986 still down on power. Thoughts?

[Morph_986]

I would like some advice about my 986 (2001 Boxster S, manual, 66k miles) that still seems down on power after completing most of the normal checks and maintenance.

First things first. Power (at wheels) is being measured by a Vbox Sport on the same piece of road with the same driver in pretty much identical conditions. Second I live at 7500ft and where the road is results in almost exactly a 20% reduction in power for a normal aspirated engine. I am therefore always going to have a weak car by normal standards. However, I am 10-15 HP down at the wheels on where I should be accounting for 20% losses in the drivetrain. Some recommended a figure of 15% for a manual and that makes things worse.

I have the Durametric software. It shows only one issue I will come to later. No pending or actual P codes. I changed the air filter, cleaned the MAF sensor and throttle body, changed the spark plugs, power cycled the engine computer, recalibrated the throttle and cleaned the air plenum. Vacuum level is appropriate for the elevation suggesting no great vacuum leaks. The old plugs did not show signs of running lean. I also did a concentrated sea foam gas treatment to clean up the injectors.

Net result, no change in HP at all, except for maybe 2-3 HP after the sea foam, but no more.

The only known issue is with the bank 2 variocam where the o rings are failing. I see green chunks in the oil filter. The Duratech software reports that both banks 1 and 2 variocam systems are working as the engine revs change as they should. The 23 degree indicated on bank 2 is steady. If I pulse the variocam in the software at low rpm, bank 1 holds steady at 24 degrees while bank 2 goes to 23 degrees for a second and then creeps back to 0 over 2-3 seconds. So that actuator is clearly going bad but is not totally failed.

I did a secondary cat delete. Absolutely no change in HP after.

This makes me think that I have a limiting fault somewhere else that is preventing maximum power. Thoughts on what this could be? No signs of worn piston bores or rings but I have not done the tedious business of a compression check. Changing the plugs was tedious enough.

The only two items I can think of is a blocked fuel filter limiting the availability of fuel under load, or garbage in the air box before the air filter. Can the Duratech software tell me if I am starved of fuel under load or do I need to squirt the fuel line into a container for 30 seconds as noted in the Bently manual? I can recheck the air flow. I noted nothing when I swapped the air filter, but I was not looking hard.

Many thanks for your time.

Philip

[Morph_986]

Just updating to basically log my situation and will update for the benefit of future google keyword searches later.

Sitrep. Checked the air box and filter thoroughly. All clean. Checked the hose from maf to throttle body. Also clean. Seems therefore engine should be getting all the air it demands.

Car is in the shop getting the bank 2 variocam mechanism changed. More testing with the Duratech showed this playing up repeatedly but no issues with bank 1. Given the part cost and the minimal labor savings, I am only getting bank 2 changed.

The technician noted some smoke on starting. I had previously checked the crankcase vacuum for pressure to see if changing the AOC was required. The vacuum was not excessive, but the tech suggested changing it anyway given his experience with that amount of starting smoke.

Will also get a fuel flow performance check when it is on the lift and to save spraying gas over my garage floor. That should rule out a failing fuel pump or blocked filter.

I will report back. In theory, any one of these three things could cause a lack of power. Fuel seems the most physical explanation. Unless I misunderstand the system, the AOC failing could cause a vacuum leak and so a loss of power, but given the modest crankcase vacuum and fairly small fuel trim values noted by the engine, this seems unlikely to me, although I am a novice.

The variocam should only cause a lack of power from 2000-5500rpm on one bank if affecting things, but reading the technical articles about how it works, above 5500rpm the cam pops back to the zero idle position. Since I measured the wheel power from 1500-7000 rpm the power should be nominal above 5500rpm irrespective of the variocam oil seal leakage rate. I saw no signs of sticking at any intermediate value. It wanted to get to zero.

Philip

[blue62]

That device does not measure power at the wheels....
It measures time over distance.
It is a very poor way to assess power at the wheels
The only way to measure power at the wheels is on a dyno.......rolling road....

[Morph_986]

Are you suggesting that the VBox Sport cannot, or is not very good at, estimating wheel power? The iPhone app Can calculate wheel power and estimate crank power from a coast down. The crank power will therefore clearly be an estimate. Interestingly, the PC software that reads the card cannot calculated power directly.

Wheel power estimate should be quite good. The device has a 3 axis MEMS accelerometer coupled to a 10Hz GPS. Using something like a Kahlman filter the inherent drift of the accelerometer is therefore cancelled. In this way accurate distance and velocity are obtained. In the app, a vehicle weight is entered (curb weight plus driver weight).

Since power is just the time derivative of kinetic energy and kinetic energy is half the mass multiplied by velocity squared, the absolute accuracy depends on the absolute mass and velocity but the relative error is just dependent on velocity errors. Unless the velocity error is large, which it is not, then the power vs speed should be quite relativity accurate.

Compared with a rolling dynamometer, there is an error associated with real world air drag, however, unless the drag changes dramatically during a run it itself is proportional to velocity squared and so easy to cancel out.

All this argument is also a bit moot. The absolute peak power from the estimate varies a bit from run to run, but the power from 2000-4500 rpm where everyone actually drives does not vary much at all. Most importantly it has not varied with any changes to the car. Finally, and least significantly in the absence of firm numerical data, the car feels only similar to two similar aged boxster base models I have driven at the same elevation (not the S as I have) and does not feel to have changed with modifications.

We shall see what the numbers and feel have to say after it gets back from the shop. The annoying thing is that I wanted to plot wheel power as a function of minor and major teaks. If there is indeed a limitation based on, for instance, fuel delivery then all improvements will instantly summed after that bottle neck is removed.

Philip

[The Radium King]

send your injectors to rce to be reconditioned. peak power is open loop so no feedback from the ecu. clogged injectors might perform ok in closed lop operation and poop the bed under higher load. ditto an old fuel filter. get a wideband o2 sensor and chuck it in the tailpipe and see what your afr readings are doing when you step on it. heck, your cat delete pipes might even have a bung for it.

[Morph_986]

The car is back. Variocams and AOS replaced.

Long story short: No change at all in power. Garage reports it drives as well as any 986 S does at 7200ft and to stop worrying.

I guess I was wrong. It turns out that the failing variocam was still performing adequately at the time and not a single thing I did, including the secondary cat delete, has made any difference in that respect.

I did discover something interesting about the oil pressure sensor from the shop. I will post that separately.

It will be a few months until my play-car budget recovers after the expensive repair, but I might get the injectors cleaned and replace the engine and transmission mounts next. I might also try the reduction pulley trick and see if that makes any difference in reality.

Now I will just drive the car since the weather is improving.

Philip

[ike84]

Listen to trk's advice. You're approaching a "tuner" type problem but without enough data. You need to know where your afrs are and what your ecu is doing. If you're leaning out at high rpms, go after the injectors. I wouldn't have them reman'd though - I wrote up a post about retrofitting brand new Bosch injectors from the later 986 into the earlier cars. It's take a few minutes and will save you a ton of $ if that is the problem.

Honestly though you haven't really proven that you have a problem. Under excellently controlled conditions with cars that have well published aero stats and other specs (that Porsche NEVER LETS LEAVE THE FACTORY) the vbox system can reproduce very accurate results. You are estimating a lot of variables that can easily skew your final results, and unless you run it at sea level for comparison then you're really just mentally masturbating and throwing a lot of $ at it along the way. Unless I missed something?

Sent from my SM-N975U using Tapatalk