ahickman

I have an 87 Runner (22RE) that was rebuilt at 200K by the PO. The stock manifold/exhaust was replaced by a custom 4 pipe manifold which is consolidated just prior to the begining of the tranny. The O2 sensor is mounted about 10" past the manifold interface and senses (bridges) only the last two cylinders. It is a single wire O2 sensor, but cannot tell if it is OEM/stock.

In order to test the O2 sensor, I tapped into the O2 signal wire at the ECU harness. With the engine warm and running, I was suprised to see that I did not get a 0.1 to 0.8 Volt swing in voltage. I throttled up the engine a couple of times, but go no response (voltage) from the sensor. I checked the continuity of the wire leading the sensor and found nothing wrong (0.2 ohms).

While this indicates the O2 sensor may be bad, I get no CEL. It was just out of curiosity that I checked the sensor. The fact that the sensor shows no voltage swing bothers me and I suspect the ECU thinks the engine is running lean all the time, hence it increase injector duration (more fuel). Since I run 35's on this bugger, I would definitely relish the idea of ensuring I get the best MPG.

Anyone with good input on the following areas?

1. Should the O2 sensor be remounted lower along the exhaust, past the point at which the four pipes merge so that the sensor is considering all four cylinders and not just #3 and #4?

2. How do I verify the 1-wire type O2 sensor is bad? Is there a typical resistance between the lead wire and the metal case?

3. Why does the ECU not throw a code if the O2 voltage is near zero all the time?

Thanks,

aaron

4Crawler

iTrader: (3)

What sort of volt meter did you use? Unless it is a high quality digital meter, it won't respond fast enough to catch the tiny voltage swings. I use an analog meter and you can see the needle on it twitching in the 0-1 volt range. You could also hook up an air/fuel ratio meter and watch the response as you drive. I run one off my stock sensor (gauge uses a buffer amplifier to allow sharing the sensor) and it works quite well:
- http://www.4crawler.com/4x4/CheapTri...FuelRatioGauge

Or you can pick up a new Denso sensor for a reasonable price:
- http://www.densoproducts.com/

ahickman

4Crawler, thanks for the suggestions. Understand what you are indicating about the DVM. I actually used and o'scope to watch the signal on the computer. Total flat line less the spurious noise (mV).

I have a 4 (maybe 3, but can't recall) wire O2 sensor from a 94 runner I had last year. The sensor is still good. Suppose I could use this one to swap out and test if I give the heater circuit 12V?

Any ideas as to why the ECU does not throw an CEL?

ahickman

Forgot to ask you about the fuel pressure guage you used. I had thought about adding a pressure guage some time back but found that they were pretty expensive (Digikey.com). Around $100 for a housing that was inert to fuel. What did you use?

4Crawler

iTrader: (3)

Yes, a scope should catch the signal, assuming it was connected and adjusted properly. A heated sensor retrofit should work, I've kicked around the idea of doing that on mine, with the old design LC header I have the sensor is in a single cylinder runner. The early ECU does have a code for O2 sensor reading open or short circuit, maybe there is just enough of a signal from yours to keep the ECU happy.

I used an Intellitronix (now Nordskog Performance) gauge and sender from Summit. I think the gauge and sender was around $50 10 years ago.

thook

That would be my guess. The ECU ignores the O2S signal so long as it is open loop and won't go into closed loop without a sufficient signal. I have read about header equipped vehicles requiring a heater circuit otherwise the sensor won't heat up sufficiently and/or cool off too quickly. The heater circuit modification was simple and successful, too. I don't remember how it was done, though. I'll have to reread.

ahickman

Thook, long time no speak! How's the herbal life?

Went ahead and removed the O2 sensor this evening to make sure it was not overly fouled. It is a Denso replacement and seems like it is rather short as compared to the factory sensor I have residual from my 94 3VZE. Read a post that said the non-factory sensors did not penetrate the chamber enough to be immersed in the air flow giving way to lower voltages.

I also ran up the road with the 02 sensor disconnected and only briefly got a CEL light (2 seconds). Have not pulled the codes yet though.

Regarding the bad TPS and open/closed loop mode...the engine runs very well at RPM's and at idle. I would think that a bad TPS would give symptoms that affect either idle and/or running speeds?

thook

Hey Aaron! Yeah, long time. All's good, man. That's almost a trick question, though....

You said it's a Denso replacement? Meaning, it's not a Denso? The only aftermarket O2 sensor I know of that are too short are the Botsched units. I just replaced my wife's (3vze) Botsched unit with a genuine Denso and the element is a good half inch longer than the Botsched.

I ran an expirament by disconnecting the sensor to see how the vehicle would respond. A "good friend" of mine (he knows who he is) suggested that it wouldn't tell me much, but it did. The mileage went up and it drove better....initially. Then, after a couple of weeks, the vehicle began losing lots of power and bucking under load. Well, I've never had the injectors cleaned (after 200k+), so I figured it must be running lean and the Botsched unit was atleast keeping the a/f ratio richer to compensate......hence I was only getting about 14mpg's on average with the unit connected. I put the Denso unit in yesterday, drove it to town that day and today, and the power has increased dramatically over when I had the Botsched unit in and even while it was disconnected. Needless to say, and to get to my point, my friend seems to think there's nothing wrong with Botsched units (ie. shorter elements) as his vehicle has passed CA smog two times already in the past four years. I think, though, the difference is that he has a post cat O2 (Denso, as I recall), as well. I'm still not 100% condeming of Bostched sensors, yet, as I still need to wait for the ECU to adjust to the new unit. I'll see what happens then and test both with a meter at the diagnostic port. Of course, right now I'm also running two cycle oil through the fuel system to clean the injectors. Then, I'm rebuilding the spare set I have and will be installing them. Anyway....lol......I do have bad feelings about the short element. It helps, I think, that my wife's vehicle utilizes the four wire, heated design. On the single wire design, I'd go with NTK or Denso flat out. The Botsched single wire units are too short for the earlier 22re's.

