S4 / RS4 (B5 Platform) Discussion Discussion forum for the B5 Audi S4 & RS4 produced from 1998-2002

OBDII Small Leak or Large Large Evap Write-up because STFA sucked..

Thread Tools
 
Search this Thread
 
Old 07-16-2006, 07:56 PM
  #1  
AudiWorld Senior Member
Thread Starter
 
EricVonHa's Avatar
 
Join Date: Oct 2003
Location: PA Philly Region
Posts: 786
Likes: 0
Received 0 Likes on 0 Posts
Default OBDII Small Leak or Large Large Evap Write-up because STFA sucked..

My S4 threw on the CEL (Check Engine Light) a couple weeks ago. It was a "small leak" emissions error code. Yeah yeah. I blew it away with the Vag-Com and it re-appeared every couple of times I would drive the car. After clearing the code and having it reappear almost consistently, it really started to annoy me. So, that's when I got serious and started searching the archives about where to start troubleshooting. The previous posts are all over the place--and I mean all over. There's talk of the line near the exhaust, N80's, vacuum canisters and check valves. And not once did I find a post where someone definitively said, "this is how I approached the problem, and this is what I did to fix it". This "ghosting" issue on this site needs some real attention and policing. We have a bunch of drive-by wannabe mechanics asking real world questions and generating great debate about how to approach things, however, it is rare that anyone says after the fix-a-fact that "this is exactly what fixed the problem"--and this one topic of Emissions error codes just flat out sucks from the archives. So, please read on hit me up on email if you have any more specific questions.

In my case, my "small leak" code developed into a "large leak" code through a couple tanks of gas. I do 400+ miles per week and if it was my exhaust rubbing through the rear charcoal canister line, I was going to fix it right away.

Before we go any further, here are the relevant components of the OBDII emissions and fuel vapor recovery system specifically on the S4. These same components exist on other manufacturers cars though their terminology may be slightly different.

Evap fuel vapor recovery continuity and theory- basically, there is a continuous loop from the N80 check valve, back to the charcoal canister, filler neck, leak detection pump and back to the vacuum test source on the driver's side. So, when "test mode" goes into effect from the ECU, the computer drops out the N80 (no voltage = closed = blocked) and the ECU also engages the LDP solenoid (12vdc). With those two valves closed, vacuum is supposed to be fully trapped in the evap recovery system plumbing and charcoal canister. If the vacuum escapes (loose or defective gas cap, cracked lines, bad N80, broken vacuum sources in front of N80 or driver's side source) - the reed switch in the LDP will drop out from the diaphragm losing vacuum and the requisite code will be set determined by how fast the vacuum is lost. If it's a slow leak then it's a "small" leak code. You know the rest. Fast leak = large.

N80 = check valve or "purge valve". Located on the passenger side directly adjacent to the Y-pipe. This is a black valve in which the evap line from the charcoal canister comes in from over top of the air-box and in one side of the N80. The other connection from the N80 exits from the valve on a 90 degree difference from the incoming evap line. (note, this valve is not grey in color- that is another electromechanical valve). The N80 is closed if the car is off. This is important to note. If the N80 is closed, it's purpose is to block all fumes (and vacuum) in the evap system towards the rear of the car. (this is the beginning of the leak detection cycle) In normal operation (not leak detection mode) the car's computer will pulse the N80 at varying rates and will allow fuel vapors to be drawn from the fuel tank and charcoal canister into the intake manifold and finally burnt as part of the combustion process. Nice little trick the greenies have worked out, eh? By default the N80 is closed. So, when the car is off, the N80 is closed. Why? This keeps the vapors in the tank and charcoal canister till the engine is started. (pressure does not build up in the system because there is a fresh air filter on the LDP pump in the rear)

Charcoal canister- accessible only from underside the trunk by removing (3) 10mm acorn nuts on the underside of the spare tire well. There are (3) lines connecting to the charcoal canister

There are (2) lines on the passenger side of the canister. The large one goes to the gas fill neck. The gas cap *should* seal this line and you should be able pull down vacuum with a Mityvac if you separate the fitting at the charcoal canister. Testing with the Mityvac in this fashion insures that the line is intact and that the gas cap seal is intact. Did you ever see the OBDII emissions testing machine that tests the gas cap off of the car? What a joke! Your tax dollars have paid for a half assed software package and half assed emissions testing package. If you test this line with a Mityvac from under the car, you'll need additional vacuum fittings and so forth to adapt up to the 15mm (or so) hose. (there's another method of testing the continuity of this line under the Vacuum canister section) The gas cap is a likely culprit, but also be aware that this vac emissions line crosses and rubs alongside another line on the passenger side spare tire wheel well.

