996TT shake down
 

My new-to-me 996TT arrived yesterday, after shipment across the country, and let's just say we have a little research, learning, and fixing ahead of us. ;-)
I took it out for a road test, and though it drove well around town, experienced the following when I hit the highway and opened it up a bit:

• The engine gave off an interesting sound which occurs only when mashing the pedal hard. I don’t remember hearing this sound when I accelerated the vehicle the day of purchase when I first drove it at the end of Feb, and did a PPI (including road test) with a local indy. Granted neither the mechanic nor I drove it that fast on open roads that day. The sound is like either sheet metal fibrillating, or possibly (guessing here) a waste gate dumping? It made me wonder if the heat shield(s) diagnosed as loose during the PPI, and repaired by PPI indy prior to shipment, were still loose.
• The max boost I could produce on the highway at speeds (as displayed on the dash readout) while driving in the 60-110 mph range was 0.4 (rarely) or 0.3 (most of the time) when I felt safe pushing the throttle hard for several seconds.
• Coming off the highway decelerating somewhat hard from 110mph down to maybe 70mph I got a couple beeps and the alternating “PSM / ABS failure”, “drive to shop” messages flashing on the dash.
• A mile or so later, I got a separate CEL and “drive to shop” msg.

The car is 2001, all stock, and generally the PPI was clean. I put my Durametric on it, and cruised through all the screens, looking for fault codes, and key engine metrics. I found the following throughout the system:

• Engine:
  • P0103 / Porsche fault code 115 – Hot film MAF sensor
  • P1671 / Porsche fault code 407 – DME relay Porsche fault code
  • P1601 / Porsche fault code 660 – CAN timeout – instrument cluster
  • P1249 / Porsche fault code 231 – Boost pressure deviation Below limit
  • P1101 / Porsche fault code 594 – Input variables charge measurement
• Air Bag: Fault code 3 – supply voltage
• Litronic Headlights HBA:
  • Fault code 1 – Engine path left
  • Fault code 2 – Engine path right
  • Fault code 3 – Axle-position sensor front
• ABS/PSM 5.7: Fault Code 5525 – Incorrect data transfer with DME master control unit.

Good news is cam deviations were quite low, and misfires zero.

The shipper mentioned the battery died during shipment (after-market radio left on was his first theory). He had to jump the battery when it came time to shuffle the load on his rig. Could any of the above codes be triggered by a battery drain event? I'm not averse to clearing codes and test driving further to see if they return, but didn't want to destroy any clues too soon by clearing codes.

A couple other minor clues or facts to share in context of the above:

• Seller kept the car on a battery tender, perhaps indicative of the after-market radio (installed at car purchase in 2001) being wired (per shipper's theory) in such a way that it isn't ignition-switched in its power supply?
• During PPI there was evidence of rodents (seeds, etc.) in the area in front of the dash, driver's side, as seen through the plastic grill with frunk open. I am wondering if "P1601 CAN timeout - Instrument cluster" might be related to any rodent damage to wiring. All systems seem to function normally in the cockpit.
• During PPI the front right headlight wasn't working (witnessed by the mechanic and by me). A few weeks later, when I had same shop do some work on the car while waiting for shipment, the other mechanic said both headlights were working, nothing to repair. Interesting. What can cause a headlight to be intermittent in operation?
• Front right outer CV boot was torn, and some early sign of corrosion on the axle, so I agreed to have shop replace it. Could "Litronic fault code 3 Axle-position sensor front" be related to that?
• Other work performed by local indy prior to shipment included new spark plugs and coils, oil and filter change, right motor mount replacement (PO's shop had replaced left mount only), new drive belt, rear brakes (rotors, pads, sensors) and brake fluid flush.

Full detail on the car is here on BaT, if anything there is helpful. I can also post a bunch of Durametric screen grabs, or log a session if that is helpful.

I appreciate any advice from the experts here. I researched a number of these fault codes on rennlist last night, as well as phrases "PSM/ABS" "weak boost", etc. and understand the PSM/ABS failures can be triggered by a number of root causes. I'm also interested in learning more about DIY methods for reasonably competent owners diagnosing boost system issues (loose hoses, failed or oily MAFS, etc.). Yes, search is our friend. But so are the black belts here. Thanks in advance.

https://cimg9.ibsrv.net/gimg/rennlis...92b8e9d9e5.jpg

PS: it has the dreaded sunroof rattle, but that's a topic for another day. Plenty of threads out there already on that matter.

Yes, a dead battery can cause all sorts of codes to be thrown. Fully charge the battery, clear all codes and then reassess.

..and try running the car with the maf unplugged ( see how it runs then ) as the maf code is common and could on its own cause those "psm/abs" codes if the maf is "bad".

the car reads as if in "limp mode ( no boost past .4 ) " but the codes lead all over the map. no pun intended. and to carlo's point. simply having the battery die can cause all manner of weirdness. hopefully that's the cause of "most" of it. GL w it.

Boost leak test too once you address those codes if you still can't boost past .4 bar and you don't have other codes you have a leak. Seems like you have your work cut out for you but I promise when it's 100% you will have no regrets on the purchase. Good luck.

Edit: I see you mentioned the Indy did the ppi but didn't drive it fast? So many problems manafest at wot you have to punch it at least once, that goes for all cars I test drive, not I'm an abusive way but you have to wind it up once or twice.

Thanks all for your rapid replies. Good to know that dead battery can wreak havoc. Today's game plan is set. And for the record, no regrets whatsoever, even with all these codes popping. :-)

I know it was mentioned above, but we all will agree on the importance of a proper pressure test. Go buy the stuff and do it yourself. So many of these cars have significant pressure leaks. Its almost not worth tracking anything else down until you do that and unplug the MAF.

Congrats on the car. I'm also in the Bay Area, and can help recommend a competent local shop should you wish to outsource some of the work.

Is there a good shopping list and how-to posted elsewhere in this form? Numerous posts about doing a pressure test (and not a smoke test), but have yet to find any clear details on equipment and methodology for the pressure test.

Markski sells a kit https://www.markskituning.com/produc...t-leak-tester/. You can make one DIY. Just google "996tt boost test" tons of hits, tons of videos.

Congrats on the car and welcome to the club!

All parts from a hardware store. Take the airbox off and cinch the clamp over the end of the pressure tester. Works like a charm.

Pro tip: roughen up the end of the pvc with some sandpaper so it doesn’t slip out under pressure


https://cimg5.ibsrv.net/gimg/rennlis...c0b6037fa7.jpg

Also - temporarily add additional clamping force to the other worm clamp so the bellows won't blow off the manifold while under pressure. Learned that one the hard way.

Very helpful gents. Gotta run to some family time, but will: 1) recharge battery; 2) clear all codes; 3) road test, including stop at HW store for DIY pressure test components. Appreciate it!

My Turbo has the stock stereo, and to prevent battery drain I keep the car on a battery tender whenever it's parked in my garage.

Do you believe (or know) that stock stereo is a known drain when parked for extended period?

not the radio per se. but for whatever reason, these cars have an incredibly high parasitic drain while simply being parked. could be the inside alarm remote sensor ( which can be shut off by double clicking your key fob until you hear it "beep" ).

but whatever the cause? most all of us have trickle chargers and leave them plugged in for any extended period of time. i recommend the cheap little ctek 3300. its like 60 bucks has a cig lighter adapter and also the alligator clips - and is the same unit porsche uses as their "branded" trickler.

it ( or one similar ) is a "must have". trust me lol

I don't know what the specific electrical draws are. I do know that keeping the car on a battery maintainer really helps.

The ECU and various sensors on the car draw a good bit of current. Besides a battery tender double locking the car, hitting the lock button on the fob twice so the horn honks and the lights flash, puts the ECU and sensors into "sleep" mode and less current is drawn.

Update: I charged the battery for hours this afternoon, then cleared all the codes from yesterday, then drove for an hour, checking out various aspects of the car. Weak boost is still there, if not worse, and the car idles rough and dies often at idle.

