Mystery power sucker?

I've posted in a couple of prior threads about the ongoing diagnosis of extreme hesitation and misfiring my '02 is doing. Most recent is the (I believe) successful diagnosis of a failed crankshaft position sensor (CPS) here . That link also has some details of the symptoms. In short, after some time of driving (half an hour?) it feels like the rev limiter is kicking in at lower and lower RPMs, finally after a few more minutes, so low that starting in 1st gear from a stop is problematic. So far I have replaced almost all of the charging system (the alternator was in fact bad, and took two remanufactured ones to solve that) and along the way replaced the battery (6 years old) and the starter/alternator cable. Voltage drops of the charging system are excellent: 0.025 V ground between alternator and battery terminal; 0.15 V positive drop between alternator and battery terminal. I've been able to probe the CPS waveforms (see the post referenced above) and the new one is working perfectly, 40 V smooth waveforms at idle and no issues I can see. Each of the above replacements has improved the symptoms a little bit. Now it takes a good hard half-hour drive before the engine starts misfiring. Durametric says it throws a P0336 code consistently (ref mark sensor, i.e. CPS). The reason for this post is that while testing the new CPS today I let the car idle for an hour with load (AC, rear defroster). Radiator fans were going strong much of the time, temp up to 200. I could not get misfires in the garage, so I took it out for a quick drive (3 minutes) and got them constantly when accelerating, especially 1st gear. During the drive I noted that the voltage was way down, 11.7 V at idle and no higher than 13 V at higher RPMs. With a new alternator (that brings the voltage right up to 14 V at cold start) and very low voltage drops it seems unlikely to be yet another bad alternator. So instead of focusing on the supply side, is it possible there could be an excessive consumer hiding somewhere? Enough to cause the alternator's capacity to be exceeded with the resulting slow voltage drop. What could this be? It would have to be an awful lot of power, I think, like 300, 500 watts, so it may not be realistic: anything like that would glow red, melt, or blow up before long. Is there a practical way to see if the alternator is constantly running at full capacity? There is never a cranking problem (and the battery is new). But I wonder if the alternator is at its limit and the car is hot and extra demand occurs, that maybe the spark plugs are the first thing to go. The car can sit for a couple of days with no battery drain so it would only be something that happens when ignition is on. For example, the starter solenoid or motor? Starter is still cranking strong so it doesn't seem it has melted internally yet.

If it was me, I'd put it right back on a flatbed to Flat 6.

The regulator of the alternator adjusts the voltage down as temp goes up to protect the battery from overcharging (it's assuming its temp is the same as the battery's temp, which is of course not true in this setup but it's true in most setup where the battery is close to the alternator). My suspicion is your engine compartment temp was very high after 1hr of testing hence the low voltage output. If you drive the car normally, does the voltage go that low? You can put a DC current clamp meter like this http://www.coleparmer.com/Product/Ex...mp/EW-26848-82 on the cable that comes out from the engine jump point (very easy to access). That cable carries all the current from the alternator to all consumers, except the starter. If this offending consumer was really there, why it only shows up when the car is hot? I'm skeptical for now, especially if it only happens with this test/idle condition and not during normal drive. I think you may have some wiring issues with the CPS that only shows up when hot, based on the error codes. The other possibility is flywheel teeth damaged. You can check the former by attaching a handheld scope to the DME CPS input then drive the car around and note the waveform when the misfires happen. The latter you can check by blowing up the CPS signal and see if each tooth is even and clear (sounds like it is based on your description). This is frustrating

Ok, one more thought. When the voltage hits that low, can you probe the back of the alternator to see if the output itself is low or the output is high but you are losing voltage somewhere else? Basically repeating the tests I suggested to you in your alternator thread. If the alternator output is really low, you can check if the current is too high by the current clamp meter. If the current is normal, the alternator is bad, else you need to track down where the excessive current draw is.

Quotes source posts: 11475311, 11475311, 11475311

It certainly was high. The fan was running frequently during the test (idling). After I shut it off the fan went on a couple of times more. I have not driven the car an hour straight. On shorter drives and starting from cold, the voltage appears to behave normally. It varies, maybe 13.5 - 13.9 or so, but doesn't go in any one direction consistently. If it goes down, it comes back up. $240.... if I could borrow one somewhere.... $200 handheld scope.... this is starting to add up! I might be able to run the Tek 475A scope off an inverter but would need a helper to run the thing and work the triggering to find a single glitch like that. It does not sound like an easy task.

