logray

Right on man glad you got it sorted!!!! :thumbup: Sorry to send you on a few wild goose chases... those videos really helped me get the timing of the light, and is usually a great remote troubleshooting aid !!!! If you get your hands on a new coolant sensor without even installing it first see if it clicks in to the connector better... :clapping:

Now why not contribute to keep the site running? :) It's well worth it !!!! The first time I had that moment I felt very compelled to send a few dollars and say... :renntech: !!!! Nice car BTW, I like the wheel crests!

I know Raby works with some other shops and I was thinking more along the lines of bringing it somewhere local and having him pick up the tab.... not sure if that would be covered by his work - for example perhaps he crimped a wire or something was damaged during the engine work. It might be worth a shot??? In any case, it sure is fun working on a 12 year old car, isn't it??? You think a 12 year old car is fun, try working with a 20 year old car like my old VW. Sooooo.... about your connector, I had the same problem with mine, and a new sensor solved the problem. I don't think you can source just the connector, so you don't want to damage that if you can help it... A few more thoughts. After watching the videos a bunch more times I don't think it is 1 second. I think it is every 2 seconds, so one of these two conditions. This is just my opinion of course. Hard to tell from the videos for sure. 1. Engine coolant level too low -- light flashes slowly (0.5 Hz) 2. Engine compartment temperature too high -- light flashes slowly (0.5 Hz) (engine compartment blower might be faulty) At this point I would think it is not the water temp sensor especially since the light didn't change when you unplugged it and the gauge went down but the light continued to flash... the 1 timing was perhaps a red herring? I wonder if it is the engine compartment temp sensor. Check out this post for the location of all the sensors: The engine compartment temp sensor #12 in the post below is very easy to access on top of the passenger side intake manifold. Dead simple to reach in and unplug it and observe the light on the dash. You might have to remove the airbox to get your arm in there. Of course, disconnecting it may or may not make the light flash, since it might have to get up to temp and if it can't read one... again kinda going back to the "just replace it" syndrome. Then again, perhaps you could blow a hair dryer or heat gun on it and monitor with Durametric to see if it behaves somewhat logically. Or even multi meter it and ohm it out. Then again, there are few specs like sensor ohm ranges avaialble to the public. http://www.renntech....post__p__206679 Another idea is to disconnect the engine purge fan on the engine lid (also dead simple to get to) and see if it flashes at the same frequency again when up to temp, and/or even maybe when trying to activate with Durametric to see if it flashes. Be careful getting it up to temp though without this fan, I would be prepared to perhaps get a fan nearby and blow it into the engine bay if you have the engine comp. purge fan unplugged. It is interesting your engine bay fan is running though... I keep going back to that for some reason. I've read that it runs on track cars that are driven hard, but for the normal driver it is unusual. I've had my car out around Sacramento area in 100+ degree temps and it hasn't ever come on, even after I've parked the car. Another lingering thought in the back of my mind is a air in your cooling system, although I'm positive Flat 6 would have used a vacuum fill tool to mitigate that, but perhaps the flashing is telling us the engine really is getting to hot.

Right on, post some pics if you have it!!!

http://eagleday.stor.../996dalisu.html http://eagleday.stores.yahoo.net/996alinrikit.html

This is a very difficult problem. Strange problems like these that don't seem to follow documentation. I read somewhere that you had your car serviced by Jake Raby???, have you tried contacting him regarding this problem? Perhaps he has some sort of warranty program. I know this is going to be difficult for you, but I would start with the two sensors. They are cheap and relatively easy to do. Replacing both those sensors are insurmountably easier and less expensive than replacing the wiring should that be the cause. Most of the evidence points towards one of those sensors. You don't really want to go down the path, but if the instrument cluster is the problem that is also a major PITA and expensive as well, compared with the sensors. edit: the DME is also quite expensive to deal with compared with those sensors, if that is the problem. Cheers man, I am hoping this is an easy fix! :cheers:

Great post and thank you for the comments. The knock sensor on bank 1 (the other side of the engine) is a PITA, because you have to remove the secondary air injection system, or at least unbolt it in order to gain access to it..

