Subject: Have you ever heard THIS SOUND inside your engine?
Nobody on Bimmerfest knows what is causing this noise and I can't use the
car until I figure out how to figure out what the noise is.
This only started after I did some work this weekend on the valve cover but
it sounds like someone left a socket inside the engine and it's bouncing
I'm pretty sure I didn't leave anything inside where this sounds like it's
in the intake manifold but it's hard to figure out where it's coming from
as it's intermittent.
It sounds like it's coming from the top of the engine somehow but it's hard
to place where.
You should not need a login to read that thread and download the videos.
Here's just an audio file:
When you download them, remove the *pdf which I added so that the MP4 would
look like a PDF to the forum upload GUI.
I can't imagine what is making that noise which is why I'm asking you for
how to debug that noise. Thanks!
The only time I've heard a similar noise is from crank throws hitting cam lobes.
It was when I rebuilt a 352 and reinstalled the spacer used for the old timing gear.
The new cast iron timing gear had the that spacer cast into the gear.
That allowed the cam to travel in the path of the throws.
It was intermittent until the cam bearings wiped.
Otherwise, your theory is as good as mine.
None of the links worked for me and my fiddling around with them. But, if it's
sounding like something's loose and damaging everything in its path, then
you'll have to get back in and re-do your valve cover work. Maybe a 3/8's to
1/4 ratchet reducer, short extension, hex bit, rolled-up tube of Threadlocker,
etc. got left behind. Even though it's a BMW, I can't imagine you accidently
and unknowingly created a problem...unless your valve cover work included
removing the PCV or slopping in RTV that clogged venting and the sound is
pressure blowing the seals.
I updated the thread on bimmerfest but I also have pictures for you below.
It's not the DISA because the DISA is intact and removed.
It's not the VANOS because I never touched the VANOS.
It's not something left in the valve area because I opened it up again.
There's something in the air intake plenum bouncing around.
But how can it bounce around?
And how can it get in the air intake plenum when I only opened up the VC?
Here are some pictures. But I don't know what the noise is.
I have a boroscope being delivered later this afternoon but that BMW M54
plenum is like a spider web.
Here are the pictures for you to see what the heck is bouncing around!
1. The DISA is the main suspect but it has all its parts!
2. I even fashioned a DISA replacement out of a piece of pine.
3. And I bolted that pine to the engine and the noise was STILL there.
4. Surprisingly the pine sealed as the engine idled fine without the DISA.
5. So I ripped off the valve cover again even though I just put it on.
6. All the bolts and washers and nuts and grommets were accounted for.
7. The valve cover underside doesn't show any nicks from anything bouncing.
8. Could a piece of the gasket have gotten from the valves to the intake?
9. I don't see any evidence of nicking on the valves or the metal inside.
10. The noise seems to be coming from the air intake plenum tubes.
I was pretty neat about buttoning things up because this is my very first
valve cover gasket job (disaster).
What I just don't get is this:
Q: How can anything get from the valve area into the air intake plenum?
What else can I look for?
(I may have to remove the plenum in which case I'll replace the CCV.)
You seem to know bimmers well. This is an E39 with the M54 engine.
I agree that the sound seems to be metal on metal.
Also the intake manifold vibrates when the sound is heard.
Maybe I broke something metal in the CCV way down below which then got
sucked into the intake? (I don't see how this could happen but *something*
is making that noise!)
The CCV, as you know, is immensely complex for a stinking PCV valve.
This is the CCV hose I broke (it's patched up now just fine temporarily):
But maybe I broke something deep down in the CCV itself, which does have
two hoses that go into the intake manifold. Here is the diagram.
The M54 CCV, as you know, is immensely complex for a stupid PCV valve:
Here are the parts placement for the CCV hoses:
It's true. I broke the hose where the air is going DOWN into the CCV.
That's it. I did nothing else. Just the VCG.
The VCG R&R steps were simple.