Here are the fault codes present after tonight’s test drive:

• P1101 Porsche fault code 594 – Input variables charge measurement
• P0103 Porsche fault code 115 – Hot film MAF sensor
• P1249 Porsche fault code 231 – Boost pressure deviation Below limit

ABS/PSM 5.7 fault codes:

• 5525 Incorrect data transfer with DME master control unit

The following codes did not return once cleared:

• Engine:
  • P1671 / Porsche fault code 407 – DME relay Porsche fault code
  • P1601 / Porsche fault code 660 – CAN timeout – instrument cluster
• Air Bag: Fault code 3 – supply voltage
• Litronic Headlights HBA:
  • Fault code 1 – Engine path left
  • Fault code 2 – Engine path right
  • Fault code 3 – Axle-position sensor front

I have not been able to pick up parts for boost leak test, but hope to do so tomorrow. Any observations regarding the persistent fault codes, and PSM/ABS warnings? Also I noticed while driving that the red perimeter lighting around the after-market Alpine radio is flickering, generally about 1Hz, but inconsistent in actual frequency. I wonder if that suggests an electrical problem. Also, I grabbed a few screens in Durametric after the test drive. Pasted here, though I confess I am guessing which fields might be helpful in ID'ing the root cause here.

I welcome any insights. I am happy to log data if that's likely to be more effective.

https://cimg8.ibsrv.net/gimg/rennlis...ab33d13f02.jpg
https://cimg0.ibsrv.net/gimg/rennlis...db8025a470.jpg
https://cimg2.ibsrv.net/gimg/rennlis...84ff562e91.jpg
https://cimg4.ibsrv.net/gimg/rennlis...acc9f6a9c3.jpg
https://cimg7.ibsrv.net/gimg/rennlis...832d2abaa2.jpg

(engine stalled at the end of this sampling session)

I strongly prefer boost pressure testing at the intercooler inlets vs. the airbox. It will identify the key leaks quickly and without chasing the phantom leaks that you will get from airbox pressurization.

You can build your own pressure testers in about 15 minutes with about $8 worth of supplies with two 1.5" schedule 40 PVC plugs and a 1/8" NPT tank valve. More info here

https://cimg5.ibsrv.net/gimg/rennlis...a261c68508.jpg

I think you need to start removing variables. When I first got my car, on the advice of several tuners, I started with fresh plugs, fresh coils, and a fresh fuel filter. If you have any sort of aftermarket air filter, replace it with an OEM paper filter. If you have any sort of aftermarket air intake, try and source a stock airbox. I know you said some of this work was already done, but I wouldn't necessarily trust it was done right. Check it all yourself. You'll probably find and fix the loose heat shield along the way.

But that aside, first thing I would do in your situation is to do a mechanical boost leak check. Critical on these cars.

You definitely have low boost, but I'm not sure if you can determine at this point if you have low boost because your car is going in limp mode or you have leaks. Build (or buy) the test plugs and TEST for boost leaks mechanically instead of trying to figure out a potential mechanical problem with a diagnostic computer and data logging. Boost leak testing is very easy to do... Now sometimes finding the leaks and fixing them isn't so easy, but you can tell in 15 minutes of testing if you have a significant boost leak or not.

After that, you can start troubleshooting other issues like a dirty or bad MAF, fuel issues, etc. It sounds like a bad MAF is a good possibility, but they are a bit pricey to just swap in to see. As mentioned earlier, try running with the MAF disconnected and see how it drives. If you do replace the MAF, many report having problems with anything but an OEM Bosch replacement. I think the Bosch part number is 0280218009

Also, my car also has a very high parasitic drain if the key is left in the ignition, even if fully off. Will kill the battery in a day or two. Simply removing the key and leaving it on the dash or floor and there is very little parasitic drain, no worries for weeks or even months.

Are there people who leave the key in the ignition when the car is at their home?

I leave my key in ashtray, car unlocked and on trickle charger. No issues, knock wood.

I never leave car/truck keys in the vehicles.

IMHO time to take the car to professional. The cause of the codes could be something very simple.

Thanks @pfbz ! Your thread on building and using leak test gear is very helpful. Heading down that path today, hardware availability permitting. Another local rennlister has a spare MAF sensor he's offered as part of the debug process, so hoping to attack both in parallel. As to your other questions, the entire car is stock-stock-stock other than after-market Alpine radio, and short shifter installed by original dealer in 2001. Paper filter, OEM air box, OEM exhaust, etc. So nothing after-market to pull out of the engine bay.

Since it's at least as much about the journey (debug, fix, repeat), as much as the destination (a fully solved 996TT), I appreciate the collective expertise of the tribe here, whether it's MAF part #'s, double-locking keys, or roughing up a fitting so it doesn't blow off at 30psi. Thanks all!

PS: one ray of sunshine in last night's test drive: I confirmed precise location of roof rattle, and think I can get to it from sunroof tilted, without having to drop headliner. Fingers crossed.

I lived for years in the mountains, front door never locked, keys in ignition of everything that had an engine inside or out of a garage or barn, never thought twice about it. And needless to say never had a problem, though it was more than a few years ago. Mountain neighbors are pretty attentive of strangers lurking about.

I'm closer to town these days, but the Turbo key would still be in the ignition when in the garage if it wasn't for the parasitic drain problem. My property is pretty secure.

I know you likely didn't mean this literally, but *do not* go up to 30 psi when pressure testing your turbo plumbing! 1 Bar is plenty for finding most leaks and really no reason to even think about going over 1.3 Bar unless you are running some highly modified setup.

When I was growing up we never locked the house doors, but we never left keys in any of our vehicles.

I would never leave keys in any of my vehicles. Not even in my garages (which are fully tied into my home security system.)

always want to go with a bosch oem maf 996-606-124-00 - ending in "124.00" part#. pretty sure its same for the 986 boxter but that has a different part number. the "porsche branded" maf ( which IS the bosch ) is $750! dont do it! ) while the same bosch#124.00 can be found for $175 new on fleabay where i got my last one. not sure about all those other codes still, but i'ts a safe bet the maf is fried - for that "hot film" code.

Time VS AMPS
 

https://cimg0.ibsrv.net/gimg/rennlis...6f3f6b236f.jpg

Update on the bug hunt:

• Cleaned the MAF sensor with MAF spray cleaner, and reinstalled. Sensor looked fairly clean coming out, and I was already skeptical this would solve the problems.
• Test drove, and same results: PSM/ABS lights popped less than 2 blocks from home, and boost an anemic 0.2ish on the highway test once warmed up.
• Codes tripping are now down to these three:
  • P0103 Porsche fault code 115 – Hot film MAF sensor
  • P1249 Porsche fault code 231 – Boost pressure deviation Below limit
  • ABS/PSM 5.7 fault codes 5525 Incorrect data transfer with DME master control unit
  • I logged the session with laptop live with Durametric. Looking forward to diving in to that data and learning more about Durametric, DME, etc.
• New MAF sensor ordered yesterday from AutohausAZ (Bosch 0280218009, $180). Probably won't see it until mid to late next week. Fingers crossed.
• Picked up PVC and fittings for boost leak DIY kit. Thanks @pfbz for the intel on that. Regardless of the outcome of this bug hunt, I want to be able to confirm zero/few leaks in boost system.
• Indy mechanic who worked on it before shipment is assisting via phone/text/email/facetime, and has a few suggestions to try while awaiting new MAF sensor (which he believes is very likely root cause):
  • Warm up engine, clear all codes, rev engine while parked and see if any codes pop.
  • Hit brakes hard on dirt/sand/gravel to see if ABS pulses pedal (determine if ABS system working physically despite fault code 5525)
  • Apply vacuum to waste gates to confirm they're functional (buying vacuum hand pump today, as I welcome any excuse to buy more tools!)
  • Inspect engine at parked idle for sounds of leaks
  • Warm up engine, shut of, disconnect MAF sensor, restart and test drive.
• The above experiments are more for my education as 996TT noob than anything. Doubtful any of them will uncover a smoking gun, but what else am I going to do with a rainy lock-down weekend??
• As to the suggestion to take it to a professional, I might as well pay them to have intimate relations with my wife. This is too much fun to hand over to someone else!

Those "vacuum" hoses (actually carry pressure) attached to the actuators can be tough to remove. First, get them to break loose and swivel back-and-forth on the nipples. Gentle grip with pliers works for this. Then heat them slightly with a hair dryer (NOT a heat gun), then grab gently above the metal nipple with needle nose piers and simultaneously push up on the hose ends with a piece of wood or plastic (to protect the metal nipples). Using this method, mine slid up and off quite easily.