I believe Macster over on rennlist has a Boxster with the same design engine with upwards of 250k miles. I'm sure there are others as well....

Thanks for the post and good to hear that you've taken care of that anxiety, real or not.

Hey Robert, I probably wont be around tomorrow but Im sure other people will be here to help. If you want to look at PET, go to www.autoatlanta.com and click 911 at the top, then 996 hard parts...

I'm not having any luck coming up with the impedance stats for the water sensor, sorry... must be internal docs. Loren seems to be really good at coming up with that stuff... let's see if he has any ideas.

Try youtube.com... free and I think the limit is bigger than that.

If you had a 2002 car you had a 3.6L already. And it has Motronic 7.8 which is compatible with any 2002-2004 engine (and rebuilt replacements after that). To find your DME version for certain you need a Durametric, PST2, or PIWIS, but based on your MY info it is a 7.8. Yes unserviceable bearing without total engine teardown. I haven't heard/read of any failures yet. It has similar load rating to the original dual row, but not as high rating as the LNE ceramic replacement. Of course the new Porsche engine design for the 911 does not have the IMS bearing in it at all.

Ok, the double "NTC" negative temperature coefficient thermistor (water temp sensor), does not appear to share that same ground point 11 that the bank 1 cylinders 1-3 coil packs use, that ground point you were asking about in your pic. On the pins for the coolant temp sensor.... Pin 1 Black/orange goes to the DME pin 74 and am guessing it is 5V+. Pin 4 Brown/pink goes to the DME and it is sensor ground. Pin 3 Black/red and Pin 2 brown/orange go to the instrument cluster. Black/red appears to be positive to the instrument cluster for the NTC H20 temp sensor. Brown/orange is analog ground for the instrument cluster. So I think the pins you could ohm out on the sensor would be 1 (pos) + 4 (gnd) for the DME and perhaps also 2 (pos) + 3 (gnd) since it appears that the DME and instrument cluster each get their own reading from that sensor - hence the term double? right? (or double meaning perhaps dual range?) I could use some help validating this BTW if anyone else is following this thread, also still need acceptable ohm ranges for the coolant sensor...

If I remember right that is a ground point for bank 1 of cylinders (1-3) spark, fuel, etc. It is possible that the water temp sensor shares that ground... even if it is dirty and adding resistance, I'm not sure it would be enough to throw that light??? You might be onto something though with a wiring problem if it is not the sensor. Going back to your efforts to isolate this problem. Since you've unplugged the low coolant reservoir sensor and the light still flashes... I am wondering that perhaps the reason why the light didn't flash or gauge move when you have the water temp sensor unplugged is because the dash is reading no resistance... meaning it's not going to budge... it needs to see resistance in order to move. You should be able to validate this by getting the temp readings from Durametric. If you remove the sensor you should be able to "bench test" it. (edit: or I know it might be difficult to do, but perhaps measure the resistance of it while it is still installed in the engine). That is, take a multi-meter and measure the resistance of the sensor between two pins. The test would be, measure ohms between ground pin and another pin on the sensor. On a good sensor it should be within a certain range. Then put the sensor in hot tap water (or if still in car warm the car up a little, but not too hot) and it should again measure within a certain range. The problem is I can't find the specs on this sensor quickly, which pins to test and how many ohms you should read out of it. Does anyone have this data somewhere? I will look at the wiring diagrams for the pins, but am stumped on the ohm ranges. For example on a 944, it should read between 400-600 ohms at room temp. If it reads above that it is bad. At higher temps, it should read lower ohms, for example at 100 degrees F, warmish tap water it should read 280 ohms, etc. etc. More examples... A typical GM coolant sensor, for example, may have around 10,000 ohms resistance at 32 degrees F and drop to under 200 ohms when the engine is hot (200 degrees). A Ford coolant sensor, by comparison, may read 95,000 ohms at 32 degrees and drop to 2,300 ohms at 200 degrees.

