[teamgomez]

Howdy all-

I found a busted exhaust hub tab during a valve adjustment last weekend and going through the learning process/parts acquisition to replace the hub and rebuild the VANOS.

First off, I must commend Lama at Biesan's for picking up where her brother Raj left off after his passing. I don't know where we'd all be if she threw in the towel and closed up shop. So many thanks to give along with condolences for his passing. The Roundel community lost a good one.

Next, I've watched many DIY's along with ECS' guide for rebuild and nobody locks the cams down during the VANOS housing installation. Sure, the cams get checked for placement with the bridge tool, but why in the h*** doesn't anybody just insert two pins in the bridge to keep the cams from moving? I've seen quite a few failed attempts with guys having to do it all over again (or worst case bent some valves) b/c the hub bolts were too tight and shifted the cams. So what am I missing here? I received a bridge tool today and by the grace of the Man on the top floor, it came with two pins which I plan to leave inserted unless I'm missing something here.

Lastly, setting torque on the hub bolts to 10.5 ft-lbs is always noted as doing by feel as there isn't enough room for a torque wrench. Well, having spent a couple seasons working on my son's CR125 Honda shifter kart engines, I picked one of these up along the way:

So it is possible to set torque on those bolts and while my German torque method usually works out fine (Gudentite), there are some places where I'd prefer to be spot-on and the bolts for the cam hubs is one of them.

Parts won't get here until the weekend so getting to spend some time cleaning up the dissy'd parts along with chasing an oil seep (picked up this 97k mile M3 vert a couple weeks ago and giving it some much needed and in some cases overdue attention). Once I get the Blackstone Labs report back, I'll either slide in some new rod bearings or just roll the bones on a new oil pan gasket to fix the weep. I'm doing the CPV o-ring as well but was surprised by the crud built up on the harmonic balancer pointer/front lower pax side of the engine. Hoping it's not the front crank seal (I loath pulling harmonic balancers...) and it's 'just' an oil pan leak. Aside from the oil seep, the engine was pretty damned spotless (it was a SoCal car its whole life and 2 owner).

Here's a shot of the new high heels the mistress got to replace the factory 19's. This is my 3rd set of Apex wheels and all the credit goes to Ryan at Apex. He's a sharp guy who really helped me with each set that went on another Roundel in the stable alongside a flat six from Stuttgart. A Koni Sport shock setup with Eibach Pro springs/Dinan Camber Plates are sitting on the workbench and will go in as well.

So as long as I don't overlook a torque value or leave out the often overlooked fetzer valve, the VANOS on this garage queen should outlast yours truly .

Fronts: 18x9's w/ PS4S 245/40/18 w/ front camber maxed out (pins removed)
Rears: 18x10's w/ PS4S 275/35/18's

 

[Sapote]

Sure, the cams get checked for placement with the bridge tool, but why in the h*** doesn't anybody just insert two pins in the bridge to keep the cams from moving?

When torqueing the hub bolts, if the cam turning then so what -- the timing at this point was already set, so if the cam moves slightly under the torque then it's OK. After set the cams to TDC then the pin should be pulled out, or you might accidentally turn the cam or crank and bend the pin.

Lastly, setting torque on the hub bolts to 10.5 ft-lbs is always noted as doing by feel as there isn't enough room for a torque wrench.

That's if you follow Beisan instruction; I don't for this reason.
Once I snugged up the 3 upper hub bolts on both hubs, remove the vanos module off the head and disconnect the LH threaded shafts, you can use the normal torque wrench to torque all 12 bolts to spec, not guessing by feel.
The other thing I deviate from the Beisan instruction: I insert the splined shafts into the hubs, then pretension the hub bolts, then install the vanos and connect the pistons to the LH threaded splined shafts. This is easier than install the vanos with splined shafts together into the hubs.

In fact, I don't even use the bridge/pin tool as I'm cheap; I used a selected drill bit for the pin and a carpenter square and a caliper to set the cams timing. Note: the pin on the cam is not perfectly 90* to the head surface, and this is why I used the caliper to set the drill bit (pin) slightly offset from the true vertical.