Hands: Remove passenger side cabin air cover large rounded tube
Hands: Remove oil filler cap
Hands: Disconeect oxygen sensor wires
Hands: Remove breather hose to the plastic cover at the CCV hose
Flathead: Remove four caps in plastic BMW engine cover
10mm socket: Remove bolts holding engine cover
Flathead: Pull up to open ignition coil retaining clips
Hands: Disconnect plugs on ignition coils
10mm socket: Remove two coil retaining bolts per coil
8mm socket: Remove center ground strap
Flathead: Unclip coil pack wire harness as a group
Hands: Mark and remove coil packs in order
8mm socket: Remove front and rear braided grounding straps
10mm deep socket: Remove 4 center stack bolts
10mm deep socket: Remove 11 bolts around outside of valve cover
Putty knife: Gently pry valve cover up
Hands: Remove VCG from cover by pulling straight up (not to the side!
Yup. I even recently bought the K-line/Dcan USB cable and downloaded and
ran the INPA software only last week, which as you may know, is the same
software the BMW dealer used when the E39 was born to diagnose the
software. I installed INPA/EDIABAS, which all the bimmer guys use. I also
have ISTA+ which is the current BMW dealer software - but it's not
installed on my Windows laptop because I don't have a newer model BMW.
All these 13 electronic control units were good when I checked for fun:
ECU: MS430DS0 (engine)
ECU: GS8604 (transmission)
ECU: DSC57 (traction)
ECU: LWS5_1B (steering)
ECU: IHKA39_5 (heating/ac)
ECU: MRS4 (airbag)
ECU: SZM38 (centerswitch)
ECU: IKI (cluster)
ECU: EWS3 (alarm)
ECU: LCM_III (lights)
ECU: ZKE3_GM1 (windows)
ECU: ZIS (audio)
ECU: RADIO (radio)
Yup. The VCG is an easy job, even on a bimmer (except for the diabolical
German connector on the upper CCV hose and one bolt in the back corner).
Yes. Definitely. Not a single part is missing. Not one is missing.
Here is a picture of them in the 36-egg Costco egg crate I use for bolts.
Spark plugs were not touched. VANOS was not touched. CCV was not touched
other than I broke the upper hose. Nothing was touched other than what was
needed to do the valve cover gasket.
Here are all the pieces taken out for example (which were hard as rocks):
I am probing with a 6-foot lit endoscopic USB camera as we speak but the
focus on the camera is really lousy.
Interesting observation. I hadn't thought of the chains.
There are two camshaft chains, one for the intake & one for the exhaust.
Certainly they are exposed.
Do you think I can safely "bump" the engine with the valve cover off and
the coils out?
I appreciate the advice because "something" is making that noise.
I can't figure out what - but it must be something - somewhere - somehow.
I could try that as I have the special waterpump wrench set like most
bimmer owners do (two pieces of long thin metal wrenches) which I used to
replace the cooling system long ago.
Like most bimmer owners, I have the same software that the BMW dealer has
which is described here:
Making sense of INPA, EDIABAS, NCSExpert, NCS Dummies, DIS/GT1, EasyDIS, & Progman
I just recently bought the #23 K-Line USB to OBD cable on Amazon.
I bought the cable because I went through a puddle and the transmission
gearshift under the floorboards got soaked so the transmission went into
safe mode (aka limp mode, aka trans failsafe mode). The switch that screws
up the transmission when it gets wet is #9 in this realoem diagram.
The only way to get the automatic ZF 5HP19 (5 speed, helical planetary,
model 19, A5S325Z 5-speed 325 newton meters ZF) steptronic transmission out
of safe mode (stuck forever in 3rd gear) was to clear the codes using
Debugging Trans Failsafe Prog - car won't shift into gear
As you probably know, that's the same software the BMW dealer used when the
E39 was born - and it was used up to about 2006 - but I also have ISTA+
which is what the BMW dealer uses today I think (but I haven't installed
Of course, it's half in German and half in English - but you know that.
The reason I was doing the VCG was not only the oil leak but I was getting
the classic lean condition codes that all bimmers get all the time.
