987.1 -> 987.2 9a1 Retrofit Build

Hello everyone, I am working on a new project that I have not seen done necessarily. I got a 2008 Cayman S Design Edition 1 about 4 years ago, from the day I bought it I did hear a slight tapping noise along with 3-5k rpm vibrations, as well as some O2 sensor codes. I knew all about bore scoring and assumed the day would come I would have to rebuild or swap the engine. I replaced the full clutch, flywheel, and O2 sensors which got me a few extra years, but I always had the oil consumption in the back of my head, planning what I will do about the bore score. I stopped driving the car much since It started throwing rich codes, blew smoke on startup, and had the signature bore score tap sound. I started looking into rebuilding the motor. Splitting the case is a relatively complex procedure, and having the engine re-sleeved, new timing chains, ims refurbishing, along with just plain parts would run almost 15-20,000$ just to rebuild the m97.21, which in itself is already a mechanically inferior engine to the 9a1 in absolutely every way possible. So cross that out. I looked into retrofitting an audi 4.2 engine, or something that has been done but is cheaper and different, but it seemed wrong given the car is meant to have a flat 6, and I would be "un-porsche-ing" a porsche. So cross that out. I finally did some research online and came across @blackknighti30 's work. Wow, very impressive work. I found a decent deal on a ma121 3.4 dfi engine on ebay the next night, pressed order. Now the method blackknight used was honesty probably the smartest way to go about this for anyone reading. Having a complete 987.2 donor car and a lift is probably better than what I am attempting, but I dont have space for another car, nor do I have space for a lift. oh well. The 9a1 uses a completely different wiring harness, dme, and pretty different canbus system. Swapping the DME requires swapping or re-marrying the immobilizer, along with various bits like the air conditioning controller, harness parts, etc. Piecing those donor parts together and making it all work absent a donor car is difficult, and just overall not ideal. My idea, Buy a 981 gt4 manual transmission, (The m97.21 engines have a different bell housing pattern than the 9a1 engines). and tie it to the 987.2 9a1 ma121 engine. Then instead of changing the dme, use the 9a1 ma120 2.9 engine intake manifold, which uses port injection, and create a hybrid swap adapting the stock harness, effectively converting the DFI 9a1 ma121 3.4 to port injection, then getting it all to run on the stock Bosch 7.8 DME. Now, in theory this is possible sure. but there are numerous hurdles. They do not share many parts, ignition coils, coolant temp, oil temp, Variocam+, cam and crank position sensors, hall effect wheels, those are all systems that will need their harnesses adapted to to the m97.21 harness, not to mention custom DME programming to make it all run. There are also systems the dme 7.8 simply does not know how to run, like the direct injection (which I will be deleting), the variable oil pressure pump system, along with the newer sump plate mounted oil level sensor. Even those who have swapped DME had to swap the cluster aswell, since the older cluster does not know how to read inputs from the newer style sensor. My intention with this project is to hybridize the 9a1 ma121 engine, to adapt m97.21 hardware and harness to hopefully run on the bosch ME 7.8 DME, if possible this would reduce the complexity of having to change over the yaw, ecu, cluster, ac controller, air con system, and psm systems. I do understand removing the direct injection would in effect be reducing one of the 9a1's "greatest hits" and would reduce its power output and efficiency, but considering the car will have a tune and headers, the power difference will be negligible. I plan on writing out the entire build process the best I can, there are alot of aspects to this build that are not well documented, if anyone has any insight or help to offer that would be greatly appreciated!

I look forward to your adventure. I think you're on the right track with reusing as much of the original 987.1 wiring as possible, and hopefully can get the new engine to run correctly on the earlier injection system. I wouldn't be overly concerned about the engine losing power by being port-injected vs direct-injected. Was Variocam+ (vs plain Variocam) introduced with the new engine? That might be a bit tricky to work around. Swapping the cluster without swapping the immobilizer may be an issue - the mileage reading may be an issue. I will be watching your thread for sure! I can live vicariously through your experience!

I have somewhat studied the Variocam systems, there previously was older "on and off" variocam systems on the early 996/ 986 , yes. but with the 997 and 987 generation cars, they brought out variocam plus, there is a cam lift and cam timing actuator. the Lift actuator is a simple on and off, the bosch 7.8 dme from the m97 engine can run this feature on both engines theoretically, they also both use cam timing solenoids to use PWM signals (modulating 12 volts) to control oil pressure reaching the cam sprockets, and thus varying cam timing. Now I know the newer Siemens/continental dme is much faster, so I assume there is a good change the PWM signals will be different. But that should be able to be coded to work, ill have to see what signals each sensor wants to make sure of this, similarly I can see both camshafts use similar hall-effect sensors, but I assume with the lack of IMS , the signal the 9a1 engine sends is different than what the ecu expects as far as cam timing. Both the m97.21 and 9a1 ma121 use variocam plus, but the ecu calibration is surely different, and the wiring is different too. I will have to adapt. For the fuel injection, the head will likely need to be ported as the port injection 2.9 and m97.21 3.4 have notches to help the spray pattern of the fuel stay, the 9a1 3.4 does not. I will have to grind specific grooves in the head to allow the injectors to have good spray patterns on the valves. As far as instrument cluster, if I can find a way to retrofit the stock oil level sensor with the 9a1 motor, via modification of the engine and sensor, or ive been contemplating creating an external oil reservoir to mount the sensor in; the stock cluster can remain. This is extremely important as my car has a factory design edition one cluster, I will not change. If I can get the dme to run the new motor, the cluster, can gateway, abs/ pasm /psm, etc can all stay. The manual transmission is "dumb" and does not require any dme coding, so the gt4 box should be a direct bolt in with some new brackets, its hard to say what needs to be changed yet with so little information on the internet.