 

[teamgomez]

Thanks for the pointers, but this one spun a bearing in my brain housing unit:

Note: the pin on the cam is not perfectly 90* to the head surface, and this is why I used the caliper to set the drill bit (pin) slightly offset from the true vertical.

I, too, was considering a drill bit/carpenter square but this is the first I've read that the cam pin-holes are not 90 deg to the head when fully retarded. I thought the whole point of the bridge tool was to ensure the cam pin-holes were 90 deg off the deck plane.

WRT the cam timing, I'm referring to the CAUTION note on pg 92 of ECS' Dual Vanos Rebuild :

The hub bolts have to be loose so they can mate with the splines and I thought this is where most of the timing errors I read about occurred. Though I do see your point about inserting the splined shafts separately. Gonna chew on that one.

 

[Sapote]

I, too, was considering a drill bit/carpenter square but this is the first I've read that the cam pin-holes are not 90 deg to the head when fully retarded. I thought the whole point of the bridge tool was to ensure the cam pin-holes were 90 deg off the deck plane.

You can verify this easily:
Ex cam slightly advanced
In cam slightly retarded relatively to true vertical line.

 

[Sapote]

I will calculate exactly the small angles deviate from the true vertical line later. These data were based on the original cam timing I measured before removing the hubs.
If you get the factory bridge tool, could you help measuring the cam pin angles relative to the steel carpenter square placed on top of the 2 cams (I believe the 2 cams form a plane parallel to the head top surface). To measure the angles, use indirect method: use a fitted drill bit or rod that not loose in the cam hole, then measure the gap between the drill bit and the square edge at certain height from the cam -- says at 4" high, the gap is 0.100" retarded side or advanced side of the square edge.

 

[teamgomez]

I did some hunting and the only data I found on the degrees off of 90 is from the Kommen tool description that says
"For setting camshaft lead angle (1.5 degree) (control diagram) when installing Vanos adjustment unit on S50B31. 2 Liter and S54 Motorsport Engine."

I did buy the Kommen tool (came with 2 pins) and while it appears to be well built, it mics out at 125mm at the base and tips of the pins when inserted so it doesn't appear to have any 'angle' built into it (photo has induced parallax due to wide angle lens...pins are parallel).

I bit the bullet and just ordered an OEM bridge and will remove this question mark from the equation. I will happily do some math to get the degrees of ret/adv on the cams after setting with the OEM Bridge tool. I thought about testing the Kommen tool on the cams b/f disturbing the timing but given your observation I believe the tool I bought is not accurate.

Speaking of some math, I have seen numerous comments about the little German dude at the whiteboard who lives in the DME taking care of 'timing the cams' so a bit of slop can be compensated for. One poster said 'up to 5 degrees off' can be accommodated.

So if the tolerance for gap between base of bridge opposite pin and head lip is less than 1mm, this equates to less than .35 degrees of error when you take the arc tangent of 1/165 (1mm gap divided by distance from pin to opposite base of tool). So just guessing that the DME won't soak up 5 deg of timing error....that's a huge number and I believe the valves and pistons would be engaged in marital relations at that point.

 

[Sapote]

Maybe they meant 5 deg crank, not cam angle, as 5 deg cam timing error would cause trouble and DME won’t be able to tolerate it.

 

[teamgomez]

A couple thoughts after more homework:

1. It looks like lots of folks get VANOS P-codes after start which sends some down the INPA rabbit hole. I don't speak INPA at the moment and hope I don't get forced to buy the Rosetta Stone translate kit (though I do have it on the laptop with the OBDII/USB adapter...just never dove in). But it begs the question about why the codes in the first place if timing is spot-on. Would it be because nobody puts the cams back how they found them b/f going 'full retard'? It seems like the DME will flag the cams on start if they're left at full retard and if we put 'em back where we found 'em (learned that in the Boy Scouts...and at the end of my Mom's fly swatter...) after torquing the hub fasteners , then the DME shouldn't start throwing codes after first start.

