Another 987.2 Convertible Top Problem

Hi All, It was suggested to me on ******** (hope it's allowed to mention that name here) to share my situation with this group in hope for some clues: Our 2010 Boxster is exhibiting the following behavior: 1. Start the car, sit in neutral with foot on the brake, open the roof handle (disconnect from the windshield frame) - windows do not drop ; now press the roof Down button - nothing happens. 2. Now press the roof Up button - the metal cover (clamshell?) raises and the windows drop. Stop quickly before bad things may happen. 3. Now press again the roof Down button - the clamshell drops back down and the windows close. 4. In order to lock the roof handle back to the windshield lower the windows "manually" with the window buttons. Roof itself does not power at any point during the above. I checked the relays above the driver foot well - swapped all 4 identical relays in all 4 positions (not 100% sure which 2 are the roof ones) and no effect on the above sequence, which I suppose means it's not the relays. My next suspect was the latch microswitch at the top of the windshield. I ordered a new switch (of which later), installed it, and.... no change. So I dismounted the latch plate and double checked that the connector is all the way in the new switch, tried again, nothing. Next I took out the new switch and swapped back in the old one, tested function, then swapped them again and tested again just to make sure I didn't get them mixed up. Nothing. Next I took both switches to the bench and tested for continuity at the connector pins with the switch open and closed (using a little clamp to hold it open). Both switches are good! (Should have tested the old one first, before ordering the new one...). The car behaves exactly the same with either switch, as described above. Any ideas for where to go next (other than to my favorite specialty shop)? Thanks for any tips.

I believe it was me that suggested that... and, yes, there is no harm to mentioning that name here, AFAIK... we're a lot friendlier here anyway, haha! I recommended that @deilenberger take a look, as I think he may have more info on the matter... I believe he has seen and helped with such issues before. But even beyond that, there are many experts here so hopefully someone can give an idea. My initial thought was that I think there may be some other switches on the top, and maybe those aren't working? I thought I vaguely remember hearing that these may be visible on some code readers as well? But I could be completely wrong as well... hopefully someone can chime in.

Thank you, Schwinn!!

Not sure if you have the top schematic or not. Here is a screenshot.

Zuch, thank you! No I do not so this is great. Studying it now... not making complete sense out of it yet. I do not understand the system enough so the following will likely be dumb questions, but I don't see: where is the control of the windows to drop prior to activation of the roof. where is the control of the clamshell, as my clamshell moves but the roof does not. I do appreciate this!

I would expect the control of the windows to drop comes from the control module... ie, it's not hard-wired in any way to the switch... again, I'm expecting this is the case. For the clamshell, good point... this might also be "remote" controlled in this manner. The schematic shown doesn't seem to show where the microswitches are, but it does seem to identify two switches in the top left - CONVTOP CLOSE MICRO and CONVTOP OPEN MICRO SW At the bottom, there are 2 relays and the motor circuit. It might be worth measuring to see if there is voltage coming to the motor... particularly since the rest seems to be moving, but the top is not? Might need more schematics to look at. I'm happy to help read them if you don't know how to...

Hey @NoIMS -- Particularly as a fellow 987.2 2010 Boxster owner, I'm so glad you posted your question. (With regards to ********, I'll quietly suggest you check out Not On ******** – #NotOnRL and will let you reach your own conclusions.) Regardless, you'll find that P9 is an incredible resource -- especially for 98X stuff and I'm really proud of our inclusive and kind community here; it's amazing what has continues to come out of our collective technical expertise. As a fellow 987.2 2010 Boxster owner -- I'm really sorry to hear you're having soft-top issues. It sounds like the programming may be messed up as the windows should indeed drop down a little bit immediately after undoing the to latch. Although I don't have the answer, what I will say is that if you intend to keep your 987 for years to come, and you're at all technically inclined, I'd consider investing in a PIWIS 2 (do a search here) box and set up a VM (virtual machine) box. It has proved invaluable for my being able to test individual parts and it has helped me narrow down problems. My friend turned me onto this and he programmed a dead rear control box late last fall -- which got my convertible top moving again. Whether it's a bad microswitch or not, I couldn't say -- but at least a PIWIS might help you get a better sense of what's going on. Cheers.

That reminds me... yes, PIWIS is certainly a good way to go. Alternatives are the Foxwell NT530 as well as other similar devices. If you're near me, I'm happy to help troubleshoot and we can use my NT530 as well... to see if that helps. I'm in central CT...