The 2nd line on the passenger side of the charcoal canister goes to the front of the car and directly to the N80. If the car is off, the N80 is closed (see above N80 section). By this time you've probably tested the N80 with a Mityvac. In any testing with the Mityvac, the tested component or line should be able to hold 20" HG vacuum for 15 seconds. If you lose more than 20%, within that time you'll fail the leak test. I haven't seen anything published about this whatsoever. However, my testing of the Checkpump (aka Leak Detection Pump or LDP) indicates that a loss of 5" vac is enough to cause the diaphragm to drop in the LDP.

The 3rd line of the charcoal canister is on the driver's side. This line goes directly to the leak detection pump (LDP). This line is also the one that can be an issue if you have aftermarket exhaust. And, let me add another editorial comment here. I have the VM3 exhaust system and this evap line has been pummeled and rubbed pretty damn hard that I could put my thumb in the gash created by the exhaust rubbing against it. Was this the problem in my case? Nope! Other lines that I've investigated are all plastic coated with a metal internal tube. Dunno if this one is the same, but again, this was not the problem.

There is one other vacuum line worth mentioning. This line has a fitting near the power steering reservoir on the driver's side of the motor. In between the valve cover and p.s. is where there is a coupling for this line. These fittings are highly fragile and easy to break on high mileage heat soaked cars. Be careful! Anyhow, this line on the driver's side feeds the LDP with it's "test" vacuum.

LDP = Leak Detection Pump = Check Pump. This really isn't a pump. However, my research into the German machinery always terms this as some type of pump. Weird. It is truly just a vacuum switch canister. This is located above the driver's side rear wheel well. There are a bunch (maybe 10) torx bolts that hold the wheel liner in place. Don't be shy. Rip that bitch out of there. There is a little bit of science behind getting it out and once you do it, you'll remember how you did it. One point to note, there is (1) torx bolt that is completely hidden and directly above the tire's outer edge (nearest the fender sheet metal).

The small line that connects to the front of the LDP with a 90 degree right angle vacuum fitting is the line that goes up to the driver's side of the motor (see above) This line always, note *always* has vacuum present when the motor is running. There is nothing blocking vacuum to this line or to the LDP whatsoever. (this is not the balance line that connects the top half of the ldp to the lower half)

The large line connected to the rear of the LDP has one of those fancy squeeze type of connectors. This line goes directly the charcoal canister. This is the same one (from above) that could be rubbed by the exhaust.

LDP 3-wire connector. The bottom two pins of the LDP are solenoid connection points. The ECM feeds 12vdc to the solenoid when it commands "test mode". You can see the "test mode" show up in vag-com while driving and watching Block 70. The best way to get into test mode quickly is to warm up the car (water @ 200) and get cruise for a few minutes @ 50mph or over. Test mode will engage below 50mph, but the aforementioned is a sure way to induce it. Also, test mode will engage multiple times for one 15 minute trip of varying speeds/load. When test mode is engaged, 12vdc to the LDP locks the vacuum from the LDP through to the other components and lines within the system. There is a reed switch indication within Block 70. When the system goes into "test", you want the reed switch to say "Open". If the reed switch stays in the open position for the duration of the vacuum test--then all is well and no code will be generated. But, you're probably reading this because you have a code. In which case, the LDP reed switch has fallen back to the "closed" position prior to the time duration of the test. The test is 15+ seconds at a time.

