Cloudsurfer

Well, here's what I have to report back. Tried the easy fix of swapping the check valve out on this side. No joy. Then to dig a bit further, realized that since the vane adjusters have a locking pin to lock the cams at their fully retarded position when no oil pressure is present, that I could crank the engine with the valve out and observe oil flow. So, I pulled the fuel pump relay and all 6 coil pack connectors and cranked the engine, first with the Bank 1 valve out, then with the Bank 2 valve out. The Bank 1 valve (the "problem" side) pukes barely a dribble of oil, while the Bank 2 valve gushes out (which is what you'd expect). So, clearly this entire problem is being caused by an oil supply issue to the solenoid valve.

For reference, so we can all learn, the drain from the solenoid valve simply goes through a hole bored into the head that lets oil run back under the area where the chains run. Nothing fancy there. The oil feed hole in the head is a pretty sizable hole, as are the holes up to to the vane cell.

So now, this brings up several possibilities. Given the size of the oil passages in these things, I find it extremely unlikely that anything could be clogging them up, especially considering that 80PSI oil has been flowing through there, any debris would have been pushed through and would be in the finger screen/ check valve. This brings up some far more concerning possibilities:

1. Bad case. Oil feeds up from the bearing carrier, through the case halves, and then into the head. If there was a crack or other damage to the oil passage, it would bleed pressure off.

2. Bad head. See #1.

3. Head gasket. Incompatibility between 3.2 Boxster S case half, 3.6 Carrera head and 3.8 997 head gasket.

4. Assembly error. I'm not really sure what I could have screwed up, but it's possible that something isn't right.

I originally exchanged my 3.6 case halves for what I was told were a set of 3.4 cases that had already been sleeved by LN Engineering, which turns out to be a pair of 2000 3.2 Boxster S halves. I asked Charles at LN multiple times if there would be any compatibility issues running 3.6 heads with these case halves and was told "no." Due to the engine's 99mm bores, a 997 3.8 head gasket must be used.

The compatibility problem already turned out to be not quite so simple, as I had to modify the case to accept the newer AOS (the older style has a fitting on the AOS itself, which fits into a boss on the case half, where the newer style has a hose nipple that connects a small hose to another nipple on the case, so I had to machine a hose nipple and install it in the case) as well as the fact that the chain housings are a bit different where the heads bolt. On one side I had to drill and tap one hold into the case, while on the other side there was simply no way to use the 4th 6mm bolt. I do not know if these differences are the extent of what separates a 3.4 (or in actuality 3.2) case from a 3.6 case. Given all that, it would seem entirely plausible that the oil feed from an older VarioCam case is different from a newer VarioCam Plus case and that on this one bank, the holes between the case and head do not line up 100%, and for all I know the head gasket could be part of that problem as well.

Unfortunately, the only thing I have at my disposal to tear down and examine is a 2004 3.2 Boxster S motor, so I can't really compare all these different possibilities.

All said, I'm really left scratching my head and dreaming of all the nasty possibilities that I don't want to be the cause of this, and I'm not at all excited to have to pull the engine out of the car and tear things apart. I'd appreciate any insight into this that anyone can share, and to date, certainly appreciate all the help I've received thus far.

Is this a Boxster - if so you posted in the Carrera section.

M96.21 67Y is a Model Year 2000 Boxster S engine. 3.2 liter 252 HP.

Interesting. This was supposed to be a 3.4.

M96/ 2167Y06741

Also on case: 29770

Much appreciated!

Does anyone know how different (if at all) the heads from a 3.2 vane cell (04 model year) boxster engine (vane cell variocam, but no "plus" valve lift control) are from the 3.6 heads?. I have the blown 3.2 engine in the stand and am thinking of tearing down the head and doing some diagnostic work to figure out the flow of oil in these things to solve this problem, as I REALLY don't want to tear apart the head on my 3.8, esp with the engine in the car.

Even better, perhaps, does anyone have a 3.6 head laying around? It can be damaged, I just need to study the flow of oil through the head.

The more I think about this, I think it may be more likely that the drain is plugged up, not the supply side, but we'll see soon enough here.

I would not be so quick to dismiss the check vavle in bank one. The info that I read describes that the function of the check valve is to assist/speed up the actuation when oil pressure is not enough to overcome the force if the vavle springs, etc.. I will try find the description and post it.