One other thing to check on the MAF is if the wiring plug is pushed on tight. The dealer didn’t push one down completely on my M5. It ran for about 15 minutes then sputtered and died. I finally got it started and nursed it to my garage. I noticed the plug was loose when I pulled the codes. No problems since.

thats the first time i have ever seen that part # for an oem bosch other than the 124.00 maf. but it appears to cross reference. learn something every day.

most unplug the maf and run to a/b it before replacing the maf, but as i've said. the hot film code is a safe bet that a new maf will cure stuff.

last bit of advice. never hire a pro to have intimate relations with your wife. it will never accrue to your benefit lol

IMO it's more fun to drive a normally operating 996 Turbo :cheers:

Have you tried just unplugging the MAF and going for a quick drive? They run surprisingly well in whatever fallback mode they go into when the maf is unplugged and might give you a good sense if the maf is an issue.

I've personally never had MAF problems, but it seems like for those that do, cleaning them never seems to resolve anything.... Guess you will know soon enough when the new sensor comes in.

Great news that you have no boost leaks! They are incredibly common and often owners aren't even aware of them if they are small as the turbos just work a bit harder to overcome them (at the expense of spool time).

Also, Have you (mechanically and visually) checked the wastegates and actuator rods to make sure they are not sticking open? .2 boost is not even close, and if your post-turbo boost plumbing is holding pressure, it is either the turbos not making boost or the ECU dumping boost... You might want to give one of the tuners like Kevin or Markski an email/call and discuss with them.

And if you didn't notice/see a profound loss of power/boost on your test drive pre-ppi, i wonder if the shop that 'repaired' the loose heat shield didn't maybe knock something else loose? .2 bar isn't going to feel right on a test drive.

I'd be looking at the turbos themselves, wastegates, actuators, actuator rods, pressure lines to wastegates, N75 valve, etc.

https://cimg2.ibsrv.net/gimg/rennlis...638c7c6e28.jpg

once you rule out a bad maf and boost leaks. if that doesnt resolve all ( after a dead battery! ) then go to wg's and n75 for potential fixes. wg's will only "stick" momentarily and there could certainly an be accompanying n75 issue ( connected bits ) which produce OVER boost codes ( usually ) not "limp mode". but first things first. maf and possible leaks is the place to start.

i realize i mentioned earlier "last advice". my bad.

Just did a test run with MAF sensor unplugged. No change in results: 0.2 bars when mashing it in 3rd gear from 3k to 6k RPM. Same "fibrillating sheet metal" sound.

You must have mis-interpreted something I said earlier. I've started pulling together the PVC and misc parts you suggest for checking for boost leaks. No leak checks done yet.

That's definitely on my short list of suspected causes.

Thanks for the diagram and suggestions for what next!

your welcome. good job. i was hoping you'd hit at least .3 on that test run - once you chased down the more obvious causes.

Bold lettering.

Yup... I evidently quickly read your "I want to be able.." as "I was able..." In that case, right back to putting a boost test on top of your diagnostics list.

If anyone has any good links to diagrams, descriptions, animations, etc., of the 996 turbo system, I would be grateful. I've spent a few hours cruising the net and haven't found much of value that isn't generic turbo tutorials.

http://www.autoatlanta.com/porsche-p..._911_parts.php

there is a copy of the service manual online someplace. if you can't find one, pm me I can send you a copy

This thread has some good information on turbos, primarily wastegates and actuator rods... Skip ahead to post #24 for info on adjusting the actuator rods, which also illustrates how the system works.

I replaced the MAF sensor this evening, and failure mode / symptoms remain the same:

• ABS/PSM failures within a few blocks of startup
• Durametric reports:
  • engine fault code P0103 fault code 115 - Hot film MAF sensor
  • ABS/PSM 5.7 fault code 5525 - Incorrect data transfer with DME master control unit
• weak boost and same "sheet metal fibrillation sound" when boost would normally kick in.

Other ideas / theories / suspicions? (beyond bad MAF sensor)
Apparently I now have two good MAF sensors. ;-)
Tomorrow the search for boost leaks begins. Even if there is one, would that trigger ABS/PSM failure at low speed in low gears, modest RPMs?

sounds like you might have a major boost leak....do the boost leak test

Leak test starting tomorrow morning, at intercooler inlet, as described by @pfbz here in Aug 2018 post.

make sure you physically inspect the turbos and wastegates while you are down there. You shouldn't have extra mechanical noises going on. Grab a video of the sound. See if you can move the wastegate arms manually. There is probably another guy with a tuned 996 around you. Maybe ask around to see if someone will come take a look. The car isn't that complicated mechanically once you get to know it. If its electrical, good luck.

Despite the fact you have charged the battery. I have a similar car ,a 2003 manual 996tt with 50,000 mi.
When I bought it the car needed a boost to start as it had been sitting. During the test drive the psm,and abs warning lights came on. The dealer replaced the battery with a new one. There are no lights since and the car can sit a week without being started and start fine.
At the very least I would pull the battery and have it load tested . There was also a battery cable update tsb related to the warning light issues.
Simple fixes often solve seemingly complex problems on these cars.

A bit of a lightbulb went on this morning in looking at the DME manual. The pages on fault codes P0102 and P0103 both state that those faults will shut off PSM. So my ABS/PSM issues would seem to be a secondary consequence of MAF signalling boost is bad/unexpected.

Thanks all for your continued suggestions. Leak jihad starts now. Is there a separate protocol (or suggested thread) on vacuum leaks? @pfbz thread on boost leaks has been very helpful.

As to suggestion that battery still not sufficiently recharged, I doubt that theory for three reasons:

1. charger went green (good/done) after several hours on the battery last week
2. car cranks fine and has been started a dozen times and driven several hours since dead battery.
3. DME manual only calls for 8v on MAF testing. That's a long ways below ~12v.

The DME diagnostic pages talk about Porsche specialty tool 9637, which would seem to be a custom wiring harness that matches the pinout of the MAF sensor. A quick search online didn't uncover a convenient price or source of that tool. One ebay listing for $455. Anyone know anything about it?


https://cimg0.ibsrv.net/gimg/rennlis...0585637454.jpg
Note text in red rectangle...

i suggested that very thing in the 3rd post of your thread. speaking from experience, i am pretty sure you are not self diagnosing this in the time honored way - given the maf would ( still is the first thing to eliminate as definitively causal - then the leak testing of course ) since that can/does cause *some* of the symptoms you've described. the accompanying and concurrent issues could well be a massive leak but as simple as a "hose".

i posted this near the top of your thread and it bears repeating ..

"and try running the car with the maf unplugged ( see how it runs then ) as the maf code is common and could on its own cause those "psm/abs" codes if the maf is "bad".

i think your problems are possibly only as vexing - as your not having had them before, as many of us have. many times lol

@k9turbo I have already run the car with MAF sensor disconnected, and the symptoms and fault codes were same/similar. I have cleaned the old MAF (see post #33), run MAF-less (see post #41), and replaced the MAF with new Bosch stock unit (see post #48). All tests failed to cure the problem. So now on to what @pfbz has been recommending for a while: boost leak testing.

I appreciate any/all suggestions from more experienced owners, and will research them to understand relevance to my current situation, but since new MAF made no difference, boost leak seems #1 suspect at the moment. And I'm convinced that having the boost leak test rig and knowledge is going to be helpful in the long run of my 996TT ownership.

My only point in posting the page from the DME diagnostics manual is to eliminate any doubt for others who might someday find this thread useful that the P0102 and P0103 MAF faults *will* disable the PSM, so don't go chasing the PSM/ABS failures until MAF faults cleared. Given that my indy replaced right front axle, I was wondering if something on the speed sensor for that wheel might be a problem. That theory now seems very unlikely.

The only thing that's vexing at the moment is shop near Chicago who did the work that I suspect introduced the problem. Not impressed by their responses thus far. They confirmed they didn't check for fault codes or push boost in the test drive after doing ~$4k worth of work. Topic for another day.

forgive me. i hadn't noticed that you had actually swapped out the maf etc. only that you later seemed to surmise that it could in fact be maf related coupled with the boost leak *probability*. it''s frustrating no doubt. wishing you luck, and ultimate resolution.

+100 on this Euros do not operate well with low charge clear the codes and charge the battery and then go for a drive.

Success!
 

Initial success yesterday on my "bug hunt", though car is still jacked up and test plugs in place, so no road test yet. Thanks to guidance and documentation by @pfbz I DIY'd some PVC plugs with quick connect, and clamped them into the hoses that run between compressor output and intercooler input. First pressurization quickly revealed a loud sound of air escaping in the left rear corner of the engine bay, coincidentally in the vicinity where one as-yet-unnamed-indy did some work (motor mount, heat shield, plugs and coils) before car's shipment west. It took me several cycles over ~10 minutes of recharge compressor to 20psi, connect quick-connect, and run hands around in tight spaces to fine the source of the wind. Lo and behold the large black hose exiting the left side intercooler was completely off the intercooler fitting.

https://cimg7.ibsrv.net/gimg/rennlis...451b67a91a.jpg


Anything look amiss here to you? We'll talk later about the shop that did this fine work.

