Trying to find VACUUM LEAK- Need advice

[Warren N.CA]

from those who "been dere, done dat."

1990 325is M20, Manual.

Motor will stumble and almost die, after running fine for a while. It's worst just above idle, and the idle will hunt from 500-1200RPM, rapidly, then die. It also happens at any speed, including highway cruise. CHECK ENGINE light comes on intermittently during these episodes.

Stomp test reports 1222 (I am told this means vacuum leak.).

Today, I went under the hood with the motor running and sprayed WD40 on all the vacuum hose fittings that I could identify, looking for the tell-tale engine speedup- but no dice. Not surprising, because the car was not acting up at that time. Maybe I'll take the WD40 with me tonight when I run my errands, and try it while (if) the car is doing the bad thing. First experienced this several weeks ago, and then nothing more until last night.

Any pointers as to how to proceed will be appreciated!

[Unregistered]

For starters, check the dipstick O-rings by pushing down hard on it while the engine is idling roughly. FPR vacuum can be toast, but as it snakes below the intake manifold, you'll not be able to spray there, which is also sight of other vacuum leak posibilities. Oil filler cap gasket? Brake vacuum assist? Idle control valve? Loose clamps on intake boots? Micro-cracks in same?

[632 Regal]

Ditto, look for the stupid stuff first, then go to the more drastic R&R methods. I searched on mine 3 times before I saw the intake boot (after the AFM) slipped off the bottom side of the intake opening...stupid stuff...

[Warren N.CA]

run low on stupid stupid stuff. Everything that I can see looks really phat. The car was gone through professionally with a fine tooth comb 4/02, and the AFM boot replaced because it looked old and ratty. I cleaned off the dipstick O rings with a rag, and wiped out the area of the tube where they seat. If only it would just break and STAY broken.



Ditto, look for the stupid stuff first, then go to the more drastic R&R methods. I searched on mine 3 times before I saw the intake boot (after the AFM) slipped off the bottom side of the intake opening...stupid stuff...

[Unregistered]

My O-rings LOOKED wunnerful. But, they weren't. Very cheap fix. Get somebody with a skinny arm to reach under the intake manifold and check the seating of the FPR vacuum tube. Better yet, get ~16" of new tube and put that on. Is your valve cover battened down properly? Oil filler cap got new gasket? If not, hardware store has neoprene big washer in exactly the right size, for ~50 cents.

[632 Regal]

mann, this sounds like a lengthy post with lots of new stuff inserted between the lines. Have you completely cleared the codes and started from a frest start? I'm tellin you its something stupid. Do the clear codes and we will start from there.

[Warren N.CA]

It ran perfectly last night. The dip stick had some grimy oil on O rings, and cleaning that off might have fixed it. I will also replace the oil filler gasket. I'd like to clear the codes so I can tell if it has recurred, even momentarily


mann, this sounds like a lengthy post with lots of new stuff inserted between the lines. Have you completely cleared the codes and started from a frest start? I'm tellin you its something stupid. Do the clear codes and we will start from there.

[Bill R.]

something has happened outside of the lambda control range...by this I mean that the dme monitors the O2 sensor and is able to do some adjusting to the fuel mixture within a narrow range to correct various deficiencies such as richening it up slightly to correct a small vacum leak or leaning the mixture out if its getting too much fuel... The range that it can correct if fairly narrow, it can't work miracles in fuel systems control... So anyway that code comes up if for some reason its gone lean or rich outside of the range of control, which would indicate a fairly good sized vacum leak or its running rich outside the range of control, a bad diaphragm on the fuel pressure regulator allowing fuel to be sucked into the intake manifold through the vacum line going to the fuel pressure regulator. I doubt that the gook on your dipstick orings was enough to trigger this code...
To clear the code you can do the stomp test and read the codes then after the codes are done flashing hold the gas pedal to the floor for 10 seconds or so. then shut the key off and repeat the stomp test to see if it cleared the code, if not you can try again. On some cars that won't work so then you have the option of disconnecting the battery for a couple of minutes or waiting until after you have done 60 starts on the car and it will clear itself as long as the original cause of the code has been fixed... So if you start the car 3 times a day thats 20 days and the code clears itself if the problem has been fixed.









from those who "been dere, done dat."

1990 325is M20, Manual.