The top pin connection of the LDP has a 650 ohm pull-up resistor in series with the solenoid voltage. No vacuum means you should be able to meter 670 ohms or so between the top pin and the lower (2) pins. If you meter between the lower (2) pins for resistance, you should see roughly 15-25 ohms. If you're slightly outside of this specification, no worries. Solenoid resistances do vary with age. The important thing is that the canister diaphragm can hold vacuum by engaging the solenoid. If you take a large magnet while metering the top pin to the one of the other two, you will be able to cause the reed switch to pull in and cause an "open" circuit (no, it does not short). This is the same thing that occurs when the test mode is activated and the entire evap system can hold vacuum. If the evap system is holding vacuum, block 71 is "open". The reed switch goes open by the movement of the internal diaphragm being pulled open by the vacuum.

Ok, so after all of this, what is causing this leak? Let the troubleshooting begin! Now that you have the above background on the components, now let's understand what this write-up is geared towards. If you have a "large leak" or "small leak" or "no activity" type of code-? this is for you. N80 codes speak for themselves, and checking that valve is easy enough with a Mityvac and 12vdc source. It either holds 100% vacuum from both sides or it doesn't. Any short to ground codes are wiring related *or more commonly* indicate a defective solenoid in the N80.

Sure, you can drop your car off at a dealer or emissions fix-it station. I guarantee that getting into this evap recovery system is a minimum of $400.00. Not that the money is too bad, but having someone getting into all of those fragile lines and potentially creating other probs is not worth the headaches. So, this is why I took this on myself. Plus, the fact that I did not have a $1,000 smoke injection machine (nor do any shops that I know of).

There is a methodical means to troubleshoot a leak in the evap system. Start with the N80. Does it leak or does it not?

Turn on the vag and go for a ride while monitoring Blocks 70 & 71. Block 70 tells you how the N80 is being pulsed. After the car is warmed up, you should see a fair amount of activity ranging from 10-25% as a regular course of action. I only saw it go to 100% full on duty cycle when I was blocking off other lines.

Block 71 indicates when test mode is engaged and the status of the reed switch during the test (or outside of the test). Beware that the status of the reed switch while the car is off is INACCURATE. The switch will not be "open" if there is no vacuum. The car will need to be started in order to show the correct status of this reed switch which is "closed" outside of test mode. I think this has something to do with the ECM not feeding the ground in accessory mode.

(all of the below testing requires that you clear the codes each time you set a leak condition)

On the charcoal canister (in retrospect) I'd remove the small vac line that is on the passenger side of the charcoal canister and runs to the N80. Plug both sides of this line. In other words, block off the charcoal canister connection and the N80 vac line with vacuum caps and go for another ride. The system failed the test again? Well, we know that this is not the problem. Or, if the system did pass the test, then you know you have a leak within the N80 vacuum line from the charcoal canister to the front of the car. This is the ONLY line that can be fully capped off without generating an erroneous code. Basically, you just simulated that the N80 is closed. The system has no way of detecting that this line is blocked off (surprisingly).

From there I would block off the larger line on the pass side of the charcoal canister that leads to the fuel filler neck. This requires a 15mm or so vacuum cap. Take the car for a ride again and does it pass the leak test or not? If it does, more than likely you have a bad gas cap. Trust me, I swapped (3) gas caps in place only to find that all (3) of them were not vacuum tolerant. Go with an OEM Audi gas cap or you'll be sorry. Yes, it appears as though other caps will work and they screw into place... but the seal tolerances are different. I learned this the hard way.

Well, if the leak still occurs, then you can go the evap line on the driver's side of the charcoal canister. Block it off and what occurs during the test ride? There is a possibility that during these tests that the system will "pass". But, because you are potentially placing a highly restrictive seal on the evap system, the reed switch on the LDP may not release or "close" (or bleed down) for a looong time. If the reed switch stays in the "Open" position--and the system requests another test while you're still out driving--an "inactivity" or "stuck open" code will appear. This could indicate that there is a defective reed switch circuit. Or, with the above troubleshooting (as I performed it) it is possible to trap too much vacuum in the system and it may not bleed out before a 2nd test is engaged (which is also a problem).

The LDP is a finely balanced vacuum canister instrument. As these cars age it would not surprise me to find out that the diaphragms or reed switches in the LDP's will become more susceptible to failure. If you're good with electronics and understand the above methodology of how the evap recovery system operates, you'll be fine!