@foodbiker and @schwinn , very grateful for you guys. Thank for the tips re PIWIS 2 and Foxwell NT350. I have an Autel MaxiLink ML529 OBD II scanner (a pretty recent acquisition). Haven't tried it at all on this car. Anybody know whether that one would be of any use for this "project"? @schwinn , so kind of you to offer help! Unfortunately we are quite far away - upstate South Carolina (although I bought the Boxster a few years ago from a very nice fellow in Northern New Jersey, not that far from you... but he didn't have the problem then...) I am not sure how deep I want and can go investing in deep diagnostics infratsructure for this car, but will certainly follow up on the PIWIS 2 and VM box to get a sense of what these are. I was wondering whether it is possible that the CAN based control software got somehow corrupted or out of sync without any permanent hardware malfunction and there could/needs to be some kind of hard reset or reinitialization which does not occur upon regular engine shut down, key out and start up. I assume/hope that some kind of OBD II device can tell. Will poke around some more over the weekend and post something... Danke Schoen!!

Window drop is controlled via CAN comfort bus. Signal goes from the rear body control module to the CAN bus (then on to the door control modules). As for the clamshell, is there more than one motor or is it the same one that runs the top? I don't know. I read a thread in the last few months, not sure if was on here or ********, where there was a list of voltage values on various signals as a function of where the top was in its opening/closing process. It might be worth digging that thread up as the guy eventually solved his problem and there was a lot of good debug info in the thread.

@foodbiker and @schwinn , very grateful for you guys. Thank for the tips re PIWIS 2 and Foxwell NT350. I have an Autel MaxiLink ML529 OBD II scanner (a pretty recent acquisition). Haven't tried it at all on this car. Anybody knows whether that would be any use for this "project"? @schwinn , so kind of you to offer help! Unfortunately we are quite far away - upstate South Carolina (although I bought the Boxster a few years ago from a very nice fellow in Northern New Jersey, not that far from you... but he didn't have the problem then...) @zuch , thank you too! No idea regarding the clamshell motor yet. The voltage mapping challenge is for another life I'm afraid, but I will search for the thread for kicks and who knows... 😳

One comment - the Foxwell (and PIWIS of course) can read data on the CAN bus and do some other tricks. A standard OBD2 scanner cannot do these, as these are specific to the CAN bus . That's why, for example, you can reprogram the comfort settings and change various other operations with the Foxwell ... which would not be available with a standard OBD2 scanner . Note that the Foxwell is also a "standard" OBD scanner as well - it just has more capabilities beyond that because it can talk on the CAN bus as well.

There is only one top motor.. it drives the two transmissions (one per side) through a cable drive. The entire mechanism moves when the gearbox turns the bell-crank that the top and clamshell drive bars hook up to. If you can put the top in service position you can barely see both of them with a bright LED flashlight. One goes forward to the front frame member of the top mechanism, the other goes more rearward to the clamshell pivot. #8 goes to the clamshell mechanism, it has a spring in it that's compressed when the clamshell is fully closed, providing tension on the clamshell sealing it against the body. The other one - going to the right in this picture (this is the port side mechanism) goes to the front bow for the top. If the clamshell opens, and the top doesn't - both of the balljoint receivers on the rod assembly going forward are broken so the top isn't being moved. This should be addressed before fiddling with the top any more, since those broken ends have been known to puncture the rubber water receiver causing leakage into the car (and eventual failure of the electronics module(s) under the seats.) There are threads here on replacing those ball-end receivers (or the entire rod assembly.)

I have roof problems also (since I got the car last year) that I had originally thought was due to changing the RCM. I rented a durametric pro cable so I could check/re-code it (amongst other things) which got me closer but not fixed. With the durmetric cable I was able to both open and close the roof without changing any coding. It wasn't until I set the car as having "convertible top - installed" that I was able to use the switch to close the roof. This also had the affect of activating the microswitch in the header rail such that when I open the latch, the windows drop a couple of inches in readiness to use the switch to open the roof. That part doesn't of the process unfortunately still doesn't work (more trouble shooting this weekend....). If I used the durametric to open the roof, the switch does close it though, so at least I know the switch works. Anyway, all that is to say that if opening the latch on the header rail isn't dropping your windows, I'd say there's a strong chance that microswitch isn't operating/communicating (Oh, I should point out mine is a 987.1 but I don't believe there's a significant difference with the latch workings). I also have a POR 2.0 diagnostic unit that can see the operation of the microswitches so your Autel may well be able to as well.