George-

Thank you SO much for posting that PDF. I am going to re-check that check valve, and try swapping it out for one taken from another motor.

Does anyone know how different (if at all) the heads from a 3.2 vane cell (04 model year) boxster engine (vane cell variocam, but no "plus" valve lift control) are from the 3.6 heads?. I have the blown 3.2 engine in the stand and am thinking of tearing down the head and doing some diagnostic work to figure out the flow of oil in these things to solve this problem, as I REALLY don't want to tear apart the head on my 3.8, esp with the engine in the car.

Even better, perhaps, does anyone have a 3.6 head laying around? It can be damaged, I just need to study the flow of oil through the head.

The more I think about this, I think it may be more likely that the drain is plugged up, not the supply side, but we'll see soon enough here.

I have a dead 3.6 head around -- PM or email me mike@lonestarrpm.com and I'll figure a way to get it to you.

Mike

PM sent!

To service the clutch or change out torque converters, you save a fair amount of time (and money) by just pulling the transmission. Does not matter if you are on a lift or jack stands, pulling the entire assembly is more involved and is usually not necessary unless there are complicating factors like all wheel drive units or a need to perform other work that becomes easier with the engine out……….

The only one I've done was a C4, so that was the only way of doing it.... just doesn't sound like much fun to split a Tip in the car....

Verdict: New motor.

The IMS bearing came apart and filled the pan with shredded metal and balls. The weird thing is that it died quietly.

The new engine is coming from a distribution center in Los Angeles. Is there a way to know if the reman has the improved IMS setup?

I have looked at many blogs re the IMS failures, it appears that 10% of these engines fail within 90km miles, many unhappy comments from victims.

My insurance is from CNA National,they have been upfront and professional but I doubt they are interested in extending the warranty (5 months left, whew). Anybody have an insurance co they like?

Thanks again,

-Zook

The "new" IMS is identifiable via a much larger retaining nut (22mm I believe). Unfortunately, there is no way to upgrade this bearing to the LN part, as the OD of the bearing is larger than the case opening, so short of splitting the cases you can't do anything to replace it. LN does, I believe, recommend removing the rear seal of the bearing to allow for better lubrication of the bearing.

Unfortunately, while considerably more robust than the earlier single row, there are still documented cases of the larger bearing failing.

At the end of the day, there are much more important things in life to stress about. If you can't live with the whatever small percent chance that this bearing may fail again, sell the car.

I don't know how much the stock battery weighs, but I know its more than 14 lbs.

BatCap 800

Albert

I think I'd be more comfortable with an Oddyssey or Braille.

Don't forget you'll need to either flash the car over to RoW software or simply live with a CEL being on due to cat efficiency codes.....

I added full power seats to a car that came with partial power seats. As long as the full power seats you purchased have the memory controller on them, I can help you wire them up. You won't have memory functionality, but all adjustments will work.

This may actually be easier to just drop the engine engine and gearbox as a unit and then separate out of the car. I've only done this on a C4, where that is the only way to do it, but keep in mind, you need to pin the torque converter to the flexplate so that the torque converter remains with the engine, otherwise you'll have one hell of a mess as the converter pulls away from the engine. To do this, you need to pin the converter through an opening in the bellhousing, and then you have to access the 6 (if I remember) bolts through the starter opening, which obviously means removing the starter. I'm sure it's possible to do this with the lump in the car, but it may be easier to just drop the whole thing.

I think that's one of the differences between the 7.8 and 7.8X DME's and their associated maps. I think you'd have a difficult (if not impossible) time controlling this in your car, as you don't have a DME output to that vacuum solenoid.

I understand that in theory, between plastic tanks and all the filtration that fuel goes through before it goes into the car's tank, that there should be almost no need for a filter, and that if a filter is present in the car, that there should be so little debris caught by it that it should last forever, but I still don't buy it. I've seen filters off earlier Boxsters and 996s that have a replaceable filter, and have seen them get clogged.

Sometime when I have a broken fuel pump from a non-return car I'd love to tear it down and see just how much crud is in there.....

Does anyone know how different (if at all) the heads from a 3.2 vane cell (04 model year) boxster engine (vane cell variocam, but no "plus" valve lift control) are from the 3.6 heads?. I have the blown 3.2 engine in the stand and am thinking of tearing down the head and doing some diagnostic work to figure out the flow of oil in these things to solve this problem, as I REALLY don't want to tear apart the head on my 3.8, esp with the engine in the car.

