Ready for a long one?
I’d like to compile everything that is known or thought to be known about the safe limits of a stock 3.0 NSX block when subjected to forced induction. It seems to me that for anyone tempted to go beyond the typical ~6psi level, the importance of this issue is second only to fuel management. I would ask that anyone who contributes include the source of their information in terms of first-hand experience, told to you by an acknowledged expert, read in a forum such as this, etc. so we can weigh each accordingly.
I’ll kick it off with some background info and what I can offer (which is not much)
The NSX block is what is known as an “open deck” design. With the heads off, the top deck surface of a “conventional” block looks like, well, a solid block of metal with big holes bored into it for the pistons and smaller holes leading to oil and water passages. In reality of course it is cast that way then finish bored & honed. By contrast, with an open deck design the cylinders look like pipes rising up out of the bottom of a large hollow cavity in the block. The ~1cm ring of open space around the outside of each pipe (cylinder) is the water jacket where coolant circulates around the cylinder and up into the head. That leaves a vertical cylinder length of ~3 ¾ inches relatively unsupported. However, conventional blocks have similar water jackets, so the important difference here is just the very top of the cylinder, which is where all the real action takes place anyway.
Conventional wisdom states that if you produce enough HP these unsupported cylinders can cause problems. One possible problem is said to be that the cylinder itself will warp out of round, which of course leads to blow-by as rings would no longer seat, and probably scoring of both cylinder and piston. There is less specific talk of the block being unable to handle lots of HP and that it can warp or twist causing the cylinders to not be perfectly straight and perpendicular to the crank (with similar results to a warped cylinder), or the crank bores losing perfect alignment. Considering the fine tolerances involved I guess either of these is conceivable.
So the solution has been to weld on a deck plate that sets down into the water jacket and connects the top of the cylinder to the block. That plate would have holes in it to allow coolant to flow through the corresponding holes in the head gasket from the head. (this is sometimes referred to “decking the block” but historically that phrase means shaving down the top of a block to true the surface and/or increase compression.) In the end it would look very much like a conventional engine block. I’ve also seen references to what I thought were additional forms of block reinforcement but never a description of what is being done. Perhaps someone can share that with us.
So, back to the safe limits. There seems to be a general feeling that the NSX 3.0 can handle somewhere in the neighborhood of 425 RWHP without modification, but I for one only “know” that from other people. However, other blocks of similar design with more history are known to suffer similar limitations. For example, it seems well documented that the open deck block of the standard 300 ZX can’t take the strain of forced induction. (The TT block was not open deck) And yet interestingly, the 944 turbo also has an open deck but people are able to bump them up pretty high without these problems. In fact, those cylinders are totally unsupported at the top whereas the NSX cylinders are at least Siamesed (joined to the adjacent cylinder) for their full length. (funny thing, the 3.0 NA 944 engine does have Siamesed cylinders) Perhaps unevenly distributed support is worse than no support at all? But the 944 cylinder walls are significantly thicker than those of the NSX so that may be a factor as well. I spoke at length to a mechanical engineer who does deck plates for 944 engines and he said they are only necessary for a serious race engine, but he still recommends them for high output street motors since it only adds ~$500 to the build up, making for cheap insurance (and more $ in his pocket). According to him part of the problem is that the cylinders kind of move around a bit under high loads so they aren’t perfectly perpendicular to the crank, but says that would not be a concern if they were just Siamesed as the NSX is.
So now I look at all the BBSC engines already pushing 400 HP. That sounds safe, but the next step is an aftercooler and even more boost. Surely that will elevate them to or above the generally accepted limit. From what I’ve read on this forum, at least a couple of these cars have already been fitted with appropriate pistons for more boost, but I have not heard of any with a deck plate. That leads me to believe that Mark B. is comfortable with running at least 450 RWHP on a stock block. Yes?
I think the first step is to better define the measure of the limit. It’s easy to say for example 425 HP, but of course that’s a peak number and an engine is at the peak for just an instant at a time. More importantly, is HP even the right measure? I would think that torque is much more relevant because it tells you how hard the engine is working and how much stress is created per compression stroke, plus you spend a great deal of time at high torque levels. (I need to run that by some engineers like the 944 guy) This may help explain why the BBSC would be safer than a turbo with the same peak HP, a more gradual torque curve means less time at the highest levels, at least until they change pulleys and blowers to get boost early then hold it there with a BOV. At that time it will look more like a turbo curve. (there is also the concern that to make an equal RWHP figure an SC engine must actually create even more power than a turbo due to inherent parasitic losses, but that is said to be minimal with the Novi)
So now for the questions:
Head gasket issues aside, does anyone know someone who discovered or proved the limits of a stock NSX block under boost, or is this mostly extrapolation from experience with other engines?
Is anyone aware of recommended block strengthening other than adding a deck plate?
It seems like I always here about new liners at the same time that the deck is added. If the bores are in good shape and at most need to be honed for the new pistons & rings (forged pistons typically need greater clearance anyway because the expand more), is there any reason to replace the liners? I would hope you could just weld in the deck, then hone for roundness and clearance, and check the crank journals align bore. No?
BTW, I have no doubt that there is a limit, but I have no solid evidence as to what that limit really is. That’s what I hope to learn without doing it the hard way.