Two other things en route to this finding that I'd love some advice on: 1) oil in the left side turbo; 2) hose "splice" on the right side turbo.

On the left side, I put a blank plug in first, before discovering the right side hose was more constrained in access for the quick connect hose, and decided to swap plugs (quick connect on left side, blank plug on right side). When I pulled the blank plug off the left side turbo, there was oil in the cup, from having been in place for maybe 15 minutes. Is that expected? See photo here of the cup post removal from left side:

https://cimg9.ibsrv.net/gimg/rennlis...28c1b20284.jpg


Regarding the right side hose between compressor output and (I presume) intercooler input, there is a "splice" or "patch", wherein a short section of larger diameter hose is clamped around the original hose as it connects to compressor output. Is that stock, or did someone decide a cheap patch was better than replacing the entire hose. There is a metal sleeve inside the factory hose in the vicinity of this junction. Here is a photo of it before disassembly:

https://cimg0.ibsrv.net/gimg/rennlis...58987e85cf.jpg


Here are left side and right side plugs installed. They remain in place this morning on the assumption that first big leak found isn't the last leak to be found, and once I solve the first leak found, I'll be in re-test cycles for a while. Note also how rusty the actuator is. Is there a test on their operating condition that doesn't involve engine running?

https://cimg4.ibsrv.net/gimg/rennlis...55832f9092.jpg
Left side plug, with quick connect to compressor 20psi hose

https://cimg5.ibsrv.net/gimg/rennlis...e11c7d4d60.jpg
Right side plug, blank.

I welcome any advice on:

1. securely reconnecting the big hose in the left rear corner, coming out of the left side intercooler. Am I pulling bumper, airbox, both, other components?
2. oil in the left side turbo?
3. is the right side hose splice factory or someone's cheap fix?
4. how to test the [waste gate?] actuators while I'm down here?

First time in this arena, in case that wasn't obvious. Thanks for any guidance here.

EDIT: for those who guessed boost leak as the root cause of the fault codes, ABS/PSM lights, etc., you win, and we have a special set of steak knives we'll be shipping out to you as soon as road test confirms root cause. For those who predicted battery depletion as root cause, the battery went on the charger during boost leak jihad, so you can claim success as well because I'll be road-testing a fully recharged battery and major hose disconnect fix together. But sorry, no steak knives for you. And for those who predicted bad MAF, I'm actually thinking original MAF may still be good, and will be reinstalling it more just out of curiosity, and checking the signal voltage as logged by Durametric now that I know that a healthy MAF sensor logs voltage values in the 0 to 5 range. In any case, no steak knives for you. Though I don't mind having a spare MAF sensor in inventory.

I'm really happy you found your problem, congratulations! I think this is a lesson of starting with the easy stuff and moving your way up. You now know your car better than a lot of people do.

As to your questions:

1. You need to remove the bumper and airbox only, I think you'll fiddle too much and it'll take longer not removing it...and I'm not sure it can be done without taking the rear appart. Its a 15 minutes thing. Word of advice, once the hose is clipped, there has to be NO gap (top or bottom) of the hose, it's a very tight fitting clamp. The shop that did your work did not follow this warning, and you see the result now.
2. Oil in the left side turbo is ok, just depends on the quantity. Have you had a recent oil change, and they would've overfilled?
3. This is factory, but those worm clamps need to be replaced from corrosion.
4. You need a vacuum pump. My advice is to fix the hose and drive it to make sure the codes are gone. You don't want to work on another component and create another problem. Fix one first and move your way up.

Cheers!

congratulations for finding the leaks and for your thorough DIY capabilities. i was always in the maf and/or leak "camp" so only half right. i'll accept your kind offer but feel i might only be deserving of a paring knife lol. good job.

Any suggestions on rusty actuator testing?

For those unfamiliar with steak knives reference, watch young Alec Baldwin here:

.
Classic.

Great progress!

A couple of notes:

1. You are testing on the wrong side of the turbo... Take the hose of the turbo exit and install the plugs there. It appears you are testing at the turbo inlet...

Testing at your current location will find gross leaks, but you will also be reverse pressurizing your #16 check valve and have a phantom leak from that, as well as likely have a phantom lea through the turbo oiling circuit.

https://cimg8.ibsrv.net/gimg/rennlis...f0bfd719a8.jpg
https://cimg0.ibsrv.net/gimg/rennlis...399c133e06.jpg

https://cimg6.ibsrv.net/gimg/rennlis...ca1bd88381.jpg

2. hose clamps look a bit beat. You might want to replace them. There are some pretty crappy quality hose clamps out there, so make sure you get good ones (or order oem).

3. You test the wastegate actuator by applying pressure, not vacuum. Some hand-held vacuum testers will also do pressure, but most will not.

Remove the hose from the back of the wastegate actuator, connect the handheld pressure pump, note the cracking and fully open pressure.

From my notes, here are my pressures before I adjusted them:

- Wastegates crack at .42 Bar / 6.1 PSI
- Fully open at .52 Bar / 7.5 psi

You can adjust the wastegates to crack at a slightly higher pressure by adjusting the nut on the threaded rod, but let's save that for another post or thread.

4. This is the tester I used, about $35 on Amazon. This tool is also very useful for testing check valves and other components that might be leaking boost.
www.amazon.com/gp/product/B01JRRIEBI

https://images-na.ssl-images-amazon...._AC_SL750_.jpg

https://cimg2.ibsrv.net/gimg/rennlis...1e393f45a0.jpg


5. I recommend changing the boost hose 0-rings if you have the connectors apart. They tighten up the connections a fair bit and can give you more positive seating of the hoses. The newer o-rings are also a slightly different design. Part #99610680106.

https://cimg7.ibsrv.net/gimg/rennlis...2c69379631.png


6. A bit of operation behind the wastegates... The N75 valve is controlled by the ECU and modulates the pressure being applied to the wastegate actuator by switching between boost pressure (from y-pipe) to unboosted intake air (from diverter valve/ f-pipe area).

7. After you re-seat the intercooler hose that popped off (and any others you might remove), give them a good pull. Try to pull them off... There is definitely a learning curve to getting those hoses seated correctly, and they often look fully seated but will blow off if not fully locked into position. I also suggest firing up the engine and trying to get a bit of boost (or re-pressure test) BEFORE you put the bumper cover back on or you might be doing the job a second time.

8. Don't be apprehensive about pulling the rear bumper. It is a pretty standard part of working on our turbos and you will be doing it quite a few times if you regularly work on your turbo. Having a second person can help at first but it's easy enough to do yourself after a time or two.

love that film. i thought it was "first prize is a cadillac, second prize is you're fired" ha. but good call.

sorry, no on the actuators. but i ( "we" - as i always have help ) did "over/underadjust" mine awhile back as i was having minor "overboost" issues ( was either that or a faulty n75 ) and it reminded me of decades ago when i would try to "synch" my old amal monobloc carbs on my bonneville. "1/4 turn this way, now 1/4 turn that way" until they were so "out of sync" until i had to start from "all the way in and turn out again" lol.

i have the 1.0 wg's actuators. if you are stock on k16's? oem setting is10.29 PSI = .7(?) so adjust to taste lol. but better still. continue to take pfbz's advice and keep up the good work!

That doesn't look like IC hose. Looks like you're pressure testing wrong hose. Can you zoom out and take a pic?

It is the wrong hose. That "hard" hose is not the one you test. You want to pull off the rubber hose from the turbo OUTPUT side so you can test into the rest of the pressurized maze of hoses, tubing, valves, etc.


https://cimg9.ibsrv.net/gimg/rennlis...1046d91f1a.jpg

https://cimg5.ibsrv.net/gimg/rennlis...82d2a08327.jpg

@pfbz and T Kono, are you sure I'm plugging in the wrong location? Here is my left side plug, rotated 90 degrees from my post above. Looks an awful lot like your left side pics, at least to this noob's eye. Also, if I found a leak on the other/far side of the left intercooler, aren't I pressurizing the boosted part of the system? I thought the input and output of the intercooler is bosted.

https://cimg2.ibsrv.net/gimg/rennlis...74b72b9719.jpg
My left side plug

You're pressurizing IC to IC. On one of the picture's you have...you have it plugged on the wrong pipe. Post #58 last picture.

