Vapor Canister Purge Solenoid - EVAP woes?

9A111009304. Does that mean anything to you? Never heard of it? That's the part number I'm likely going to have to shell out over $200 for, for my 987.2 2.9L Boxster -- with apparently an annoying installation process. Yes, it's pay-to-play with our cars, but I really just wanted to get out and drive! I went out for an amazing Sunday drive yesterday, as the cold and rain appears to have finally taken a breather here in New England. Had some wacky Food Biker adventures (which I'll post separately on social). The day started out with a 1 hour highway blast. Unrelated to all this, the car felt ever so slightly squirrelly at high speed, and I figured it was just my nerves given that I haven't been able to drive much these past few months post-surgery. What was up? Checked the warm tire pressures, which were hovering around 37-38psi -- which I think explained a lot. Dropped the pressures down at a huge parking lot in NH, and the car felt drivably improved for the rest of the day -- and my confidence quickly came back. However, at a stop light, I could have sworn I smelled a slight bit of gas and wasn't sure if it was from my car. Didn't think much of it but kept it on my radar. At some point a bit later, I was at another stoplight and the car blew a CEL at idle. The car seemed/felt fine, but I headed back to the barn modestly and got home okay. Car seemed to run smoothly on the way home. Well, I threw the PIWIS 2 on it, and saw a P0441 code, which upon looking up is the Vapor Canister Purge Solenoid. Hmmm! Something to do with the Sunoco 91 fill-up earlier in the day? I was already down about 1/8th of a tank at that point when the code blew. Today, I internally wiped the gas cap down (2 years old?) off with a paper towel dampened with 71% isopropyl alcohol, including the metal filler neck, and then re-attached the cap -- as apparently a bad seal can cause that code. Started the car 15 minutes later, and it didn't seem quite right (hunting up/down) and maybe a bit rough -- but it smoothed out within 3 minutes - guessing it had something to do with the newly installed Pedal Commander (more on that later). Anyhow, post PIWIS read and clear, I found the following: https://www.planet-9.com/threads/how-to-replace-purge-valve-the-easier-way.251905/ ...but this was for a 987.1 I'm wondering if it's the same install instructions, as the MA1 (9A1) engine often has more in common with its 981 brother. @jjrichar - Any chance you know? No codes thrown again (yet) per some local driving today, but would love to plan ahead before tariffs kick in per Euro parts. I figure if it blows the code again, I'll order the part. I'm fine with fiddly installs (tiny 8mm nut in a bad spot) and have my QuickJack lift. I just can't afford to have the car stranded though with an install I can't complete -- I just have too much going on right now. I also just don't want my rental apartment and garage to fill with gas fumes in the process -- that's the most important issue of all. Any thoughts, feelings, advice, please let me know. Thanks, everyone.

I don't have anything for you unfortunately. Being in Australia the emission controls aren't quite as heinous as in the US, and so the system installed in the cars I have seems to be different to that of a US spec 981.

I believe you can just change the valve on it's own: Purge Valve

Thanks, @jjrichar - I'm guessing it's the same or similar in the 987.2/981 but likely missing for the Australian market as you say... @Drabbs - That would be cool, but the hose is a hard plastic and likely bonded to the valve. I had hoped to do something similar for my wife's MINI when I damaged the line, but I don't think one can easily reinstall the valve -- even if I cut it out of the plastic - the plastic is a very hard and brittle material. Go figure.

Before you start throwing the parts cannon at the car, take the solenoid off and blow through it. Like, literally put it up to your mouth and blow. Now put 12V to it, and blow again. I can't remember which way is which, but if you can blow freely one way, and then it's completely sealed the other, than the solenoid is fine.

I would do as theblooms has said and if faulty try to remove from the hard lines - worst case you're where you first started - whole assembly, but if not, just the valve

@theblooms and @Drabbs - Thanks for the suggestions. Very much agreed about solenoid checks - I've done that many times before with solenoids for valved exhausts (vacuum). If that were the case, this would be a no-brainer as I tend to bench-test them with +12V. Thankfully, I'd like to think that my days of "parts cannon" firing are long gone. The first key problem is that the EVAP hard lines to the solenoid are manufactured out of brittle and molded plastic, that are literally molded onto the solenoid. This means, you can't remove them for love or money -- and the molded hose is the expensive part. As this is a common part that fails on many vehicles, one tricky solution out there is to literally hacksaw the solenoid off and then trying to remove the broken barb out of the hose with a needle nose pliers. Then, apparently one can gently heat up the hose with a heat gun (on low) to try and soften it enough to then reinstall a replacement Bosch solenoid. Given the gas fumes, I'm not too keen on that but it's doable. Although technically one could replace that garbage molded plastic line with fuel line, I haven't really seen anyone do that and I admittedly wouldn't want to take a chance in this case RE: fire. I've had silicone tubing that was supposed to be fuel proof deteriorate on our Vespas and I'd want a one-and-done solution. The other tricky part is very much the entire hose removal process. Given the link above, apparently it's a difficult to remove and reinstall -- and I don't even have a confirmation that the 987.1 is the same or similar process as on the 987.2. I think my next step here is now that I've cleared the code, I'm going to drive the car a bit and keep my eyes peeled for additional codes and my nose out for any gas fumes (incorrectly vented gas fumes). If it doesn't blow a code again, hopefully it was just the gas cap and a one-time thing. If it blows the code again (which I'm guessing it will), I'll remove the hardtop... put the Boxster on the lift... and try to inspect to see if that's the similar hose/valve as in the 987.1 and plan for the repair. Anyone else have this issue (especially on the 987.2), please do chime in! Thanks, everyone.