Some more points.....

Single wire, non-heated units typically are only really good for 30k miles. I guess that also depends on other factors, such as how well the vehicle is running (too rich, or lean) and if one uses nefarious fuel system agents that might shorten the life span. You probably have no idea how old the existing sensor is in your vehicle?

The sensor needs to be placed wherever in the exhaust that it can detect the exhaust from all four cylinders....not just two. Being that far away, in the case of headers, it does need to be heated. If the sensor doesn't reach atleast 600*, it won't begin to generate a voltage signal. On your vehicle, the ECU it seems can atleast detect there's not an open circuit, so it wouldn't throw a code. But, the sensor is also basically not functional so it never goes into closed loop and ignoring the O2. I believe this is what's going on with yours.

A bad TPS will produce poor idle and/or running symptoms. If you have an electronically controlled automatic tranny, there's even more problems. You'll find shifting issues popping up.

thook

The spark plugs and exhaust smell say the opposite, though. Smelled very rich with the Botsched unit in. Smelled better with better mileage with it disconnected. Even better smell with the Denso unit, but I hadn't had it in long enough to judge mileage. Also, the mileage, with it disconnected, only ever went up and not down. I was getting nearly 20mpg's. Were it eventually running richer, the mileage would've dropped accordingly, I would think. No?

thook

Nevermind that last question of mine......I know it would've.

Here's my interpretation of what I'm seeing....

(BTW, sorry Aaron. Don't mean to hijack, but maybe this will be somewhat informative and atleast closely related to the topic)

With a properly operating O2S and the system being able to properly operate in closed loop, the ECU can monitor a/f ratio and make the minor necessary adjustments in injector pulse duration (less or more).

In open loop, the ECU can only make those adjustments based on basic operating conditions..... air volume and temp, rpm, coolant temp, throttle position, and battery voltage.

Given that all of the pertinent components are in proper working order on my wife's vehicle, the ECU won't make any a/f adjustments outside of those parameters. IOW, only the normal amount of fuel or injector duration for normal conditions would be given........which, from what I've read, is darn near close to ideal a/f ratio already. The O2 only makes minor adjustments to this trying to keep a/f in balance due to fluctuations in the basic operation with the other components........particularly if one is beginning to wear or malfunction.

So, theoretically, without the O2, and everything else working right, the vehicle should run just fine......just not getting the mileage and power that it possibly could with a good O2. Now, if close enough to ideal amount of fuel is being called for, and the injectors are being pulsed as they should, the only way I figure it could be running poorly while still getting good mileage is if the fuel wasn't adequately reaching the cylinders. Any unused fuel would returned to the tank....hence no drop in mileage ever while disconnected. I figure the Bostched sensor was atleast keeping the a/f ratio rich enough to not run lean, but was too lazy to keep it from being too rich.....the injector pulsing open longer than really necessary. Once the Botsched unit was disconnected, the ECU basically unlearned the previous patterns cumulatively and the lean condition revealed itself.

Again, though, my verdict's really not out, yet. Still need some more testing.

thook

Yep. Approx. 35k on it with a new wire.

Oh! I think I see where you're going. Go on....

thook

Lol.....you mean if it was running too lean while disconnected it would cause the major drop in power and start bucking?

thook

Would you mind rephrasing that and being a little more specific......pleeeaazzzzze.

thook

Wait......I see what you're getting at. I have to admit, that's some info I haven't entirely studied, yet......only briefly. Go on......

thook

I understand that. I'm just not sure exactly what you're trying to say in relation to the O2S issue. Are you simply trying to say when the sensor was disconnected it ran lean, so the KS retarded timing and that's why it started bucking?

thook

Is that same as saying with the Botsched unit in it stayed rich. With it gone and the ECU eventually reverting back to basic fuel trim or unlearning the sensor patterns, it was then running too lean? Hahaha.....well, I think that's was just saying in different words. But, maybe you're just adding other factors to clarify?

thook

I think you are, and I understand what you're saying. Thank you. Very good point. Running lean, combustion temps were high under load, so timing was retarded. Probably doesn't help that I have no EGR on this thing. But, that's never been an issue until this.

thook

Okay, what I think you suspect is that the Botsched sensor was working, but being that it was running lean (from possibly/probably clogged injectors) the sensor was continually telling the ECU to keep it rich, keep it rich, keep the injectors open and it was eating up the gas. Disconnect and it could no longer do that. That's why I need to test. The Denso unit seems to make it run better, though. But, it's only been a couple of days. I'm waiting to see what happens and then I'll know better. Mileage could get worse, again.

But, if this was the case, why wouldn't it throw a code? It was still able to keep a/f in balance though it was taking a lot of fuel to do so?

thook

It didn't seem to be running unevenly. On the other hand, one injector is much, much newer than the other five....hmmm. I shall need to do some meter testing.