Once I've got my 996TT dialed in, I'll be focusing on synching these from the '54 Chevy 3600 I picked up recently:

https://cimg2.ibsrv.net/gimg/rennlis...cc69561f6.jpeg

Yes, everything stock here on my 996TT.

If three of you are all saying so, clearly I did it wrong. I guess that means I don't win the Cadillac, but my beginners luck found a big oops by indy, so I think I at least deserve the steak knives, and not the 3rd place pink slip, right?

youtube markski 996 turbo boost leak test. There's a few video's of what to pressurize.
A lot of guys on here boost leak test regularly as many are modified and check routinely. Many run more boost than stock and develop leaks over time due to age/increased pressure etc. I just did my annual boost leak test 2 days ago.
K konos pic post #65 is exactly how you check.

This is the one that doesn't look right...

https://cimg8.ibsrv.net/gimg/rennlis...ceb8df6cbc.jpg

It looks to me like you are connecting to the hard-pipe on the turbo intake, red arrow below?
https://cimg9.ibsrv.net/gimg/rennlis...766ef524ce.jpg

Thanks all. Yes, I plugged the wrong location on the right side. I corrected that (see below), and pressed the left side post-intercooler hose back into the intercooler. It seemed to snap in well, and per @pfbz guidance, I tugged on it fairly firmly and it seemed to stay "clicked in". I then pressurized system again, and heard a fair amount of air coming out of the openings on both turbos created by disconnecting the pipes that go from turbos to the intercoolers. Is air from the turbo outputs expected? I taped them over with duct tape just for grins, and then heard air primarily in the air box opening. At that point I decided to road test the refit pipe from left side intercooler to rest of intake system. Note I had not pulled rear bumper or airbox to insure a secure fit of the hose into the intercooler. I put the fully recharged battery back in, cleared all codes and went for a drive. Car behaved MUCH better than previously, idlng smooth and steady, accelerating without hesitation, etc. No CELs or PSM/ABS errors. This confirms for me the root cause was the hose popped loose, and not battery or MAF sensor. After a thorough warmup, on the highway I started pushing boost a bit. Around 80mph and 0.2 or 0.3 bars boost the hose apparently popped, and I suddenly heard the sheet metal fibrilation sound, ABS and PSM errors popped up on the dash, and weak boost ensued. I pulled over, checked engine bay, and sure enough the same hose had popped again. I hand pressed it back in, cleared the codes, and drove 15min home keeping boost no greater than 0.2 bars, and apparently without popping the hose again. All of this was logged on Durametric, for future reference.

Next step this week is pulling bumper and air box and doing whatever it takes to secure the output hose to the left intercooler. Then downstream from there is resuming leak testing. Thanks all for your advice on the journey. Getting there....

https://cimg3.ibsrv.net/gimg/rennlis...0d5d544082.jpg

Right side plug situation corrected. Steak knives for all who pointed this out to me.


https://cimg5.ibsrv.net/gimg/rennlis...df25fad590.jpg

It was nice to see no errors from on-board diagnostics, or Durametric, after hand-fitting the disconnected hose back on to left side intercooler.

i've always kept a few of these around - because they are well known for losing their retentive/clasping functionality - and at 4$ each? why not.

Porsche Retaining Spring Clip: Part# 996-106-802-12

https://www.pelicanparts.com/catalog..._pg1.htm#item1

following with interest and empathy. you'll get through this easily.

There are also at least two versions of the boost hoses themselves... The early boost hoses had hollow 'ears' (the part that the spring clip holds against), and on those early hoses, sometimes bending them out a bit with a small flat head screwdriver helps.

Newer versions of the boost hoses have solid 'ears' that don't get bent up.

Tweaking the ears a bit, along with fresh o-rings and clips mentioned a few times already, plus getting the hose seated properly, will keep the hose in under boost. The good news is that if the hoses are going to blow out, they almost always do it on the first drive, not a week or a month down the road.

This isn't the specific hose you are having problems with, but it shows the ears I'm talking about.

https://cimg3.ibsrv.net/gimg/rennlis...cd9e9e50df.jpg

Also, when pressure testing, if you hear air coming down the small hard pipe that leads into the larger hard pipe that is on the driver's side turbo inlet, that indicates a bad #16 check valve. Unplug that line (it is a quick disconnect) and verify if air is coming down that line. It SHOULD NOT bleed air under pressure if the check valve is working. It is a very common failure, inexpensive replacement part, but a complete b1tch to get to for replacement. (p/n 996.110.135.70)

Other than that, you shouldn't be hearing any air leaks in or around the turbos if you are pressurizing at the intercooler inlets, nothing down there is pressurized!

This thread shows the parts I'm talking about... https://rennlist.com/forums/996-turb...pics-help.html

Thank you both. I will order new clips and O-rings, and send the snake cam in for a detailed look as to which type of ears are on my hoses.

EDIT: I should check the boost hose connection on right side, and compare it to the left side.

Snake cam isn't really going to show that much. Bite the bullet and take your rear bumper off... It's turbo 101 maintenance.

to add to pfbz's mention of the #16 check valve. agreed, it is a common source of leaks. very hard to get to ( at the plenum area ) and if cracked, you will actually hear it leak as air escapes under pressure testing at 20 +/- psi. it's cheap plastic part.

New guy question: why no access from car interior?

By the way, thanks @pfbz for the list of cheap-while-I-am-in-there parts. Great approach.

I've been reading @markski thread on 6 speed re boost leaks. Great thread. I'm sold. Favorite quote thus far: "it's free HP!"

Interior? Or anterior?

If interior then uh that enginr is noisy, lots of foam padding and nvh stuff there. Also engine is pointing the wrong way.

Anterior well you can... Take bumper off airbox etc.

It's difficult enough to clean the inside of the rear window. I can't imagine how hard it would be to reach/work on what little of the engine would be exposed if there was an access panel cut into the body back there.

did you get it put back together? with the quarantine, all I have is to read your story and wait for the ending. Why make us wait so long?!?

@jdexter sorry to keep you in suspense. Maybe I should reach out to Netflix about making this a mini-series? "As the Torque Wrench Turns"?

I pulled the bumper cover last night to get a better look at the left side intercooler-to-boost-hose connection. Planning to test drive this morning and see at what boost level it pops off again. Do the spring clips wear out, or the O rings or other details inside the connection? Curious what it takes for that connection to hold 1 bar consistently, in the face of age, road vibration, temps, etc.

Any advice from anyone on extract a bit of oil due to overfilling (by Chicago indy)? Obviously I could do a controlled drip from loosened drain plug. I have a siphon intended for fuel use, could possibly send that down the filler tube. What are the consequences of over-filling? The owner's manual is very clear about NOT overfilling.

Planning to stop by my trusted indy of 10+ years to get his eyes on it, as well as other work performed by Chicago indy. Frankly, I am wondering what else was done sub-par, as they also replaced front right axle and did rear brakes. I truly regret having let them touch the car at all.

And in case it wasn't clear from previous episodes, and refuting remote guessing by Chicago indy: (1) MAF sensor operates fine (logging expected voltage values on signal line); (2) DME looks fine (enjoying logging and graphing!); (3) there are no "blown fuses" as remotely relevant to the original symptoms (embarrassing guess by Chicago indy); (4) full battery charge did not change symptoms of boost hose blow off leading to PSM/ABS lights and codes; (5) transport driver did not f*ck up the car by having to jump it due to battery drain during transport.

More later today...

you don't need to test drive it. Just work on the pipe until you think its on there well. Then pressure test it to 20psi. If it holds you are good. Its likely you'll find more leaks. Track those down. Pressure test again. Find more leaks, repeat.
If you want to drain some oil, just open up the engine crankcase plug. Drain that out. Refill with half of what comes out. Most of the oil is in the sump.

@wkf94025 Make sure you hear a very audible click when assembling.

If your blowing hoses off on a car that didn't have the problem before, I think it's almost certain that you just aren't getting the hoses fully seated. Replacing o-rings and clips certainly helps, but as much as anything, I think it just helps to make it more obvious when the hoses are not completely, fully, and correctly seated. It's a bit tricky to get them in there properly.

...and when you have the system pressurized, reach in and give the the boost hoses a good wiggle. If they go bang, you didn't have it seated fully, or some combo of your clips/o-rings/ears might need replacement/upgrading.

