I recently started taking ADHD medication and now that I can actually focus on things I feel confident enough to try to undertake a really crazy project.
I acquired an L28E engine from a 1980 280ZX in good condition with only 67k miles on it to try to build something crazy. After doing a LOT of forum reading about dual overhead cam swaps over the past month I ran into 4 options:
1. Weld two four-cylinder heads together
2. Cast your own head
3. Head off a Mercedes M104 inline 6
4. Head off one cylinder bank of a Mercedes M120 V12.
Welding two heads together or casting something custom are both things I've seen videos of working but aren't possible for me. I have heard tell that the M104 and M120 engines have the same cylinder bore and head bolt pattern, but I haven't been able to 100% verify any of that. The M104 head has the intake coming in on the same side as the stock L28E, but the exhaust comes out on the other side. The passenger side (right) cylinder head off the M120 is the same, but the driver side (left) cylinder head has the exhaust coming out on the same side as the L28 with the intake on the opposite side. This is preferable because I won't have to relocate the oil filter or starter or anything to accommodate the exhaust since it will be in basically the same place, and fitting the intake on the other side shouldn't be too hard, especially with what I'm planning. (More on that later.) With all of that in mind, I am pretty set on using the left side head from an M120.
There are obviously a couple of big issues with sticking a 90s Mercedes head on an 80s Nissan block. I am leaving out cams and timing chain assembly for reasons that I will get to later. The main issues all come from measurement uncertainty in a lot of places. The cylinder bore for the M120 is 89mm and the bore on a stock L28 block is 86mm. That is obviously enough to cause issues. I've heard that boring an L28 block out 3mm is possible but is a lot like playing Russian Roulette. I'm fairly certain that the bolt pattern is the same, but the oil and coolant channels definitely aren't. I considered making a spacer between the block and the head which could make the channels flow into each other, but that would change compression and require two gaskets. I have no machining experience so I have no idea if it is possible to mill the block and the head to get the channels to fit. I know that external oil lines are doable, and I can't think of any reason external coolant lines wouldn't work. I have a set of 6-2 headers and I was thinking that I could probably get a spacer fabricated that would act as an adapter for the presumably different bolt patterns, but full custom headers probably wouldn't be a bad idea. The intake manifold will definitely have to be custom. There's some fuel injection stuff I'll have to figure out with that at some point because there is absolutely no way carbeurators would work for this at all.
Let's assume that the bolt pattern is the same and works, I can bore out the block, and i can run external oil and coolant lines. The head is now bolted on to the block with stock L28 crankshaft and connecting rods but 89mm pistons to fit the bore. I'd have to measure the compression with the new head, but I've heard tell of fairly stock L28s surviving very high compression ratios, so I'm not too worried about that.
By this point, everyone is probably wondering why I haven't talked about camshafts or timing chains, since I've heard those are the hardest parts of a DOHC swap. Here is where my idea goes from simply crazy to quite possibly delusional. I don't plan on having camshafts or timing chains. I'm sure some people here have seen THIS video. With my CAD experience and programming experience, as well as access to Wesley Kagan's files that he used on the Miata build, I believe it will be easier and less expensive for me to build a Freevalve system than it will be to make custom camshafts and a custom timing chain assembly.
All I will have to do for the Freevalve system to work will be to fabricate an adaptor that can bolt on to the Mercedes head, fabricate holders for the valves, figure out where Wesley got solenoids that can fire fast enough to run an engine, and add a bunch of sensors for crank position and some other data the Freevalve system needs to time itself correctly. I can use his file as a guidline and pretty much just adapt it to have the right hole patters and angles and such, in addition to adding two more cylinders. I want to go to a Megasquirt ECU system, and I am curious as to whether or not that could run a Freevalve system or if I would have to run it off an Arduino like Wesley did.
I was thinking that for the intake I would just run carbeurator style velocity stacks with holes for the fuel injectors attached right onto the intake ports of the head. With full programable control over the valves, I should be able to program them to function in place of throttlebodies by only letting in a specific amount of air. It would probably be a lot easier to just have a traditional intake manifold, but this would be very unique and take up very little space so I want to at least try it.
Anyway, I'm 17 and anyone who knows me describes me as being simultaneously the smartest person they've ever met and the dumbest person they've ever met so please feel free to tell me exactly how stupid my ideas are, as long as you throw in your two cents on how to make it better. Specifically I would really appreciate any 3D models of an L28 block, diagrams of a stock head gasket with measurements, and advice on machining since I have no experience with that.
Thanks for reading this excessively long post! I'll be posting pictures of the teardown of my L28 and of the Mercedes head once it arrives. I plan on making all of the CAD files and software for this project completely open source so the community can improve on the design and build their own versions.
