996 Calibration: I'm not smart enough to read what PIWIS is telling me

[Keith U]

My 2001 C2 top has not worked the entire five years I've owned the car, full saga below.*

 

Without all the back story, here's where I am today. When I click "Calibrate," nothing happens and PIWIS 3 returns "Function inactive, end position not reached."

 

The front latch is reset per the service manual, and the clamshell potentiometer was installed at as close to 6.2k ohms as I can get it (about 6.24-6.26). [I rebuilt this engine in my dining room with my cat, 12,000 miles ago—I may not know what I'm doing, but I can follow directions.]

 

Top is in service position.

 

Actual Values / Input Signals

Latch locked  Catch not locked

Latched unlocked Catch not unlocked (can't be both, right, has to be one or the other?)  EDIT: RESOLVED, please see follow up post.

Convertible top Convertible top raised

Convertible top compartment Convertible top compartment lid unlocked

Right side flap up Right side flap not up

Right side flap down Right side flap down

Left side flap up Left Side flap not up

Left side flap down Left side flap down

Speed Speed >5 km/h (3 mph) (that's not correct)

Handbrake Applied

Latching hook Latch hook not actuated

Engine-compartment lid Engine-compartment lid closed

Convertible top button Close convertible top button actuated

Supply voltage 14.45 v

Warning light on 36044.3 min (!)

Actuation counter 1 1746

Actuation counter 2 65536

Calibration counter 2

Pot 1, convertible top 0.92 v

Pot 2 (conv. top compartment. lid) 2.51 v

Calibration: Pot 1 val 1 0.80 v

Calibration: Pot 1 val 2 1.27 v

Calibration: Pot 1 val 3 1.88 v

Calibration: Pot 1 val 4 2.29 v

Calibration: Pot 1 val 5 2.96 v

Calibration: Pot 1 val 6 3.22 v

Calibration: Pot 2 val 1 1.55 v

Calibration: Pot 2 val 2 1.82 v

Calibration: Pot 2 val 3 2.00

Calibration: Pot 2 val 4 2.71

 

These aren't even actual calibration values, are they, they're stored values from the last time it successfully calibrated.

It's my latch isn't it. There's a microswitch I've overlooked, a stupid simple on off switch, and because of that, it never starts the calibration process.  Guess I'm going out there one more time this weekend and taking that off again.

 

Insight invited, thanks for letting me ramble. 😄

Keith

 

*The Complete Saga

I bought the car like this in January of 2001. Tip cab with a top that had been manually put up via emergency procedure. Dash light for the top was on.

I found a matching hardtop an hour away that same week, manually put the top down, slapped the hardtop on, and drove it. Turns out, the left rear window regulator had failed and someone had manually pulled it up (the motor was missing) and then drove a drywall screw through the regulator frame close enough under the bottom of the window bracket that the window stayed up. I rebuilt both rear window regulators and in October of 2021, started working on the top. I probably didn't reset the latch correctly at that time, or the clamshell pot. Still, I told Durametric to calibrate and it started to. When the top moved doubt 1/2 way up, I could hear the pump motor straining and—since the rear interior panels were still off—I took a look: both hydraulic pistons were jetting fluid down the rain drains. I put the hard top back on. Then the engine failed, then (unrelated) I moved, and I spent all of 2023 rebuilding it.

July, 2024, I sent two cores to Cab Hydraulics in Sarasota to rebuild. Put them in, new hoses too, and this time I reset  the latch and the clamshell pot correctly. I think. But there was no longer a top warning light on on the cluster. I told Durametric to calibrate it, and it wouldn't. Opened a dialogue with the developers, sent them logs they asked for, did everything the guy said, and the last I heard was that maybe their software wouldn't calibrate it unless it was in an error state. Then (not their fault) I took a job and started the process of moving. Hard top went back on.

 

July 2026, time to get back to it. I knew I was going to have to have access to PIWIS, so I followed that thread on some other forum ("under $75") and got the image running on my shop computer. Which, I understand, isn't the same as pulling up to the dealer, so I'm aware that's a possible failure point. I 100% believe the latch and clamshell pot to be correctly reset now. No dash light. When I click "Calibrate," nothing happens and PIWIS 3 returns "Function inactive, end position not reached."
 

 

Edited July 27 by Keith U
Update

[Keith U]

OK, fixed the latch / catch not unlocked issue. I had not fully advanced the hook after lining up the dots as I thought I had. Microswitch now engaged, correct result reported.

 

Same result when calibration attempted: nothing, followed by "Function inactive, end position not reached."

 

 

Edited July 27 by Keith U
Extra word

[Loren]

Must not be seeing one of the microswitches (dirty, misaligned, or broken).

 

 

[Keith U]

Thanks, Loren—I had not noticed there are three microswitches in the latch, I'll look at that 3rd one next. 

[Keith U]

The dash lamp for the top comes on during power on (bulb check) then goes out. And stays out. Always. No matter the state of the top. Latched, unlatched, service position, down and stowed. No light. That doesn't seem normal. I took the controller out and magnifying glassed it for obvious solder issues but came up with nothing. Not really a surprise, visual check isn't worth much.