Also, there is no way to circumvent the system. Yes, you can block off the N80 line and be done with it... but then you are leaving potentially unvented fuel vapors in a pressurized condition. If I ever found that this line was d.o.a. under the car, I'd just put a small fresh air filter on the vacuum canister and tell the greenies I proudly did so. These OBDII emissions systems are so complicated, that's my waiver!

If you try to bypass any other part of the system--the ECU will tell you that you are causing problems and will generate another code to cause you to fail emissions testing. It will also fail the readiness mode. ? So there you have it, OBDII is on your car and the government and greenies have complete control over your ability to legally drive your car. Isn't life just great?
Old 07-17-2006, 04:24 AM
  #2  
AudiWorld Super User
 
lemieux@curbside's Avatar
 
Join Date: Dec 2002
Posts: 17,772
Likes: 0
Received 0 Likes on 0 Posts
Default

DAMN ! ..
Old 07-17-2006, 05:17 AM
  #3  
AudiWorld Super User
 
curbside crackdealer's Avatar
 
Join Date: Oct 2005
Posts: 3,843
Likes: 0
Received 0 Likes on 0 Posts
Default

Wow. And you wonder why I had no answer when you asked what the problem was? Good info.
Old 07-17-2006, 06:12 AM
  #4  
Senior Member
 
S4Poor/deux@Curbside's Avatar
 
Join Date: May 2006
Posts: 6,131
Likes: 0
Received 0 Likes on 0 Posts
Default

Jebus Eric. Hell of a write up, great info.
Old 07-17-2006, 07:06 AM
  #5  
Head Cat Herder

 
Kris Hansen's Avatar
 
Join Date: Oct 1998
Posts: 66,550,451
Received 85 Likes on 48 Posts
Default

Excellent! I'd like to add this to tech, unless you have any objections.
Old 07-17-2006, 07:14 AM
  #6  
AudiWorld Senior Member
Thread Starter
 
EricVonHa's Avatar
 
Join Date: Oct 2003
Location: PA Philly Region
Posts: 786
Likes: 0
Received 0 Likes on 0 Posts
Default Thanks, I was really frustrated with the lack of info out there and the documentation.. so...

without buying or borrowing a smoke injection machine, I went at it by just blocking off the various lines and watching how the VAG reacted. I got into it a lot more than I needed to, but it was worth it. I'm now a certified OBDII technician. Ha!
Old 07-17-2006, 07:19 AM
  #7  
AudiWorld Senior Member
Thread Starter
 
EricVonHa's Avatar
 
Join Date: Oct 2003
Location: PA Philly Region
Posts: 786
Likes: 0
Received 0 Likes on 0 Posts
Default No prob on adding.. feel free to edit as necessary. There are some cracks ...

at the greenies and "big government" lurking in the write-up, but please feel free to change accordingly. -E
Old 07-17-2006, 08:13 AM
  #8  
Head Cat Herder

 
Kris Hansen's Avatar
 
Join Date: Oct 1998
Posts: 66,550,451
Received 85 Likes on 48 Posts
Default

Thanks
Old 06-11-2008, 07:23 AM
  #9  
AudiWorld Senior Member
 
cparke's Avatar
 
Join Date: Apr 2008
Posts: 687
Likes: 0
Received 0 Likes on 0 Posts
Default Now it's linked within the B5 A4 FAQ too.

(where EVAP has been noteably absent)<ul><li><a href="https://forums.audiworld.com/a4/msgs/2820654.phtml">Message within the B5 FAQ thread</a></li></ul>
Related Topics
Thread
Thread Starter
Forum
Replies
Last Post
Kaotiic_S4
S4 / RS4 (B5 Platform) Discussion
2
11-17-2015 11:41 AM
Iron_Mike
A8 / S8 (D3 Platform) Discussion
4
10-09-2012 04:07 PM
ChooChooNongNongHead
A4 (B5 Platform) Discussion
2
11-30-2007 05:54 AM
How Bizarre
A4 (B5 Platform) Discussion
5
05-25-2006 06:38 AM
S42000togo
S4 / RS4 (B5 Platform) Discussion
7
04-30-2006 06:50 AM



Quick Reply: OBDII Small Leak or Large Large Evap Write-up because STFA sucked..



All times are GMT -8. The time now is 07:13 AM.