HEGO stability problems, not normal switching behavior

[Jdubes]

Hey guys,

Just did my first datalog. Looks like I have a few issues to work out. The TPS, ECT, and ACT don't seem to be working right. Just thought I'd post it and see what you guys think. I basically started the truck and let it run for a few minutes. It's about 20 degrees out and I let it run enough to hit about 140 water temperature. If I let it try to idle it goes too lean and wants to die. 1500-1800 rpm high idle it'll have a half decent a/f near 14-16. But it seems like one HEGO isn't responding properly. I do notice a pop in the exhaust as well almost like its skipping a little. Looks like I got my work cut out for me.

You can see where spark changes, that pretty much follows my throttle input. I moved the throttle occasionally or held it at certain times. But since the TPS voltage didn't change you can't see that. I'll have to look into why these sensors don't seem to be reporting properly.

[Jdubes]

Update:

Well I've gotten some progress. I had to upgrade my binary editor program to BE2010, get the drivers setup, and re-registered. Now that that has been done and I datalogged again...I've found that the TPS, ECT, and ACT are working fine now. But my left HEGO still doesn't work.

I checked the wiring for it and I'm going to check the sensor.

More updates to come.

[Jdubes]

Ok so I replaced my O2 sensors to no avail....not the solution.

I'm getting AFR's all over the map on my wideband. The regular O2's are giving me weird voltages that I can't make sense of.

The HEGOs are all over the place and seem to sometimes be working and other times not be. I'm having a hard time figuring out if there is infact a wiring problem or if the HEGOs are working fine and actually are representing what they are seeing for a/f.

I drove for awhile with HEGO 2 unplugged and it registered voltage between .0125 and .0225 the entire time. Meanwhile, HEGO 1 (still plugged in) registered the same type of voltage until about 45 seconds into the log, possibly indicating time for warmup. For about 3-4 seconds it registered around .7-.8 for voltage then pegged on .0025. It spent the rest of the log (about 15 minutes of driving) showing .0025 with a few spurts of voltages in the more normal .4-.8 range. When HEGO 1 was showing between .4-.8, most of the time the AFR was in the 13-16 range. When it was at .0025 the A/F was off the chart...way lean at 18+.

I went and plugged HEGO2 back in and did another log. You can see down the log that both 02's will drop to .0025 and stay there for long periods of time...and at other times represent normal voltages between .4-.8. So it can't be a wiring problem as far as I can tell.

Anyone ever run into anything like this before? If an unplugged sensor shows .0125-.0225 as a voltage and not constantly showing .0025 like the other...that means to me that .0025 doesn't mean lack of signal it means the input is way off.

I'm thinking this may be due to a mechanical problem, any thoughts?

[cgrey8]

There are a lot of points that don't make sense. The RPM, ACT, ECT, TPS, MPH, and a few others are "stuck". That's an indication that the tune doesn't have these listed in the payload list. Try redownloading your tune to get those parameters back into the datalogging list.

At the very top, there's a time when HEGO1 is reporting in the 1.3v range. Typical HEGOs shouldn't do that. Most will report 1v at the highest and most only report voltages no higher than .85v. My HEGOs, even at WOT, rarely report .8v. So that makes me suspect as to the accuracy of that value as it is datalogged.

At this point, there's more confusion than usefulness in this datalog. So even the things that look like they could be legit, I have little confidence that they are reporting actual conditions. Perhaps someone else can glean something from that log. But I'm afraid I just can't offer much.

[Jdubes]

How about these? One is short, and not warmed up (see ECT), but shows a little of what I'm talking about.

The 2nd log is a little longer, and is at full operating temp. You can see that sometimes it's just stuck on .0025 on both sensors, other times it's showing voltages.

I believe that first datalog I posted, was too inaccurate and should be tossed out. Things changed when I switched to BE2010.

I've noticed that it just doesn't seem to be consistant. The wideband gives off weird readings sometimes and I'm not sure what to think at this point. I still don't get RPM on the normal digital gauge either and I'm not sure why. I have RPM in my datalog on another payload but it doesn't register on the primary screen.

The other sensors seem to be working ok, so that is good, but the HEGO's are acting very weird.

[cgrey8]

Zip the logs before posting them. Zipping will compress the file. If you use Winzip, set it to the highest possible compression.

[cgrey8]

The longer log shows the engine running bad lean. Both HEGOs and WB are confirming that either the engine is leaning out or that you got some serious overlap on your cam letting unburned intake air through to the exhaust which is being picked up by the HEGOs and WB. But the behavior isn't typical of overlap cam syndrome since the log is showing leanout at cruising RPM/Loads, not just idle.

