When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Our '94 with 3K miles on its "new" engine is creating gas fumes from the charcoal canister system.
I need to understand more thoroughly the operation of the charcoal canister system and the path from the fuel tank to the intake manifold. Attached is a scan of page F-129 of the 1994 RX-7 Workshop Manual as a reference. I have added an arrow to the check valve between the Purge Control solenoid-controlled valve and the intake manifold.
Reading the text below the picture, I note that it seems to address a normally-aspirated engine. But if that were the case, the check valve would not be needed. The purge control valve, which is sent a signal from the PCME, would be all that's needed to operate the system. But in a turbo'ed engine under boost without that check valve, the air pressure in the intake manifold could blow back through the open purge control valve and pressurize the charcoal canister with gas fumes, and vent them into the atmosphere.
Terminal 3H of the PCME (ECU) is the signal that controls the purge valve solenoid. Is that an analog signal that can actually adjust the opening of the purge valve, or is it a discrete "open/shut" signal? The picture's text says "Sending a large volume of evaporative fumes at one time into the air intake system deteriorates the air fuel ratio; thus, the PCME uses the solenoid valve (purge control) to regulate this volume." And, "The volume of fumes drawn depends on engine conditions." What conditions? What controls the terminal 3H signal output of the PCME? It appears to be an "open-collector" switching transistor type of circuit that just "grounds" the cold side of the purge control solenoid winding when active... is that right? If so, then the purge control valve is either closed or open, and no in-between. But I can't tell for sure.
Can anyone shed a little more light on this? Thanks! The charcoal canister system, from the 1994 RX-7 Workshop Manual, showing the blowback-preventing check valve.
The purge control solenoid seems to be on/off, but the ECM may cycle it frequently. I have found that if my FD sits for a month or 2 then some fuel smell escapes, after driving it the smell goes away after some canister purge occurs. Does your car sit for extended periods by any chance? If so, a good long drive with some steady state operation out of boost may resolve the problem.
i'm under the impression that the Purge Solenoid is duty controlled, like an injector or the boost control.
the check valve is there to make sure everything after it doesn't see boost, like most of the solenoids, but now that you mention it, the solenoid should do that job.
Evap solenoids are duty-cycle controlled on most cars that I'm aware of. (similar to the wastegate control solenoid and prespool solenoid and idle control valve solenoid). I haven't measured the stock ECU control of the Evap Purge solenoid, but I can confirm that leaving the solenoid unplugged for a long time (months or years) will result in fuel smell. I recently connected the Evap Purge solenoid on my car and activated it in my standalone ECU, and that has helped with the fuel smell. There was some more discussion here: https://www.rx7club.com/3rd-generati...ative-1130423/
I agree the ECU could turn the solenoid off based on manifold vacuum / boost, but mechanically the solenoid might only flow well in one direction and not be able to block 10psi of boost pressure. If you've got an air compressor and some spare time it might be possible to test this.
Question about the check valve pointed out in my attachment: If it's stuck open, then during boost when intake manifold pressure is higher than ambient, air will flow back through that valve into and through the purge valve, even if it is duty-cycle controlled, and thence into the hose/catch tank/charcoal canister. Not good. If the check valve is stuck closed, opening the purge valve will have no effect at all, and the evaporative fumes will always be dead-headed at the check valve. Also not good. Which is the worst case? Will either condition cause fumes to be vented into the atmosphere?
I think the 93 service highlights book is online. That breaks down a lot of the FD's systems and explains how they work. That may have some clues.
Also if you look at the ECU troubleshooting chapter, they show each pin and what voltage reading you should get under what circumstance. That will also give some clues. That will typically show if it's duty cycle, 12v on/off, etc.
This is one of those systems that isn't well understood and generally seems to work.
Going back to the original problem, you state that you are getting fumes - how? when?
The solenoid is duty controlled, which is confirmed by checking page F-72 of the 93 service highlights. Page F-77 says that the failure condition of the evap purge valve is to set duty to 0 (stop flow), but that would be if you have a trouble code/check engine light on the stock ECU. As you have pointed out, a working check valve is needed. Page F-131 and 132 of the 93 workshop manual (different than the document I just referred to) shows troubleshooting of the solenoid and one way valve. The way the EPA certification test works is that the vehicle drives the "city cycle" (FTP 75) with the canister pre loaded a certain amount. The solenoid has to operate under some conditions of the test or the vehicle will fail and can't be certified and sold.