When I upgraded my intercoolers, I bench-pressure-checked the intercoolers. This is where it became apparent that new 0-rings really helped... With the old o-rings, there was a bit of wiggle/play in the system and it was more difficult to tell if they were 100% engaged. I also was able to get them to pop at least once with wiggling. With new o-rings, everything tightened up. I personally didn't see any issue with the spring clips, but I might replace them as well next time I have the system apart.

An online oem parts shop like fcpeuro or rmeuropean can get you the clips and 0-rings in a few days, seems like you should just do it to eliminate variables. They are inexpensive parts.

As for the oil overfill, how are you determining this? What does your electronic gauge read? If you do need to drain oil, I'd just dump some out of the oil pan drain (not the sump drain) and then refill to the correct electronic mark. Trying to siphon out oil seems like it would likely not work and be a fairly big pita.

Thanks all. Pelican has O rings and clips in stock, so I'll likely order a pair this afternoon after meeting with local indy. Audible click detail noted.

Oil overfill concluded by the electronic gauge showing all bars lit, including the bar above the full mark. Am I misinterpreting the display? Plus there was a bit of oil in left turbo, when pulling hoses off for boost leak testing, which one of you surmised might be over-full oil. I like @jdexter method of draining a limited quantity.

Poco a poco. What else are you going to do during lock down besides deep dive into your new toy?

No. Definitely overfilled.

If the oil gauge is over max bar, you can drain from crankcase. That'll remove about a qt or so and should drop the oil gauge within min-max range. Now if it's still over max bar, you have a major overfill

hurry wkf - pelican only has 5 in stock! :) my money is on new clips along with properly fastened/connected hoses.
any time you re-attach boost hoses ( y pipe/ic's etc ) seriously tugging on them afterwards is recommended. i changed my spark plugs a couple months ago. i still popped an ic hose entering a fwy onramp immediately after buttoning it all up. it happens.

https://www.pelicanparts.com/catalog..._pg1.htm#item1

Victory! Pulled the left side boost hose completely off, and spring clips off, checked O rings, reinstalled everything (except bumper cover), pushed and pulled each end to insure solid connection, and now the car holds 0.7 bar boost in strong pulls in 3rd or 4th on the highway. No sheet metal fibrillation, no weak boost, no PSM/ABS error codes. Thanks all for your guidance on the journey! New clips and O rings en route, though they'll probably just live in the parts bin until hose pops again. I suspect there may be other leaks in the system, but for now, I'm thrilled to enjoy scary (to me) acceleration without codes popping.

After about 7th hard pull on the highway, I got a dash alarm for battery / generator, as shown:

https://cimg6.ibsrv.net/gimg/rennlis...4c6af4601f.jpg

What is curious to me is that Durametric shows no fault codes in the wake of this warning. Any suggestions?

Before I put the bumper cover back on, any "while-I'm-in-there" chores recommended?

I drained 0.7 quarts from the crankcase, and display now measures as follows:

https://cimg2.ibsrv.net/gimg/rennlis...037ce99f91.jpg

Thinking I'll put in ~0.4 quart and re-check.

I'd personally run it where it is now and skip the top off. You can add a full quart when it reaches one bar above the minimum.

Will do. The run-it-near-full seems to be best practice for the non-Mezger 996 engine.

one of my best friends as a kid was in poco. he later joined the eagles.

congrats on the latest improvements.

why not then also for the mezger motored m96 turbo? i never even have a chance to run more than 1/2 a qt low. i love my cars motor.

Why would running more than half a quart below maximum harm your engine?

Awesome! Welcome to the Turbros. Nothing like getting elbows-deep in your Porsche to make it your own...

Curious to see what, if any, leaks you find next. Compared to say replacing the #16 check valve, getting a boost hose to properly seat is a proverbial piece of cake! :roflmao:

On the oil level, dry-sump motors with high oil capacity (like the Mezger) are much less sensitive to slightly low oil levels since your not relying primarily on scavenging the oil from the bottom of the engine to lubricate... The sump provides a reservoir for lubrication and the scavenger pickups just refill the sump.

I think that theory is to keep the IMS submerged in oil or something like that

go back and pressure test again before you put it all together. valve 16 is a pia, but so are dv and f pipe leaks. You'll want to zip tie every line you can find. Alternator and voltage regulators are common failures. Not sure how to tell whats up with yours. I'd take a look at your ground before doing anything. A ****ty ground can cause all sorts of havok and is an easy fix.

High on the to do list next is:

• getting Kenwood double DIN installed
• get Torque Pro running in real time on that display (phone mirrored via Android Auto)
• fix roof rattle, ideally without pulling headliner
• learning more about logging via Durametric, and what parameters logged indicated weak/good/great boost, fuel ratios, etc.
• probably new tires, despite reasonable tread depth, as mfg date is 2013 on current tires

Can't speak for others, but for me, lock down is not boring.

Did this warning go away once you stopped and restarted the car?

i never said it would or did. i simply said that "i never even have a chance to run more than 1/2 a qt low. i love my cars motor."

and that would be because i choose to top it off whenever it runs a 1/2 qt low. the fact that "i love my cars motor" doesn't mean necessarily that any harm will come to it by running it 1/2 qt low. but i can see how you might have inferred that was my belief. it is not. i simply prefer to run with full, or nearly full oil according to the "oil gauge estimator" in our cars lol

No. The voltage needle is exhibiting behavior that suggests the alternator is dying or dead. Whereas it usually reads somewhere in the mid 13's, it has been dropping throughout a couple drives today, currently around 11.5v. Interestingly, when I goosed the throttle a bit, around 4,000 rpm, the needle would jump up, and sometimes oscillate in the mid 12's, but dropped quickly back to high 11's as soon as I let off the gas. I recharged it with 110v battery charger between drives, raising battery voltage from high 11's to low-to-mid 12's, but driving for ~20 min it eventually ends up back in the high 11's, suggesting charging system not doing its job.

EDIT: my indy confirmed symptoms suggest dead/dying alternator, and he discouraged installing reman units if new are available. Just ordered a new one from AutohausAZ. $318 all in.

Probably the regulator. It is a common problem with these cars.

You can remove the alternator and replace just the regulator or the whole alternator w/ regulator. Just be aware that there are two different types of alternators. One for a Tip and a different one for a 6 speed manual.

Info in these threads.

https://www.6speedonline.com/forums/...e-replace.html

https://rennlist.com/forums/996-turb...battery-4.html

Thanks Carlo. Yes, aware that it's a different alternator for MT vs Tip. I will try replacing regulator only first, and if that doesn't solve it, all new alternator.

Oil does more than just lubricate in our cars. It also keeps temperatures lower and consistent. I've never seen any data to know the effect changing oil levels on cooling rate or lubricating capacity, but I'd hate to find out the hard way.

$318 just put the new alternator in. I replaced the regulator in my 99C4 once and it got by for about 4 months before it started doing weird things like the lights and voltage needle were pulsing. Just do it.

And miss the opportunity to scrape my knuckles twice???

you just nailed the reason *most* techs will advise that although a 40$ vr might be the issue, it's wiser to just swap in a new alt. saves the probable near term future labor cost.

Fascinating to me the value we put on our own time in opting to go DIY, whether smaller lighter tasks, or the heavier jobs. Some days I am okay with $0/hr net value of my time, because the learning and doing are fun, end to end. Other days the decision is driven by actual cash cost to have a pro do it. Net net, I think most of my projects end up at < $5/hr true value of doing it myself, if I'm honest with myself, so I rationalize the decision by focusing on the fun and the learning. Long way of saying I don't mind the risk of pulling the alternator twice.

my 40/45 years of near continual porsche ownership have made necessary that i know as much as i can, and actually do about diy maintaining my car as possible. that said, i don't wrench myself and wouldn't if i could. was never my thing. but i am *present* whenever possible ( always ) for all "one day" work so, yeah.

but while it's not fun for me, per se. i do anything and everything i can with help, on my own. it helps maintain my car's baseline.

all that said? i once swapped a clutch on a rented lift - with beer and chicken and an hour long misaligned clutch fork. that ain't happenin' again, anytime soon.

but yeah. make friends with your mechanic if you haven't already or aren't one. good news btw on your .7 sustainable. cheers.

I'm pulling the alternator to try a new voltage regulator. I have removed air box, Y pipe,, MAF sensor, and various other sensors and vacuum lines. Removed the idler pulley in front of the alternator. Removed long bolt and short bolt that hold the alternator in. Alternator loose and wiggling on the left side where short bolt is, through tight on the right side. Any tips for getting the alternator loose?

i loosened the bracket to the right of the alt

I've read this several times trying to make sense of it.

