The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

Please Support Morrison Fabrications
Please Support STM Tuned

2G Front o2 sensor reads 4.41v

This site may earn a commission from merchant
affiliate links, including eBay, Amazon, and others.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

spooln4fun

15+ Year Contributor
638
10
Jul 9, 2005
BELPRE, Ohio
I know a proper sensor should cycle between .02 and I believe 1v but mine stays steady at 4.41. Would any thing cause it to read this high or is it simply no good and I'm thinking to much into this?
 
When did it start doing this?
A narrowband O2 sensor won't output more than 1v normally. Your ECU is configured for an AEM wideband on the rear O2, are you sure of it's connection.

You also have a problem with your idle speed (1200RPM) and your idle switch being inoperative. Either correctly adjust the TPS or simulate it in DSMLink
 
Could you post a longer log that shows varying operating conditions, like different amounts of throttle and so-forth?
Do you have any place in a log where it thinks it is in Closed Loop? (Closed Loop = 1).
That would be interesting to see because the Front O2 should be cycling up and down there, and the range of numbers would be interesting to see. Shouldn't be just a constant value like it shows in this log. And it should always be less than 1 volt.

I notice that in the Rock Auto listing for Oxygen Sensors for the '99 GST, they show the upstream and downstream sensors all together on the same list. Be sure to have an "upstream" sensor for your front O2. I see Bosch, Denso, and further down the list I see NTK, upstream sensors. Those brands should be good. MITSUBISHI > 1999 > ECLIPSE > 2.0L L4 Turbocharged > Exhaust & Emission > Oxygen (O2) Sensor.

The rear O2, if it has a wideband on it, we should see varying numbers there too. In this log I see 0.51 volts constant. AEMWB shows 9.1 to 1. I don't know if that jives with what AEM would say because I don't have any AEM wideband manuals that old (the 30-2301) and don't see one on AEM's website anymore! But it's surely wrong.

The AEM 30-2301, from what I read, has 2 outputs - a 0-5 volt and a 0-1 volt. "0-5v calibrated output with 0-1v secondary output." Be sure it is the 0-5 volt you have hooked up to rear O2.
I wouldn't be surprised if the analog output of your AEM is bad though.
 
The rear O2, if it has a wideband on it, we should see varying numbers there too. In this log I see 0.51 volts constant. AEMWB shows 9.1 to 1. I don't know if that jives with what AEM would say because I don't have any AEM wideband manuals that old (the 30-2301) and don't see one on AEM's website anymore! But it's surely wrong.
He also has the rear O2 value locked so the 0.51v is faked by DSMLink.
 
He also has the rear O2 value locked so the 0.51v is faked by DSMLink.

Yeah, that's a case where I get really confused by the Help you get in the application when you pick the ? button.
It gives the impression that you should lock an input when you have an aftermarket sensor on that input. A sensor that is doing something different than what the OEM one would do on that input.
Hovering over the lock button doesn't help me either because that says "Locks rear O2 voltage seen by factory ECU code at 0.5 volts." Geez.
So I don't know how it should look in DSMlink and can't verify it one way or the other because I don't have a 2g. 😄
 
Oh I see RawAEMWB reads 0.02 volts in the log. That's probably the actual voltage that's on the rear O2 input, from the wideband.
I found a manual for the AEM 30-2301 and according to the calibration table in there, 0.02 volts would be an AFR of about 8.42 on the 0-5v scale. That would be way way off I think for the conditions in the log.

0.51 volts on that same table would be an AFR of about 8.86, so that would be no good either even if it was the actual voltage from the wideband.
 
Last edited:
The rear o2 is the least of my worries. All the info is greatly appreciated. I bought it used, completely installed it and found out the sensor was shot.

I'll grab a longer log of the car for you.

It could have been doing this since day 1. I believe the sensor is upstream as the length was right and plugged in correctly. At first I was hoping I accidently plugged it into my cruise but my cruise functions as normal.