[Keith U]

Just going to keep reporting on things that haven't resulted in a successful calibration.

 

Fuses checked.  Battery on charger and showing 14.34v.

 

Mini switches in the latch cleaned, and all report anticipated actual values in PIWIS 38.25 with one exception.  The switch depressed by the fork arm when the top makes contact with the windscreen frame reports results the opposite of expected values: when in the service position, it reports Convertible top raised. When manually depressed, engaging the switch, it reports Convertible top not raised.  Back and forth, 1:1 with depressing and releasing the "fork."

 

The one other anomalous actual value reported is speed which continues to show > 5kmh.

 

Parking brake is engaged and that is reflected in actual values.

 

Latch bench tests as expected: deploys under power, retracts when polarity is reversed. Then latch was reset per the FSM before reattempting calibration.

 

Side potentiometer arm (reminds me of a large paperclip 🙂) is intact (unbroken) and captive by pin on hydraulic ram so that pot moves freely when the ram moves. (Not captive as in bound.)

 

Side flaps lubricated and behave as expected (up and down) when actuated via drive links, with anticipated actual values reported. (Up is up, down is down).

 

Hydraulic pump has fluid, no leaks, and when power is directly applied, goes up and down (when polarity reversed) as expected. 

 

Clamshell motor has been disassembled and relubed and bench tests as expected.  Installed, it seemed to be binding, so I completely disassembled the clamshell transmission. No broken cables or gears, and I cleaned out all the crud before relubing everything.  With everything reinstalled, when power is directly applied to the motor, clamshell smoothly goes up and down (when polarity reversed). Potentiometer set to 6.24k ohms on bench; tests 6.28k ohms installed onto clamshell transmission on car.

 

All points of the top frame that move have been lubricated with teflon dry lube.  When the hydraulic fluid relief valve is open, the top can be raised and lowered by hand with no binding.

 

When I depress the #3 microswitch (latching hook, windscreen frame) with my finger, the rear windows will go up when their respective buttons are pressed. When released, they will not. As expected.

 

When attempting to calibrate, nothing moves and "Function inactive, end position not reached."

 

Control module replaced with same part number: when attempting to calibrate, nothing moves and "Function inactive, end position not reached."

 

I'm running out of ideas. It's as if every single component works exactly as I'd expect it to work, but in concert, the software is like, "nope."  On a lark, I even asked Durametric to calibrate it. "ECU returned a negative response. Sub function not supported / invalid format."

 

 

Edited August 9 by Keith U

[Keith U]

Following up, though the top isn't 100% sorted yet. POTs 1 and 2 were on the car when I got it, I didn't change them. I changed the arm on POT 2, before my first post here, because mine was broken, but not the pot itself. I assume at some point these worked on the car.

 

Looking at the actual and stored values from above (and copied below), Pot 2 (conv. top compartment. lid), when reset to 6.25k ohms, was showing an actual value of 2.51 v.  The stored value at fully open was 2.71 v.  So, following this thread (which presented the same issues I was having), I removed POT 2 and the motor from the clamshell transmission and advanced it using a 9v battery in tiny increments, then plugged it back into the control module to read actual values, until the wide open actual value was 2.71 v.  Then the top actually opened fully and closed to the point that the latch would latch. With the clamshell fully closed, I got an actual value of 1.55v, which is identical to what had been stored.  POT 2 is close, and apparently close enough to not need recalibration: stored value 1 is 0.80v, and I'm reading an actual value when closed of 0.83.

 

Along the way, I managed to strip one of the side flaps, so that's off the car for a 3d print solution, but I believe that's the last thing I need to do: top seems to open and close as expected.  Without software calibration. 

Will post again once the entire system is 100%. I have not tried calibrating again.

Keith

 

From above (I can't unsee it, now: 2.51v at fully open is not 2.71 v):

 

Pot 2 (conv. top compartment. lid) 2.51 v

Calibration: Pot 1 val 1 0.80 v

Calibration: Pot 1 val 2 1.27 v

Calibration: Pot 1 val 3 1.88 v

Calibration: Pot 1 val 4 2.29 v

Calibration: Pot 1 val 5 2.96 v

Calibration: Pot 1 val 6 3.22 v

Calibration: Pot 2 val 1 1.55 v

Calibration: Pot 2 val 2 1.82 v

Calibration: Pot 2 val 3 2.00

Calibration: Pot 2 val 4 2.71

Edited August 16 by Keith U

[Keith U]

Side flap threaded insert had split, so the rod was spinning in it at the top most position. The microswitch wouldn't engage, so the clamshell wouldn't close.

 

I made a replacement.

 

https://makerworld.com/en/models/3232512-porsche-996-cabriolet-side-flap-repair-insert#profileId-3661479

 

https://www.thingiverse.com/thing:7401985

 

EDIT: With my part made out of polycarbonate and installed, the top opened completely as expected until the windows went up, then closed completely as expected until the windows went up. Then my part failed: threads gone. I sent my design to a shop to make me some out of nylon, PA12, I'll see if they hold up better. Regardless of the stable solution,

The top works.

Edited 14 hours ago by Keith U