2. If cams are properly set with the bridge tool and crank is pinned, then my current understanding of the only way to booger the timing (outside of chain slack/slop/a few thou of mesh slop here and there) is to allow the pistons to extend (or slide aft) during hub pre-load/tighten/torque. So with the VANOS unit on the bench and fully assembled, why not press the transfer gear/pistons all the way forward/in then pop the cap off and mic the amount of piston extending from the housing so you know the 'full fwd' measurement? Then after installing the unit to the engine, pop the cap off (maybe leave the o-ring out and only snug a couple cap fasteners down) and observe pistons while torquing the hubs. If the piston extension distance remains constant, then the cams should not move and timing should be dead-nuts-on (chain slack/etc/aside...cheating a couple degrees on the crank setting or individual cams a diff discussion). At least if you run into timing issues, you can rule out piston creep during torque set.

And 5 deg at the crank is still 2.5 deg at the cams; will hunt for a book answer to DME timing tolerance. Never want to anger the German dude at the whiteboard in the DME...his intolerable cousin lives in the SMG programming and he's the last guy I want to piss off.

When we're talking a degree here and there, its becoming evident that it is imperceptible to the eyeball. The 54 mm pin thru the bridge would only be deflected .9 mm at the tip vice the base (54 mm pin with 35 mm extended thru the bottom of the bridge shaft so tan1.5 x 35) if the bridge tooling was designed (and milled) for 1.5 deg of cam adv (EX) or ret (IN). I will compare the Kommen bridge tool with the OEM once I get it in 4 days. The folks at Kommen said they've been selling theirs for years and nobody has had any problems (reported anyhow). I'll be removing that question from the equation here shortly.

Appreciate the help.

 

[Sapote]

1. It looks like lots of folks get VANOS P-codes after start which sends some down the INPA rabbit hole. I don't speak INPA at the moment and hope I don't get forced to buy the Rosetta Stone translate kit (though I do have it on the laptop with the OBDII/USB adapter...just never dove in). But it begs the question about why the codes in the first place if timing is spot-on. Would it be because nobody puts the cams back how they found them b/f going 'full retard'? It seems like the DME will flag the cams on start if they're left at full retard and if we put 'em back where we found 'em (learned that in the Boy Scouts...and at the end of my Mom's fly swatter...) after torquing the hub fasteners , then the DME shouldn't start throwing codes after first start.

Never happened to me, and I set both cams at max retarded after done the timing. Engine fired right up with no codes. Engine when shut off might park the cams at the optimized positions, but it also is OK when they are set at max retarded.

2. If cams are properly set with the bridge tool and crank is pinned, then my current understanding of the only way to booger the timing (outside of chain slack/slop/a few thou of mesh slop here and there) is to allow the pistons to extend (or slide aft) during hub pre-load/tighten/torque. So with the VANOS unit on the bench and fully assembled, why not press the transfer gear/pistons all the way forward/in then pop the cap off and mic the amount of piston extending from the housing so you know the 'full fwd' measurement? Then after installing the unit to the engine, pop the cap off (maybe leave the o-ring out and only snug a couple cap fasteners down) and observe pistons while torquing the hubs. If the piston extension distance remains constant, then the cams should not move and timing should be dead-nuts-on (chain slack/etc/aside...cheating a couple degrees on the crank setting or individual cams a diff discussion). At least if you run into timing issues, you can rule out piston creep during torque set.

People might screwed up the timing for some reasons:
a. turn the crank CCW to nail it at TDC, causing chain slask.
b. the pistons might not be at the max forward (retarded) when bolting down the vanos; but this is the rare case as with the solenoid plate removed, there is no pressure to resist the pistons. This could happened when the vanos was not torqued to the head evenly -- cock eye and jammed the pistons.
c. the pretension bolts are set not correct, so when the vanos bolted down did not push on the hubs hard enough to take out all of the slack/plays.
d. the chain pulling side was slack with the crank at TDC, leading to error after timing.

 

[Sapote]

I will calculate exactly the small angles deviate from the true vertical line later. These data were based on the original cam timing I measured before removing the hubs.
If you get the factory bridge tool, could you help measuring the cam pin angles relative to the steel carpenter square placed on top of the 2 cams (I believe the 2 cams form a plane parallel to the head top surface). To measure the angles, use indirect method: use a fitted drill bit or rod that not loose in the cam hole, then measure the gap between the drill bit and the square edge at certain height from the cam -- says at 4" high, the gap is 0.100" retarded side or advanced side of the square edge.

