1994 454 TBI runs rough, misfires, backfires, dark exhaust

[kennyrodgers]

Hey Guys,
I was under the MH today doing the air bags (job from hell) and I got a slight whiff of gasoline and I noticed that one of the lines to the evap canister has a great big bloody hole in it. I'm pretty sure from a quick visual that it's the line from the tank and not the one from the throttle body. Even if this is the case would it cause a vacuum leak being on the tank side.
Cheers.

[SLEETH]

need pic of hose/unit in question before answering question

[kennyrodgers]

No worries buddy, I'm there again tmrw so I'll post one up when i get home.
Cheers for now.

[SLEETH]

usually 3 lines going to a charcoal canister=some times 4
might even  have a purge solenoid on top of it=these do go bad and cause ruff running & running rich condition

[kennyrodgers]

Just 2 on mine.
One goes from the base of the throttle body to a solenoid then to the canister, the one with the split goes back towards the rear.
It was full on air bag mode today so I didn't get chance to trace the line. I'm just putting it out there to see if it would cause any issues other than the smell of petrol.
Thanks again Bud.

[M & J]

No need for a pic mate. If the line is from the tank normally it only collects the tank vapor and stores it in the cannister until the ECM opens a valve and purges the canister (draws it in to the throttle body to be burned).
Patch it, but if you can start the engine and feel for a vacuum first, plug it and see if the engine sorts itself.

[SLEETH]

check the operation of that solenoid = they do burn up =and stick open

[kennyrodgers]

Updates and more mystery challenges.
Hi folks, it's been a busy few weeks so still not had chance to take her for a decent run yet.....However.
I got the original injectors serviced a couple of weeks ago and managed to fit them a couple of days ago. Right of the bat the dark smoke has gone, she's didn't throw a rich code when going over to closed loop and the idle was nice and low.  I'm going to get the replacement injectors flowed and compare the results between the two so I can check to see if they are indeed the correct ones.


Move forward to yesterday.
I had to move her round the yard to do some work on the interior and the dash park and brake lights did not light as they usually do and the idle was way up. When she went in to closed loop she threw a code 36 idle fault. I'm not sure if this is connected but there is a wiring fault under the dash that we've not managed to get to yet in that the electric mirrors and mirror heater switch sometimes work and then sometimes they both don't. I had my hand under there at the back of the radio and wonder if I loosened the bad connection. Also, the levelling jacks havn't worked since I bought it. They're showing not in park. I was leaving this one till I've sorted through everything else, just thought I'd mention it just in case.

Do you guys know if there's a common earth for these things, brake light, park brake light, Mirrors, specifically the IAC etc that would cause these problems and if so where would it be.
Cheers guys.

[Rickf1985]

Well, Grounds are the root of all evil when it comes to computer controls. all it takes is .10 volt resistance and you will have problems. The controls in the engine will mostly be grounded at the thermostat housing and at the back of the head. Look for a group of black wires all coming to a common terminal and then take them off and sand the area clean whether it looks like it needs it or not. Find the ECU and the ground wire for it and be sure it is grounded directly to the chassis battery. Also be sure there is a good ground from the engine to the chassis battery besides the heavy ground wire. In the case of the motor home I would run a ten gauge wire from the same mounting as the group at the thermostat housing down to the battery ground. This way you are bonded all the way through for the computer. The mirror problem "should" not affect the engine computer but if the ECU ground is compromised then all bets are off. Be sure that loose wire is not going to short out and cause a fire.

[Froggy1936]

Hi Rick Remember me mentioning That after installing the TBI & 4L80E I did not see any increase in MPG The original set up was single exaust with the O2sensor in the single pipe after both sides came together I installed the O2 sensor in only the R/S exaust Do you think that control unit is reading lean mixture for 8 clys (due to only reading 4cly) And increasing richness ?? Other than low MPG all is good (does not have the power of the Quadrajet) Frank

[kennyrodgers]

Hi Frank.
Thanks for getting back.
Fired her up again today and the idle code did not show although i think the idle is still too high. Mirrors and heater are working again  D:oH! for now.
By way of an experiment i deliberately created a vacuum leak and the idle went up but the IAC pintle didn't look like it moved at all. I then disconnected it and there was no change in the idle speed, re connected and still no change. Iit's a new IAC as well that the previous owner installed less than a year ago.
Time for some IAC research.
Thanks for the tip about running the earth direct to the battery, that can be Sundays mini project.
Froggy, not sure if your reply is meant for me but the o2 after where the 2 downpipes join.
Thanks again guys.

PS, the park brake light is still not lit in park.

[bluebird]

Do you have the emergency brake set? Most have an emergency brake switch. What system do you have,HWH ?

[kennyrodgers]

Hi Bluebird.
If you mean for the jacks then yeah, I've tried it all ways. park on the selector and the footbrake set. They are HWH Jacks and according to their literature the Blue wire is not getting a signal if the red "not in park" warning light is lit.
I've traced the blue wire as far as possible, it's spliced into the park brake switch at one end but it disappears behind a box that holds the fusebox at the other. I've tried unscrewing the box but there's that much crap in front of it I can't get to the back of it.
Fun fun fun.

[bluebird]

That blue wire is a switched ground according the the wiring diagram for the pump solenoid and control. If you have a 200 series system. There should be a red wire that is fused for the hot. It should have power in the acc or run on the key. Depends on how they wired it.  Should be a white for ground.

[Rickf1985]

Somewhere a long ways back in the posts I sent you a Idle relearn procedure. Might be good to go through that. set base idle also. If you can't find it just Google Idle learn procedure for your year. Be sure to check all of the grounds though or you could be chasing your tail.