Hey @Kablook , Thank you for the compliments on my build. Much appreciated. As to your build, I think it sounds like a really cool idea. I do have a couple of questions though. First, why choose a GT4 transmission? Nobody has been able to confirm that it will fit in the 987 chassis and it has much different gearing that basically makes it suffer a bit in I think 2-3 gears. A 987.2 6-speed would actually be preferable and is more than capable on track in my experience. The hardest part of my swap actually wasn't the wiring (well, once I figured out what needed to be done), it was actually the parking brake cable modification and maybe a bit of the PSM stuff. I would reach out to Scott at Softronics and see if he is able to modify the CAN version on the 9A1 DME to 987.1 before I would discount the idea of just swapping the 3.4 DFI as-is. I wanted everything to be a 987.2 in my build, but in theory, you could get away without swapping some modules out. I think if I hadn't swapped the PSM module, gateway, cluster, yaw sensor, etc. and kept everything else 987.1, it would have worked fine with the 987.2 DFI DME. You wouldn't likely get some of the features of the 987.2 that I got, like hill hold and the shift indicator on the cluster (minor things for sure). It's possible that you could get everything to work fine with just making the DME wiring changes. If you do decide to go the 2.9 intake on the 3.4 DFI engine idea, you may be able to just pick up a 2.9 wiring harness and use it since my understanding is that the 2.9 still uses the same DME 7.8 as the 987.1 Cayman S engine. That way you could treat it like a bored out 2.9 engine, in theory. I have never seen a 2.9 in person, so I'm not even sure if the intake will work on a 3.4 DFI engine, but if you've verified that, I think that's your simpler option. One problem you may have is the A/C compressor as the compressor is the same type as the 3.4 DFI, which is what required the change over of the HVAC controls and CAN bus stuff, as well as the addition of the temp sensor on the HVAC box behind the dash. I don't think you can bolt the 987.1 compressor to the 9a1 engine at all. To be honest, it doesn't really solve all of your issues with not wanting to swap the other modules. While you don't need to get a whole donor car, I have listed the things you would need to piece-meal the conversion. As for mileage on the cluster, I have an X-Tool that makes mileage correction quite easy. That shouldn't be a problem. That said, you should also be able to swap the gauge faces to the 987.2 cluster if you want easily and the X-Tool can fix the mileage. PIWIS can correct the VIN and everything as well. Cluster also has nothing to do with the immobilizer. The immobilizer only needs the key, immobilizer unit itself and DME to match. It's not hard to match the new DME from a 2.9 to your car if it's needed using PIWIS and an email to your local Porsche service department. I'll be watching this closely, so if you have questions on your way, let me know.

Thank you for your response @blackknighti30 , I truly enjoyed reading how you went about building your cayman, It helped give me the confidence to start such a swap, considering I have previously only done engine swaps with equivalents. I appreciate your insight and help on my build as-well. I spoke to a couple local porsche shops, they have done complete gt4 swaps into a 987 boxster spyder and have verified to me the drivetrain is basically a direct bolt in, part of the reason I chose the gt4 box is the price is basically the same as the 987.2 manual box, are less rare and hard to find, Part of what makes the gear ratios on the gt4 so long and miserable is the fact they run 295/30/20 rear wheels, that tall wide tire is what contributes to the long gears, reduce that down to a 275/35/19 which is what I run and that should reduce the "long ness" albeit, yes I will still be stuck with the longer gears, but it does have strengthened synchros, better cooling, it made sense in my case given I am ordering all the pieces online individually. The only real downside is longer gears, yeah. I did actually speak to Softronic before, we had a long conversation where this idea was actually somewhat pitched to me by him after hearing my goals. He is a pretty brilliant guy, walked me through the fact that in the end these motors only need air, fuel and spark to breathe and run, and in the end the 9a1 is essentially a higher flowing m97.21 once I ignore the dfi system as a whole and treat it like a better breathing m97. All things custom tuning can solve. And yes he did mention that most of the can bus issues with the DME swap can actually be solved through software rather than parts, but in the end the work, hours, and thus cost is so significant compared to either doing it my way and attempting to retain OEM dme, or your way where you just gave it full 987.2 hardware, and thus functionality. You and I essentially have the same goal, reduce the amount of problems by keeping things as close to factory as possible, albeit through very different approaches. I don't actually have the 987.2 dme, nor do I even have the right harness, the one included on my drop out motor is a 987.2 pdk harness, I think the motor came from a cayman R. Keeping my car as stock as possible is my only way of having any real cost effective-ness. The 2.9 might use a similar DME with very similar wiring, but there are just about enough differences between the 9a1 and m97.21 that trying to attempt to make the 2.9 harness work, would actually make things more complicated in my mind at least, the m97 is a pretty simple motor, both do speak a similar language too considering they run the same family of bosch sensors, with mild differences. The motor I got does have its complete serpentine package, thankfully since you are absolutely right, the power steering pump, and compressor are very much not cross compatible. But if I am keeping the wiring oem, I could really just wire the 9a1 compressor to trigger with the oem .1 wiring, I am pretty sure they just trigger off a ground wire that completes the circuit to engage the clutch. The physical hardware doesn't seem to be part of the canbus system. I know in your work you mentioned the style the power steering connectors attached changed from threads to a quick connect, did you solve that by creating a new flanged connection? or did you just swap over the complete 987.2 lines. I am still figuring things out as I go, but in theory if I can just get the dme to accept conversion files and run the 9a1 with the car thinking its an "m97", it should work, the only parts I know the 7.8 will never be able run are the variable oil pressure system, but that defaults to max pressure anyways so leaving it ignored should be fine, and some custom work should get the oem oil level sensor reading too, negating the need for a different cluster or dme, the rest of the cam position, crank position, variocam+, oil pressure dummy switch, coolant temp, knock sensors, o2 sensors, technically with some wiring changes should work as oem. I don't have your knowledge with this, so I hope what I am saying makes sense, thank you again!