Motor will stumble and almost die, after running fine for a while. It's worst just above idle, and the idle will hunt from 500-1200RPM, rapidly, then die. It also happens at any speed, including highway cruise. CHECK ENGINE light comes on intermittently during these episodes.

Stomp test reports 1222 (I am told this means vacuum leak.).

Today, I went under the hood with the motor running and sprayed WD40 on all the vacuum hose fittings that I could identify, looking for the tell-tale engine speedup- but no dice. Not surprising, because the car was not acting up at that time. Maybe I'll take the WD40 with me tonight when I run my errands, and try it while (if) the car is doing the bad thing. First experienced this several weeks ago, and then nothing more until last night.

Any pointers as to how to proceed will be appreciated!

[Unregistered]

When I mentioned the dipstick O-rings, I should have also noted that you should go with OEM O-rings: I put some 50 cent fatter hardware store neoprene O-rings on mine, and it sealed up perfectly. However, after a few weeks these O-rings sorta got fried in place, making the dipstick very hard to pull out.

I have not checked yet, but there is probably another O-ring or similar seal at the bottome of the dipstick tube, where it fits into the oil pan. That will need attention/replacement too, as well as cleanup of collected crud down there. BTW, if there IS oil leakage at the base of the dipstick tube, then air can leak IN via the same path the oil leaks OUT. Fix it.

Also, on the oil filler cap: I used a hardware store neoprene washer, which was exactly the right size. I first removed the stock washer, then put on the new one, then put the stock one back on top of that. That way, the stock washer is still mated with the valve cover lip, and won't get fried on, but has the new washer behind as a shim for a tighter seal.

Valve cover: Yes, I know it's supposed to take a dry gasket, without sealant. But, it was leaking, so I put a thin layer of gasket sealant on top and bottom surfaces of the old gasket, replaced the valve cover, and re-tightened to spec torque. The excess sealant oozed out the sides, so I waited overnight 'til it congealed, then trimmed excess with a razor to make it look purty.

Replaced your FPR vacuum tube, and similar tubes, yet? Mine LOOKED fine, but was literally toasted through in places not easily seen, so leaked bigtime. That meant the FPR was not doing its job properly, so injectors had wrong pressure and air/fuel mixture was off = poor performance.

BTW, the prior owner and BMW dealer (who is very well paid to know better) had totally neglected all of this easy fix. So, the car idled and accelerated poorly, the prior owner despaired of even bigger BMW dealer repair bills, so fire-saled the car for $3,250. New O-rings, vacuum tubes, and gasket sealant cost me ~$6.50 to remedy these problems. If you can think of a cheaper way to get more bang for your maintenance buck, let me know.

[535ise]

For starters, check the dipstick O-rings by pushing down hard on it while the engine is idling roughly. FPR vacuum can be toast, but as it snakes below the intake manifold, you'll not be able to spray there, which is also sight of other vacuum leak posibilities. Oil filler cap gasket? Brake vacuum assist? Idle control valve? Loose clamps on intake boots? Micro-cracks in same?

Sorry for the dumb and slightly off topic question but what's the connection between an (intake) vacume leak and a crank case leak(dipstick o-ring's) ?

The reason i ask is because i use an oil catch tank rather than venting the fumes back into the intake below's and was wondering if I was doing some thing wronge ?

(This picture was taken before i tidied the wiring up)

http://homepage.ntlworld.com/a.law47/Pictures/eng%20bay%202.gif

Cheer's

Andy

[Bill R.]

bmws and any openings such as the orings on the dipstick will create an unregulated vacum leak that will have an effect on the idle., In a case like yours where you have installed a mass airflow sensor instead of the afm I would guess that you're probably using a hotwire mass air flow sensor, if thats the case then if you check on most cars equipped with hotwire sensor they have the tube from the valve cover downstream of the mass air flow sensor, this is to keep any oil mist from contaminating that hot wire sensor... If it were mine I would move the sensor closer to the air filter and then connect the hose from the valve cover as far downstream as I could get it, maybe a couple of inches in front of the throttle body....












Sorry for the dumb and slightly off topic question but what's the connection between an (intake) vacume leak and a crank case leak(dipstick o-ring's) ?

The reason i ask is because i use an oil catch tank rather than venting the fumes back into the intake below's and was wondering if I was doing some thing wronge ?