I have a decent OBDII scanner in addition to the INPA BMW dealer software:
I have lots of the dealer software other than INPA but haven't used it yet:
I will check if there are codes with INPA, but there should be a lot since
it was running without the CCV hose for about 30 miles.
Yes. You are correct. There *is* a chance the sound was already there.
It couldn't have been there a long time - but I only hear it when I have
the hood open. I don't hear it with the hood closed. Certainly I don't hear
it when I'm in the cabin driving (I don't hear anything inside the car).
SO it might not even be related to the VCG job - but it still has to be
I hadn't thought of that, where I did recently install the BMW dealer
software of the era when my car was new called INPA which allows me to read
all the electronic control units and reset the codes for all modules (not
just OBDII modules):
Some of this BMW dealer software can also program each module but I didn't
run any of that, and it can reset something called the "adaptations",
again, which I didn't touch to my knowledge.
I should be able to get the spark advance out of the dealer software though
My OBDII reader shows spark at 6 degrees advanced at idle and then it jumps
to 20 and 30 degrees in the live data as I drove - where I don't actually
know what it's supposed to be. I don't even know where that data is.
I have heard spark knock before where it sounds like marbles under load.
This doesn't sound like marbles. It sounds more like one marble.
I hadn't thought of the chains.
I'll look more closely and try to turn them somehow.
I did try the USB endoscope today but I can't get it to focus on anything:
I also tried the classic flexible magnet tool for picking up objects:
I even tried a powerful shop vac - but this plenum is very complex:
Even with a tiny flexible hose as the vacuum hose I couldn't do much:
Sorry in advance, I couldn't resist with a discussion so on point. I don't
know why anyone buys or even puts up with German crap. How many decades,
models and makes does it take. Truth about BMW e46 M54 engine
Yes. I heard the noise initially but had to be somewhere.
That's when it also had no vacuum so I had both problems at the same time.
Then I fixed the vacuum (by patching the CCV hose) and the idle came back.
But the noise remained. I haven't driven it since.
The vibration, with my hand, does seem consistent with something bouncing
inside the intake manifold. I can't check now and I didn't check extensively,
but it seemed when the noise was heard, a small vibration was felt.
I agree. I'm grasping at straws. I can't see how the CCV sucked something up
either. The DISA is the most likely but it's not missing anything.
I'm ok with that premise because how could anything get in the intake
manifold and what would make it bounce around at idle anyway. But then,
I have to ask my self what is making that sporadic knocking sound?
I put all my tools away, and no nuts, washers, or bolts are missing, but
maybe "something" is in the deep nooks and crannies of the intake plenum.
I'll take another look today at least to knock that out - if it turns out
to be that, I'd be ecstatic - but I don't see why it would knock about.
I understand. Many things "sound" like they come from place 1 when they
are really coming from place 2. The engine is vibrating when it knocks though.
Ever so slightly.
150K. That's not a lot for an M54 engine.
I'll see if I can do that. I won't have a help handy for most of the day.
I'm pretty good with my tools, in that I always put them back.
I'm also good with nuts/bolts as I put them in the 36-egg egg carton.
And I always put back any bolt that I can even when the parts are off
so that I don't forget where it goes.
Given I was working there, and given that it's bouncing around, my first
assumption (after DISA pin) would be something is bouncing in the valve
area. I will see if I can turn the engine with help later in the day.
It took a while after I cleared the codes for the lean condition codes
to come back. I suspect the engine wouldn't idle well after I patched
the upper CCV hose if a valve were stuck open. And even so, it would
throw, I would think, lots of codes and cause an immediate fuel shutoff
after just one in 200 RPM (which is the fuel shutoff limit).
So I don't think so. But even if it did stick a valve open, would that
make a sporadic knocking sound?
Yes. I will look today with INPA (which is that BMW dealer software), and
I'll report back. Given I ran with the bad CCV hose and then cleared it
but ran some more, I'd expect a bunch of engine ECU emission codes, but
I'll check all the 13 electronic control units again.