Interesting and glad you're making headway. One thing about the oil level sensor is that the GT4 one is even different than the 987.2 one, so if you plan on making the oil level check in the cluster work (would need a .2 cluster anyway), you would need to replace it with a .2 oil level sensor. I found this out once I got my GT4 engine running in my car. Did you swap over the front cooling system bracket behind the water pump and the thermostat stuff to the .2 stuff? Your coolant hoses and such won't match up and there are additional sensors for the electronically controlled thermostat and such that your .1 DME won't be controlling. Also, for the high pressure fuel pump removal, you can just get one of these "lids" to block it off: 9A1-104-403-00. It's what the 2.9 would have instead of the high pressure fuel pump . Another thing, what did you do about the oil cooler under the intake? Did you swap it for the one with the power steering cooler type? Since 981s have electric power steering, they use a different oil cooler that doesn't have the fittings for the power steering lines to connect to. I just swapped mine from the 3.4L, but what did you do?

Oh yeah, total oversight for the HPFP plug , ill just order one of those instead, should help open up some space in the engine compartment too, good catch. The engine Im running is the ma121 3.4 from a 2011 Cayman R / S, just the transmission is gt4 . the oil cooler is already connected directly to the power steering pump , making it a pretty simple plug and play deal for the steering lines that go to the front of the car, very happy about that. I ordered the two main engine hoses from the 9a1 series cayman since yes, the two main coolant hoses that go to the main tubes are very different looking parts. In the meantime I took your advice and opened up as much space as i could with that parking brake cable and did some engine bay cleaning while I was there. The main issue is the signal the 3 pin oil level sensors even send is just too different to effectively translate to work with the old 4 pin connector the stock ecu expects, neither sensor will ever really communicate right, I think I may end up actually removing that sump sensor and using it as a template to design a dipstick adaptor that also doubles as an AN line connector to a external small reservoir that the oem "dipstick" style sensor can go in and be happy, well unless I can come up with a better idea in the meantime. Those newer oil level sensors output signals only newer cluster and ECU can read. On a side note, the 981 gt4 transmission is advertised as an OEM LSD unit, but i noticed turning one output flange, the other rotates in the opposite direction, just like an open diff, and I can even hold one flange with one hand and rotate the other... just like an open diff. I validated I do infact have a gt4 box with the factory LSD , but are they just that weak? I do hear the gt4 stock lsd is bad but I am somewhat confused.

Nope! the ecu does not seem to release spark no matter what I do (unplugging cam sensors to trigger wasted spark fallback program) nothing, the engine gets fuel but the 987.1 ecu does not seem to cooperate! I opened the bin file from a .1 and a .2 ecu , The reference crank sync values are similar, really similar. Both cam sensors were unplugged as a test and it still did not command spark, so I’m no longer convinced this is simply cam/crank correlation. The DME clearly knows the engine is cranking and is fueling, but something is still preventing ignition release. I compared the 987.1 ME7.8 / ME7.8.1 binary to a 987.2 2.9 PI ME7.8 .2 binary. The obvious cam/sync values appear very close: likely 987.1 cam reference angles around 121/301/481/661° crank and 987.2 around 126/306/486/666°, with phase windows roughly ±10° vs ±12°. That does not look like an obvious massive offset. Current plan is to introduce a 987.2 2.9 PI Bosch ME7.8 .2 DME , ideally 997.618.608.00 / Bosch 0261201349, and see how far I can get with immobilizer transfer/coding and PIWIS coding to make it talk to the 987.1 chassis systems. I’m especially interested in whether ME7.8 .2 can be coded/adapted enough for 987.1 cluster, HVAC/A/C request, fans, ABS/PSM/ PASM , and CAN behavior. Do you think PIWIS will get me far as far as coding the immobilizer and cluster/other chassis modules and equipment on a .2 bosch efi ecu ? You mentioned you probably could have gotten away with not replacing much of it, how so? I still think the .1 ecu is theoretically capable of running the new motor, The simple fact is finding definition files and figuring out specifically what values are wrong and why the ecu is upset and wont release spark seems like a backup plan if a 7.8.2 ecu can fix my issues "easier". Starting to understand why nobody has tried this route bahaha.