From my stock engine, I measured the cam timing angle based on the cam timing hole and the head top surface:
EX: 96* (6* advanced to the head vertical line)
IN: 89.3* (0.7* retarded to the head vertical line).

It's a mystery why couldn't BMW drill the cam timing hole to true 90* to the head top surface.

 

[teamgomez]

What condition are these angles? i.e. where they sit after engine turns off or are these the TDC 'timed' angles? A bit confused as the bridge tool (Kommen) is as close to 90 deg for both as I can measure. OEM tool arrives tomm.

I measured mine where they sat after engine shutdown and the EX was 20 deg adv, and IN was 90 deg.

I was heads down under the hood yesterday and made a few mistakes along with learning a few things not covered in any of the videos I saw.

1. Using the Beisan Transfer Gear tool to install anti-rattle kit, I let the socket slip under me and the weight of my SnapOn 1/2" electric gun and bent the exhaust shaft about 5 deg. I'm sure there's more than one shop who'd bend it back and call it good, but not so sure I could bend it to 90 perfection nor do I know any of the metallurgy to understand if the metal can hack it. Thoughts b/f I spring for a new one? The worst part is there wasn't any perceptible axial play and I was putting in the anti rattle kit as a 'while you're there' thing. But the old adage 'if it ain't broke, don't fix it' comes to mind and I wouldn't be in this predicament if I stuck to that sage wisdom...

2. Retarding the cams. I thought that after 'breaking it free' by rocking back and forth that the cam would rotate somewhat freely. Was having a hard time getting it to move (and found out that my 24mm open end is actually 25.2mm...damned Pittsburg garbage...) until I noticed a squirt out of the vanos solenoid mating orifice while rocking. Turns out the cam was rotating, but only by about a degree or two at a time. Who'da'thunk?

3. The electrical connector to the solenoid didn't want to disengage...using a bit of a 'persuader' only shattered it so now need to track one down or safety wire in place. Dammit!

Not every landing is perfect and we only learn by the bad ones, right? Old enough to know better on some of these missteps but we'll get 'er done...

 

[Sapote]

1. Using the Beisan Transfer Gear tool to install anti-rattle kit, I let the socket slip under me and the weight of my SnapOn 1/2" electric gun and bent the exhaust shaft about 5 deg. I'm sure there's more than one shop who'd bend it back and call it good, but not so sure I could bend it to 90 perfection nor do I know any of the metallurgy to understand if the metal can hack it. Thoughts b/f I spring for a new one? The worst part is there wasn't any perceptible axial play and I was putting in the anti rattle kit as a 'while you're there' thing. But the old adage 'if it ain't broke, don't fix it' comes to mind and I wouldn't be in this predicament if I stuck to that sage wisdom...

More headache for nothing. There is no reason to mess with the roller bearing if there is no perceptible axial plays. Even with free new parts, to lap the bearing spacer to achieve zero axial play is not a trivia job, and most of the result of doing anti-rattle turned out worst than before, is my conclusion. For the bent shaft, I would get a new one for peace of mind instead of nightmare of a ruined head with bent valves due to a broken shaft and cam out of timing.

 

[Sapote]

What condition are these angles? i.e. where they sit after engine turns off or are these the TDC 'timed' angles? A bit confused as the bridge tool (Kommen) is as close to 90 deg for both as I can measure. OEM tool arrives tomm.

I measured mine where they sat after engine shutdown and the EX was 20 deg adv, and IN was 90 deg.

Cam timing after engine shut off are not parked precise, as the DME has no good information as the crank slowing down. I consider these angles closer to random data, so no, my angle data are for when the cams are at TDC compression stroke.

IN was 90* advanced also? Not clear to me.

 

[Sapote]

2. Retarding the cams. I thought that after 'breaking it free' by rocking back and forth that the cam would rotate somewhat freely. Was having a hard time getting it to move (and found out that my 24mm open end is actually 25.2mm...damned Pittsburg garbage...) until I noticed a squirt out of the vanos solenoid mating orifice while rocking. Turns out the cam was rotating, but only by about a degree or two at a time. Who'da'thunk?