I will try to help out a little. I am most impressed by your electrical analysis of the CPS and that seems to be your entire focus, but you have a new CPS, new battery, new alternator. By process of elimination you have replaced the parts most in question yet still have hesitation and low/bad idle with no power curve for acceleration after the engine is hot and under load. You suspect a power draw because that is what you most understand. You have to ask the question of yourself, what is happening after the engine gets hot? Why doesn't it happen when the engine is cold (I assume it is not happening when the engine is cold)? Why does it keep throwing the P0336 even after you have replaced the CPS? Have you checked and cleaned the connection points for the engine ground wire? Did you carefully inspect the CPS connection and wire harness, particularly for chafing or any melted wires? You need some real time information so you use Durametric to get more information about what is happening to the engine as it warms up. Hesitation, bad idle and rough running are symptoms of a lot of other things that need to be checked...fuel trim, vacuum, AOS, MAF, air leaks in the intake, clogged cats, O2 sensors, etc. I think you need to begin by getting some good base line engine control readings from Durametric with engine cold and then hot, or take it to somebody who has the knowledge and equipment (a dealer) to help you diagnose what is happening when your engine is hot. It is frustrating to self diagnose, patience is required. Expand your diagnoses to include other reasons for what is happening.

Quotes source post: 11475286

+996 assuming it's not a installation error. OP, didn't you just have this engine rebuilt or "reconstructed" by Flat6? I know they say they rigorously test their motors afterwards. Wonder why these things weren't picked up in testing.

DBJoe is correct that Mike needs to think in a broader sense about what causes power lag under load (as well as voltage drop). If I had written the check to Flat 6 for their engine, I'd let them figure it out. But that's me. Good luck OP

Since you already have a scope, you can use this this $81 but it's less handy and it won't be effective for measuring parasitic current draw (say < 1A) for future use if needed. DSO scope is more like $170 http://www.ebay.com/itm/like/121200521852?lpid=82 I agree these are not cheap for one time use so I think invest them only if you plan to use them in the future. To answer your question about current drawn from the alternator though, this is the only way I know how to measure it. I'm sure you have your reasons not to ship your car back to Jake for now but I agree with Joe that you need a setup that can monitor/log the engine vitals in real-time when the problem occurs. You already have Durametric so if you have a laptop, you can get a lot of info on what's going on in the engine when the problem occurs. Since your problem seems to be not bank specific, intake and fuel supply are also prime suspects in addition to the CPS. Would be ideal if you can also measure the fuel pressure when driving but that requires more rigging and investment on a pressure gauge or a pressure sensor.

Shipping the car back has its own set of issues -- downtime at the least. Jake has been extremely helpful and responsive (to a fault IMHO), and shipping it back is not out of the question but is low on the list of best actions to take. Yes the car was thoroughly tested before shipping, both in extended driving (under heavy load, and some stop-and-go) and on a dyno. It showed absolutely no issues at the time it was put on the truck back to me. He has data logs from the dyno and extended drives through the mountains. I don't think this is a Flat6 issue. That said, I am going to take the advice here and do some intensive data logging on the Durametric. I am new to this so it may take a day or two to get useful data logs. I plan to monitor a large number (50+) sensor items throughout the car at the fastest rate I can for at least half an hour to an hour. Grand total that's only a few MB of data. Then some correlation analysis.

Just a suggestion: check the 02 sensors for proper operation (voltage dithering) to be sure there isn't some lean condition.

Just note that Durametric or any other scan tools have sampling frequency limits so the more sensors you log, the less resolution each one has (figure about max of 10 samples per second to share across sensors you log). I would recommend the following to experiment with for BOTH banks where applicable. I so wish that our car had a fuel pressure sensor that Durametric can read. RPM MAF Throttle position Coolant temp Intake air temp RKAT FRA Precat O2 voltage Postcat O2 voltage Cam deviations Misfire counters on all cylinders DME voltage

Below is a chart of Cyl 1 misfires and three other data items: RPM, ignition angle, and throttle actual value. RPM is on the right side scale and all three of the others are on the left side scale. The X axis is just the Excel line number but there are roughly 6 samples per second on average and I was sampling 20 data items. These four seemed to be the most interesting but I will work on more detailed analysis later. The chart shows just over 2 minutes of driving time. It looks to me like misfires happen over longer periods of time, but I don't notice them unless I'm putting load on it. See lines 4700 through 4900 where cyl 1 misfires rise from 0 to 25. That is about 32 seconds of driving time, with varying RPM and throttle. I also monitored Cyl 6 misfires and it doesn't exactly match Cyl 1 but the trends are very close. So I think both banks are misfiring the same. There are 6 minutes from line 2400 to line 4500 where I was on cruise control so a very constant RPM and just slightly moving throttle. I was going 78 by the speedo. There were no misfires this entire time. (Only the last part of those 6 minutes are shown in the chart) Misfires start when I punch the throttle at 4550, and RPMs go over 4k. I started speeding up and braking a few times (you can see the RPM variations in the next 200 lines through 4700). A few misfires start on the very first throttle punch around line 4550, but then stay steady for the next couple of throttle punches. They they start rising without any throttle action at all from 4700 - 4900. The rate of misfires increases; it's slow around lines 4550-4600 but a lot faster starting at line 4800. But that faster misfire rate is happening without throttle and (I presume) without load. The highest misfire rate appears to correspond to high ignition angle and high RPM. misfires.png

