2000 C2 Intermix > help needed in diagnosis

Hi, I recently purchased a 996 Carrera 2 manual. Engine has been rebuilt by a reputable rebuilder in the Netherlands 5 years ago / 3.5kkm. PPI was inspected and apart from a few items, they found nothing wrong. After driving the car home however, I found oil in the coolant. (No obvious signs of mixing the other way round). I asked the company that did the PPI and they indicated they found nothing. Symptoms - Performed a coolant pressure test, and it drops about 1 psi in 30 to 45 minutes when cold, when warm it drops a bit faster. - Oil in Coolant, not the other way round. Also... online I saw quite a few pictures with the coolant looking like mayonaise, nothing like that in my case, just a bit of oil on top of the coolant and after having run for a while, the oil on top of the coolant looks like mayonaise, but it doesn't look like the whole system is contaminated. What I already did Based on this information I asked around on Facebook and to a few specialists. The advice I got was to; 1. Remove the head cover / spark plug tubes and inspect the holes for cracks; did find a few hairlines in cylinder 4, but very hard to figure out whether these are cracks or scratches. The other holes looked fine 2. Pressure test the oil / coolant heat exchanger; did so and no leaks. Next steps? At this point I'm wondering what the best next stage of diagnosis would be. - A faint hypothesis I have is that the oil might be a leftover from the previous rebuild (in which case the valve seat of cylinder 3 was dropped). But I do have slight pressure loss. Just wonder if 1psi drop every 30-45 mins when cold, is problematic, or normal? - Should I just drop the engine and visually inspect the heads/get them pressurised, or are there any other ways for narrowing down the problem? Thanks in advance for your help!

That picture of the spark plug hole looks more like a crack. Clean it up with brake cleaner, wipe if off really good, and run your finger nail across the lines to see if that is a crack or just stain.

If you want to test your hypothesis that the oil in the reservoir is just leftover from the rebuild then you could drain the coolant and refill with fresh coolant to see if the oil comes back. I'm doubtful that is the case though. Those marks in the cylinder 4 spark plug hole look like scratches to me but hard to tell. Typically the first place the head cracks is on bank 1 (cylinders 1-3) following a water pump impeller blade breaking off and getting lodged in the coolant passageway causing a hot spot. If it were me I would pull the valve cover off bank 1 to inspect for cracks.

Quotes source post: 20282756

Well, imho that type of slight drop your seeing in not endemic of a cooling system leak. But. thats on most other cars. The type of head crack the 996.1 gets in its heads oil gallery passage ways that only leak oil in the coolant system not coolant in oil, are very small. However I'm not sure just a 1ppsi drop over 45 minutes denotes that type of leak. If it were me, I'd flush the coolant system and collect the old coolant and inspect it to get an idea of how much oil is really in it. Depending on what those results are, refill coolant system with mostly distilled water & maybe 1-2 gallons of coolant (depending on where you live and its current temperature) for now and see what happens...? 2) OR,,, call Len Hoffman the 996 head guru thats expert on fixing cracked 996.1 heads, see what he says... I'd do both.

Performed another pressure test overnight, coolant is mostly out of the engine, so it was just air in the system. Pressurised it to 16psi, checked about 20 hours later and was down to 2psi. I think this definitely indicates a leak(?) Next step is top remove the valve covers & camshafts. Would you do this with the engine in the car or outside of it? In theory I should have enough space, and the engine is dismantled quite a bit already. Just wondering what you'd do? I guess one of the advantages that I'd have with the engine still in place is that I could still pressurise the whole coolant system and perhaps see if anything comes out of the crack(s)?