Even better, perhaps, does anyone have a 3.6 head laying around? It can be damaged, I just need to study the flow of oil through the head.

The more I think about this, I think it may be more likely that the drain is plugged up, not the supply side, but we'll see soon enough here.

Todd-

Thank you, that is extremely helpful. So, after looking at this schematic, since this is a hydraulic system, that has the issue of performing slowly at low RPMs, I would think the problem could be either in the supply or drain side, as either one would cause too slow of a movement at the vane cell adjuster. Is it reasonable to suspect that the non-return valve on one side is not functioning, and that is why this issue presents itself only at low RPMs? Where exactly is that valve? I know there is a check valve inside the the finger screen that sits directly behind the solenoid valve, is this the non return valve in the diagram?

I suppose the only evidence working against the culprit being the non-return valve, is that this problem actually seems to get better as the engine warms, and the oil thins, so that would point away from the non-return valve being the issue.

Back to the idea of using air pressure to blow out these passages, that seems relatively easy to do to blow out the supply side (P) by blowing air back down that way from the valve port, but to blow out the returns, I imagine you'd have to blow air up from the sump?

That is going to be complicated…………only way I can envision is to push oil from a pressurized container thru the system from the oil pump side with each actuator alternately removed, as you obviously cannot do this with the engine running. Another possibility would be to push compressed air backwards through the system; I'm trying to think of any downsides to doing this (like how the pressure control valve and spring in the oil pump would respond), and cannot think of any problems pushing air backwards might cause, but I cannot think of anything problematic. Before you assembled the engine, did you blow out the passage ways? Sometimes machine work and block cleaning leaves small bits of trash in these passage ways……… There is also the possibility you may find that one bank has a leaking oil passage way that is bleeding off the oil pressure on one side, but usually that shows up somewhere else, like oil in the coolant.

I'm afraid that none of these "Easter Egg hunt" diagnostics are going to be easy or quick………..

This is exactly what I was thinking last night. Thing is, I'm not entirely sure what the exact flow schematic looks like as to how oil gets from the oil pump, up through the case half, through the head, through the lifter assembly, and into that solenoid valve. Everything was meticulously cleaned during assembly, so the only thing that seems plausible now is that either something, I'd guess likely a small piece of extra sealant, broke away and is now plugging the supply to the valve, as I just can't see anything plugging the small bit of plumbing from the valve to the vane cell adjuster, or there's something wrong with the actual vane cell adjuster and/or cam (which I still think is unlikely, but hey, stranger things have happened). I'd ideally like to find some scientific way to measure both pressure and flow volume at the solenoid supply and compare this side to side (4-6 works fine, 1-3 is the issue, so if there's a significant difference in pressure or flow that would indicate the problem- of course next would be solving it). What's maybe easier, would be blowing air backwards through it, as if anything was plugging the passage it should get freed up, though who knows if said piece of debris would wind up in the sump and get picked up by the filter, or get pushed back into the same place it was and re-plug the passage, or maybe worse plug something else.

What's perhaps most frustrating in my mind, is that I never ran this engine before it came apart to get sleeved, so for all I know either a case half or head simply has a bad passage and is bleeding pressure off internally. For the record, there is no intermix going on, and I'm inclined to agree that if it was a bad passage somewhere that bled off oil would likely end up in the coolant, so perhaps that's somewhat reassuring.

Hypothetically, would any harm come from running the engine with one of the valves removed at a time? The cams simply default to zero advance without pressure, so I can't see anything bad coming from that (shoudln't be remotely possible to punch a piston to a valve). I had thought of pulling the spark plugs and just rolling the engine with the starter with one valve removed at a time and compare readings.

Does it seem likely that if something is plugging the supply side that blowing air back down it backwards would free it up? That seems like the quickest thing to try, especially since I did discover (for everyone's future reference) that these valves do, in fact, fit through the valve cover, so pulling them out literally takes 2 minutes (in my case, since I'm running B&B Caymans S headers the O2 sensor is right in front of the phasing valve, so must be removed). If you did go down this road, would it maybe be wise to pull the pressure relief from the oil pump housing first? Ideally, I'd like to get whatever is plugging this up (if that really is the problem) out of the engine, not have it floating around to get stuck somewhere else worse (like a piston squirter).

