There are two basic types now in use; the narrow-band Lambda sensor, and the broadband sensor.

The narrow-band Lambda sensor only detects the presence or absence of oxygen; its' output signal varies from roughly 0.1v to roughly 0.9v. It's basically a "yes" or "no" type of thing, and the ECU is constantly seeking back and fourth from too rich to too lean in order to keep the engine running near stochiometric (14.7:1 air/fuel ratio by weight).

The newer broadband sensors can accurately report the actual air/fuel ratio detected in the exhaust. They are not interchangeable. They sort of work opposite to the older lambda sensors.

 

Define "Air Fuel Sensor", where is it at on the vehicle? Is it up by the intake (MAS/MAF), On the exhaust, two spots (O2), or other?

It is before the catalytic converter. It is a v6. So there is one on each bank. The ones before the cat are a wide band type sensor. There are also 2 after the cat to monitor cat problems. They are narrow band sensor.

The o2 sensor I am working with is a 5 wire ion pumping sensor. I know it has 5 wires a 5v ref a heater control a heater ground a ecu air fuel input and one another ground.

The O2 sensor in the exhaust pipe ahead of the Catalytic Converter is incorrectly called an "Air Fuel Sensor". It gets very hot so its life is not long.

Yes I am having a lot of 2001 to 2005 model years coming in with codes for failed "air fuel sensors".

Actually, the oposite is true. The emissions system on a American vehicle must be warrantied for 50,000 miles. Most "One wire" or non heated Lambda sensors would fail before 50k because they were constantly being heated and cooled as the car would go in and out of idle. They couldn't handle the stress and the auto makers didn't want to handle the warranty claims. 3 or 4 wire heated sensors keep the sensor at one high temperature greatly prolonging their life.

I am seeing them fail around 100k miles.

I guess my point is that if he's a mechanic, he should know this.

I think i provided the info in this post.

Look, our OP just wants to use this for testing, and is a professional mechanic.

The way to approach this is to post something that will be quickly destroyed if the engine is running, which is just about anything that isn't designed with the high peak transients that will be experienced when the engine is running.

It could be powered directly by the battery, for all that matters. The low dropout really doesn't apply - as long as the circuit can output from near 0v to near 5v and at 0mA-400mA, it'll do the job.

It's not that hard.

I can't keep answering all of these questions by myself. Someone else come up with a viable circuit with parts that can be obtained from a local Radio Shack - which is what I have in mind. It can be done, simply and cheaply.

That would be fine but I would like to build it myself. Spoon feeding me helps no one. I just dont understand what is wrong with my set up? I am just confused on the math. Maybe this will look a little better. Amusing the ecu is 0 ohms ( I dont think it is) are my max values to the ecu at the bottom of the pic true????? As I spin the potentiometer I should see less volts and amps on the ecu data list.

A modern O2 sensor is electrically heated so it warms up fast and is able to do its emissions control job sooner. I remember adding a 25W heatsinked resistor to "fool" the ECU that the heater was intact when it and the O2 sensor were removed for more power (and lots of emissions). I think an oscillator replaced the O2 sensor because the ECU expects the O2 sensor to oscillate between too rich and too lean.

I had a failed O2 sensor with a big hole burned through it.

I dont want it oscillating I dont need it to this is not a replacement for the air fuel sensor. Just a test tool. I will not be in the car long or with the motor running.

There are two basic types now in use; the narrow-band Lambda sensor, and the broadband sensor.

The narrow-band Lambda sensor only detects the presence or absence of oxygen; its' output signal varies from roughly 0.1v to roughly 0.9v. It's basically a "yes" or "no" type of thing, and the ECU is constantly seeking back and fourth from too rich to too lean in order to keep the engine running near stochiometric (14.7:1 air/fuel ratio by weight).

The newer broadband sensors can accurately report the actual air/fuel ratio detected in the exhaust. They are not interchangeable. They sort of work opposite to the older lambda sensors.

This is not a yes no sensor.

 

I got this from the Megasquirt boys. They do not limit the output to 4.88 volts, they just use a simple 10k pot.
Make R1 240 ohms, eliminate R2 entirely. It is bypassing a large part of your 5 volts to ground. Make your adjustable pot 10000 ohms. You will now get 0 to 4.88 volts to feed to your tester. Tolerances of the 5 volt power supply, resistor R1 and the pot can affect the output voltage. To compensate for that, replace R1 with a 1000 ohm trimpot and adjust to suit. The load on the 5v regulator is a few miliwatts.

If you stick with the 25 ohm pot, R1 will need to be 0.6 ohms to limit the output to 4.88 volts. The load on the regulator now becomes`1.37 watts. A heatsink might be a good idea.

For any other combination:

I = E/R = [Vsup/(R1 + Rpot)] = Vout/Rpot
Vsup x Rpot = Vout(R1 + Rpot)

 

See the attached for what I had in mind; very similar to what JaguarJoe's talking about.