[kennyrodgers]

Hi Rick.
Yes, I saved that info and it's on the list for Sunday / Monday. I'm hoping my buddy can come over then we can take her for a run. He can monitor the scanning software on the laptop whilst I try and keep her pointed in the right direction.

[Rickf1985]

OK, But the software will not tell you if you have a bad ground since it does not know the reason for the readings. The readings you will be seeing will look normal but be off but .10 volt reference to ground. I am just pulling the .10 out of the air for an example. So if things look good but are not still check the grounds. You could do a running voltage drop test but that is best done with a DSO scope so you can review it later. It is very hard to pick up stray drops or spikes due to bad grounds by physically watching live data.

[kennyrodgers]

Hi Rick.
I'm definitely going to run that extra ground from the Thermostat to the Battery and clean up the T'stat terminals whilst I'm there, it was a great tip and it's the first thing I'll be doing on Sunday.
We cleaned up a bunch of other grounds a while back but not the T'stat one and I'll look into the DSO scope test you mentioned.
Many thanks once again, I'd be stuck without ya.
All the best,
Pete.

[kennyrodgers]

Hi again Rick.
Just had a look into a DSO and wonder if you think this might be up to the job. it's affordable and readily available.
it has an oscilloscope and can will warn you if the circuit is + or - .8 of a volt.  It looks quite handy for other stuff as well.
Please let me know what you think.
Pete.
http://www.sealey.co.uk/pdfs/instructions/PP100.pdf

[Rickf1985]

Pete, That is not a DSO, it is a regular oscilloscope but it is a neat little tool for less than 150.00. I might look into that just due to the ease of setup. If you do not have a scope that would be ideal for a lot of testing.
What I have is a Snap-On Vantage MT2400. It is obsolete now but works just fine for this kind of stuff. A DSO will store a scope pattern over several minutes to several hours of time so you can scroll over it later to see what was going on. What you are looking at there is real time and even though it has an alarm at .8 volt you have to remember that .9 volt is the entire range of an O2 sensor from full lean to full rich! An O2 sensor works in millivolts so a bad ground of .1 volt is huge to the reading on the O2 sensor and can make a BIG difference in how rich or lean it runs. .450 is dead center and stiociometric ratio for idle on the O2 sensor. Now if you are off by .1 volt it will actually be seeing .550 or .350 at the computer and you will be rich or lean accordingly.

Again, I am using the .1 as an example, it never works out that easy.

[Rickf1985]

MT-2400, Can be used for any kind of waveform testing or DSO testing.

[DaveVA78Chieftain]

Problem with hand held scopes like the PP100 is that the screen is to small to really see the trace easily.

I have a Velleman HPS10 handheld.  Usable for some things (e.g. crank or cam sensor, O2 sensor, etc.).

I also have the Hantek 8 Channel Automotive O-scope 1008B (EBAY) I use with my laptop (larger display).  https://www.youtube.com/watch?v=f7dgZJ3mM_o

I can use the Hantek HT-25 ignition probe with either device.  I can view the spark plug traces on my 92 Lumina 3.8L V6 van OK but I never got the 94 Honda civic 1.5L engine to display properly on either device.

[Rickf1985]

Problem with hand held scopes like the PP100 is that the screen is to small to really see the trace easily.

I have a Velleman HPS10 handheld.  Usable for some things (e.g. crank or cam sensor, O2 sensor, etc.).

I also have the Hantek 8 Channel Automotive O-scope 1008B (EBAY) I use with my laptop (larger display).  https://www.youtube.com/watch?v=f7dgZJ3mM_o

I can use the Hantek HT-25 ignition probe with either device.  I can view the spark plug traces on my 92 Lumina 3.8L V6 van OK but I never got the 94 Honda civic 1.5L engine to display properly on either device.

I used to be able to link the Vantage to the laptop using Snap-link software but they have stopped supporting it and it calls for a password to override the one time use disc.

[kennyrodgers]

Hi all,
It's been a while but time for an update.
The good lady wife and I both had a few days of work last week so as we had a few days of sun forecast we decided to go for our first shakedown run and spend a couple of days away. We chose a local site about 30 miles away just in case things went south.
There were a mixture of road types, a fairly short Motorway / Freeway section and some good 40-45 mph single carriageway roads and a couple of semi steep hills in the mix.  She ran great on both journeys, plenty of power on tap, all the gears worked including O/D, temps were very stable and no fluid loss.
I was worried about one section of the return journey where you join the M/Way on an uphill grade. I drive heavy trucks for a living and when loaded it's a hard pull getting up to a good speed to join the traffic flow. She performed great though, we hit the traffic at 50mph and there was a little more on tap if I needed it so I was really pleased about that.


We still have a high idle issue and I've not had chance to do a re-learn procedure yet and we now have a hard code 36.
By chance I happened upon a thread in a Chevy forum re the code 36 and in some heavier vehicles they fitted an Idle control actuater that works in conjunction with the IAC. If you remember a while back I posted about the mystery object on the side of my TBI and that's what it is. It's basically a vacuum operated throttle kicker for when switching on the AC or when shifting from Neutral/Park. I wound mine in thinking it was the cause of the high idle but no.
I've found info on how to set it so I'll be doing an idle relearn then setting the ICAM correctly and we can take it from there.
One other slight oddity is that there's a relay somewhere in the drivers foot well that's clicking on and off every four seconds or so. It'll do it for a minute or two then stop.  No other codes thrown so that's a great result from what was a great first trip out.
Also, just a quick note to to say thanks to you all, especially Rick, without you guys we'd it wouldn't have happened.
All the best for now,
Pete.