I'm still not sure if the rate of pressure loss is 'normal' or 'exceptional'/indicative of a crack. 1psi an hour is not a whole lot, and I guess a system can leak. If anyone can offer some perspective/experience on this that'd be great. My brain is going between just tearing out the engine, removing the heads and getting them checked vs. running some more diagnosis and perhaps assembling it to see how it holds. Things that indicate the engine might be fine; - No mayonaise intermix, there's just oil floating on top of the coolant. It's not like the whole system looks like mayonaise. Perhaps a remnant of the old rebuild - No visible cracks from the spark plugs, other than cilinder 4, which are in an odd place (usually they crack towards the exhaust valve) Things that concern me - There was still oil in the coolant and it was quite a bit. - There is pressure loss when testing (just not sure if its problematic) For now I'm thinking along the lines of what @allcool suggested; vacuum fill the system with some cheap distilled water & a little bit of coolant, pressurise it again and leave it overnight. I've already drained the engine oil, so I could remove the sump plug and see If water comes out. Just wondering if that'd be enough to be observed?

I would not condemn 1psi loss in an hour. The fitting at the coolant tank cap can easily do that. If this were my car, I would drain it well, clean the coolant system, fill it back up and drive it some more to see if this is a phantom. Lacking a positive failure point, I would hesitate to fix anything.

You are just kidding yourself......the oil in the coolant is a big problem and probably not from a previous rebuild...( unless the previous rebuild is due to an Intermix as well ) 1PSi per hour is a substantial leak....Think of one of your tires leaking 1 PSI per hour....it would be flat overnight .....Although 1 psi per hour could be leaking from anywhere, like your test equipment ( I'm pretty sure you didn't soap test your equipment to see if it air tight)...it could also be from the Intermix point .....Testing procedures and accuracy are important.. Although you probably don't want to hear this... but the intermix is most likely a crack in the bank 1 head.....My advice is to remove the valve cover and cams and then pressure test and use soapy water spraying at the spark plug holes...

Before you pulled vacuum did you drain the existing coolant into a clean bucket(s) and determine how much oil was actually there? Did you bubble test your vacuum tester? That is a tough spot to get a solid vacuum link without taking off the hinge. Check all of your hose clamps. I've seen minimal loss on my vacuum lift gauge, but I never left it over night. I would first see how much oil was there and how much comes back after 100-200 miles.

I never said it was NOT a problem. I just said it could be your test rig and that a positive ID of the leak is needed before I would fix anything. My Airlift kit can leak 1psi per hour at the coolant cap. I never test it more than 15 minutes.

Quotes source posts: 20286941, 20286467

I did do that. I reckon about 0.5 to 1L of oil, but hard to assess as it's floating on top of the coolant. Is was clearly separated from the coolant, not the mayonaise like stuff I've seen on here. Tho I'm not sure if that makes any difference. You might be right, I might be kidding myself, I mean, I sure want this to be an easy fix But obviously I'm realistic enough to know that wishful thinking is not a proper solution. ​​​​​​The previous rebuild was with the PO (5 years back / 3.5kkm) and because of catastrophic engine failure; a dropped valve seat in bank 1-3. The cylinder head was replaced as a result, together with having all cilinders re-lined and all bearings replaced (done by a reputable engine builder in the Netherlands and proven by an invoice/pictures). I guess intermix could've been present, but I don't know for sure. 1psi is something, but, on the other hand I really don't know what's normal. This is my first time having an issue like this (rebuilt a few engines before tho). I guess a system can always leak a bit of air here and there, that's why I was wondering what others with more experience in this area, have measured before. I did actually perform a 'calibration' test on my equipment by pressurising a blank plug that I 3d printed and screwed in the tester itself, that showed no pressure loss. I'm contemplating the following steps; 1. 3d print another set of adapters & blanking plugs that allow me to isolate the engine from the rest of the the cooling system and pressure test that only. 2. If the above yields the same sort of pressure loss, I will remove the valve covers and visually inspect for bubbles / cracks. I'm contemplating doing so with the engine still in the car, but would like to hear your thoughts on that?

Okay, so not a few drops. A quart is roughly 3/4 inch thick layer of oil in a 5 gallon bucket. If it were me, and it's not. I would fully drain, even the front rad lines. Refill with some proportion of antifreeze and water. Go and drive it. You're not making a milkshake with the lubrication system. Give it a few hundred miles and let it sit overnight. Drain it off and see if you have a clear separation line. If you do then it's time for deeper inspection. It sucks if this was just rebuilt a few thousand miles ago. But, I wonder if they actually did the heads.