Don, thank you very much for this! Just one question: Why would the windows not drop a bit upon pulling open the center latch handle? "How do they know" whether the balljoint receivers are broken? Thanks!

Did a bit of Googling this AM to see if I could find a diagram of what's happening with the mechanical bits as the top goes up or down - no luck, but I did find: Repair: Porsche Boxster Convertible Top Repair - 986 / 987 (1997-08) - Pelican Parts Technical Article - pretty good repair summary with some decent photos from Pelican Parts. Some troubleshooting on Renntech of a 986 top (which is very similar to ours): Boxster hood problems! - it shows the clamshell mechanism quite clearly.

No attempted progress today either. 😩

You just need to unclip and manually move the roof open to the point that you can reconnect the end link

I see the top is not in "service mode" - I'd suggest putting it into that mode before trying to work in that area, it will make access easier. It should be in the owner's manual, if not - there are videos on YouTube and drawings and explanations in the AllDataDIY 987 service information. Service Position mostly involves disconnecting the retaining cables from the bottom hoop on the top (these limit how high up the bottom hoop - and hence - top - can go when the top is closed.) To get to those you have to release the trailing edge of the water guide (runs across the car, sits under the joint of the top and the clamshell when the top is closed - and guides water to the two water troughs on each side, and thence to the drains.) It just unclips by pulling it down. The cables are popped out of the ball fittings at the top of the cable to the top-hoop. They have to be pulled just right and they'll pop out. More fun putting them back in. Finally - after all that is done (which only takes maybe 5 minutes) - there is a strap attached by velcro to the bottom rear edge of the top assembly. That strap is released and has a loop in it that is taken up to one of the alignment nubbies of the forward edge of the top front hoop (the nubbies go into the windshield header to align the top when closing it.) Putting the strap there folds the top up enough that you can work on the car engine.

Don, thanks a lot for this tip!! Sounds annoying but probably better than trying to fight the ball sockets without it... Just ordered the plastic ball sockets on Amazon, $11.99 + tax for a pair (free shipping with Prime). I'm now thinking that my windows don't drop when pressing the roof Down button because, as you can see from the position of the links, the transmission is already in the Down position, and the controller thinks that the roof is down, so it has nothing to do with faulty sensors, switches or relays, but purely due to the broken sockets. Now, to pop the sockets back on the balls I'm envisioning powering the transmission with the Up button to an intermediate position so the links can reach the balls with the roof partly lowered by hand. Therefore the main question I have now is can the roof, while its links are not connected, be folded partly down by hand by pulling it back by its leading edge to get the balls to where the links can reach them without damaging anything ? 😱 Thanks!

The answer is yes.. and a big trick in replacing the sockets so you don't need to adjust the length of the rod/socket assembly. Wrap some masking tape around the threads behind the socket so you know how far the original ones were screwed onto the rod. Then install the new ones to the same depth. Tightening torque from memory on the 30mm bolt that holds them to the bell-crank is 7nm. You may be able to replace the balls without removing the arms, which would be the best case scenerio. Good luck. Take some photos. Let us know how it comes out, perhaps we can make a DIY out of it.

Just going back to the sensors for a bit. I was in my car and had the Foxwell plugged in, so I poked around a bit. We all know there is a latch switch at the front... but there appears to be a sensor elsewhere in the mechanism as well, because there is another sensor I found that changes state once the top comes "back" off the front. If I weren't a dummy I would have taken a picture of it, but I can do that again if needed to help get the actual name. It was in the rear module and I saw it by looking for the live-data there. My theory is that if this sensor is not actuating then the car may think the top is always-up or always-down... hence the windows aren't dropping, etc. I don't think you need to worry about that yet, as you need to fix the ball joints first... but just something to keep in mind if you continue to have issues...