But all the sudden, the log shows normal AFRs. Log FLAGS1 so we can get a glimpse at when the EEC goes OL and CL. I don't see anything that shows why all the sudden the AFRs go normal. Something happened, I just can't see what since there isn't enough being datalogged.

[Jdubes]

The engine has a custom camshaft. There is a question about the valvetrain and whether or not it is working properly. I am about to do a compression test.

There is a possibility it has a valvetrain problem, I'm hearing valvetrain noise.

I will log that, is there anything else you'd like me to log in particular?

[cgrey8]

Here are the things I datalog:

• MAF
• ACT
• ECT
• TPS
• Load
• RPM
• TS
• ISCFLG
• ISFLAG
• FLAGS1
• LAMBSE1
• LAMBSE2
• KAMRF1
• KAMRF2
• PW1
• PW2
• HEGO1
• HEGO2
• AFR
• ISCDC
• Spark
• BAP
• VBAT
• DASPOT
• AEFUEL
• EFTRFF
• IPSIBR

Damn having a QH is nice. I remember when this list was limited to 16 things with the TwEECer.

[decipha]

wow that datalog looks familiar, i've been there and done that, I had to deal with this problem personally and before the 1/4pony was released it had me chasing my tail

you have a faulty ground feeding the o2s, im willing to bet there a bit of resistance in the wiring, t-tap into the ground at the o2s and ground it to the transmission bellhousing, both o2s at the same location to reduce interference

also change these scalars...

fuel closed loop max afr - change it to to 16, you don't need anything higher than that
hego min switches before warm - change it to at least 4, that way the ecu doesn't pull fuel unless the o2 actually works

system mil delay time - change it to .1 or so that way you can see if the o2 is faulting


that should have you at least mobile

[93cobra1928]

I can't speak for the tuning changes because I am just getting started on it myself but I have diagnosed lots of driveability and agree 100% on the HEGO ground loop being the problem. That was the first thing that came to mind when I read the OP

[Jdubes]

Thanks for the reply deciphra,

I will look into that. I can't/won't use a T-tap because it won't last here in New England. I will simply re-do the ground wires and make sure they are clean and well connected.

There still has to be a valvetrain issue because I hear ticking sounds but I wouldn't be surprised if you're correct. Hopefully you are, thanks again.


And BTW....how the hell are you not blowing that M5R2 up?! That tranny is weak...I dropped that in favor of the ZF-S5-42 and am looking for the S5-47 version.

[decipha]

the m5r2 for the supercoupes is a specific transmission, its not the same as the m5r2 for the trucks, our transmissions don't have the inherit problem that the truck transmission have. Although they have the same model name they are very different transmissions and are not interchangeable.

theirs a handful of guys running 10 seconds on these transmission in the supercoupe club of America http://www.sccoa.com with no problems, just have the bad blocker problem

oh yeah i drive mine like a baby its my DD so i don't beat on it

[Jdubes]

That explains a lot,

[Jdubes]

OK we need to dig a little deeper.

I went ahead and ran all new wires for my O2's. I connected both grounds to the battery on a common lead. I ran new signal wires and connected them to the appropriate terminal on the ECU. I ran a new positive wire from both o2s and ran them to the power distribution box and solder/spliced them to the o2 relay circuit.

Runs differently, and seemingly better....but still getting .0025s. I didn't get a good chance to look over my work yet, or change my datalog settings, but I ran it quickly to see how it was and got a short log.

Check this out....

[DRHaulsee]

OK we need to dig a little deeper.

Check this out....

Hey, I'm a novice but that looks really bad to me. Is your wideband actually working? And if so, how is your car even running? You got AFR's either dead rich or totally lean.

I'm not sure you got the engine warm enough for the O2's to work at all. Wasn't you ECT less than 50 degrees?

[Jdubes]

Engine temp was low, but it doesn't matter...you can have it at full operating temp and still get those readings.

The o2's seemed to start showing readings within 20 seconds or so of it running. My wideband takes less than a minute to go through the heater warm up and achieve normal temp.

Wideband seems to be working fine, notice when the O2's read between .4-.8ish the AFR comes into a normal range? It seems to react right. The truck runs when it's in a normal range but when its in the high 20s or low single digits it wants to die.

It'll run but it runs like shit. You can drive it down the road and its very quirky and what not but it does move. It runs somewhat ok when the AFR's are in a normal area, naturally.

