What's new
CK5
Register an account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members.

TPI Failed Emissions

Hossbaby50

Hossbaby50

3/4 ton status
Joined
Sep 1, 2001
Posts
8,969
Reaction score
1
Location
Peoria, AZ
I need help figuring out why my truck failed emissions. They said the truck is dumping to much fuel. I replaced the O2 (1 wire) with a 3 wire O2 sensor yesterday and it seemed to make my truck run worse. The idle is all shifty and it is getting alot of pops from the exhaust when I am slowing down.

I am going to reinstall the old sensor and see if it helps. What else causes a TPI to dump to much fuel? It is a MAF TPI setup on a stockish 350sbc. I am not getting and SES codes.

What do I need to check and look for? What can cause excess fuel to dump into the system?

Harley
 
THats a really broad and stupid statement on their parts. Did they give you a readout of what you scored? To simply say "its dumping too much fuel" is a failure on their parts to communicate. Is it too much CO? Too many HCs(most likely)?

You are correct, running in open loop will cause it to dump too much fuel. A 3 wire will help get out of open loop sooner. The fact that its not running right after the install of one says to me you have a bad O2 signal wire connection, I could be wrong though. Popping on Decel also sounds like Open Loop operation. Some PCMs are setup to detect decel and go into Open Loop, namely to back off the fuel, but maybe yours isn't doing this properly, I don't know. Any idea what fuel pressure you are pushing? Stock TPI injectors? Catalytic converter(s) installed? Remember, a stock Camaro will have the huge GM cat and likely AIR injection, both of which will lower HC output. You may just need to get a new chip tuned for leaner operation, if you want to not run cats.
 
You running a y-pipe, O2 after the Y so its seeing the whole exhuast stream?

Any exhaust leaks?

I'm dual from the headers back, drivers side bank is completely "blind", exhaust leak on that side (haven't welded pipes up yet) and I've got a couple of injectors that aren't in good shape, coupled with an O2 sensor that (at least scanning at idle) isn't operating correctly. I get some "burbling"/popping out of the exhaust I shouldn't and MPG isn't where it should be either, so I know I'm dumping more gas than I should be.

VSS in place/working? (sorry, can't keep track of everyones setup)
 
Also, scanning for your setup (I hate this 4 letter/number search feature BTW< it excludes TBI, TPI, etc) I found a post that said you passed for two years in 2004. Why are you testing again??
 
I just reinstalled the 1 wire O2 sensor and it is running better with less pops on decel and smoother less jumpy idle.

I am running dual exhaust. I have a cat on each pipe. My sensor is in the driver side pipe. The passenger side is blind. The passenger side also has an exhaust leak. The E-test guy only put the probe in the drivers side pipe.

I am getting E-tested again because I moved to Colorado and I am trying to get my truck registered here in Colorado.

Here is the readouts:

............IDLE...............2500rpm...........Limits..........Results
HC PPM...401................103..................400............Fail
CO %.....5.92................4.39................1.50............Fail

That is the only things I said I failed.

With the Idle being better I imagine that the HC PPM's will pass on the next test but why are my CO% so high and how can I bring them down?

Harley
 
My truck is not running a VSS sensor. It is running in Park/Neutral all the time. It has been that way since day one. I passed the AZ Etest which is harder with flying colors. It passed so clean the last time that they gave me a 2 year tag instead of the 1 year like they always did with my carburator.

My truck was up to operating temp. I ran it around town to warm the motor and the cats up and left it running while I waited in line to get tested.

The fuel pressure should be fine. I just installed a new Walbro fuel pump and new fuel filters yesterday also.

The injectors are stock 350 injectors that were flow matched and rebuilt about 3 years ago.

Harley
 
Chart from my textbook:

chart.jpg


CO is the result of incomplete combustion. Richer the mixture the more CO, until of course the mixture is so rich that it won't combust and misfires, then CO will drop off. Red line is stoichiometric, left is rich, right is lean. I'd say you are running rich. Sounds like maybe you simply were not getting out of Open Loop operation and were running rich all the time. Bad O2 sensor or sensor signal(wiring) could be the problem there.
 
Only thing I can see influencing an emissions test (in this case) regarding VSS is a dyno test where the throttle follower function (tied into VSS) would be evident. I'm assuming what you are doing now doesn't have a roller test.

Even then, here anyways, you don't get a graph of what happens AFAIK, they just give you an end result which averages it or something. I don't emissions test luckily. Not yet anyways.

Is it running poorly other than how it tests? Its possible that now your plugs are fouled badly. Weren't you having some odd ignition problems as well?
 
I sure am glad we dont have those POS emission test here where I live. Probably will be manditory before to long though. :(
 
No Rollers here. There were in Phoenix but not in the CO test that I am doing. My old ignition problems ended up being a faulty pickup coil. It would make the module act like it was bad. I haven't had that problem for awhile.

The only problem I have had was when the truck had been run for awhile and was let cool down to 150* or so it would run funny until it got back to operating temperature. It would just surge some and once it hits operating temp it was fine.

I changed my oil last weekend and when I did I pulled out 2 plugs from the drivers side and they were perfect. Nice and tan. They looked so good I didn't even bother to pull the rest out. The gap was perfect and the plugs were looking great. I also cleaned my K&N last weekend.

Harley
 
Hey, wassup duuude, glad to see you made it there good.

Just curious, you said you had it warmed up when you ran it through emissions,,,what exactly is the temp at the time?