I acquired an L28E engine from a 1980 280ZX in good condition with only 67k miles on it to try to build something crazy. After doing a LOT of forum reading about dual overhead cam swaps over the past month I ran into 4 options:
1. Weld two four-cylinder heads together
2. Cast your own head
3. Head off a Mercedes M104 inline 6
4. Head off one cylinder bank of a Mercedes M120 V12.
Welding two heads together or casting something custom are both things I've seen videos of working but aren't possible for me. I have heard tell that the M104 and M120 engines have the same cylinder bore and head bolt pattern, but I haven't been able to 100% verify any of that. The M104 head has the intake coming in on the same side as the stock L28E, but the exhaust comes out on the other side. The passenger side (right) cylinder head off the M120 is the same, but the driver side (left) cylinder head has the exhaust coming out on the same side as the L28 with the intake on the opposite side. This is preferable because I won't have to relocate the oil filter or starter or anything to accommodate the exhaust since it will be in basically the same place, and fitting the intake on the other side shouldn't be too hard, especially with what I'm planning. (More on that later.) With all of that in mind, I am pretty set on using the left side head from an M120.
There are obviously a couple of big issues with sticking a 90s Mercedes head on an 80s Nissan block. I am leaving out cams and timing chain assembly for reasons that I will get to later. The main issues all come from measurement uncertainty in a lot of places. The cylinder bore for the M120 is 89mm and the bore on a stock L28 block is 86mm. That is obviously enough to cause issues. I've heard that boring an L28 block out 3mm is possible but is a lot like playing Russian Roulette. I'm fairly certain that the bolt pattern is the same, but the oil and coolant channels definitely aren't. I considered making a spacer between the block and the head which could make the channels flow into each other, but that would change compression and require two gaskets. I have no machining experience so I have no idea if it is possible to mill the block and the head to get the channels to fit. I know that external oil lines are doable, and I can't think of any reason external coolant lines wouldn't work. I have a set of 6-2 headers and I was thinking that I could probably get a spacer fabricated that would act as an adapter for the presumably different bolt patterns, but full custom headers probably wouldn't be a bad idea. The intake manifold will definitely have to be custom. There's some fuel injection stuff I'll have to figure out with that at some point because there is absolutely no way carbeurators would work for this at all.
Let's assume that the bolt pattern is the same and works, I can bore out the block, and i can run external oil and coolant lines. The head is now bolted on to the block with stock L28 crankshaft and connecting rods but 89mm pistons to fit the bore. I'd have to measure the compression with the new head, but I've heard tell of fairly stock L28s surviving very high compression ratios, so I'm not too worried about that.
By this point, everyone is probably wondering why I haven't talked about camshafts or timing chains, since I've heard those are the hardest parts of a DOHC swap. Here is where my idea goes from simply crazy to quite possibly delusional. I don't plan on having camshafts or timing chains. I'm sure some people here have seen THIS video. With my CAD experience and programming experience, as well as access to Wesley Kagan's files that he used on the Miata build, I believe it will be easier and less expensive for me to build a Freevalve system than it will be to make custom camshafts and a custom timing chain assembly.
All I will have to do for the Freevalve system to work will be to fabricate an adaptor that can bolt on to the Mercedes head, fabricate holders for the valves, figure out where Wesley got solenoids that can fire fast enough to run an engine, and add a bunch of sensors for crank position and some other data the Freevalve system needs to time itself correctly. I can use his file as a guidline and pretty much just adapt it to have the right hole patters and angles and such, in addition to adding two more cylinders. I want to go to a Megasquirt ECU system, and I am curious as to whether or not that could run a Freevalve system or if I would have to run it off an Arduino like Wesley did.
I was thinking that for the intake I would just run carbeurator style velocity stacks with holes for the fuel injectors attached right onto the intake ports of the head. With full programable control over the valves, I should be able to program them to function in place of throttlebodies by only letting in a specific amount of air. It would probably be a lot easier to just have a traditional intake manifold, but this would be very unique and take up very little space so I want to at least try it.
Anyway, I'm 17 and anyone who knows me describes me as being simultaneously the smartest person they've ever met and the dumbest person they've ever met so please feel free to tell me exactly how stupid my ideas are, as long as you throw in your two cents on how to make it better. Specifically I would really appreciate any 3D models of an L28 block, diagrams of a stock head gasket with measurements, and advice on machining since I have no experience with that.
Thanks for reading this excessively long post! I'll be posting pictures of the teardown of my L28 and of the Mercedes head once it arrives. I plan on making all of the CAD files and software for this project completely open source so the community can improve on the design and build their own versions.