When you have a catastrophic failure such as a valve seat dropping of a rod breaking that can also cause "Intemix" the oil and water mix will happen so fast and the engine will stop so fast that any " Intermix" will remain in the engine compartment and will be eliminated when the engine is removed...

Thanks all for the advise and helpful perspectives. The engine is already disassembled quite a bit, so I'm leaning towards biting the bullet, dropping the engine, taking the cam covers off and inspecting the heads. There's a slight risk that ill find nothing, assemble it and still don't know what caused it, but it's the most thorough approach. Before I do this however, I will perform one last pressure test; I have designed and am currently 3d printing a set of adapters that allow me to plug the engines cooling hoses. I've also desgined an adapter that allows me to hook up my pressure tester to the engine cooling system. That way I can disconnect all coolant hoses from the 'chassis' (have to do this anyway to drop the engine) and test the engine in isolation. These adapters will also help the engine is out and on the stand, as I can then still pressurise it, and hopefully see some coolant/bubbles come out of any potential cracks.

This morning I tested the pressure adapters. Most worked, but the one that I attached the pressure tester to had a slight leak. (air pressure dropped faster than I'd expect from such a small leak tho, but that's not a very scientific observation) I therefore decided to push ahead and lower the engine. Not a difficult job per se, just some work to do. Next step is to print a new adapter and set of camshaft locking tools. I will then take of the valve covers, cams and the rack that holds the valve lifters, pressurize the cooling system and look for leaks. To be continued. Heads definetly look recently rebuilt.

I wish I could print tools ...

Quotes source post: 20289097

Just buy a printer and wing it. I started 3d printing because of my 924 Rebuild; that thing needed so many plastic parts renewed... The businesscase for a printer was a no brainer. So anyhow, here's the tool I printed, and that allowed me to continue the disassembly. Hypothesis has been confirmed, there is a leak in the cylinder head. But in a bit of a weird way; the head seems to have been cracked before and repaired, the repair is a bit of a strange job and is leaking. I will contact the company that performed the repair, to see what they think of it. (That's also why I will refrain from posting pictures for now, some people know which company it was and it might reflect badly on them without the full story.)

Excellent Testing and the use of special tools to help in the process !!! Glad you finally found the Root Cause on the Intermix .... I'm very curious about this "unusual crack" ....most of the time these cracks follow one of two predictable patterns,,,, but this head had a valve seat drop, which itself is unusual for a 3.4 .. Valve seat failures are more common on the 2002 3.6... it this is curiously out-of-the-norm

Congratulations on finding the root cause. I am not sure what the "repair company" is going to do after 5 years. I suggest you just buy a head from the factory and complete the repair and move on with life.

Very few Shops can repair those head cracks successfully .....Being as the original head was destroyed beyond repair. according to your previous post.... The Shop must have sourced a used/repaired head for the engine .....Good Used uncracked bank 1 3.4 heads are almost non-existent nowadays ...

Quotes source post: 20290075

That's exactly what happened. A replacement head is also on the invoice. That head cracked in cylinder 3, going from the spark plug hole to the pressed in coolant plug. It was just repaired in a bit of a strange way and that repair failed again. If I were in the US, I'd send it to Hoffman without even thinking about it, but I'm in the Netherlands / Europe. I searched and could not directly find a shop that specializes in these repairs. If anyone is reading this from Europe and has experience with a good shop that can weld cylinder heads, please share.

I think these guys r only a short boat ride or chunnel away. Its where i would have a european motor repaired. https://www.hartech.org/porsche-engi...s/introduction

Quotes source post: 20290196

Thanks, I also considered them, physical distance is nothing. However, since UK is separated from the EU it'll be a hassle with import duties. Came across this company the other day. They seem legit; https://www.redhead-zylinderkopftech...chwei%C3%9Fen/ Might send them an email if I cannot find anything closer to home. ​​

R there boarder checks for entering the uk? Can u just drive it there and declare it both ways. Worse case is u may have to pay the vat fee on the repair? I just dont know how the uk and eu play. I would have to imagine there r a few german shops that can do it. I know Manthy will not work on the m96 motors anymore. By their web page it looks to be a compentant shop. Maybe there is a european/german facebook 911 group and get some 1st hand reviews.