Two 470 Ohm resistors, one capacitor, one 10k pot, and one 5.1v Zener diode, all available at your local Radio Shack.

R1 allows ~15mA current through the Zener. The wideband sensor really won't output much current to speak of, so there is really no point in having the capacity to output more than microamperes. If the signal is being pulled low or high, there's a problem that needs to be repaired.

 

Cool thanks. I also found this http://www.allaboutcircuits.com/vol_1/chpt_7/1.html
it will help me with my math next time I do a series parallel set up. I see why my set up will not work now.

 

the other thing is that a linear regulator can't output all the way down to 0V, the minimum output would be about 1.25V, which I think someone mentioned above.

that resistor thing posted above will work fine (and go down to 0V), assuming the ECU input is high impedance. If not, could you stick a little cap in series with the output to prevent the DC loading? obviously the frequency response of this circuit can suck since its a DC application.

@sgtwookie, he mentions a 100mA max for the ECU input, right? we don't know the impedance of that input so how can you build the circuit with resistors? if it outputs more than 100mA he might fry the ECU, which I think is why he wanted a linear regulator.

also @audioguru stop being such a stickinthemud. he's clearly not some kid with a ricemobile, he's new to circuits, and drew up a nice diagram and explained his approach before asking for help. even if he was some obnoxious kid illegalizing his car, we don't know that, and he didn't come in saying "hi guys i want 2 mod my car can u bild me a circuit to do that pls?"

 

the other thing is that a linear regulator can't output all the way down to 0V, the minimum output would be about 1.25V, which I think someone mentioned above.

A 317 will go to 0 volts if the adjustment leg ties to -1.25 volts (a little trim helps if you need an exact 0). If you need long-lasting performance, use a reference diode and an op amp buffer to invert the level.

 

the other thing is that a linear regulator can't output all the way down to 0V, the minimum output would be about 1.25V, which I think someone mentioned above.

that resistor thing posted above will work fine (and go down to 0V), assuming the ECU input is high impedance. If not, could you stick a little cap in series with the output to prevent the DC loading? obviously the frequency response of this circuit can suck since its a DC application.

@sgtwookie, he mentions a 100mA max for the ECU input, right? we don't know the impedance of that input so how can you build the circuit with resistors? if it outputs more than 100mA he might fry the ECU, which I think is why he wanted a linear regulator.

also @audioguru stop being such a stickinthemud. he's clearly not some kid with a ricemobile, he's new to circuits, and drew up a nice diagram and explained his approach before asking for help. even if he was some obnoxious kid illegalizing his car, we don't know that, and he didn't come in saying "hi guys i want 2 mod my car can u bild me a circuit to do that pls?"

I dont understand half of what you just said but thanks...... I printed the book of this site and am reading it now. So maybe soon I will understand this all a little better. I have read a few books on the subject but it was just simple stuff.

Trying to simulate 2 wide bands and 2 narrow bands On a van nun the less would be pointless and a pine lol. The new ecus are smart and would probably figure out the signals are bogus in a few seconds.

 

i dont know much about the automotive end of things; im just using my knowledge of circuits.

what I am saying is that if you use the resistive circuit, the ECU pin might "load" it and cause the resistive circuit to not output the voltage or current that you want. Thus you need the linear regulator.

Do you actually need it to go all the way to 0V, or can you use this to check at some mid-scale value? for example if the ECU is meant to see 0-5 V, and you are just checking to see if it works, you can use any value from 0-5. so even if the circuit cannot output down to 0V you don't need it to, just set it to something like 3V to check if the ECU is working.

if this would suit your needs its a very simple circuit.

if you actually need it to go to 0V you might want to use a switching regulator (buck converter) with a filtered output to convert the 12-14V car battery voltage down to 5-0V.

 

I just need something that has a little window of change below 4.8 volts and less then 100ma that all.

Gathering the stuff to make ^^^^^^ circuit.

 

A modern O2 sensor is electrically heated so it warms up fast and is able to do its emissions control job sooner. I remember adding a 25W heatsinked resistor to "fool" the ECU that the heater was intact when it and the O2 sensor were removed for more power (and lots of emissions). I think an oscillator replaced the O2 sensor because the ECU expects the O2 sensor to oscillate between too rich and too lean.

I had a failed O2 sensor with a big hole burned through it.

So your car made the 'Stink' you complain about?

 

So your car made the 'Stink' you complain about?

I helped my son modify and soup up his Eagle Talon all-wheel-drive with turbo car. It produced 300hp or 400hp with terrible pollution. All little kids modify their cars so they "stink" with high pollution. They also "lower" them so the shock absorbers are destroyed (bounce, bounce, bounce) and tint the glass black so nobody behind can see ahead of them (causing a rear end collision).

My car was unmodified and clean. But I hate when the engine stops running (its governor) when I exceed the speed rating of the tires. Especially when I am passing on a two-lane road and a big truck is coming towards me.