises. I'll get to fix the cam issue in a short time...

[cygnus x-1]

However, I had to switch back to wasted spark because I couldn't get rid of a sporadic loss of synchrony. I'll say 2 loss of sync in a 15 minutes trip.


Hey Buzo, sorry I missed you question until now. I assume you are still using the 36-1 toothed wheel for primary ignition input? If so then I think the only reason it would lose sync is if the primary input has a problem. The secondary input (cam sensor) is only used to signal whether it's on the first or second half of the engine cycle. In a sequential setup if the cam sensor is lost completely the MS will revert automatically back to semi-sequential, e.g waste spark mode for ignition. You shouldn't notice when it switches. If the cam sensor signals too often (noisy) I think it should do the same thing, but I'm not as sure in this case.

So the sync problem only occurs when you add the cam sensor?





OK, the question. Could you please tell me how is your cam sensor installed in your car? I am doing something wrong, I know it, but after I applied all my knowledge of signal filtering I just couldn't fix that noise that causes the loss of sync. Thanks.


I'm using a hall effect sensor with a magnet buried in the head of one bolt on the cam pulley.


http://bluegreenlabs.com/Prelude/Mods/SequentialCOP/CamSensor.jpg



The bolt with the blue on it has the magnet. This picture was taken awhile ago, and looking at it now I'm not sure if the bolt is in the right place. I'll have to double check when I get home later. I've never had a problem with loss of sync so it's been pretty reliable.

I'll try and read through your old threads on msextra.com and see if anything jumps out that could be causing you sync issue.

Using a fuel injection distributor would work for sequential ignition, but since you already have a crank mounted wheel it would be a bit of a step down. Crank based signals are more accurate because they aren't affected by cam belt stretch and vibration.


C|

[defiance]

Wow I like what im seeing here ! So you guys are running megasquirt in place of the stock ecu, and using extra sensors to increase gas mileage ? I like where this is going. any updates ?