Awsome project! Keep up the good work, I think you're close! I think the me7.8 .2 conversion feels like much more work from where you are at the moment. I believe the 9x7.2 uses a different alarm/immo, and can messaging and other modules are different. The immo emulator I have explicitly states it only works on 9x7.1 dmes, and I've tried alarm sync with me7.8.1 on a me7.8 9x6 and it doesn't work so I'd prob not bother with that path given how close you are. From experience an me7.8.x doesn't seem to need much to fire the coils, and I've not seen a situation where it fires injectors but not the coils (but then I've never tried what you're doing!). Not even sure if the DME uses cam sync on crank - on standalones you generally set to crank in wasted spark (for a faster crank). Obvious I know, but when you say the dme will not release spark, you've scoped the dme output or a trigger pin on a coil? Have you counted teeth on the 60-2 wheel at TDC? I'm not sure how much you've tested, but some suggestions of what I'd be looking at if you haven't already: -physically check trigger wheels are in the right place (even though I'd expect a spark on crank regardless, just not nec at the right time!) Here are the 9x7.1 trigger angles (from maxxecu manufacturer basemap and known good on my 9x6 ): crank: 89 degrees BTDC (check the missing teeth are ~15 teeth off sensor at TDC) intake cam 1: 75 ATDC intake cam 2: -283 ATDC For info, on an me7.8 the coils will fire even if the injectors are disconnected or if there are general eskonf issues. -check live values of inj pulsewidth, spark angle, RPM and cam positions on cranking from the DME , this will hopefully indicate if the DME can read the sensors/trigger wheels - in particular interested in whether the cam angles are plausible on cranking. I'm not sure how good data is on cranking, but worth a try! -scope sensor outputs to make sure they're reading the 9x7.2 trigger wheels correctly Worst case if you're really stuck, an off-the-wall suggestion could be that you could do similar to what I do and run a maxxecu as a piggyback inside the dme case, even if just temporarily and it ultimately ends up just being helpful for troubleshooting purposes (you could use it to properly decode and scope your cam/crank signals as it has awesome tools built into the software). Unrelated, how are the intake runners (dia/length) on the 2.9 intakes vs the 3.4? Have you plugged the DI ports on the head or left the injectors in?

Thanks — focusing just on the spark/no-start side: Current ECU is still the original 987.1 ME7.8 / ME7.8.1 DME and original 987.1 chassis/immobilizer. Engine is 2011 MA1/9A1 3.4 converted to port injection. What we know: DME wakes up normally. Tach moves and Autel shows RPM during crank. Throttle/load data is present. Fuel pump and injectors are active. It is getting a lot of fuel; exhaust/garage reeks of raw fuel. No cough, pop, kickback, backfire, or combustion at all. No DME codes on high-level Autel. CEL is on key-on, goes out when cranking starts, then comes back after roughly 5 seconds of cranking. Coil 12V constant feed is present with ignition on, We checked the coil control/trigger side with basic tools and did not see any 12V pulse/activity while cranking. I have not yet scoped the actual DME coil output, so I agree that a scope check at the DME/coil trigger pin is still needed. Cam sensor test: I unplugged both cam position sensors. It still cranked, still fueled, still no start/no combustion. First attempt sounded slightly different, then went back to the same dead crank behavior. This makes me question whether the issue is simply cam phase/cam sync, since with no cam signals I would expect ME7 to at least attempt crank-based/wasted spark. Physical trigger observations: The 9A1 crank and M97 crank look very similar, but the crank dowel/index position appears 180° different. The crank sensor mounting position is also very different between the engines, so I do not yet know the net missing-tooth position relative to the sensor/TDC. Cam trigger wheels appear very similar, maybe identical. I have not yet physically counted teeth at cylinder 1 TDC. Your 89° BTDC / ~15 teeth off sensor reference is probably the next big check. Question: Given ME7 usually fires coils with very little required, what would make it fire injectors and show RPM but apparently not command coils? Does the CEL going out on crank then returning after ~5 seconds suggest it is initially entering crank/start mode, then losing/aborting some reference/scheduler state? Next checks I plan: Scope DME coil output pins during crank, not just test light/meter. Scope crank sensor signal at the DME. Set cyl 1 to true TDC and count missing-tooth position relative to the sensor. If I can get PIWIS working, log RPM, injection time, ignition angle, cam positions, and any sync/reference status during cranking. Regarding the immobilizer - digging through petka, i see both have the exact same part number for the ignition starter switch, now where it gets weird is the 09 cayman base calls for the 99761817216 or 99761817215 part number, the .1 calls for a 99761817214, 99761817215, or 99761817216 part, this heavily implies the new one is just a updated part number or something as the old part number is discontinued, rather than an actual hardware difference they will require a different DME If the missing-tooth position is not close to the 987.1 reference angle, I’m wondering if the DME sees RPM and allows fuel but never gets to a valid ignition scheduling state. I am working to get my hands on a PIWIS 3 laptop, having issues getting that set up, but. I am kind of at a loss of where and what to do next, i have spent dozens of hours trying to create a DAMOS or definitions to compare bin files, but nothing seems to exist between comparing base maps, I did get my hands on a .2 ecu, but you seem to be confirming my fears of a whole level of can bus integration issues. I spoke to promotive and softronic, first softronic said my stock dme can do it and its simple coding, but then he ghosted me and I never heard back, promotive seems to think I should get a .2 ecu, code the immobilizer, then see if it starts on that, then pay to have both ecus shipped in to have the rest changed over to resolve integration issues, and nobody else in the US or even europe seems to know anything about these ECUs to my research.... A truly weird situation. Any ideas?