No. I only cleared the 13 ECU's stored codes.
I saved the old and new files so I have it all.
A lot is in German but I didn't think it abnormal that there are some codes
as the "pending codes" in BMW ECUs is different than for OBD codes.
The OBD codes have basically 3 federal test procedure (FTP) drive cycles.
The BMW codes have 40 drive cycles. If a code happens within any of those
40 drive cycles, it never goes away. It only goes away if the code never
appears in all 40 drive cycles. So, if a code appears on drive cycle #1
and then it appears again on drive cycle #39, it remains in the Electronic
Control Unit module's memory.
I might install the current BMW dealer software because ista+- gives more
information on a code like the description, mileage when code happened,
frequency, .... basically the freeze frame.
INPA gives just the code but INPA works with all BMWs.
ISTA+- only works with the newer BMWs.
I did not do anything outside of the VCG job except break thing like
I broke off the rectangular metal 4-inch cover plate in the back over
the wires and hoses that snake around the engine. So many people break
that plate that they don't even cover it in any of the DIYs! It's
always gone - but this is the original valve cover so my plate was
original. It's a plate that everyone breaks when trying to remove it for
better access to the rear passenger side corner valve cover bolt where I
have the broken plate so it can't be that.
I also broke the CCV hose. And one small clip on the coil-wire combined
harness case. Surprisingly the SAS/SAP (secondary air system pump)
vacuum hose and pipe didn't crumble in my hands, but that's only because
it did long ago like it does on everyone - so it was replaced long ago.
So I did "just" what needed to be done to do just the valve cover gasket.
I'm going to do more work today and then put it back together.
Thanks for all the help because I'm at my wit's end and the engine is at
UPDATE for today...
I put the valve cover gasket back on and ran the engine for 30 miles with
the DISA blocked off by the pine block.
Even with the DISA bypassed by the pine block, the sporadic hard knocking
sound was heard but it seems to be *deeper* in the engine now that I'm told
it's the CCV doing something weird (i.e., I no longer think it's in the
So pretty much we've eliminated:
1. It's not in the valve area
2. It's not in the timing chains
3. It's not in the intake manifold
With Pinewood DISA bypass, since the electrical connector was just
dangling, I got an expected P1 code, which turned out to be:
P1512 DISA (Differentiated intake manifold) control circuit signal low
If anyone wants to make their own Pinewood DISA, I measured the dimensions:
Length = 15.0cm
Width = 7.5cm
Thickness = 1.8cm (1.6cm for the rear upper hole - where a better thickness
would have been ~1.5cm for the entire block)
Hole diameter = 1.9cm (mine is tapered because I used a cone drill bit)
Hole 1 centerpoint = 2.7cm from the left side horizontally, 1.9cm from the
Hole 2 centerpoint = 10.8cm from the left side horizontally, 5.4cm from the
Thanks again for your advice as I really do not know what the noise is yet.
Even though I don't know what it is, I've eliminated a few things.
The videos on the net of the E46 M54 sound are the exact same sound though.
Some concluded it was the CCV while others did not.
None explained how the CCV could be making that sound though.
But - this is the good news - if it's the CCV - then I can drive the car -
where the major risk is hydrolock (which would be fatal, of course).
A PCV overhaul is, unfortunately, a big job on the M54 - and not cheap -
but at three hundred dollars or so in parts, the engine won't be destroyed.
Thanks for bringing up the ICV again.
I *do* very much have an air leak as my codes are classic on the last drive
which are the same codes (save for the extra DISA codes) I had before the
1. P0171 System too lean bank 1
2. P0172 System too lean bank 2
3. P1083 Fuel control limit mixture too lean (bank 1 sensor 1
4. P1085 Fuel control limit mixture too lean (bank 2 sensor 1
5. P1512 DISA (Differentiated intake manifold) control circuit signal low
As you know the bimmer, there are MANY hoses which can leak, so what I
really need is a good smoke machine. A friend and I built one out of a can
and a glow plug where it burns glycerin to smoke and where we put a
compressor set to as low as it will go (~2 to ~4psi) to push the smoke.