My gas cap was unremarkable in appearance, it tightened up as it always did from new. It threw an evap code last year right before it was going for tailpipe sniff, ‘twas quite an adventure for data repopulation. Anyhow, my point is this: Just my opinion here but about a decade is a good time to preemptively replace the fuel cap . We don’t live in the era of “metal twisties” on a hinged clamshell mount where “ good” “meant still mounts tight and doesn’t leak fuel”. They aren’t expensive, so especially if I lacked the durametric for resets I’d replace it in or around that timeframe. Also, after long sits, ( a couple winter months or more) removal and resetting of the gas cap will only help. Weather cycles in the storage area can have an effect. In the south tire pressures and our caps are susceptible to dramatic shifts from this phenomenon.

Good points , @Boristhespyder - Per my previous posts above, I had replaced it with a brand new OEM one about 2 years ago, as the original (from the previous owner) had lost part of its end-hook at the end of its tether. However, it did sit for two months during recovery during surgery, so I think I'll keep moving it around. I definitely tighten it down to the clicks each time. I swear - I just hope that's the case and it was simple as something as this and a $16 gamble is nothing as compared to the stupid hose replacement. Great idea about open/closing to keep them in better shape. Yes, these are definitely a far cry from the metal ones. A little while back, I replaced a cork gas seal on my (soon to be up for sale) BMW R75/6 gas cap, which won't dissolve in alcohol. The one thing that I do wonder about is that here in New England (and in northern US states), there's a high alcohol content in gasoline during the winter months. The alcohol has degraded many a seal over time (at least in our Vespas) and I've had to go back in and replace them. Wondering if this gas cap might be another victim. In simulation mode on the PIWIS 2, I think I found the EVAP valve and will likely throw it on the car today to see if it can activate it or not in the "drive link check" setting. Would be good to know if it goes on/off and if I can actually hear the solenoid move.

Keep this in mind: just because the solenoid clicks, doesn't mean it's sealing. I've been down this exact road with the exact code in my wife's old Mazda5 minivan.

Great point, @theblooms - Really glad you pointed this out. Thanks! If it's not clicking at all, then we definitely have a culprit. If not, then it might merit replacement anyhow.

Well another point, if it's not clicking, it could still be good! There could be a break in the wire or some corrosion somewhere. You live in New England, and salt plays hell on wires, harnesses, and connectors. A tool like a Power Probe could help diagnose that.

Plot thickens ever so slightly... (TL;DR = Original CEL code was P0441 “Evaporative emission system (below limit value). Ran PIWIS 2 for entire DME, and all checked out okay… but did this testing cover what could have gone wrong?) In the DME (CAN), I ran all of the following “drive link” tests sequentially… all of which ran successfully. I could audibly hear each activate without issue: Valve lift adjustment 1 and 2 Tank vent valve Secondary air pump Exhaust flap (NOT APPLICABLE TO MY CAR) Fuel pump relay 1 Fuel tank leak diagnosis pump Fuel tank leak diagnosis valve Engine purge fan Coolant…something for PDK (NOT APPLICABLE TO MY CAR) When I had the PIWIS earlier in the day, during one of the tests, there was a brief hiccup with the "Fuel tank leak diagnosis valve” (Negative control unit response: Incorrect conditions”), but that quickly away — which I expect wasn't a big deal particularly as the car wasn’t running and there was nothing to measure? Before running the PIWIS live on the car, I had run the demo earlier in the day, which was for a DME (DFI) engine. Interestingly enough, some of the options that showed up in PIWIS 2 demo mode didn’t show up when I connected it to the 987.2. This seems likely as the options are directly dependent on what the P2 sees in any given connected car. Again, my 987.2 2.9L is an EFI car, so the options were obviously different. You’ll note some 987 DFI demo differences below: Bank 1 intake camshaft Bank 2 intake camshaft Fuel pump 1 Valve lift adjustment, bank 2 Valve lift adjustment, bank 1 Exhaust flap Engine purge fan, level 1 Fuel evaporative valve Engine purge fan level 2 Quantity control valve So although some of the items were the same, you could see that the wording was sometimes slightly different. I have often found this between different models (the PIWIS options for a 981 are sometimes different than in an 987.2 or 987.1, etc.) Interestingly enough, the “Fuel evaporative valve” was missing in my EFI version of the DME — which I figured would be the faulty solenoid and there were no other clearly similar options in mine. I doubt "Tank vent valve” was the same as that allows you to fill your car. Both "Fuel tank leak diagnosis pump” and "Fuel tank leak diagnosis valve” sound similar, but I believe that’s separate EVAP setup that slightly pressurizes the system to monitor for leaks. So, I’m kind of puzzled as there were no other options to check/test in my system that seemed remotely similar. Meanwhile, surely that solenoid is in there? Any thoughts, let me know. ** @theblooms - Another good point. Thankfully, although I drive year round, I'm pretty religious hosing the car down during the winter. Plus, this was a southern car for almost its entire life and has thankfully been spared electrical corrosion.