I have been having issues with drivability and after diagnosing with the PST, I have a "laggy" valve advance on Bank 1-3, where it would take considerably longer than Bank 4-6 to advance, and while it was "taking it's time" to advance on acceleration the car would cough and sputter. Today i changed the valve phasing hydraulic solenoid out, and while it was out, checked the finger screen behind the valve for any debris and found none. With the new valve in place, the problem is exactly the same, so I am left scratching my head a bit as to the cause of this.

Since the valve was replaced and the symptoms are the same, I feel it's safe to rule out a bad hydraulic valve, which was my first suspicion. It would not seem to be an electrical issue, as the solenoid is clearly getting the command to advance or retard from the DME, just doing it too slowly. That leads me to think that this must be an oil pressure/ flow issue somewhere, either before the solenoid valve, or after the valve in the actual vane cell adjuster. When I pull the valve out it is full of oil, so clearly oil is getting to the valve, and as to how to measure the actual pressure there, I really don't even know where to begin on how to do that. On the other end, the vane cell adjuster just hydraulically floats, and takes position from wherever oil pressure pushes it to, so I can't really figure what could go wrong in there.

What is surprising, is that the car does not throw fault codes for this. One would think that since the DME knows what degree of cam advance or retard it wants, and also knows that there is a difference in values from one bank to the other that this would throw codes, but it does not.

I'm really stumped on this one, so any and all ideas are welcome.

Knowing that this is a recently assembled engine, are you sure there isn't anything obstructing oil flow (I have seen problems when people put too much sealant on surfaces and it ends up where it does not belong over time)? These actuators need both pressure and a level of flow volume to work correctly...what kind of oil pressure does the engine have?

I've completely thought of this as well. Oil pressure is good, 4-5 BAR, the question is, how can I verify sufficient pressure and flow at the hydraulic solenoid port, which would be the oil entering the valve?

I have been having issues with drivability and after diagnosing with the PST, I have a "laggy" valve advance on Bank 1-3, where it would take considerably longer than Bank 4-6 to advance, and while it was "taking it's time" to advance on acceleration the car would cough and sputter. Today i changed the valve phasing hydraulic solenoid out, and while it was out, checked the finger screen behind the valve for any debris and found none. With the new valve in place, the problem is exactly the same, so I am left scratching my head a bit as to the cause of this.

Since the valve was replaced and the symptoms are the same, I feel it's safe to rule out a bad hydraulic valve, which was my first suspicion. It would not seem to be an electrical issue, as the solenoid is clearly getting the command to advance or retard from the DME, just doing it too slowly. That leads me to think that this must be an oil pressure/ flow issue somewhere, either before the solenoid valve, or after the valve in the actual vane cell adjuster. When I pull the valve out it is full of oil, so clearly oil is getting to the valve, and as to how to measure the actual pressure there, I really don't even know where to begin on how to do that. On the other end, the vane cell adjuster just hydraulically floats, and takes position from wherever oil pressure pushes it to, so I can't really figure what could go wrong in there.

What is surprising, is that the car does not throw fault codes for this. One would think that since the DME knows what degree of cam advance or retard it wants, and also knows that there is a difference in values from one bank to the other that this would throw codes, but it does not.

I'm really stumped on this one, so any and all ideas are welcome.

One of the simplest things is to swap the stock seats out. The spare tire is an easy thing to lose. After audio, carpet, and small things like that, the process gets expensive in a hurry.

phillipj & cloudsurfer,

Thanks for your responses. The car is at an indepent repair shop and all signs are that they know their stuff. They told the insurance co it needed a new engine, but an agent has to verify this.

It amazes me Porsche would design this shaft with one end supported without a bearing, and the other a sealed bearing awash in oil.

I hope not, but if it turns out to be fixable should I install the British made ceramic bearing?

Thanks,

-Zook

The ceramic bearing is available from LN Engineering. If the motor is savable, that is absolutely what you want in there.

As to the design, it's been debated everywhere, but the long and short of it is that the company that brought you the 996 and Boxster is not the same company who brought you the 930. The modern day Porsche is run by the accountants, just like almost every other profitable, publicly traded company.

To further entertain you, consider the fact that these engines share the same head castings for both banks (again, to save cost), and this is why the IMS shaft has chains at both ends, vs all on one side (like all previous Porsche flat 6s).