Afraid I don't know the answer on what would make the injectors fire and not the coils - given they're all switched on in eskonf (which they are), you've got signals coming in (which you probably do) and no DTCs- your guess on signal timing would be my bet. I'd have thought cam trigger timing issue would cause a DTC, but sometimes the DME isn't very good at logging DTCs on cranking. Datalogging the cam angles should help. Also not sure on CEL behaviour, it also varies by how the error class is defined in the flash which varies by region (I'm more familiar with EU calibrations as I'm in the UK). You shouldn't feel at a loss and I don't think it's that weird - you more or less know it's either cam or crank sensor timing assuming it's picking up the cam trigger wheels. If it were me I'd first focus on this - what's wrong with the sensors/hardware side - you know if you fix that it will work and it shouldn't be too hard. Start with the physical measurements, and if all good, whether the dme can pick up the cam signals and what they look like. Once you know exactly waht's wrong, you can figure out best course of action. Hopefully it's just a case of sorting the trigger wheel on the flywheel - with the sensor bellhousing vs crankcase etc I wonder whether it's just this (and this is easily enough overcome). If it's the cam sensors it might need a bit more head scratching on how to fix You also should try to get the 7.8.2 dme to work given you've got one - what's holding you back from trying it? Reading your last message, maybe it's more similar than the me7.8 and me7.8.1 and at least the immo will work. copy your me7.8.1 immo eeprom dump onto the m27.8.2 dme and see if it starts! Don't touch your alarm module (unless you've taken eeprom backups etc) or do any alignment with piwis on your existing module given you have a working sync'd dme /alarm/key setup. If you've got a dump of the flash from the me7.8 .2 I wouldn't mind a look - if you've got the immo eeprom too I can compare to a 9x7.1 . I saw your nefmoto post too, I think looking at the dme side of things is going to be a bit of an undertaking (and I say that as someone who's quite familiar with them). If you want to log cam syncing, take a look at RAM variables and see if there's anything you can log. I'm not sure whether you can alter the trigger wheel parameters in the calibration or not, I've never looked at this. I don't have any definitions for the 9x7.2 dme unfortunately, they're not that common.

I compared the cam trigger wheels, the sensors are both Camshaft position sensor - 9A1 and m97.21 = Bosch 0232103092, So the sensor itself is identical, the trigger wheel looks similar, should not be a major issue. The crankshaft position is a different sensor, but uses a very very similar bosch hall effect design and an identical connector, The m96 -m97 engines have the crank sensor on the bottom of the engine, different then the 9a1 which has it mounted closer to the intake, on the bell housing of the transmission, around 10-o-clock. I have checked the 12v constant, the coils are getting power, i see no 12v pulse on crank, but I tested with a standard bulb tester, not an oscilloscope. I presume all engine grounds are good as the throttle body, starter, injectors, all fire and use the same grounds as the coils. the primary difference between the m97 and 9a1 coils is an additional ground, even which so, by function a coil pack is a very simple device - I would not expect there to be an issue at the coil pack itself. I have only been able to test with a autel scan tool, which verified the ecu sees rpm, but on the Autel , while cranking camshaft specified angle, deviation, all of it just said 0 KW. Maybe the ecu just did not have those values out on crank, maybe the autel was not reading them right. My next course of action will for sure be getting PIWIS running then from there, I can hopefully determine if It is a hardware issue, or the software in the 7.8.1 locking me out. I do not own any eeprom coding tools or dme bin file extractor tools, and I really don't want to brick my dme , or the 7.8.2 dme , since both would be really hard to come by - With that being said, I have dug online, alot. and have found what seems to be a 7.8.1 bin dump, and a 7.8.2 bin dump, from a m97 motor and a ma120 motor, I tried using a me 7.8 definition file I found from a 996 (the 7.8 is more different from the 7.8.1 then it seems), but it did not get me too far, from what I decoded, the crank signals are not that different? one of my previous posts had my specific findings, these are not bin dumps from my dme , but they might have information that can show what is and is not compatible to the right eyes. It seems @blackknighti30 has successfully programmed the .1 immo, and even psm to work with the siemens unit, which is notoriously far more difficult to work with than any Bosch ME unit, only really leaving the AC and Sport mode can lines as "majorly" different, as it seems the dme uses a different protocol to trigger the compressor (despite both models having a very very similar 12v clutch trigger simple mechanism) for the compressor on. - I am starting to think the best of both worlds would be finding out if the 7.8.2 ecu can retain its original "engine values" whilst adopting the rest of the 7.8.1 data, easier said than done, if even possible. Cayman Files - - Google Drive drive.google.com The right eyes might get far with those, for me on winols, the farthest I got was infinite hex values hundreds of German names