Why did you want to rotate the cams (using the 24mm wrench)? To set them at timing at TDC? Did you remove or loosen the solenoid plate?
I don't follow Beisan instruction. The best way to set the Cams timing at TDC for the purpose of vanos timing is to remove the hubs bolts, get the vanos and splined shafts off, then it is easy to turn the cams to perfect timing at TDC.

 

[teamgomez]

...The best way to set the Cams timing at TDC for the purpose of vanos timing is to remove the hubs bolts, get the vanos and splined shafts off, then it is easy to turn the cams to perfect timing at TDC.

Well, that makes perfect sense. Both ECS and BS say to put the cams at TDC b/f pulling VANOS unit off (solenoid plate already removed so when trying to rotate the cam, you're working against the oil in the VANOS head). I figured the steps were in that order for a reason, I guess they aren't :-(

IN was 90* advanced also? Not clear to me.

Sorry. The IN cam was 90 deg off the head, or 0 deg RET or ADV (give or take a few tenths).

I can't get my head around your EX timing settings as 6 deg ADV is well off the bridge tool that I have (Kommen) and will get the OEM one tomorrow to double check.

As for the transfer gear shaft, headache noted and unless I can scrounge one up, looks like Dr V is the only one I can find that sells the whole piece. Am I missing a supplier? And FWIW- looks like that damned piece is cast. I barely leaned on it to see if I could even get it close and it made up my mind for me...

Oh yeah- I did grab the ECS VANOS o-ring kit and one of them comes up well short of the plate hole diameter (circled in red). Going to bolt it up later this week after I get my hands on a new transfer shaft and see if causes any grief...not sure if the PSI will create any issues here.

Thank for the assist!

 

[Sapote]

As for the transfer gear shaft, headache noted and unless I can scrounge one up, looks like Dr V is the only one I can find that sells the whole piece. Am I missing a supplier?

I would look on eBay for a good used unit.

2001-2006 BMW S54 E46 M3 Z4 Intake Exhaust Vanos Unit Hub Splined Shaft 3.2 Gear | eBay

Find many great new & used options and get the best deals for 2001-2006 BMW S54 E46 M3 Z4 Intake Exhaust Vanos Unit Hub Splined Shaft 3.2 Gear at the best online prices at eBay! Free shipping for many products!

www.ebay.com

 

[Sapote]

I found a busted exhaust hub tab during a valve adjustment last weekend and going through the learning process/parts acquisition to replace the hub and rebuild the VANOS.

This is a simple case: I would just get a hub to replace the broken hub. Done, nothing else to be replaced but to remove all 12 cam sleeve bolts, clean all threads on bolts and holes, then add a drop of red Loctite.
I would also have the vanos pump disk drilled with 2 new smaller holes for less plays on the EX hub tabs to avoid breaking again. I continue to use the old o-ring seals on the solenoid plate, if there is no leak. Why open new can of worms?

 

[wsybert]

That sucks on the gear shaft. I just finished doing mine last weekend. Took my time and it went smoothly. I read the Beisan procedure a few times and watched several videos to understand the process in detail before I stared.
Ended up doing it all... anti rattle, seals, pump disk, solenoid, upper chain guide, and bolts.
I did notice very slight axial play in the gear shaft assembly. Also, when I disassembled it, there was noticeable wear on the ID of the sleeve. I didn't have to lap any parts. When I torqued and checked it all, no play and spun freely. My car as 57k on it.
Good luck!

 

[teamgomez]

Got the OEM cam bridge tool yesterday and quick gander reveals the Kommen tool is inferior to the OEM. The pin is a very snug fit in the bridge holes on the OEM versus a sloppy fit on the Kommen that allows for pin tip deflection when fully inserted. Quick swag also shows OEM has ~2 deg adv on the EX side and no discernible diff on the IN side. I'll take some snaps/better measurements later this week after some work turbulence smooths out.

Also got helped out by Chris at DrV. He got a new shaft in the mail yesterday. Really helped out!

 

[Sapote]

I did notice very slight axial play in the gear shaft assembly. Also, when I disassembled it, there was noticeable wear on the ID of the sleeve.

What sleeve -- the ID of the spacer that set the bearing axial plays?