The 03 Boxster S that I have, which I have stuffed an LN based X51 3.8 into, had the IMS bearing shear its retaining bolt. When I separated the gearbox and engine, the end of the bolt (with nut still on) was sitting in the bellhousing. Shockingly, the cams are still in time and it doesn't appear that any valve to piston contact occurred, so at some point I'll likely go through the engine and re-build it. Many others I've seen have had more disastrous ends.

From the sound of it, yours may be saveable. Pull the flange off the IMS, and inspect the bearing. Hypothetically, all the metal that was created should be inside the tube, and NOT elsewhere in the engine. If there is metal throughout the engine, well that's game over. If not, see if you can remove the outer bearing race from the IMS shaft, without damaging the shaft (if it's not already damaged). If you can't do this, it's game over. There is no way to remove the IMS shaft without tearing the engine down. If you can cleanly remove the outer bearing race, put the engine at TDC (using the front pulley and fixing pin), and then check the cam timing by pulling the green seals in the heads. Visually, you can see if all 4 cams are lined up straight, as the marks should be vertical. You don't necessarily need the timing tools to at least eyeball this. If all the above is good, hypothetically the engine could be saved by installing a new, upgraded bearing from LN, and verifying (with the timing tools) that the engine is in time.

As to a new motor from Porsche, dealers do not rebuild these. They get a reman motor from Porsche. Depending on luck, you may get a new motor, or it may be a reman. I'm not sure there's really any difference as to the internals, as they will have the newer IMS. You can tell a reman motor by the presence of "AT" in the serial #.

For those that don't know, the IMS shaft is geared at a 1:2 reduction from the crank and drives the cams and oil pump (at the front of the engine). It has nothing to do with balancing the engine. Unfortunately, in the M96/M97 engines it's a horrible design, with the front of the shaft supported by the oil pump housing (with no bearing) and the rear (which you see the flange of under the back of the crank where the flywheel attaches) sitting in a "sealed" ball bearing. Contrary to everyone's belief, this bearing does NOT receive pressure lubrication. It was designed to contain sealed bearing grease, but what happens is that engine oil washes the sealed lubricant away, and this is why the recommendation of more frequent oil changes does hold some water. The bearing in any motor after 05 is MUCH larger and uses a much larger bolt and nut to retain it to the flange. With the Series II cars that use the A96 engine, the IMS shaft was eliminated entirely.

You mean other than the fact that it is illegal and contributes to excessive air pollution.

If you lose the cats, you either need to flash the car to RoW software or you'll run into all kinds of CELs for cat efficiency and secondary O2 readings.

Well, as far as I can tell, the part has superseded to the newer #, which is shared with a CGT (oddly enough). Should have it in in a week or so to solve this little issue.

Thanks for all the input gentleman.

996.105.303.05 is for a 996 engine

996.105.303.06 is only for the Carrera GT(according to where used in PET)

I was told that it superseded to the 303 06 part for a 996?

Well, for all those interested, there is no way to test the camshaft solenoids on a VarioCam Plus engine (though I suspect there is a way to do this on a VarioCam car), as I would guess the difference lies in the all or nothing advance of the early system versus the continuously variable nature of the later system.

What I did do tonight was hook up the Durametric and log actual values for MAF readings, expected and actual cam values for both banks, and O2 voltage for both banks while driving the car around with an "assistant" to monitor. The evidence supports a bad hydraulic valve on the one bank. The problem improves to the point where it's almost not a drivability issue as the motor and oil warms, but the one banks cam readings lag significantly when the DME commands a large change and lags less when a small amount of advance is needed, which is further verified by O2 reading differences when this happens and the car coughs. The MAF is working fine.

My hypothesis is that the valve has not fully failed, but is failing, as it improves as things warm up. Further, since the car runs great from 3k up, I suspect that the difference in oil pressure (more pressure as the RPMs rise) allows the valve to work properly at higher RPMs, but not at lower.

Since one side is working fine, and one is not, would it be prudent to just replace both (which I'm really kicking myself in the *** for not just doing when I built this thing)?

Can someone verify that the part # for the valve is 996 105 303 06? Just want to make sure I get the cam adjustment valve and not the valve lift valve, as the PET doesn't make it very clear which is which. Thankfully, the quoted part is only ~$120, but this is not going to be fun to swap out.