Where on these graphs is unhooking the fuel tank from the cannister?

George

 

worded kinda funny, I started it with the gas tank unhooked. With it hooked up it has negative trims as well.

 

I do not know how the canister is constructed internally but I suspect that the port for the fuel tank is adjacent to the port for the purge valve without going through the charcoal portion. The evap diagram and what you describe suggests this. When the truck is off the vapors from filling the tank go through the carbon on the way out the vent. When the engine is running and the purge valve is open the fresh air s pulled through the charcoal originating from the vent and fuel vapor goes into the engine. If there any leaks in the gas tank and related evap lines then some vapors or air are pulled into the purge valve. When you remove the fuel tank vent hose from the canister and plug it, all the purge air comes through the canister charcoal. When you take your finger off that port fresh air with no vapors goes straight into the purge valve.

George

 

Decided to go ahead and replace the charcoal in the canister. $14 for the replacement charcoal isn't too bad. I already have the other filter media for the ends and black RTV to put it back together. I was hoping to find more of a sign that it had been flooded though.

Replacement media should be here by Wednesday (8/10) and I'll get it all back together.

 

Did the existing media show any signs of being saturated?

Is the gas tank connection at the cannister right next to the purge valve connection?

George

 

Existing media had some big white chunks that broke apart while I was dumping it out that you can see in the picture. Had a smell of gas, but I suppose that's not crazy for it's purpose.

I've done my best to illustrate the layout of the charcoal canister.

2002 Blazer charcoal canister illustration

Edit: So it looks like the vapor from the gas tank can bypass the charcoal and the purge valve pretty much pulls straight from it. Saturated charcoal is starting to make more sense now. When I removed the gas tank line and plugged the port, it started going rich. This would make sense if the charcoal was saturated right?

 

This confirms my theory in that the purge line and gas tank lines are essentially connected together at the cannister. When you pulled the gas tank line off the purge valve pulls in pure air through the adjacent gas tank port. When you plugged that gas tank port then the purge valve has to pull in outside air through the vent solenoid and all of the carbon. If that causes the truck to run like crap then:

The carbon is saturated

and/or

The purge valve is opening too much

and/or

There was a problem being rich to begin with

and/or

The O2 sensors/computer are struggling at the limits of fuel delivery compensation

Unfortunately more sophisticated problems like this require a higher end scanner to fully diagnose. That's why I suggested that low end BT/smart phone app because you can monitor purge valve opening on the cheap.

I like the Danny Draw graphic.

George

 

24 short hours away until I know if this fixes my problem. I was hoping the RTV would squeeze out more around the edges.. Maybe I should have got a bigger tube, but I did a pretty thick bead all around the edge so it should be water tight.

 

The plot thickens as the story of the great purge (valve) is revealed.

So it turns out that the control of the purge valve is quite complicated, especially once you start driving. At idle its pretty straight forward. The purge valve is closed until around closed loop then it opens immediately to low to mid teens (percent open) than ramps to around 20% open, holds there for a few minutes and then ramps back down to around mid teens and stays there.

Once you start driving however all hell breaks loose. The open percentage ramps up with load and/or speed and stays at 100% during a steady high speed cruise. It drops to zero percent during decceleration and really flies all over the place under varying load conditions. My take is that the valve is opened by the computer as much as possible without effecting driveability and continues indefinitely for maximum cannister vapor extraction. I monitored this behavior on a three hour highway cruise and the only thing that changed is that after a couple of hours the open percent dropped to 82% at high speed cruise. During all of this trims stayed under 10% with very few exceptions so clearly this whole time the computer is accounting for a presumed volume of this unmetered air. This was on my 2006 GMC Sierra during a 30 minute drive in the city and on the highway.

George

 

Ladies and gentlemen, WE GOT EM!

Jammed the canister back in with all new filter material and activated carbon, reconnected the evap system, and ran it for 10 minutes. Trims did NOT nose dive to -10.



I wanna give George a HUGE thanks for sticking with me through figuring this out. Advising me what tests to run and what the results could possibly mean. I don't think I would have got this figured out without his immense help and knowledge.

Now I'm curious how the can got flooded in the first place. I never do the multiple clicks at the pump. I'm going to be a little extra cautious now and maybe stop it before it reaches the first click. I have a pretty good idea how much to put in based on the gas gauge and being a gallon or half gallon short of absolute full won't be the end of the world.

I've got a little bit more buttoning up on it to do (welding new cat in and clear coating headlight housings) then I'm gonna take it for a put around the neighborhood.