I don't think altering CAN comms is as simple as setting addressing/copying over from the 7.8.1 in the calibration data, I expect you'd have to disassemble and mod the code. If the CAN architecture is significantly different (I really don't know enough about the differences between 9x7.1 and .2 to know the answer to that) it would likely be a much bigger task than fixing signals to run your original dme or using something like a Feather to do can address translations to keep other modules happy. If there are rolling codes and checksums etc CAN translation can be a bit of a task, hence why I think it's easier to just fix your trigger wheels/signals so your original dme is happy - this shouldn't be too bad and the end result would be nice and factory - if you diag with porsche tools everything will look original etc. I would be surprised if there's something 'locking out' the dme , given it worked on the old engine I'd put money on it being just a case of giving it the signals it wants. Having said that if i) the me7.8 .2 CAN addressing is largely the same and ii) it works with your alarm module, that would be a great solution, so I think esp seeing as you have one it's got to be worth a try to at least get sight of what the challenges would be. Shame we're in different countries, I'd be happy to copy your 7.8.1 dme (immo) eeprom onto the 7.8.2 dme so you can try to start it. By copying the eeprom you'll keep your original chassis no, spec info, operating hours, overrevs etc as long as the eeprom data is the same format and you'd be able to use both on the same alarm box. Obviously all contingent on the eeprom data being in the same format between 7.8.1 and .2. On other bits, the crank sensor is a VR sensor. I'll check whether my car shows cam angles on cranking when I have some time. Before anything else I'd def do some closer looking into what the dme doesn't like about your trigger wheels, that has to be the easiest bit that you've not bottomed out - just pop the sensors out on both engines and check the positions of the trigger wheels. If that doesn't reveal the issue, do some measuring with a scope.

I am following your thread as a bystander as honestly this topic is way over my head However, on another forum that I am also part of it, someone posted the link to this guy, he had a good experience and was able to solve his problem http: //www. carpro. si/ Hopefully it will help, regards!

Well, my string of luck is a special type of nonsense. Got piwis working, pulled a bin dump of my DME with durametric cable, paid a professional Porsche tuner to pull the programming codes and immo codes, which they were in a Hex format, so I used AI to decode them, got something like ME7.8 .2 donor/current DME : 34 17 52 00 00 00 -> 341752 67 34 20 CE 17 00 00 00... -> 673420CE17 ME7.8.1 original car DME : 30 CB 4A -> 30CB4A 5C 30 61 82 CB 00 00 00... -> 5C306182CB went into piwis , went into the dme coding, put in manual, usa, coupe, just like the piwis manual said, went to the screen where it wanted the old and new values, put them in, annnndddd nothing. it said programming failed before even getting past 0%. to make things better? the engine fans started running on full max, with the cluster showing a whole bunch of brand new horrible errors with a bing noise ive never heard the car make before, amazing. truly,. I cycled the ignition, now Piwis would not even recognize the dme and would not find its hardware or software number, and the car in ignition on had the fans on max until the ignition was tuend off, seems like Piwis bricked its very own dme , the 7.8.2,. And yes the car was on a factory approved battery charger, and yes all the other parameters were good. so yeah,. going to mail out the dme to a professional to re-flash to hopefully remove the brick and fix the immo, this car is pretty ridiculous, and PIWIS seemingly bricking its own ecu , is just the icing on the cake. It seems going in @blackknighti30 direction of putting in the .2 harness, with the .2 siemens unit is most certainly the better choice at this point, unless I later find out there was some kind of wiring mishap or some hardware issue that was my fault causing the no spark, but rest assured I checked the wiring diagrams 20+ times, the .1 and .2 caymans have the same exact pinout and wire colors for the coil packs, the only additional wire is for another ground. It seems the .1 ecu also pulled that 5.25 kw crank angle too, it probably just can't reference it to the cams. idk. The 7.8.1 idea on efi converted ma121 idea was originally pitched to me by Scott from Softronic, when i called in asking about how I should go about getting a siemens unit, i was told that the bosch dmes are far easier to work on and work with, a custom tune is all that will be needed, aside from converting it from dfi to efi, low and behold now that the engine is in, cranks but no start, all my calls to get his help tuning and emails get ignored, not once can I get through, His only response being "try getting in a 7.8.2." All the race shops in my area say the same thing, he plays games like that, then ghosts you when you need him/help. Now I am down the rabbit hole deep enough I arguably couldve just jammed in an rs5 motor, or an ls. regarding your comment on the trigger wheels, the 9a1 crank position sensor is in an entirely different location, and the wheel is the same tooth count, but the -2 notch is in a different area too, otherwise the crankshafts themselves are near identical, and the motors themselves share very similar architecture, its pretty ridiculous that you can't change cam/crank sync values in the stock dme, well thats at least what I am told. Ive spent more time trying to figure out these DMEs than anything else, and turns out nobody seems to know what to do in my case.