If it is me, I would march down to the dealer, plunk down a credit card and pay $3500 US for a populated head, and spend 16 hours to install it into the car. You are halfway there with your engine out and free labor to boot. Really, that is cheap when it comes to Porsche ownership.

Quotes source post: 20290266

Not sure where the 3.5k comes from, I'd definitely consider that. Cheapest I can find is 6k https://teile.com/en/porsche-parts-s...linder-1-3/266

Rose passion is a very good suppliier. 99610400305

I was just reading Hoffman Head response to Riggard on Facebook about his head, they won't touch his head or any previously welded head at their shop. time to find a used or new head.

The price of the brand new 3.4 head has jumper up recently from ~$3k to ~$5K.........The price of Porsche parts never goes down as they get older like for Chevy or FoMoCo parts...

Last year, that was $3.5K at Redwood City Porsche.

I'll see what my options are. Right now I'm thinking about a few scenario's; First of all I've contacted the shop that performed the previous head repairs to ask for their opinion/advice on this. Although the work is out of warranty, I think the way this engine was assembled, is not how it's supposed to be done. So I'll see what they say. Perhaps we can come to an arrangement. If the above doesn't work out, I'll contact a few shops that do engine repairs and see if they have replacement heads and/or whether it's a possibility to trade this head in. (Consider purchasing a replacement engine for 4k, take the heads, get them rebuilt/repaired and sell the rest) If all the above fails... pay 6k for a Porsche head.

I thought I have heard somewhere that Slakker (or LNE) have used heads for their engines ....

This evening I continued to inspect the 4-6 cylinder head. I was really happy to see it looked immaculate. Only setback is that my torque wrench somehow quit working. Luckily I always test mine by manually 'clicking' the head on a low torque setting, so no bolts stripped (yet ;') ) I'm also in contact with a repair shop that might be able to provide me another cylinder head for the damaged side. So I'm seeing some light at the end of this tunnel.

Great work and good news regarding the replacement head! What is the lift table (?) you are using?

https://www.motea.com/nl/motorfiets-...zwart-114968-0 This is the table I'm using, it's a motorcycle lift, but should be able to carry 500kg and is very compact when not in use. A hydraulic table would be a bit better, but I simply don't have the space, and it's more expensive too.

Quotes source post: 20292966

Thank you!

Had contact with another company that's into 911 engine rebuilds. He stocked quite a few heads (some of them welded, but by a professional company and thoroughly inspected.). He's able to deliver a rebuilt head at about half the price of oem. Id have to hand mine in for the valves and all that. Have 2 more inquiries outstanding, but I'm tempted to go with this offer.

So that means he wants $3k for the rebuilt head ????

Quotes source post: 20295613

nah, I exaggerated a bit, more like 2,2k. Btw, I did not post the pics of the left head yet, but here we go, this is what I found in cylinder 1-3 head. The engine was pulled out by Shop A and then 'rebuilt' by another shop (B) about 3,5k /5y km ago. Total bill: 12k, of which 8,5k for the rebuild. I cannot quite assess which of the 2 has supplied this head, but in my honest opinion a botched up head like this should have never been anywhere near the car. Unfortunately it was 5y ago and by the previous owner, so taking legal steps is probably the road to nowhere. As you can see the head has been welded before, but in a crappy way, causing it to leak again. Furthermore; they seem to have cut backthe face of the cylinder head to the valve lifter rack for cylinder 1 back a bit further than 2-3 and to compensate... they just cut the hydraulic lifter rack in 2. Hard to keep faith in the specialists; the one doing the PPI didn't catch 1+ liter of oil in the coolant, and the one that did this rebuild �5y ago, installed a head like this. I guess I'm happy that I'm kind of able myself, just hope that when I spend some money on getting a new head, it'll be better. (Also appreciate all feedback/comments here!).