I’d like to compile everything that is known or thought to be known about the safe limits of a stock 3.0 NSX block when subjected to forced induction. It seems to me that for anyone tempted to go beyond the typical ~6psi level, the importance of this issue is second only to fuel management. I would ask that anyone who contributes include the source of their information in terms of first-hand experience, told to you by an acknowledged expert, read in a forum such as this, etc. so we can weigh each accordingly.
I’ll kick it off with some background info and what I can offer (which is not much)
The NSX block is what is known as an “open deck” design. With the heads off, the top deck surface of a “conventional” block looks like, well, a solid block of metal with big holes bored into it for the pistons and smaller holes leading to oil and water passages. In reality of course it is cast that way then finish bored & honed. By contrast, with an open deck design the cylinders look like pipes rising up out of the bottom of a large hollow cavity in the block. The ~1cm ring of open space around the outside of each pipe (cylinder) is the water jacket where coolant circulates around the cylinder and up into the head. That leaves a vertical cylinder length of ~3 ¾ inches relatively unsupported. However, conventional blocks have similar water jackets, so the important difference here is just the very top of the cylinder, which is where all the real action takes place anyway.
Conventional wisdom states that if you produce enough HP these unsupported cylinders can cause problems. One possible problem is said to be that the cylinder itself will warp out of round, which of course leads to blow-by as rings would no longer seat, and probably scoring of both cylinder and piston. There is less specific talk of the block being unable to handle lots of HP and that it can warp or twist causing the cylinders to not be perfectly straight and perpendicular to the crank (with similar results to a warped cylinder), or the crank bores losing perfect alignment. Considering the fine tolerances involved I guess either of these is conceivable.
So the solution has been to weld on a deck plate that sets down into the water jacket and connects the top of the cylinder to the block. That plate would have holes in it to allow coolant to flow through the corresponding holes in the head gasket from the head. (this is sometimes referred to “decking the block” but historically that phrase means shaving down the top of a block to true the surface and/or increase compression.) In the end it would look very much like a conventional engine block. I’ve also seen references to what I thought were additional forms of block reinforcement but never a description of what is being done. Perhaps someone can share that with us.
So, back to the safe limits. There seems to be a general feeling that the NSX 3.0 can handle somewhere in the neighborhood of 425 RWHP without modification, but I for one only “know” that from other people. However, other blocks of similar design with more history are known to suffer similar limitations. For example, it seems well documented that the open deck block of the standard 300 ZX can’t take the strain of forced induction. (The TT block was not open deck) And yet interestingly, the 944 turbo also has an open deck but people are able to bump them up pretty high without these problems. In fact, those cylinders are totally unsupported at the top whereas the NSX cylinders are at least Siamesed (joined to the adjacent cylinder) for their full length. (funny thing, the 3.0 NA 944 engine does have Siamesed cylinders) Perhaps unevenly distributed support is worse than no support at all? But the 944 cylinder walls are significantly thicker than those of the NSX so that may be a factor as well. I spoke at length to a mechanical engineer who does deck plates for 944 engines and he said they are only necessary for a serious race engine, but he still recommends them for high output street motors since it only adds ~$500 to the build up, making for cheap insurance (and more $ in his pocket). According to him part of the problem is that the cylinders kind of move around a bit under high loads so they aren’t perfectly perpendicular to the crank, but says that would not be a concern if they were just Siamesed as the NSX is.
So now I look at all the BBSC engines already pushing 400 HP. That sounds safe, but the next step is an aftercooler and even more boost. Surely that will elevate them to or above the generally accepted limit. From what I’ve read on this forum, at least a couple of these cars have already been fitted with appropriate pistons for more boost, but I have not heard of any with a deck plate. That leads me to believe that Mark B. is comfortable with running at least 450 RWHP on a stock block. Yes?
I think the first step is to better define the measure of the limit. It’s easy to say for example 425 HP, but of course that’s a peak number and an engine is at the peak for just an instant at a time. More importantly, is HP even the right measure? I would think that torque is much more relevant because it tells you how hard the engine is working and how much stress is created per compression stroke, plus you spend a great deal of time at high torque levels. (I need to run that by some engineers like the 944 guy) This may help explain why the BBSC would be safer than a turbo with the same peak HP, a more gradual torque curve means less time at the highest levels, at least until they change pulleys and blowers to get boost early then hold it there with a BOV. At that time it will look more like a turbo curve. (there is also the concern that to make an equal RWHP figure an SC engine must actually create even more power than a turbo due to inherent parasitic losses, but that is said to be minimal with the Novi)
So now for the questions:
Head gasket issues aside, does anyone know someone who discovered or proved the limits of a stock NSX block under boost, or is this mostly extrapolation from experience with other engines?
Is anyone aware of recommended block strengthening other than adding a deck plate?
It seems like I always here about new liners at the same time that the deck is added. If the bores are in good shape and at most need to be honed for the new pistons & rings (forged pistons typically need greater clearance anyway because the expand more), is there any reason to replace the liners? I would hope you could just weld in the deck, then hone for roundness and clearance, and check the crank journals align bore. No?
BTW, I have no doubt that there is a limit, but I have no solid evidence as to what that limit really is. That’s what I hope to learn without doing it the hard way.