(This picture was taken before i tidied the wiring up)

http://homepage.ntlworld.com/a.law47/Pictures/eng%20bay%202.gif

Cheer's

Andy

[Warren N.CA]

When I mentioned the dipstick O-rings, I should have also noted that you should go with OEM O-rings: I put some 50 cent fatter hardware store neoprene O-rings on mine, and it sealed up perfectly. However, after a few weeks these O-rings sorta got fried in place, making the dipstick very hard to pull out.

I have not checked yet, but there is probably another O-ring or similar seal at the bottome of the dipstick tube, where it fits into the oil pan. That will need attention/replacement too, as well as cleanup of collected crud down there. BTW, if there IS oil leakage at the base of the dipstick tube, then air can leak IN via the same path the oil leaks OUT. Fix it.

Also, on the oil filler cap: I used a hardware store neoprene washer, which was exactly the right size. I first removed the stock washer, then put on the new one, then put the stock one back on top of that. That way, the stock washer is still mated with the valve cover lip, and won't get fried on, but has the new washer behind as a shim for a tighter seal.

Valve cover: Yes, I know it's supposed to take a dry gasket, without sealant. But, it was leaking, so I put a thin layer of gasket sealant on top and bottom surfaces of the old gasket, replaced the valve cover, and re-tightened to spec torque. The excess sealant oozed out the sides, so I waited overnight 'til it congealed, then trimmed excess with a razor to make it look purty.

Replaced your FPR vacuum tube, and similar tubes, yet? Mine LOOKED fine, but was literally toasted through in places not easily seen, so leaked bigtime. That meant the FPR was not doing its job properly, so injectors had wrong pressure and air/fuel mixture was off = poor performance.

BTW, the prior owner and BMW dealer (who is very well paid to know better) had totally neglected all of this easy fix. So, the car idled and accelerated poorly, the prior owner despaired of even bigger BMW dealer repair bills, so fire-saled the car for $3,250. New O-rings, vacuum tubes, and gasket sealant cost me ~$6.50 to remedy these problems. If you can think of a cheaper way to get more bang for your maintenance buck, let me know.

[Warren N.CA]

I should know sooner if the problem persists.

Last night I got caught in creeping traffic and was amazed at how smooth the car ran in 1st gear, creeping along, under 5mph at 800 RPM- didn't need to use the clutch at all, as long as the car in front didn't come to a dead stop.

I might have fixed the problem "temporarily" just by jostling things around. In that case, it will probably show up again, likely at the most inconvenient time.



something has happened outside of the lambda control range...by this I mean that the dme monitors the O2 sensor and is able to do some adjusting to the fuel mixture within a narrow range to correct various deficiencies such as richening it up slightly to correct a small vacum leak or leaning the mixture out if its getting too much fuel... The range that it can correct if fairly narrow, it can't work miracles in fuel systems control... So anyway that code comes up if for some reason its gone lean or rich outside of the range of control, which would indicate a fairly good sized vacum leak or its running rich outside the range of control, a bad diaphragm on the fuel pressure regulator allowing fuel to be sucked into the intake manifold through the vacum line going to the fuel pressure regulator. I doubt that the gook on your dipstick orings was enough to trigger this code...
To clear the code you can do the stomp test and read the codes then after the codes are done flashing hold the gas pedal to the floor for 10 seconds or so. then shut the key off and repeat the stomp test to see if it cleared the code, if not you can try again. On some cars that won't work so then you have the option of disconnecting the battery for a couple of minutes or waiting until after you have done 60 starts on the car and it will clear itself as long as the original cause of the code has been fixed... So if you start the car 3 times a day thats 20 days and the code clears itself if the problem has been fixed.

[632 Regal]

Valve cover: Yes, I know it's supposed to take a dry gasket, without sealant. But, it was leaking, so I put a thin layer of gasket sealant on top and bottom surfaces of the old gasket, replaced the valve cover, and re-tightened to spec torque. The excess sealant oozed out the sides, so I waited overnight 'til it congealed, then trimmed excess with a razor to make it look purty.

the reason not to use stuff on gaskets is that stuff that oozed outside the gasket also oozed inside the engine. as the heat and oil contact this stuff some usually ends up in the oil pan and in the pickup screen and never breaks down small enough to be trapped in the filter. altho I saw a rather extreame example of this it ended up frying the bearings, since i saw that I wont use that crap anywhere it has the chance of entering the oul pan.

