I've made an air/fuel gauge that will give me a red, green, or yellow light depending on if the O2 sensors are seeing the engine running lean (red), OK (green) or rich (yellow).

This isn't really any different than anything else on the market except that it uses just 2 LEDs (one for each bank), and each LED can change color independently, depending on the voltage it sees from each O2 sensor. So it will be a very small setup installed just below the sun shade so I can't help but notice the indications (these LEDs are REALLY bright, even if direct sunlight is hitting them).

But how to test it, especially the "lean" indication? Is there a safe way to make the engine run lean for a short period of time? Even a very short period of time would do.

Thanks,

John
1987 XJ-S V12

 

 

hand held programmer

accurate AFR digital wideband readout gage the best way!

i have made an average MPG by steady cruise of 18 MPG US gal. with my PreHE V12, 1978! close to some of the HE consumption numbers!

but i have much more power, in traffic or passing, then back to ECO button, for steady cruise!

pic of control box, and cockpit control in center console ( hard to see, gage is on left of center stack, i should move it to center of dash instruments),someday?
ron

 

unbeknown to many you can run PreHE very lean without problems, at light cruise and hi/vac numbers!
ron

 

Hmmm. Good question.

Apply higher, artificial vacuum to the ECU? Fueling would be reduced but actual incoming air to the engine would remain the same.

That might work....if the O2 sensors don't force an off-setting compensation.

Cheers
DD

 

 

o2 sensors work by measuring the amount of remaining oxygen. remove the o2 sensor from the exhaust pipe and it should read extremely lean. you could probably dangle the o2 sensor in the port of the exhaust and move it outwards slowly until it triggers lean.

 

I think lack of 02 sensor input would simply put the system into open loop; in other words, 'base fueling map'....which isn't lean.

Cheers
DD

 

unless i misread he is trying to make his car run lean to test his a/f ratio led light gizmo he made. in witch case all he needs is a lean signal from the o2 sensor the car does not actually need to be lean and it doesn't matter what the ecm mapping does. the sensor would need to be plugged in. just that the voltage output on the sensor shows lean should trigger his light if his device is working properly. and yes the ecm mapping will try to compensate by supplying more fuel making it run rich until you move the sensor out far enough out of the exhaust that the sensor is reading so lean that it falls out of the range the ecm is looking for then it will fault and go to a baseline map.

 

 

Easiest would be to introduce a vacume leak. For example removing the rear balance pipe...

 

A vacuum leak = more incoming air but the pressure sensor in the ECU, which has no idea where the air is coming from, will just 'think' the throttles have been opened ....and adjust fueling to maintain the correct mixture.

Well, that's how I see it, at least

I think reducing fuel pressure may be the best bet, as mentioned.

But, wait, in thinking about this, if 'rich' and 'lean' are defined here as simply the amount of trim being provided by the O2 sensors then EZrider's method should work.



Cheers
DD

 

Thanks all,

I'm going to try EZrider's method as soon as I can get the O2 sensor unstuck, I've got some work to do in that area anyway so might as well make a party of it.

I'll report back.

Thanks,

John

 

my center control panel in center glove box,, rich or lean while driving, gage readout in AFR or Lambda, pre V12 runs lean and stumbles at say 60/65 mph, just kick it up a notch, and smooths out !
for traffic and passing just atouch of button and full power ,(Rudolph).LOL.. it has a 100% range either side of 0%.

it is a very educational device, and fun to play with , with our TOY cars!

 

Firstly what are trying to achieve?

If it is tuning AFR doing it with a narrow band O2 is a waste of time as a NB O2 only pulses between rich and lean. Rich being about 0.9V and lean 0.1V. At idle or cruise in closed loop the AFR will pulse between rich and lean about 3 times a second.

If you need to tune you will need to use a wide band O2 sensor and an appropriate controller.

 

warrjon,

Not sure if that question is addressed to me, but all I'm trying to do is have a quick indication of if I'm running lean, rich, or normal. So the thing I made is simply two multi-color LEDs, one for each bank. If the voltage is between 0 and .25 volts (running lean), the appropriate LED will show red. If it's between .25 and .75 volts, it will show green, and above .75 volts (rich) it will show yellow.