Looks like a fueling or air issue. Would help once we see the rest of the sensor values I mentioned above. BTW, are the cam deviation values on both banks SOLID (constant) throughout the whole drive? I hope they are. The MAF sensor will be very suspicious but you said you had already unplugged it before and the car behaved the same. If you see the FRA, FRAU, or FRAO larger than 1 (say 1.1 to 1.3), definitely suspect the fuel pump.

Quotes source post: 11477576

I got 20 sensors but only charted 4. The whole file (CSV) is here: Dropbox Link . The cam deviations are solid, yes, do not change except a very small bit: bank 2 changed from 0.77 to 0.78 to 0.80. Bank 1 was -3.56 start to finish. MAF, correct, disconnected it and no change occurred. On the drive home I will set it up to log all the Adaption values and give up on cam deviation, O2 sensors, and all but one cylinder misfire counter to get more data points.

I've checked the whole file. Good news is you don't have a voltage problem. The fuel trims and O2 sensors also look excellent! I don't see anything obviously wrong (except the misfires of course). One value missing is the coolant temp (you did log the engine compartment temp though) but I expect it to be inline due to the excellent fuel trims. Hope Jake would chime in if he's reading this thread.

If the issue were mechanical the car would be coming back here. Nothing indicates that it is and all values are where I'd expect them to be. Its been 5 years since we had a gremlin show up after our work was done and a car was shipped home. This is why I personally put 155 miles on Mike's car before it left, just like all the rest. ZERO faults, no pending codes, perfect fuel trim and all was documented the day the car left here. No matter what we do these cars are aging and the gremlins will become more and more of an issue moving forward due to this. The time has come when we will no longer accept vehicles, and do all of our work with the engine system only. Quantifying all this work using my engine dyno and a day of run time prior to shipping. I implemented this today. Nothing ever stays the same. Mike has been awesome. He has done an exceptional job of diagnosing the issues and carrying out some tests to give us more info. When he asked me if he should take it to a shop to have the work done, my response was "I doubt the technician will be as good as you have been". Thats truly how I feel.

Mike, you have been a lot calmer about this whole thing than I would be. I know you've spent a boat load of money to get the car fixed, but it seems since it was returned, nothing but headaches...that would drive me bananas. 2002 Turbo - Basalt Black Metallic / 2003 Carrera 4S - Speed Yellow / 1955 356 Continental 1500 - Rust Red

Dave, I understand what you're saying but here's the situation. The engine repair is what was contracted. It was my responsibility to get the car there and back (and they even helped -- a lot -- with that.) As far as I can tell that work was done to absolute top notch quality, based on evaluation of engine performance before these misfire problems start. Dealing with Flat6 was one of the most pleasant, professional, and smooth transactions I have been party to. I have absolutely no beef with Flat6, Jake, or anyone there and would do it again in a heartbeat. The car is (aside from the current problem) in better shape with a better engine and higher reliability than when I bought it a year and a half ago. The current problems appear to be due to, as Jake says, aging. The car is 13 years old and a number of components are likely marginal. As buyers of a used Porsche we go in knowing that. It makes sense that the battery, cable, alternator would have failed or been close to it, especially given that it spent its first 11 years in Minnesota. CPS appears to have been an infant mortality issue and given the many parts Flat6 replaced, one sensor goes bad... OK. They shipped out a new one overnight. Contractually Flat6 is not even obligated to do what they're doing now (helping diagnose). Many shops would probably have me have it checked out, at my expense, and only if it can be proven to be their fault would they pay for repair. That has not even been vaguely close to the case here. I've turned to Rennlist because it is pretty clear to us that this is not strictly an engine issue but something feeding or controlling the engine, and other expertise is needed. Yeah it drives me bananas too but I think this is the best approach in terms of time and money, and I appreciate the board's advice on this so far. Hopefully the thread becomes useful in the future as well.

I'd take it where you get the most thorough scan possible - either through your (good Indy) or take it to the dealer for diagnosis and then to your Indy or do the repairs yourself. These cars are so technical when it comes to the situation you have explained that it needs that professional review before moving forward.

Well put to what Mike said........