[Jazzman]

You are right. Code 1222 does not necessarily means vacuum leak. On my 740iL with the same problem, it ended up being my in-tank fuel pump. The pump is ready to quit and it is not pushing enough fuel to the system that it leaned out enough to cause the fault. Check your fuel pump.

[Bill R.]

engines last so long today is due to the crankcase ventilation system which vacums any combustion byproducts that get past the rings into the crankcase out and into the intake manifold where they are burned in the engine and out the exhaust... This reduces harmful acids, moisture and sludge buildup in the engine... By removing this and having only a catch tank you're actually shortening the life of the engine...






Sorry for the dumb and slightly off topic question but what's the connection between an (intake) vacume leak and a crank case leak(dipstick o-ring's) ?

The reason i ask is because i use an oil catch tank rather than venting the fumes back into the intake below's and was wondering if I was doing some thing wronge ?

(This picture was taken before i tidied the wiring up)

http://homepage.ntlworld.com/a.law47/Pictures/eng%20bay%202.gif

Cheer's

Andy

[Unregistered]

Jeff, you're certainly right about keeping crap outta the oil pan, etc., and normally I'd go with a new gasket. But, here in East Bumphuk we ain't go no stinkin' BMW gaskets at Billy Bobs' Parts 'n Bait shop. So, until I can get a fresh new one from BMA, I used the tube of stuff that's worked flawlessly on my other cars for the past couple of decades: After careful study, I've not yet seen it to crumble off and go elsewhere, so I'm gambling that the risk of errant sealant is less than that of leaky gaskets, for the next little while at least.

Warren, if you jostled the dipstick, etc. and "temporarily" fixed the problem, then you're probably getting pretty warm as so its source(s). In all likelihood, you have SEVERAL small leaks which collectively skew the electronic input to the fuel injection. So, keep after the basic stuff as mentioned previously.

The best advice I ever got regarding vacuum leaks came from a service guy at the Nissan dealer where I got parts for my '78 280Z. That car had L or J Jettronic Bosch fuel injection, so was essentially similar to our cars. After years of chasing hesitation, stumbling, erratic idle, sluggish performance, etc. , this guy said "90% of the time it's a vacuum leak." So, I used to carry a razor blade in the glove box, and trim 1/4" off the end of offending vacuum tubes, then just push them back on the barbed vacuum fitting. (I was amazed that a leaky tube for the heater control could also cause trouble with engine idle, but it did.) Trimming the tube tip off worked every time, cuz the leak is usually where the stretched rubber tube splits at the barbed nipple. Not always, though, so just buy several feet of new vacuum tubing at Billy Bobs and spend tomorrow afternoon checking things out. BTW, these tubes are pretty good as stethoscopes: Stick one end you ear and hold the other end at various spots that may be leaky--you can hear if it leaks. Intake manifold, brake vacuum assist, window seals, etc. etc.. Keep some coiled up in your tool kit or with the spare tire for such purposes.

[535ise]

bmws and any openings such as the orings on the dipstick will create an unregulated vacum leak that will have an effect on the idle., In a case like yours where you have installed a mass airflow sensor instead of the afm I would guess that you're probably using a hotwire mass air flow sensor, if thats the case then if you check on most cars equipped with hotwire sensor they have the tube from the valve cover downstream of the mass air flow sensor, this is to keep any oil mist from contaminating that hot wire sensor... If it were mine I would move the sensor closer to the air filter and then connect the hose from the valve cover as far downstream as I could get it, maybe a couple of inches in front of the throttle body....

Thank's for that Bill. :) So the connection between the vacume leak and crank case leak is that it will change the level of vacume in the inlet manifold and will have a knock on affect as the fuel reg will be getting a false/wronge reading which affect's the A/F ratio on a standard set up ?

I thought the crank case was at a slight pressure rather than a vacume because of the blow past from the piston ring's ?

So i'd better change the oil more often to counter the blow by contaminant's ?

I'm useing a stand alone programable ECU that uses a throtle poteniometer(sp?) rather than an MAF or AFM.The ECU was mapped with the oil catch tank fitted and run's well but is the fuel reg going to be causing me any problem's ?