Also I forgot to say I like to pour fresh oil about a little less than half way up the filter housing before I screw it on. I'm probably being anal doing it that way but I dont think it hurts doing either way. Maybe with oil already in the filter housing it reduces the chance of oil not getting to parts as fast as they should.

Interesting JFP as always thank you! It is also interesting that the original 996 workshop manual says zero +/- 6 degrees.... of course that was a certain number of years after the Boxster had been produced.

Well I'm having trouble locating those parts as being available separately from Porsche. You might need to replace the pump reservoir if that's where the hose is going into. P/N 99731422000 MSRP $82.87 Edit, you might also have to replace the return line that goes into the pump reservoir, if that is damaged 98634743906, MSRP $55.20. Or are you talking about the line that goes into the pump itself, not the pump reservoir?

Those animated pictures in my post are just representations. The timing of them might not be 100% spot on to what you see in the car. I would trust that the 1 second/1 hertz intervals (on or off) means the condition where that temp sensor is malfunctioning, which is what the documentation is telling us. "4. Temperature sensor at water outlet faulty -- light flashes rapidly (1 Hz) ; pointer on the right" This is just a guess, but it might be possible that both conditions don't have to be present in order for that light to flash at 1 second intervals. In other words EITHER the sensor could be malfunctioning AND/OR the pointer of the gauge could be all the way to the right. Following our advice earlier, once you tackle the known, if it is still not fixed then you can start tackling the unknown. The unknown part could be the instrument cluster and/or gauge itself malfunctioning, etc. Since the coolant temp sensor is inexpensive and easy to replace, much easier than the coolant level sensor to get to, especially after Loren's info - that would be my next step. You will need to drain the coolant from the block because the water sensor tip does protrude into a water cavity. I recommend a 5 gallon home depot bucket for this job. To refill it is best done using a vacuum system, however others have been successful just following the manual bleed procedure. Look up Loren's post on adding a third radiator for those instructions.

Page 69 of Diagnosis to 45 for Boxster 986 says +/- 4 degrees. Interesting.

Based on the pic your clamp looks far enough down on that fitting to grab... just don't overtighten... but certainly you can snug it up. Only case might be if that fitting is broken or cracked DUE to over tightening (the hose would might break first), in which case you would need to replace that drain hose that goes from the tank to the water pump (with a second tee under the engine that goes to heater core loop).

Hey JFP, isn't the Boxster +/- 4 degrees, or maybe I'm not remembering right... I thought the 3.4L and above was 6 degrees. Could be wrong of course. edit: grammar corrections.

13.8v should be good, but a real test on an alternator is current draw (amps). not voltage. It could be the regulator on the alternator. But I would just charge the battery, and take it for a couple long drives. It should be OK. If it happens again I would suspect battery, alternator and/or voltage regulator, or corroded wires.

Just tighten the screw a little more. It should be OK. That small hose goes to the AOS. The fitting that the hose goes over is molded into (if I remember right) to the larger hose in the picture that goes from the coolant tank to the water pump, and is also tied into the heater core loop. The larger hose is a pain to replace since you have drain all the coolant and then refill using a vacuum system.

Ok, USA uses 315mhz for the key transponder, which is option code m535. The rest of the world uses 433mhz, or option code m534. Do you happen to have the P/N on your alarm unit? I am guessing it is going to be m534 option code. I am wondering if the pill antenna (and/or other parts) are setup for 315 mhz (USA) and therefore ignoring the 433mhz signal coming from the key. Just a theory here on that one, but maybe even though the alarm unit is now 433mhz there is something else in the car that needs to be changed out as well to receive that signal, perhaps something near the ignition switch or the length of the antenna needs to be modified for that frequency??? Also using your Durametric are you able to connect to your airbag module? Do you see any fault codes in there?