I used to have a car with the 350 TPI, and installed a lower t-stat, and cooling fan switch, to help keep the engine cooler, but found out when i drove it to Flagstaff area in the cooler weather, it ran in "cold mode",
Thats the other question, do you have the cold start injector setup on that thing?
 
Not really pertinent it sounds like, but any chance that 3 wire was a bosch unit? I'm always hesitant to jump on an entire product line for something, but the Bosch O2's really seem to have problems from the get go.

Any idea if the engine temperature the ECM seeing is right? If idle quality doesn't change hot/cold, I'd be leaning towards that as not an issue.

How about cold start injector? Those too can leak, are you running one? I guess even the temp switch on those could go bad.

When I broke a rocker and long story short, the engine went rich, the O2 sensor saw the problem, and tripped the appropriate rich code. If you aren't seeing a code and you know it's running rich, (as opposed to ignition or hardware failures) then somethings preventing the ECM from looking at the O2 sensor.

I don't have my manual here, I'm sure there is a section on problems like this.
 
The stat is a stock 195* unit. The temp gauge says that it is right at or just about that temp when it was tested.

The O2 sensor was a Bosch 13077 that I found the part number on here for. The sensor I put back in was a stock 1 wire Bosch for a TPI Camaro.

The thing that I wonder about with the Bosch 13077 that I put in was it only had about 1/2 as many openings for gas to get to the sensor compared to the stock 1 wire unit which I wondered about possibly being part of the problem. But I don't know for sure. I may try to get Autozone to warranty the 3 wire sensor but I had to cut the pigtail off to make it work in my truck so I may be SOL.

Harley
 
the O2 sensor is basically a pyrometer for the computer anyways, it doesn't "sniff" gas, just determines the heat and adjusts the air fuel ratio accordingly,

You're runnin a stock t-stat, i would check or even just replace( because they are only like $7.00 or so) the water temp sensor,,(for the computer) to make sure it is working properly to ensure the computer is reading that the engine is in fact warmed up and get it out of the choke/add more fuel mode.

If you're dopping too much fuel, SOMETHING is telling the computer to do so.

Timing is out of the question, everything is correct with the distributor after you replaced the pick-up coil?
 
4xcrazy said:
the O2 sensor is basically a pyrometer for the computer anyways, it doesn't "sniff" gas, just determines the heat and adjusts the air fuel ratio accordingly,
Click to expand...

Incorrect by far. An O2 sensor is a Piezoelectric device. It creates its own voltage through a reaction. IN the case of an O2 sensor it reacts with the oxygen in the air. More oxygen less voltage. An O2 sensor needs to be at 600*F to operate. His problem when it "cools to 150 and runs crappy" is that the O2 sensor also cooled down, out of its operating range, and the computer had to switch into Open Loop mode. An O2 sensors, does just that, it senses oxygen, thus the name. It can't pick up Hydrocarbons, so it has no idea you are running that rich, it just senses less oxygen as a rich condition. Problem is, if you have a cylinder misfire, all it sesnses its tons of oxygen, it can't pick up all that raw fuel. Thus it richens the mixture cause its thinks you are lean(lots of oxygen). Not saying thats whats happening here, just telling you how an O2 sensor works.

Open loop is when the PCM ignores the O2 sensor and just runs its own presets for fuel. This is usually a pretty rich mode, the PCM doesn't want to risk burning down your motor. A heated O2 sensor just helps you get out of Open Loop mode quicker. It really sounds like you had a bad O2 sensor or bad wiring with the three wire. There is no reason it should run anything but better with one.
 
Even in open loop the motor should run good. The only time thats not the case is if something is wrong. Whether that be a hardware problem (leaky injector for instance) or a tuning issue RELATED to hardware. (cam that isn't compatible with the stock open loop tune)

You mention Autozone. They scan for free. If they use one of the data scanners, instead of the code readers, then your engine temp sensor for the ECM and your O2 sensor operation can be checked. It's not worth replacing things just because they COULD be the problem, you'll end up out $500 in parts with the same problem.

I know wiring up O2 sensors can be tricky. I'm still not completely clear about what the problems are, but the wire to the ECM IIRC has steel or some other non-copper metal in it, which means solder doesn't stick well, for one thing. Any sort of resistance on that wire is likely to cause problems, and connectors/bad splices could easily do that. We're talking about 0-1V, which isn't much at all.

If you can watch the O2 sensors operation, you will know its staying in closed loop at idle (which is important for the idle portion of the test obviously) and you will be able to see if the sensor is bouncing between rich/lean as it's supposed to, or if it's "fixed" at one end of the spectrum or not. I watched my scanner the other day as the engine warmed up, and as the engine temp reached what SHOULD have been the temp closed loop started working right, I saw that it was still in open loop, and that the O2 voltage was not switching voltage, instead staying around .45mV indicating it wasn't getting hot enough.

AFS-74 is the recommended Delco heated O2 sensor for TPI IIRC.
 
not to steal the thread or anything, but, i just got back from emissions testing and i passed! not only that but by looking at the readout i passed rather well. i feel as if i cheated the system or something..... :laugh:
 
Yeah, but you gotta watch what it's doing to know if it's bad, or if it's just cold, or if it's seeing a real problem. (which in this case it doesn't sound like, since there is no rich/lean code)

Unless of course you just like throwing $20+ at a problem hoping it will go away. I think my scanner cost $100.
 
You must log in or register to reply here.

Latest Posts

Top Bottom