The inlet piping has changed a bit since the last picture.The standard elbow pipe from the air filter has been replaced with an 100 mm Samco elbow.The 75 mm black pipe has been replaced with a home made glass fiber pipe that start's at 100 mm to connect up to the new elbow and then taper's down to aproximatly 75 mm at the TB end.This has then been covered with foam pipe tape to hope fully insulate it from heat.

This was done to help improve the air flow from the air filter and into the TB.As you can see from the picture the wiring is still untidy and i've got to get two BIG hose clip's for the new elbow.

http://img12.imageshack.us/img12/6310/intake.jpg

Sorry for rabiting on.

Thank's

Andy

[PTClark/'95-530i-5spd]

So this topic came up a year or two ago on this message board -- and the guy was losing his wits, so he brings the car to some mad scientists (okay, not mad scientists, but think-outside-the-box guys nonetheless)... Their solution? -- mounted a smoke machine onto the intake/airbox.... it worked. -- the smoke leaked out of some crevice that could not be seen from visual inspection - ...would make for a great patent if they could develop a universal mount for something like that...
Best,
Peter in CT
'95 530i/5spd

[Unregistered]

Warren, I recant and repent: Instead of gasket sealant on the valve cover gasket, use good quality electrical tape along the joint. That oughta work pretty good, and won't get sucked down into the oil strainer. Might use it around the dipstick interface, or similar applications.

I'm also considering using some good clear or black 3M silicone bathtub caulk to seal around the brake vacuum, etc. Won't start the engine until AFTER the caulk has fully set up, though, to keep it from being sucked into the innards, and to prevent its chemical out-gassing (if any) from damaging the O2 sensor.

Anywhere you find factory one-shot compression clamps (like those at the PS pump reservoir, which always seem to leak), replace them with SS marine grade clamps that don't cheese-grate the rubber hose.

[Warren N.CA]

vacuum line I could find, looking for something amiss. So, I haven't really isolated it to the dipstick area. Now I need the car to cooperate by going into the failure mode again (Or else decide that it's fixed.).

Thanks for the additional tips. I may just start replacing hoses, if it happens again. I'll be clearing the code soon, so I can start fresh.

BTW, I carefully torqued the valve cover over its new gasket, to the specified value, about 2 weeks ago, after valve adjustment.



Jeff, you're certainly right about keeping crap outta the oil pan, etc., and normally I'd go with a new gasket. But, here in East Bumphuk we ain't go no stinkin' BMW gaskets at Billy Bobs' Parts 'n Bait shop. So, until I can get a fresh new one from BMA, I used the tube of stuff that's worked flawlessly on my other cars for the past couple of decades: After careful study, I've not yet seen it to crumble off and go elsewhere, so I'm gambling that the risk of errant sealant is less than that of leaky gaskets, for the next little while at least.

Warren, if you jostled the dipstick, etc. and "temporarily" fixed the problem, then you're probably getting pretty warm as so its source(s). In all likelihood, you have SEVERAL small leaks which collectively skew the electronic input to the fuel injection. So, keep after the basic stuff as mentioned previously.

The best advice I ever got regarding vacuum leaks came from a service guy at the Nissan dealer where I got parts for my '78 280Z. That car had L or J Jettronic Bosch fuel injection, so was essentially similar to our cars. After years of chasing hesitation, stumbling, erratic idle, sluggish performance, etc. , this guy said "90% of the time it's a vacuum leak." So, I used to carry a razor blade in the glove box, and trim 1/4" off the end of offending vacuum tubes, then just push them back on the barbed vacuum fitting. (I was amazed that a leaky tube for the heater control could also cause trouble with engine idle, but it did.) Trimming the tube tip off worked every time, cuz the leak is usually where the stretched rubber tube splits at the barbed nipple. Not always, though, so just buy several feet of new vacuum tubing at Billy Bobs and spend tomorrow afternoon checking things out. BTW, these tubes are pretty good as stethoscopes: Stick one end you ear and hold the other end at various spots that may be leaky--you can hear if it leaks. Intake manifold, brake vacuum assist, window seals, etc. etc.. Keep some coiled up in your tool kit or with the spare tire for such purposes.

[MBXB]

Unlit Propane Torch
A 'Vette buddy uses an unlit propane torch and passes it near every suspected area as the engine idles. When the rpms increase, you've found a leak.
He claims to be able to isolate every one.