Hello, I received conflicting emails on whether the car was running or not... Some from what I thought was your wife or girlfriend said you had it running and she wanted to buy the tune from me for you as a surprise. The other from you said you had issues... I had called and no answer and sent the email you refer to or one of them and said you could if you wanted as the IPAS is the same get a 7.8.2 DME on EBAY inexpensively and change the IMOB codes and check your signals. I asked as I just mentioned that someone said it was running though? Was it or not? I didnt get a reply until I noticed this post that it isn't running and the 7.8.2 DME was bricked. From what I'm reading it sounds like an aftermarket PIWIS was used and that will brick many different modules if it's one that uses the FTDI DLL and not the SAMTEC ones as the factory does. Basically it went in a second and scrabbled the data.... Generally those DME 's with a factory tester wont brick for that change and you simply then go back in on an actual bricked one and select DME and then the only other option shown is Programming and write back to it. I would assume you still have the original one you're having issues with still though. The DME code is 6 digits alphanumeric and remember there are no O's in binary and the IMOB code is also the same type in 10 digits. Should you put the wrong codes a real PIWIS will not do anything and give you a code its incorrect as it checks them first in which yours didnt. The new and not the old wrong code even entered will just not allow the car to start yet the fans won't default to high. The fans do this when the other control units can't find the DME in the CAN com. This would generally happen in a failed tune writing, bad checksum etc yet again you were only changing the IPAS data and not programming the MAP region. There had been another article done and all noted back in 2018 or so on this type of conversion using a DFI lower and the fuel injection from the .1 car yet I can't find it. Ive also done it for several shops... The 7.8.2 was used in the 2.9 and the 997.1TT. Let's not confuse ghosting with conflicting and confusing emails as I mentioned as I'm very busy and have people months later expect me to remember everything we discussed. I have at least 10 in any given day of these from people, shops and tuning outfits and also manufactures asking similar questions on files and conversions etc. This has been for well over 20 years with emails and many years prior on the phone . Lastly keep in mind I spent a lot of time on the phone discussing and emails about the whole idea and you haven't purchased anything from me.... So ghosting people as normal to whom I have been helping almost would imply you bought a product and when it went south I didnt help you.... You didnt. Best of luck in the project and if I can find the post on the conversion I will post it.. I also worked with Paul aka blackknighti30 on his with some questions etc. Best, Scott

Hi Scott, It does seem like this response is more here to save face then it is to actually resolve any issues, so lets just clear up a few things. 1- I have no idea who you are mentioning as far as a wife or girlfriend, I am the only one who contacted you regarding my car. 2- The car never ran. I will say, I was initially very impressed. You called me back promptly and we talked about the project for a bit, I asked you about potential CAN issues with stock chassis components and the Siemens unit, you said they can be programmed to work, but its arguably far easier to simply run a .1 dme , with the ma121 running an adapter plate or the 2.9 manifold to allow the .1 ecu to fire the motor up. The phone call ended with me specifically asking you "what tune should I purchase" with you responding "I have some 987.1 conversion files on my website, but the files for this will be different so call me once you get it together and we will go from there" You never answered since. Not one phone call, not one voicemail, nothing. Why give someone an idea you claim you have made work, then when they ask for your help finishing it, vanishing from thin air. I specifically stated over voicemails "which tune should I purchase off your website so we can get started." This went on for quite literal weeks. As a matter of fact, some of my messages showed you left them on opened, that is not the same as being too busy to even see them, that's ignoring them. Finally way later, you answer "You may have to do that with the DME yet I do not recall an issue with the last one.... Just get a 7.8.2 and program the Immobilizer for it and try it." -> directly the opposite of your original advice of "we can make the 7.8.1 work." I even contacted professional race shops in my area to ask if they had me 7.8.1 tuning experience, or maybe a direct contact to you since the Porsche tuning community is so small, and I needed to buy your tune and get your help. Some specifically told me that they've worked with you in the past, and now don't due to the exact issues of ghosting amongst others. Actually not a single race shop around here even works with you. I was referred to another Pro tuner, who by the way responded to me in minutes, told me which package to purchase, and has been amazing since. Its easy to sit down here and write down excuses, it's even easier to keep your word and support your customers, and sell some tunes in the process. If you wonder why I didn't purchase anything, maybe answer your phone, or email, or something and tell me what to buy next time. You could have easily returned one of my calls, spoke to me man to man, Id have left your name out, instead you chose to put more effort into reading my thread, and writing that thread response covering up your name and reputation, rather than actually helping solve any problem, your true intentions show, its not great. Truthfully, I am not here to put down anyone's business, I am here to make a ma121 run in a .1 Cayman S chassis. Apologies if this message isn't what you wanted to see. This thread is full of useful information, with everyone you see above being an amazingly kind, smart person pitching in to help, and I am documenting the entire process so in the future others with bore scoring can have more options. Please allow us to keep the thread focused on the car, and refrain from muddying it with other matters within it's contents. Thank you to everybody else who has stayed tune to this thread so far, and to those who have offered advice or help, I truly appreciate it.