I believe he is saying knowledge is power, and though he is not physically able to wrench his own car, that knowing at least as much as his mechanic is very beneficial.

that would be my next step, although one video on YouTube said it was not necessary. That same video said there are three bolts holding in the alternator, but he never pointed out where the third bolt is. Another YouTube video showed only the two bolts and did not remove the bracket to the right.

First video (see around 9:15 re 3rd bolt)


Second video (no mention of 3rd bolt, nor removal of "oil bracket")

I'm enjoying getting in to the engine a bit more. Starting to appreciate all the comments by others on boost and vacuum lines, fitting, check valves, etc. Not sure I want to tackle much in either boost or vacuum department during this alternator jihad, but good to get a look at what's in there.

https://cimg6.ibsrv.net/gimg/rennlis...b48a1dfe67.jpg
Air box and Y pipe out, drive belt released, idler pulley out of the way, and two alternator bolts removed.

https://cimg0.ibsrv.net/gimg/rennlis...4d968e1ec5.jpg
Back bolt on the left side was fun. Perhaps removing "oil arc bracket" is next step, though some say not necessary.

https://cimg5.ibsrv.net/gimg/rennlis...b22efc75a4.jpg
First look up the throat of the beast. Nice to get a better sense for hoses, check valves, etc in this neighborhood. Is that N75 to lower right of throttle body?

Yes

It is wedged in there pretty well by that bracket. I did mine years ago. I renumber using a pry bar to remove it.

Pry bar it is. Plenty of pry bars and hammers on 6speed thread on alternator removal. Also, post #49 there talks about backing out the long bolt 3 turns from tight, then smacking it toward the front of the car, to press the steel bushing back into alignment with the surrounding structure. Anyone familiar with that situation/method?

Thanks T Kono for confirming N75.

Also, during tear down last night, I discovered that one of the three Y pipe mounting bolts was MIA. Great. More gifts from most recent indy? Who knows. Regardless, will match up with the two bolts present, assuming all are same length.

Just had mine off for a TB cleaning, pretty sure the bolts are all the same (2004).

The factory service manual page 643 confirms the "back out three turns and hammer forward lightly" method:

Note: Germans translate "alternator" as "generator". ;-)

https://cimg8.ibsrv.net/gimg/rennlis...b4ec082a8d.jpg

In Germany when all else fails hit it with a hammer (aluminum mandrel).

The alt has 3 point of contact. L(short bolt) and R(long bolt towards the head of the bolt and the other end of the bolt). You loosen the long bolt and smack it gently to loosen the other end of the bolt(it's at the back of the long bolt that you can't see). That 3rd point of contact has a "nut" that tightens when you tighten the long bolt....it's hard to explain but once you have it out you'll know

"threaded bushing" is the term i was looking for

zip tie all the hoses....easy access to a lot of them. All the rubber hoses are old. If you aren't going to replace them, tie them as they are going to leak soon or later

Thanks @geetee . Very helpful description of the long bolt protocol. I understand exactly what you're describing. First experienced something like that on my Schwinn bike back in the mid 60's.

Will do re zip tie everything I can get easy access to this week.

1 other gasket i would replace is the TB gasket since you are already there. Mine on stock boost was leaking at 52k. Next tuned at 120k ish... or so. It's a somewhat common leak. Ask 911tuning markski who tunes/works with quite a few of these cars on a regular basis. I believe 99611031802 is the part number but double check

Alternator is out, with the help of a little bit of pry bar leverage. The threaded bushing at the far end of the long bolt seems very frozen in place. I put the long thread rod back in, minus three turns, and "three gentle taps" per service manual didn't do squat. Larger hammer and greater force did nothing either. I threaded the shorter bolt into the bushing all the way, thinking maybe hard torque would free it up. No joy there either. I think I can get it back in as is, but would be nice to free it up. Liquid wrench tonight, one more try tomorrow.

GBox 2nd gear detent arrived. What a ridiculous price for this widget. ($206 incl tax). If it's not patented, would someone please machine an alternate source? Yikes.

haven't heard mention of that detent part in awhile. if i may? are you hoping to mitigate an existing 2nd gear popout situation - or you hope it will work as a preventative?

I've experienced pop out twice. My indy says the more it happens, the more it's going to happen, so get on it asap. Welcome any/all advice on this.

Your mechanic is correct. The pop out issue was corrected by the factory in 2002 by changing the shim stack position of 2nd gear inside the transmission. The gbox detent might work might not depending on how much the syncros / dog teeth have rounded off already. I didn't know you were looking for one I have an unused gbox detent which I bought for my 2003 before reading the problem had been fixed .

you can also do the ball bearing "fix" https://www.6speedonline.com/forums/...do-pics-8.html
both of these fixes won't help a trans that already pops out. Time to look into a rebuild. 997gt2 gears here you come.

oh man, that sucks as it obviously can't get better on it's own and he's right. if your 2nd gear synchro has "rounded" too badly ( it's actually the dogring iirc = the cause of it ), i've heard that the detent can slow, if not fully prevent it.california motorsports also makes a "shift arrester". before i got my g'box repaired on my '02, i used to literally hold onto the shifter when going downhill.

add: per jdexter's suggestion just posted,.. i've actually heard the ball bearing fix works better than the detent. but have no direct experience with that.

here's the CMS part https://californiamotorsports.net/co...shift-arrester

Send the detent back. You can accomplish the same thing with a $.50 ball bearing from Porsche.

Thanks for the guidance on the G-Box detent. I will return it, likely suffering their 20% restocking fee. Embarrassed I didn't do better research before hitting Buy Now. Looking forward to the 50 cent fix.

Someone somewhere, perhaps on another alternator thread, suggested putting some heat on the frozen threaded bushing. I assumed that should be with my [propane?] plumbing torch? As opposed to my heat gun, soldering iron, or hot air reflow gun? Should I apply heat to the alternator body surround, the steel bushing itself, or both? Then try to loosen it while it's hot or wait for things to cool down? Haven't thawed a frozen part with heat before.

I did make a modest financial sacrificial offering to the Snapped Bolt Gods in the form of purchasing four small Cobalt bits and corresponding extraction tools, hoping that these will keep said gods from f*cking with me on this 20 year old engine. Though obviously my time is coming in that regard.

Use a plumbing torch. Heat just the outside of the surround, not the bushing itself. Do it fast, hot and direct and then immediately work the bushing to free it.

Tim on 6speed has an excellent write up of installing the ball bearing on the detent spring which effectively places pressure on the detent spring. It's even a Porsche part. I think I paid like 40 cents for it years ago. It's a little tedious install, but I did it without removing anything but the rear wheel. You just need to get creative with extensions, wobbles and bits.

good advice. it is a highly informative thread.

https://www.6speedonline.com/forums/...n-do-pics.html

@Carlo_Carrera thanks for guidance on the torch. It worked. Threaded bushing is loosened up, and hammered toward front of car enough to be flush with the alternator arm surface. I believe that will make re-install simpler/easier.

@s65e90 and @k9turbo thanks for the 6speed link. I saw that thread a few days ago, and good to know that's the best source for the 50 cent mod.

Voltage regulator removed from Alternator. Check out the difference in brush lengths between old (right) and new (left) regulators. Fingers crossed this explains the low and wobbly charge voltage, and that nothing else is bad on the alternator. Anyone an expert in decoding Bosch VR codes?

https://cimg5.ibsrv.net/gimg/rennlis...db4bb91c1f.jpg

Here is an old thread on Pelican forum that does a nice job describing alternator removal/replacement, and also discusses the various Bosch part numbers for voltage regulators that appear identical in form factor and similar/identical in performance, on his Porsche. My VR part numbers aren't an identical match, though I suspect the new one is going to work just fine.

EDIT: both AutohausAZ and Pelican web sites confirm the new part number works for the 996TT (and other non-TT 996's). If the internet says so, it must be true, right?

New regulator is correct.

And if you're comfortable soldering, you can attempt to turn a $40 repair into a ~$14 repair by buying new brushes only, something like . Or cheaper still, something like $5 brushes. Assuming bearings seem solid and freewheel pulley is smooth in driven and freewheel modes.

EDIT: how do you disable the annoying Amazon graphic and newline when inserting a link to something on amazon? Just want a plain text hyperlink.

There is probable more things wrong with the old regulator than just brushes. Replace the whole thing.

Exactly. Brushes are pretty simple and yours look like they were probably be fine ... I think the endemic fault in our alternators is the regulator circuitry itself and $40 is a gift compared to the $3,000+ your local friendly porsche dealer would likely charge you for a full alternator replacement!