Of course I can buy one at an auto parts store, but those I've seen are at least 2 inches in diameter, and would seriously affect the inside appearance of the instrument cluster. My device is 1 inch wide and 1/2 inch tall, so I can fit it just below the sunshield and will blend in nicely. Anyway...

Ok, so I tried EZrider's method, no luck. Not that his method isn't valid, but I suspect I have a bad O2 sensor. I won't go through all the troubleshooting, but I ended up pulling the sensor, hooking the heater wires to a bench power supply, and confirmed that the heater worked. I then reinstalled the sensor, hooked up the signal lead to a digital voltmeter, hooked up the heater leads and confirmed that they are getting power, and started the car. I got no voltage reading at all on the O2 sensor. Since I bypassed most of the wiring (I cut into the wires just a few inches from the sensor), I hopefully eliminated a wiring issue, but maybe not...

Out of curiosity, I measured the voltage on the signal wire that leads to the ECU (doing this with the O2 sensor completely disconnected). For whatever reason, I'm seeing about .5 volts on that line. It seems that I shouldn't be seeing any voltage on it, but none the less it's there.

So my questions are...

1. Is there anything wrong with my methodology of testing the sensor by measuring its voltage when it's not connected to the vehicle wiring?

2. Since I spent the better part of two days getting the d**m thing unstuck, is there a compelling reason to replace it (them) with wide-band sensors? I'm quite happy with how it's running and don't want to complicate issues, but if a switch is recommended I'd rather do it now than later.

Thanks!

John

 

These devices are not all that useful in checking lean/rich running because the ECU will cycle the engine from lean to rich 3-4 times per second, then it becomes difficult to see the flashing LED's.



Anytime you disconnect an input from a measuring device like an ECU the input can float because there is no load, I would disregard this test.

Be aware that the voltage from the O2 sensor is not a stationary DC voltage, it is actually a pulsing DC voltage which is in fact AC. If you have a DVM on DC the meter will not have the response to show the high and low volts. Best way to do this is with a scope, next best is a DVM on DC volts with peak hold function (on my Fluke called min/max hold).

So my questions are...

1. Is there anything wrong with my methodology of testing the sensor by measuring its voltage when it's not connected to the vehicle wiring? O2 sensors generate their own voltage as long as they are HOT so this ok.

2. Since I spent the better part of two days getting the d**m thing unstuck, is there a compelling reason to replace it (them) with wide-band sensors? I'm quite happy with how it's running and don't want to complicate issues, but if a switch is recommended I'd rather do it now than later. You can not just replace a NB O2 with a WB O2. WB sensors require a controller to operate, if you replace NB with WB and a controller that has NB output this can then be sent to the Jag ECU.

Unless you are tuning there no point installing a WB O2 sensor at the cost of 200-300USD each with controller. Replacing the O2 sensors with OEM units is probably a good idea as they do have a life expectancy. This would be a lot easier than trying to diagnose a sensor that is not performing to specification. A failed sensor is easy to diagnose one that is working but not to spec is difficult.

 

post #3 ,i said a Wide band system is the best way to know what your AFRs would be, basicly it would only be more useless information, unless you are doing some fine tuning!

i did 10yrs vocational teaching college level, and knowledge is useless until it is applied!

ron

 

Glad to hear that answer warjon since it saves me a lot of money! I don't plan on any tuning so I'll just go with a 3 wire replacement.

My Fluke has the min/max feature, so I'll try that when a get new O2 sensors.

By the way, what performance change should I notice since I have at least one O2 sensor that doesn't work at all? Does the ECU fall back on the other sensor if one fails?

Thanks,
John

 

warrjon can you tell me how to test the 02 sensors? I believe they are the original sensors; the car has 70K miles (MY 1990, V12). I'm still sorting things out, but am disappointed in my MPG. Wouldn't inoperative sensors effect gas mileage poorly?