A great and intellectually stimulating thread. I hope it works out for the OP. My only question, given the op�s knowledge, why you didn�t view the car yourself before purchase? PPI�s are good but not worth the paper they�re written on, because there is no recourse if they get it wrong. With your knowledge and this forum/facebook, you�ll get there and get to enjoy that C2 - it looks good.

Quotes source post: 20296127

Boils down to basic human stupidity I guess; I inspected the car and found it to be in great condition. The only thing that I didn't thoroughly inspect was the coolant; I visually checked the reservoir without removing the cap, and it looked good. The coolant was pink (in my previous experiences the coolant would always be mayonaise/brownish). As the engine was rebuilt 3,5k ago by a reputable company and the general condition of the car was immaculate, I didn't expect any issues here and figured if anything would be wrong, the PPI would certainly catch it. Expensive assumption.

That is one of the two common places these heads will crack ....I see he welded the freeze plug hole completely shut ....lol.....the other is from the spark plug hole to the exhaust valve guide. ... It is a even harder place to fix... I'm wondering if the Old School "Pinning" might be a better fix process for these heads instead of welding ....I have seen a lot of high dollar $$$ aluminum heads fixed using the Pinning process.... BTW,, that's a pretty good pic of the area, but could you mark the crack ?? My old eyes can't quite make it out ...

Quotes source post: 20296149

The freeze plug being welded is something I see with other welding jobs too, so that wasn't immediately alarming to me. Pinning could well be an option, but this head is shot. There is no crack visible in the pictures, so your eyes are fine. The crack however propagates into the bolt hole to the right of the spark plug. That had been welded and tapped into, but the crack is still there. If all is right one of the pictures is a gif/movie and shows bubbles coming out.

Pulled the head this evening, cylinders look great, no scoring or whatsoever. One thing I don't quite get is the parts numbers for the heads; the part that is on the car is 996 104 139 1r and if I look that up online its like 1.5 to 2k. However, if I look up the PET, that number cannot be found and the head is refered to as 996 104 003 05, which is almost 6k. Anyone happens to know the reason?

The number 996 104 139 1r is a "casting number" of the raw casting that can be machined to fit the Bank 1 or Bank 2 after proper machining and threading for the chain tensioners and the " machine matched" cam cover ... I truly doubt it could actually be purchased in this "raw state' as I have never seen or heard of it being done....The only thing I have ever seen that be actually purchased is the bank specific head with numbers matching cam cover... There are a lot " ghost parts with prices" in the Porsche system that don't actually exist......ask me how I know.....lol....

Quotes source post: 20300395

Interesting... Would it be possible/reasonable, to convert/machine a 'bank 1 head to a bank 2 head,, or 2 to a 1..?

Quotes source post: 20301797

I checked on this many moons ago.....it is entirely possible, but was more reasonable to just source the correct Bank head, and I doubt that will ever change.....

The casting number makes sense! I also observed that they are quite similar, yet also different in a few area's. I can well see a 4-6 head being modified into a 1-3 head, but otoh, I guess having a crack repaired, is probably about the same amount, if not less work. But it provides some hope, might the 1-3 heads ever run out. Interestingly enough, if the casting is the same and the heads typically crack in head 1-3, it's less likely to be a casting thing, and more likely to be an environment variable causing the part to eventually crack. I havent heard/read much comments about what specific environment variables could cause this, but I'm going to give it some thought. Update about the project is that I took the head off, and the cylinders look great to me. (The first one is a full wet liner, the other 2 have been relined, all with iron sleeves) The head will be shipped this week, and hopefully I have a rebuilt spare returned soon. Another thing

Yeah, that's a weird welding job for sure. I wonder if the welded over freeze plug could cause the head to crack. Could it have cause a hotspot due to restricted flow?

If I remember right, the bank 2 runs hotter since the coolant goes there last after it has already being heated up by other parts of the engine.