Is that what a dealer would charge? Has anyone had their local Porsche dealer replace their alternator? Very curious.

I'm sure it would be ridiculously expensive...

I'm showing list price for a Porsche-boxed alternator at ~$1,800 (997-603-012-07), local dealers almost always bump it up 10% or so above 'list', plus 3+ hours of labor at $180, shop supplies, etc. and I bet it would be close to $3K at most dealers.

It's really shocking how affordable it is to maintain our Turbos if you do most of the work yourself, fix rather than shotgun parts changing, and buy OEM parts instead of the identical part repackaged into a Porsche labeled box.

The alternator is about $1500 plus $250 for the pulley. Labor @$150/hr at my local pcar dealer I would guess about 3 to 4 hours plus tax. How many hours will you have in it including diagnosis ?
locally I can buy a new replacement unit for something like $300 no exchange needed. I seem to recall it's the same as a vw part.

agreed, dealers are insane and usurious. i replaced mine with a reman bosch ( about 400? ) a few years ago and *we* did it in about 2 or 3 hrs in my driveway.

If you want to LYFAO get a quote at a local dealer for a four-corner full brake service with rotors, pads, flush, etc....

i hear ya. my FP failed on the fwy awhile back and even though i knew exactly what it was, had to have it towed since the local dealership was only a cpl miles away and i couldn't get to my pal's to DIY it. my pocket is still recovering but they did give me some weird dealer coupon discounts since i actually argued with the service tech about. you guessed it - "book time" vs "real time". saved a cpl bucks, but not much. iirc? a fp parts and labor. $1100+! criminal lol

Something like 5.5k at porsche south orlando.
The "premier" Central florida porsche dealership.

On a side note i have the generator replacement from po records... Something like 2k installed in 2011 i think. 997 part.

Eh its not bad, fp is 500$ venturi lines 300$ "$pecialty" tool (chizel and hammer) 100$ gasket 12$ box of rags and cleaner 20$. Diy. Thats how much it cost me minus the tool.

Sorry, what is FP?

EDIT: fuel pump. Duh. That question was pre-first-coffee.

Agree with all the above on DIY economics. Partly why I was searching price of brushes only -- wanting the DIY out of pocket to be an extremely small % of the bend-over-at-the-dealer cost ($4 / $3,000). That said, I am replacing the voltage regulator with new.

100% as to the VR ( not to be confused with the FP lol ) some here have had success with just r&r'ing it ( and yeah only 40 bucks ) but i've long been a believer that the labor to redo the alt in case it was not just the vr makes a one time r&r of the entire unit ( particularly if you buy the bosch "reman" ) more prudent.

just my take.

Just minimal wrenching skills can save you a lot of $$ with these cars.

Over the years I've come to realize routine maintenance parts like brake pads filters etc are usually available aftermarket at reasonable prices. The outrageous prices parts are the items not usually replaced.
In1987 I bought a 1986 911 with a missing cigarette lighter. The dealer wanted $1200 for a new one. I discovered the lighter from a 1975 VW beetle was identical for $15 ,same knob etc but the picture of the cigarette was slightly different.

On another topic I noticed a picture you have showing a rear tire size of 315. The recommended size is a 295. If you have a different diameter front to back you will overheat the viscous clutch between and it might also have contributed to your abs ,psm fault warning lights

Thanks for pointing that out. Will investigate. Maybe need to go 2WD until I can DIY a temp sensor into the viscous clutch unit. ;-)

Anyone else running differential diameters front to back?

EDIT: I recall my date codes on the front are circa 2013, so perhaps something to be addressed when I re-shoe the horse next.

I don't know what size you are running on the front I saw a recommendation is for less than 3% variation in revolutions per mile front to back.
No need for a temp sensor better to size your tires appropriate to the cars needs. It's pretty universal to use the same size tires on all 4 corners of a 4x4. With staggered sizes the required sizes aim for the same revolutions per.mile. Your car set a warning related to psm or abs. My point was there may be an underlying problem created by a conflict between relative speeds of the front and back wheels.

in case the lockdown continues well into the next month as i'm sure it will in CA. here's a lengthy debate on tire ratio aspects and the potential for harm by using unmatched tire sizes ( if still awd ) or as you say wkf simply go rwd and enjoy "mismatched" tires as i do :) there is no obvious consensus as such but it's is generally agreed upon ( wait, you mean it's not?! lol ) that as long as one stays with a 3% differential in rolling diameter ratio front x rear. you'd be fine.

after continual beatings however, the horse died though before anyone's viscous coupler did. i kid.\

https://www.6speedonline.com/forums/...ire-se-up.html

Yes, the front/rear tire diameter ratio is critical, but that doesn't mean you need to run a 295 in the rear. I'm running 245/40-18 fronts (25.6") and 315/30-18 rears (25.5").

That gives a perfect front/rear ratio, with the fronts being just a tiny bit taller to correctly bias the viscous coupler.

I couldn't see the front tire size when I posted. You can select larger tires and maintain the correct split as posted above
just thought you should be aware and check it out

Quick and dirty string method around front and rear tires shows 1" difference over ~77" circumference, so I think I'm within the 3% ballpark pretty handily. If lockdown boredom continues, I'll drive up and down the driveway ten revolutions on each wheel with index mark, and tape the entire distance covered in 10 revs for more precision. Then repeat on passenger's side for additional data points. ;-)

Sorry I am too lazy to scroll back and find it. What are the sizes of your front/rear tires?

:corn:

Front: 225/40 ZR 18
Rear: 295/30 ZR 18

^ isn't that the stock fitment? forgive me, it's been a decade lol

I think @johnsjmc might have seen one of my pictures of boost testing earlier in the thread and seen the 315 rear?

Perfect, case closed.

Starting to log boost and other metrics during 3rd gear pulls from 3,000rpm to 6,000rpm. Here are five pulls, showing speed, rpm, ignition advance, and boost. In calculating boost in bars, I am taking the "Boost pressure of sensor" field in Durametric, minus the "Ambient Pressure" reading at same sample time, divided by 760 mm Hg per Bar. Sound right?

First three pulls I was logging other parameters, and not boost. Last two pulls had the proper boost params enabled. RPM on the left axis, everything else on the right axis, with boost * 100 to achieve reasonable scaling. Curious what the cup shape in boost near the top of the pull is. Anything look amiss on these?

More work (and fun) ahead with analysis of the MAF and lambda numbers....

EDIT: OAT around 72*F. Altitude de minimus. 91 octane (likely with 10% ethanol). Totally stock intake, exhaust, and tune.

EDIT #2: I found the 760 mm Hg conversion on 6speed, but I'm thinking the proper denominator is 980, not 760. Hpa, not mm Hg?

https://cimg4.ibsrv.net/gimg/rennlis...d1dbbf2d06.jpg

Sample rate is average about 5 per second, despite setting Durametric to highest sampling rate. Pretty anemic. Reasonably modern Sony VAIO laptop with solid state hard drive, so no excuses there. I have not experimented with process priority, or disabling virus checker, or reducing number of parameters logged. All TBD.

Bumper removal is 101. Bunch of hex screws and a a couple of nuts. Mine in the operating room right now. Easy access once opened up.
https://cimg4.ibsrv.net/gimg/rennlis...ce75fd542.jpeg

Thanks. Yes, bumper removal merit badge has been earned in first month of ownership. And a few more as well. Snake cam gets used early and often, partly because I like good cheap tech, and partly because I find it helpful to learn about the car when i have 20 free minutes, and am curious. Yesterday when replacing the airbox at the end of alternator replacement, before cinching up the main air box bolt near engine bay latch, I took a look at the two posts on the bottom of the air box that seat in the grommets on the frame. Previous mechanic botched the right one, folding the grommet over as the post went around it. Peace of mind confirming that didn't happen again. I also took a look at the far side of the N75 valve where I had zip tied the hose on that forward nipple, confirming hose was seated all the way, and zip tie was in the right spot. A lot easier with snake cam. But I'll confess, some days it's a solution looking for a problem. Next task for super snake is to revisit the rattle in the roof where the steel bar is loose from the plastic mounting piece on the right side, aft of retracted sun roof. Looked at it a month ago, with photos and videos making very clear what's wrong, and showing someone's previous [unsuccessful]attempt with some black goop-like substance. Question at the moment is do I remove headliner, or attempt a McGyver band-aid without removing headliner? Additional reconn with the snake is step 1.