But it doesn't generate enough smoke.
What all bimmer owners need is a smoke machine in the shop!
If you know of one in the hundred dollar range - I'd buy it!
I agree with you completely. I'm *sure* there are still air leaks. It's an
E39. There are always air leaks. Either where you say (which I don't
doubt), or somewhere else (like in the CCV hoses buried in the engine).
I did put back the DISA last night, using the FIPG as a gasket instead of
the $10 6-1/2cm o-ring, which I didn't have and you can't just easily get.
I figured I'd let the FIPG cure overnight so I'll clear the codes in an
hour or so and then see what comes back. What I really need is a smoke
It's amazing to me that something so simple doesn't easily exist at a cheap
enough price to be affordable to all shade tree mechanics.
There is metal inside the CCV but I agree, not much (a spring, I think).
The ICV makes more sense, I agree. Plus it's *easier* to access! :)
I will look up how to R&R the ICV. I don't have the gasket handy though, so
that's one reason for not wanting to take it off just yet.
The Pinewood DISA is just a diagnostic experiment to eliminate "rattle"
from the flapper valve. I agree, it's not the DISA so with or without the
wood block, it's still not the DISA. With the FIPG on the DISA round mating
surface now, there should not be any air leaks at the DISA.
So I've put back the DISA as it's not an issue (although EVERYONE always
says it's the DISA just like whenever you have a cooling system problem,
everyone always says it's the thermostat!). :)
Yup. Nasty DIY. Nobody likes it. Expensive too. Hundreds of bucks in parts.
All for a stinking PCV valve.
Funny you mention that because I didn't respond to the guy because I
thought the same thing. I actually tried but I can't even *see* the CCV and
I know where it is. It's in the middle, like the heart of the engine. It's
just not something I'm going to get my hand on while the engine is running.
I agree with you.
Sounds like Usenet, only everyone here is a curmudgeon! :)
Yikes. Lots of current when that touches ground.
I find the forums are iffy - but sometimes you get a good hit like the guy
who found the E46 M54 videos of the exact same clicking sound.
At least I know from that there is nothing left inside the engine from the
VCG R&R job. It's likely that the clicking sound came only after I fixed a
massive vacuum leak (the oil was smoking up the engine it was leaking so
The theory is that the CCV started working when I fixed that one massive
air leak. Of course, it could be that I broke the CCV when I shoved in the
repair hose also. Either way, the CCV is original so at 15 years, it's well
past time to replace it.
It's not. The electronics on this car are SUPER sensitive. Why risk it?
For a twist of the 10mm nut on the negative battery cable terminal, which
is nicely designed to access with a ratchet and a 10-inch extension, it's
just not worth the risk.
I've replaced the alternator and the two mechanical tensioners, and the
idler pulley, so I agree. I had to rip out a lot to get to it - but that
was a few years ago. It's hard to get the "B" terminal back on, as I
recall, because it has a lip protecting it, as I recall, but it's not
something I'd ever risk with the battery in.
I agree with you.
Thanks for sticking with me. At the moment, I'm compiling all the o rings
and hoses I have to get to do a CCV job. Most people say it takes them all
day so it will take me all weekend as I take lots of pictures and go slowly
and put all parts in their respective places and put all tools back each
time I use them so they're always in the same spot all the time.
I'm not those guys who ask and leave. I'll be back - but there may be no
news until the parts come in and I have a weekend to work on it. My biggest
danger now is the hydrolock potential (I'm not sure I understand how
The next thing I need to do is get the vacuum leak sorted out which I'm
sure is in a hose somewhere (maybe even the CCV hoses).
My biggest lament is that there should be a cheap (hundred dollar range)
effective smoke machine out there. All it has to do is blow thick smoke at
three psi which doesn't seem all that hard to do.
In article ,
Now that wasn't nice! ;)
We do have a mechanic that works for us at the plant that is very good
with a hammer and grease gun, maybe he'll suffice! ;)