[cgrey8]

Without more things datalogged, I can't get a good feel for what the engine is doing. No datalog should go without having RPMs at least and I didn't see RPM or N being datalogged.

The HEGOs show blips of being rich, but that's it. Also the LAMBSEs don't seem to be responding to the HEGOs telling me you aren't in Closed Loop. But since you didn't datalog that either, again, I'm just guessing.

Also keep in mind that the ground wire on the HEGOs is not for sensor ground (what you care about), it's the voltage return line for the HEGO heaters. Any ole ground for them is just fine. The HEGO sensor ground is actually the case of the sensor which uses the exhaust pipe, which uses the engine block, which uses a ground strap to the frame, which should have a ground strap to the battery. That's a long and convoluted ground for HEGOs to have to go through with lots of opportunity to crap out and get ground loops and whatnot. So what I did was had a piece of metal welded to both pipes of my exhaust pipes, then bolted a ground line directly to that piece of metal, which was in fairly close proximity to the HEGOs. I ran that ground line up to the ground lug at the EEC. I also solidified a number of other ground issues I saw around the truck. But your HEGOs are showing they are capable of hitting rich voltages well above .6v. So I don't think grounds are your problem. I'm thinking there's issues in the tune. But without more being datalogged, there's no way for me to say that without more certainty. Revisit the list of recommended things to datalog above. If you have a Quarterhorse, there's no good reason why you can't get all of those.

[teal95]

If the HEGO has 4 wires one of them is ground (1 signal, 2 for heater and 1 for signal ground). I have put 4 wire HEGOs in several of my cars to cure the grounding issue. On the 2.3t cars the HEGO ground for the EEC is on the compressor side of the turbo. I just ran a wire from the extra pin of the HEGO to the same mounting stud as the EEC HEGO signal ground. (The EEC has a separate pin for HEGO signal ground.)

steve

[cgrey8]

Really? While theoretically 4-wire sensors should be usable on an EEC that expected the older 3-wire, I've heard on at least 1 occasion that they couldn't get those newer 4-wire HEGOs to work on an older GUFx EEC that expected 3-wire HEGOs even when the wiring was done as perfect as they knew how to do it. I wasn't heavily involved in investigating that so it's possible those efforts just gave up too early before figuring it out. I don't know of any technical reason why 4-wire HEGOs wouldn't work, but if there is one, I'd guess it has to do with difference in AFR detection range...it is possible 4-wire HEGOs have a different AFR transfer and thus would require a different switch-point voltage or something. That's purely a guess not based on anything I've seen or heard. The guy I'm thinking of that tried to "upgrade" to the 4-wire HEGOs wound up replacing them for 3-wire versions and finding that the 3-wire sensors worked where he could never get the 4-wire versions to do right. Again, I wish I had more details...just passing on what I remember from a few years ago. But if you have done it with success, then I chock that old report up to some unknown issue that he "accidently" solved at the same time he replaced the 4-wire sensors with 3-wire versions.

At this point, I say this guy needs to get a better/more complete datalog so we can see what's happening. At a minimum, get RPMs (aka N) and FLAGS1 (which includes OL/CL status). Being that the engine is obviously running lean with what should be rich LAMBSE values, I'm going to say to either boost your fuel pressure or drop your Injector Slope values a bit. This may not be the right thing to do long term, but it should get the fuel mix at least closer to what the LAMBSEs what them to be, which will allow you to get better driveability and thus better datalogs that you can then work with and hone.

[Jdubes]

Ok I think we're making more progress.

I datalogged very parameter again and should have all the info you need to see. I don't know why, but ISCDC wouldn't show up and isn't in the datalog list. It was when I did my first logs on BE2009. The other thing is I am not getting the RPM displayed on my digital dashboard but it's in the log, so the info is there.

Ok, so I tried zipping the file and hopefully you can see it now. I took it for a drive and after looking through the log, one thing I picked up was that any time the o2's get to .0025....I'm off the throttle. Notice the .94-.96 TPS voltages every time the o2's are at .0025.

It runs really lean when I'm off throttle. Part throttle it seems to get close...WOT it's a little off too. I'm starting to sense that the HEGOs are working properly.

I'm not very familiar with all the other stuff, still at the basics stage and learning everyday.

Thanks for all your help so far!

[Jdubes]

So I checked for vacuum leaks and found one. It runs a little better, idles a little better...and the AFR's are a little more stable but it's still basically the same senario.

It still goes way lean at idle...and under WOT. Cruise stays between 13-16ish mainly between 14.5-15.3.