Hey guys, good news to report. The new plastic ball sockets did the trick. A few observations/lessons: 1. I used the service mode - thank you, Don! I found the stupid strap which is supposed to hold the bottom rear of the top up worthless. Instead I used a large styrofoam molding I have lying around to support the bottom rear of the top. Anything light, stable and smooth about 15-20" will do. 2. I did use the masking tape method to mark the edge of the threads, but only after confirming as best as possible that the lengths of the broken vs. new sockets from the center of the socket to the threaded end are the same. Mine are clearly not from the same mold and I am somewhat concerned because the new sockets are a bit thinner with less of a gradual relief at the base of the socket, so they may break sooner... Maybe they are of stronger material? 🤞 At least now I know what to do... 3. To pop the new ball sockets back on the balls I used a crawbar reacting against a piece of wood held in place by masking tape. Quite easy. 4. I popped the sockets back on the balls first before attaching the other end back to the bell crank. The screws at the bell crank end in my car are Torx T40. There are two rubber washers which go on both sides of the link flange at the connection to the bell crank, one between the screw head and the link, the other between the link and the bell crank. The latter may drop down when disassembling - don't forget it! 5. (added point in edit) Strangely both T40 screws were "hand loose" to start with - I was able to undo both of them by hand... When tightening I could not get enough clearance for my torque wrench and my drive adapter so I tightened them as tight as I could with my relatively short ratchet handle. 6. To line up the link with the bell crank hole to put back the screw use brief touches on the Up and Down buttons. 7. Before getting out of service mode it's a good time to shop-vacc the cavity and water drains. Happy ending (for now, I'm sure I'm jinxing it...). Thanks a bunch to y'all for the great tips and encouragement! Oh, by the way, anyone in need of a guaranteed good front latch microswitch?... If so please PM me. Thanks!!

Well done! I agree that the aftermarket sockets might be less sturdy than the oem ones, but that's not altogether bad news. It's an intentional design for that part to fail before something more expensive would break. I had one broken one on my 2005 when I first purchased it and it's possible that the reason it failed was simply 15 years of plastic fatigue? I'll never know, but the aftermarket sockets have worked fine now for a years worth of up/downs. You know a lot more about your Boxster top and how it works now.

I had one broken on my '09 - and the top still worked. The rubber washers on mine mostly perished - one on each side was laying in the bottom of the water tray, the other was stretched out of shape. Porsche wants to sell the screws, bushing, and washers - about $20/side. I found an $11 kit of fiber and rubber washers on Amazon that had ones suitable for the use - and actually a bit better (larger OD adding a bit of strength to the rubber washer.) Easy fix. I now have a spare set of the ball-sockets in the center console cubby - just to make certain it never happens again (Eilenberger's Law#2 - You never need the part you have.. Laws ) BTW - have to agree on the strap. Mine never seems long enough so I end up using a bungee from its loop to the nubbin sticking out of the front hoop of the top.

Great job and writeup! So, we know the top is moving properly now, does that mean that the car is detecting the movement and the windows are dropping down properly as well? If so, my guess is that the sensor in the mechanism (wherever it is) is finally working, so the computer now knows whether the top is up/down (latch sensor plus this other one, since the top is moving properly now)... but I just wanted to make sure that it's all "happy" on that end as well. Again, great work!

Thank you for the kind words, and yes, all is normal. As I mentioned in a previous post, before the fix the windows did not drop when pressing on the down switch because, while the top remained up due to the broken sockets, the transmission was already in the down position (crank pointing just about straight down), so pressing the down switch did nothing, as it does nothing when the top is down. My suspicion would be that the additional sensor you found with the Foxwell may be a position sensor of the transmission? Thanks for the follow up!!

Forgot a curious detail which I will edit into post #26: Strangely both T40 screws connecting the links to the bell cranks were "hand loose" - I was able to undo both of them by hand. When tightening I could not get enough clearance for my torque wrench and the drive adapter so I tightened them as tight as I could with my relatively short ratchet handle.

The next common issue with these tops at the age of the cars is that when you are closing them, at about the 3/4 ways closing point, the "bead" at the edge of top is supposed to be precisely tucking into the matching sized channel of the hard plastic trim above the windows. But as the elastic fabric piece that "pulls" on the canvas top, in an effort to line that bead up in the channel, begins to lose its elasticity, the bead will miss the channel and become oversttessed. It can break that bead.

Oyyyy. All I can say is if it weren't for it being my wife's near-daily-driver, part of a pretty long tradition of roadsters, I would have chosen a Cayman... 😱

Hey, I like the Cayman , so don't get me wrong, but that top down option is priceless when it's priceless.