I'll adjust the tps and get an updated log with in a few days. I won't be able to drive it as it's blocked in the garage but I'm able to give different throttle inputs. I'll let it run long enough to get into closed loop as well and upload that as well.
 
Here are more up to date logs, apologize for the delay again. Just hard to be very prompt due to my work schedule.
On "long drive" it idles for about 60 seconds before i drive around. its in town for a few minutes than i get on the highway.
If the "cruise weird" log is an issue for another thread and unrelated to this front o2 than i'll make a new thread once this issue is resolved. I just wanted to post it as well just to have as much information as possible.
 

Attachments

  • small drive hesitation.elg
    75.9 KB · Views: 18
  • long drive.elg
    225.2 KB · Views: 16
  • driving around.elg
    78.5 KB · Views: 13
  • cruise weird.elg
    16.1 KB · Views: 16
  • cold start.elg
    52.1 KB · Views: 16
On "Long Drive" the O2 voltage varies hardly at all, 4.37 to 4.43 is what I see. It's going in and out of closed loop at about the right times. It's going into open loop when load factor exceeds about 1.0 and is usually in closed loop at lower load factors. Looks like it's obeying the "OpenLoopThresholds" tables in ECU direct access, so that's good.

I can't tell but maybe your worst running is when it's in closed loop?
In ECU config, Misc, you can lock it into open loop. Have you tried that? Might be interesting.

I think what I would do is buy another new O2 sensor, a different brand than the one you have. Probably I'd get the Denso 2344026 or the NTK 23558, or get one straight from a Mitsu dealer (which would cost about $100 more though).

If you still get voltage readings from it in DSMlink that are over 1 volt, then I would check the accuracy of that input on your ecu. Because that's another thing that could be wrong. It might just be reading voltages wrong on that input.

You have to be careful to do that in a way that is "safe" for the ecu. The way I've done it before is with a small battery of the 1.5 volt type, like a AA or AAA cell. Preferably one that has been used a lot so it has less pizzaz than a new one. Probably it will put out just slightly over 1 volt. Anyway use a multimeter to measure the voltage of that battery with no load on it, so you know what it is, and then use it as the O2 signal voltage source instead of the O2 sensor and log that. It's a lot of words to describe well how to do that so let's save it for if it's needed. But you would do that with the engine off, and ignition key "On" so you can log.

I always thought a nernst cell could never put out more than about 1 volt. But I read something to the effect that if there is current leaking into it from somewhere, like maybe the heater circuit, then it can put out higher voltage.
 
Last edited:
On "Long Drive" the O2 voltage varies hardly at all, 4.37 to 4.43 is what I see. It's going in and out of closed loop at about the right times. It's going into open loop when load factor exceeds about 1.0 and is usually in closed loop at lower load factors. Looks like it's obeying the "OpenLoopThresholds" tables in ECU direct access, so that's good.

I can't tell but maybe your worst running is when it's in closed loop?
In ECU config, Misc, you can lock it into open loop. Have you tried that? Might be interesting.

I think what I would do is buy another new O2 sensor, a different brand than the one you have. Probably I'd get the Denso 2344026 or the NTK 23558, or get one straight from a Mitsu dealer (which would cost about $100 more though).

If you still get voltage readings from it in DSMlink that are over 1 volt, then I would check the accuracy of that input on your ecu. Because that's another thing that could be wrong. It might just be reading voltages wrong on that input.

You have to be careful to do that in a way that is "safe" for the ecu. The way I've done it before is with a small battery of the 1.5 volt type, like a AA or AAA cell. Preferably one that has been used a lot so it has less pizzaz than a new one. Probably it will put out just slightly over 1 volt. Anyway use a multimeter to measure the voltage of that battery with no load on it, so you know what it is, and then use it as the O2 signal voltage source instead of the O2 sensor and log that. It's a lot of words to describe well how to do that so let's save it for if it's needed. But you would do that with the engine off, and ignition key "On" so you can log.