In your case there could be leaky or damaged hoses. A service facility (maybe even a dealership if you have one you trust) with a smoke tester would check for leaks that way. On a modern vehicle the car can diagnose its own leaks (hence the classic loose gas cap check engine light). It does that in some cases with a built in vacuum pump, but many vehicles just use a pressure and fuel tank temperature sensor in the line to see how tank pressure changes when the vehicle cools down after being shut off.
Fumes are always vented to the atmosphere at a very small rate; ie some leakage is assumed at the various connection points. Only the very newest "zero evap" vehicles are basically leak free (virtually undetectable with lab grade instruments), but in your case yeah you shouldn't smell anything. My stock '95 never smelt like gas.
If I were you, here's what I'd do:
1) Inspect the 1 way check valve and make sure it's installed facing the correct direction (towards the manifold so vapors can flow in).
2) Inspect the solenoid, applying voltage to it and make sure the air is flowing.
3) Inspect hoses under hood, if necessary take the vehicle to a service station you trust with a smoke tester (due to the tank leakage points being hard to check)
4) Inspect the fuel temperature sensor. It's possible the sensor is no longer accurate (but not totally dead) and that is affecting evaporative purge duty cycle signal. So maybe the ECU thinks the tank is super cold and this won't activate the solenoid. This seems like a long shot though.
These cars are so old that you can't plug in a Mazda tester (which all the dealerships have thrown away) and read the actual solenoid duty. You'd need an oscilloscope or other science-projecty stuff.
Thanks for all the good information; I'll have a look at the Service Highlights (haven't read that yet). Dale, please sell me one or more of your Viton valves in case my aluminum OEM valve is stuck either closed or open... don't want to get underhood until I have a replacement handy!
digging a little deeper into the workshop manual, page F-158 says that the solenoid operates from 1500-3300 rpm. So you can assume that would likely be in a higher gear (4th or 5th), with at least some vacuum to pull the vapors in. Since much of the EPA city test is in the 20-30 mph range, that's where I'd expect it to work (3rd-5th gear, out of boost).
At least one potential source of the gas smell issue has been corrected. Yesterday I received Dale Clark's shipment of his Viton valves and replaced the aluminum ball-and spring valve (pointed to in my post #1). I found that I could blow through the OEM valve in both directions (although with somewhat more effort in the back direction). Dale's Viton valve was a solid check valve... sealed perfectly in the back direction and open when it should be. The Viton valve fits the tubing (also Viton) well and should perform perfectly. Thanks, Dale!
If the defective OEM check valve was the only problem, hopefully the charcoal canister system will purge the canister and remain free of fumes vented to the atmosphere. If there is another problem, like the purge valve itself, I'm sure it will show up soon. We''ll drive the car and see what happens.
As of today we still have not smelled any evaporative fuel fumes from our FD. Replacement of that check valve with Dale Clark's Viton valve seems to have done the trick.
Seeing as this is about the charcoal canister and instead of making a whole new thread , So do you remove or keep the canister? Im going single turbo and need to know If i can remove the canister or keep, and if i DO remove it, what will happen with all the fumes? Want the car not to be a bomb ...
Seeing as this is about the charcoal canister and instead of making a whole new thread , So do you remove or keep the canister? Im going single turbo and need to know If i can remove the canister or keep, and if i DO remove it, what will happen with all the fumes? Want the car not to be a bomb ...
keep it, if you delete it your car will just smell like unburned gas all the time.
I'm not sure this is true. Single turbo cars delete most of this stuff. Mine doesn't smell.
See this is the thing, People are saying different things. How have you vented it off? Do you just put the vent hose pointing somewhere else or have a vented gas cap or what?
I removed mine entirely and I don't smell raw gas like that. I smell a lot of other things when my car runs but gas isn't one of them. Single turbo of course
I removed mine entirely and I don't smell raw gas like that. I smell a lot of other things when my car runs but gas isn't one of them. Single turbo of course
So is it venting straight from the tank or are you using a vented cap??
I'm asking from ignorance, of course, but why the heck would having one turbo instead of two have anything to do with a raw gas smell?
it doesnt... just figured I'd mention
Originally Posted by GrapeSoda
So is it venting straight from the tank or are you using a vented cap??
Vented cap and venting straight from the tank. I just removed the can and left behind whatever was connected to it.
Originally Posted by neit_jnf
Do stand alone ECUs (PFC, Adaptronic, Limk, Haltech, etc) control the evap system or just ignore it?
Pfc ignores it as far as i know. With the modern standalones, you can set them up to do evap things like stock but unless you need it for emissions, which i don't think you can pass with a standalone, whats the point?
So with the PFC the Charcoal Canister is there just adding dead weight doing nothing? Might as well eliminate it and use the line as second feed, 2 feeds 1 return. hmmm
07-25-21, 12:19 PM