One other follow-up. It just dawned on me to try simulation mode for the 981 , which is basically the 987.2 (MA1) engine. The PIWIS put in another DFI engine, but the entries were similar to what I saw for my 987.2 . Interesting. I think at this point, I'm going drive it for the next few weeks and just keep my eyes peeled for another blown code -- and keeping that gas cap really clean. If it blows the same code again, especially on the day-of a fill-up, clearly something is wrong and the question will be where. If anyone has anything else they want to add, I'm all ears. Thanks.

Yesterday, literally at 259,999 miles and just moments before Cherry ticked over to a milestone 260K miles at a stoplight, it blew the CEL again (again, it happened while waiting at another stoplight). Also noticed faint whiffs of gas when starting the car and figured it's my car and not someone else's. PIWIS this morning confirmed the return of the 0441 code. I hate to say it, but I think it's time that I just order that $$$ hose and replace it as a next step. Assuming the PIWIS was likely able to activate/deactivate that solenoid, I'm guessing that the valve seal itself may be leaky. But before I order the part, if anyone wants to chime in with any last-minute suggestions, I'm all ears. Again, I'm fairly certain that the newish gas cap (2 years old) isn't the problem and I'd replace that as a second resort -- especially with the occasional and likely gas smell upon start up. Thanks.

Here's an update, especially for any 987.2 owners: Instead of shelling out $200+, I replaced the hose assembly from a used 2009 on eBay. The awesome news is that is MUCH more easily replaceable on the 987.2 as opposed to the 987.1 - no 8mm bolt to fiddle with - it's all on the top of the engine and not under the left runner! The item on the hose assembly that looks like a canister is likely a filter, and not a one-way valve as I had thought. I did some pressure and vacuum testing, and both my original and the replacement part seemed to work just fine. I tested using both +12V on the bench, and more thoroughly via the PIWIS. (I had already closed up the bottom of the car, so I didn't do testing from that end of the replacement hose with the PIWIS.) On the 987.2, the "Tank vent valve" is the item you want to select and test in the PIWIS for this solenoid per the "drive link checks." If I had to do it again, I would have just put the car up on the lift (QuickJack) and removed the rear bottom plastic tray. That would have made snaking the hose, plus disconnecting and reconnecting the hose assembly much easier. However, it is doable through removal of the left (starboard) wheel well and front wheel liner. Yes, definitely tie a string to the bottom of the hose to help correctly re-route the replacement when reinstalling. MAKE SURE you reconnect the electrical connector to the solenoid and you close things up. Lucky for me, I realized in bed overnight that I had forgotten to reconnect it before I had restarted the car. Definitely a PITA to take everything apart in the Boxster and then put it back together, but it gave me the opportunity to go back in there and do some double-checking. So, post-installation - as both hoses seemed fine, I'm guessing this isn't not going to solve the problem. However, I'm trying to be methodical about this and I installed the replacement hose anyhow. The car did stumble upon a cold start-up again, so I'm still puzzled - but that could have been from the last time I turned it off. TBH, I'm expecting the stumble to return at the next cold start. I took it for a solid half-hour freeway run, with some start/stop city driving. No error code, but I didn't quite expect it yet. Fuel tank is still half-full. I think the next step is ordering a replacement gas cap anyhow (as they're so inexpensive), despite the fact that I replaced mine roughly two years ago and it looked just fine. But then, it's means question of moving forward in the car to the charcoal canister area. The only other stuff I'm wondering is "what's the other most recent stuff I've done to the car" that could have caused this? Last fall, I had installed and uninstalled a vacuum valved exhaust. However, I made sure to carefully reinstall where I had tapped off the vacuum line -- and I doubt there's a leak there. I did do the 981 exhaust tip hack and OEM exhaust gap hack as well, and I've left it on a VERY tiny gap that's effectively size of a screw hole. Given that's so downstream, I doubt that's the culprit either. Again, if anyone wants to chime in, let me know. I'm hoping to do a really good long drive tomorrow (assuming the sun finally comes out) and give it another test. Thanks, everyone.

Sounds like you're going the right direction. And thanks for the update