I always thought a nernst cell could never put out more than about 1 volt. But I read something to the effect that if there is current leaking into it from somewhere, like maybe the heater circuit, then it can put out higher voltage.
Thank you, I'll pick up a good brand front o2 and also inspect the wires while I'm connecting to be sure they're ok. If it still reads over 1v, I'll grab a use battery and give that test a try.
 
Are you 100% sure you didn't connect the WB to the front O2 sensor pin?

There aren't too many ways for the front O2 to be reading > 1v. One is that you're actually feeding the ECU a higher voltage (which a narrowband sensor won't do) and the second is that the front end conditioning circuit for that input is damaged. I have seen that happen one or twice on 1G ECUs but not yet on a 2G.

Given that neither O2 sensor are behaving the way they should I suspect a wiring issue more than ECU problems.
 
Last edited:
I can see how it would be easy to hook it to the wrong pin, especially when the wires are the same color.
 
I can see how it would be easy to hook it to the wrong pin, especially when the wires are the same color.
Yeah and right next to each other.
It's so hard anyway to work on those wires, at least on the 1g it is because they are so short.
That's why a few years ago I put about 20 inches of extra wire in there on every wire going to my ECU. So now when I take the cover off of it, I can spread that whole mess out on the floor of the passenger footwell where it's easy to see everything and work on it.
That white wire way over to the right that is draped over the ecu and has a quick disconnect in it - that is my wideband wire. From when I was fooling around with it when my wideband was acting stupid. And I used that break in the wire as a place to connect a multimeter and to connect a variety of old small batteries to log their voltages as a check on that ecu input.

Also as you can see, I've taken a liking to green painter's tape with blue Sharpie for flagging things 🤣

You must be logged in to view this image or video.
 
Last edited:
This 4.41 volts thing sounds a lot like what I got a few months ago when I cut my wideband wire a few inches from the ecu. So that ecu input was connected to nothing. Air. I logged it that way with engine running, and the logged voltage on that input was 4.45 volts. (that was on my EGR input, and the logged voltage value shows as "RawLinWB" in the screen shot below).

At the same time, I had my Fluke multimeter reading volts on the wire from the wideband, and it read 0 volts. That wire was just out in the air connected to nothing but the multimeter.

I wanted to quit using the EGR input anyway. So my next step was connecting the wideband wire (even though it appeared to be dead) to the IAT input, with quick disconnects in the line for easy checking. So I put that together, and with the wideband wire connected to the IAT input, I alligator clipped the Fluke meter to the line. Logging with engine running, Fluke measured 4.72 volts and the log showed 4.69 volts (RawLinWB in the log). I could show a screen shot of that too. I actually have videos of all this stuff too, showing the Fluke and the wideband gauge and the laptop.

So with the OP's case here, it seems to me like he has either a dead thing connected to the ecu, or nothing connected to the ecu, on the O2 input.


You must be logged in to view this image or video.
 
Last edited:
If it came to testing ecu input #76 for voltage measuring accuracy, there'd be an easier way than cutting into the wire near the ecu.
Instead of that, you could just work with the O2 sensor plug in the engine bay.
4 terminals in that plug. Find out which one goes to pin 76. Hook up the + end of your small test battery to that terminal. Hook up the neg end of the small battery to any handy ground. And there you have it.
Although that way of doing it would give you nothing if there is a break in that wire somewhere, so you'd have to check that out then.
The other 3 wires in that connector, probably 2 of them are grounds, and 1 would be 12v power.
 
Well it appears I have my wideband wired into the front o2 spot. I clearly got confused/turned around when doing this. I'll get it switched to the correct wire and report back. I appreciate everyone's input and patience in my delay to respond.

You must be logged in to view this image or video.


You must be logged in to view this image or video.
 
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Innovation Products Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com RixRacing RockAuto RTM Racing STM Tuned

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Back
Top