I'm not making any excuses and have no reason to, or even to post to this....Ive been doing the first conversions in the US since 1999 and still make files for the new cars. I didnt sell you anything or tried to and you didnt buy anything. I simply was answering questions you had on my own time , before,during and after you installed the engine. It sounds like whatever you have isn't working and why you made a post about using the 7.8.2 DME . I suggested checking the sensor as it shouldnt be an issue as the 7.8.1 has worked fine... The only muddying up the waters was your two posts. This was my last email to you that I didnt get an answer back on about the confusion on buying the software as I called that number back also.: "It looks like I have two people doing the same thing and his works and yours doesn't..... Or all about the same car? Do you know of a phone number that starts with 507?" This was in regards to the conversion that was done and running and wanted a tune for the car with the same specs . This one was done on a 2006 Caymans S we used the 2.9 intake system as I mentioned on it and was Matt W's car in 2018 and he had an extensive article on this site or perhaps another... : This was a note to me in his order after we had it running as I found his emails. 2006 Porsche Cayman S with MA1.21, 3.4L DFI engine (Block, heads, exhaust ). MA1.20, 2.9L Intake using EFI injectors. M97 3.4L DME and wire harness Engine runs well and passes emissions, but fails readiness condition on O2 Sensors , Evap system, and Secondary Air. The Secondary Air pump is not installed but will be eventually. Am looking to correct the engine programming and pass inspection. Had to add this as I dug it up on one of the conversions in my emails. ....Have a Happy 4th... I will not post further and good luck. Best, Scott

@tim_s @blackknighti30 Big Update!!! Car runs! Ended up going to the 7.8.2 DME, I made a complete oversight in my original work, the 9a1 uses entirely different coil pack fire strategies than the earlier coil designs, that was the no spark issue. I assumed there was only an additional ground for the 4 pin v 3 pin coil harnesses, which is true, but the reason there is that second ground is the 9a1 coil packs are "smart coils" unlike the older style coil packs on the .1 cars. This is a pretty common issue people encounter when attempting to use audi R8 coils on n54, or other models. To conclude: the 7.8.1 dme is fundementally incompatible with the 9a1 engine, unless you put in a translator for the spark signals, and even so, the timing values may fundamentally be different. The 7.8.2 natively supports those newer coil packs. New dme was programmed, turned on first crank. I initially had a really rough idle and stumble along with these codes 1- P011 (A camshaft position) Bank 1 signal implausible 2- P1517 Throttle adjusting unit adaptation canceled due to environmental conditions 3- P0660 Intake manifold tuning valve no signal bank 1 4- P0663 Intake manifold tuning valve no signal bank 2 5- P0197 Oil temp sensor below limit value 6- P0523 Oil pressure sensor (Exceeds limit value) 7- P1152 Oil pressure regulating valve activation no signal I also have all the abs and psm lights on ofc, thankfully in the .bin of the .2 file I can see there are options for the old and new abs and psm and air con modules, I just dont know how to program the dme to look for the older units to fix the issues. I solved the rough idle and got things running smooth by fixing a twisted wire on the intake bank 1 cam actuator, that solved codes 1-2, super easy. Car runs good now, idles smoother than before The intake manifold tuning valve no signal are just since they are connected to different pins on dme connector C, simple change of pins and those should go away. The last three codes are the more complicated ones, the 987.1 harness never had provisions for a pressure/temp sensor, a pan mounted level sensor, and mainly a variable displacement oil pump. I can retrofit those into my harness, well mostly. Weirdly on the 987.2 the oil pressure sensor sends signal to the cluster, which sends the pressure signal back down the can bus to the dme, weird idk. and my cluster is a .1 cluster, so it does not even support any of that. The temperature part of the sensor and the regulating valve can be wired directly to the dme. From what I hear the oil pressure valve is mainly a tool porsche put for efficiency, and is a pressure solenoid designed to counteract a spring in the pump to reduce pressure, thus unplugged it should give the engine max pressure (probably a good thing?) See page 4 of this pdf, its a great explanation on the 9a1 oil system https://automotivetechinfo.com/wp-content/uploads/2018/11/Evolution-of-Porsche-Oiling-System-Part-2.pdf?utm_ Anyways, I can wire in the valve, and even without the cluster sending the pressure signal down the CAN the DME has a backup protocol to run the valve without any pressure signal, sure it isnt the most ideal, but every 986 and 987.1 never even had an oil pressure sensor to begin with, maybe ill add in a aftermarket proper gague, not sure, undecided. Trying to mess with a new cluster when I already have chrono and pccm+ etc is really not fun, but I guess it would also let me bring back the new level sensor. I haven't decided if i should wire in the oil pressure regulating valve in or not, if its sole purpose is mild efficiency gains, seems like a waste of time, otherwise if the engine needs it, I will. Other than that, it all feels pretty good, the gt4 box feels heavenly compared to the stock shifter, gears are longer for sure. I haven't gone past 4k since it isn't tuned yet, but ill smoke test it and see what else we can iron out to get things 100% the only other issue is I hear a tiny little rattle noise when I give it hard throttle and let off immediately for maybe a half second, not sure if that is normal and just since I have all the engine covers off, or if theres something unhappy with the motor, otherwise no knocks or ticks or anything,

Nice job figuring out the issue!

Ah great job! Glad you got to the bottom of it, sounds like a good result -you must be relieved! In terms of running the 987.2 dme with your existing alarm module, does that presumably mean the SDI dme from the 3.4 would too - so if you re-wired you could run the SDI3 from the 3.4 and keep direct injection? How does it feel in general running port injection?