As a reminder, in part one of these aluminum block series, we mostly covered the process of reverse engineering the factory Nissan VG30DE cast iron cylinder block. We went over some features, looked at material distribution in cross section views and quickly went over things that will eventually need attention.
In part 2, we will discuss what changes need to be made to address the issues we currently have and will try to visualize potential solutions to those problems.
Let's get to it! I want to say that with all my work, I do not mean to discredit what Nissan engineers did when they developed this block. Quite the opposite actually. It is a fantastic, compact and reliable block and worked well for a huge number of production cars. I have seen multiple cars that are running these blocks clocking in hundreds of thousands of miles. Granted, those were naturally aspirated vehicles, but still the block works well for what it was designed to do (220-300hp) and then some. As a matter of fact, the block is strong enough to support close to 900whp in some of the heavily modified Z32 cars that I know. Obviously that involves replacing pistons, rods, studs and many other components, but the block itself holds up.
So when does the block start having issues? Most of the issues start happening when you are not careful with your torque numbers. Inherently, these are very torquey engines. Bottom end of the powerband is great and you barely have to rev these engines to get your car going. That is great for drivability, but when you install huge turbos and turn up the boost, all that torque becomes an enemy at high rpm. To put this in simple terms: block flexes from high torque numbers and since cast iron is not very elastic, it cracks in its weakest spot.
My personal opinion is that using excessive high grade aftermarket main studs (L19) actually hurts the block. You are technically pre-loading the material all around the anchoring in the webbing and as a result take away from the limited elasticity that the material has. Given the fact that you are doing nothing to address the actual flex, when that flex happens, you have even less elasticity before the block gives out and cracks. If you study blocks that suffered a catastrophic failure, it is easy to see a pattern that most failures happen in the webbing around stud anchoring (not surprising given the lack of material that I pointed out in part 1) or cylinder walls crack because the casting goes out of alignment and while combustion pressure is doing its part. The "out of alignment" part also contributes to spun bearings which often happen way before the block fails.
Interestingly enough, factory cast iron girdle does not fail all that often. Even when it does, it is the connecting rails that crack (again, think of what happen to that girdle when block starts flexing and where the load goes). Things go out of alignment, clamping force is reduced, girdle starts moving around, bearings/rotating assembly goes out of alignment and things start to fail. It is beyond me when people start trying to address issues that are caused by block flex by replacing main girdle with caps made of a different material. Upgrading main caps to billet steel version is done in applications when factory caps or girdles fail by cracking down the centerline. Perfect example of that is the Nissan RB engine. When you are installing billet caps on your VG block, you are addressing a problem you never had and ignoring the problem that remains. You may not share my opinion and that is ok, but notice how when billet caps became available for RB and JZ, their power levels went up significantly. When billet caps became available for VG, it did nothing to average power levels, but only put a dent in your budget after doing a ton of additional machining work. More on that later.
At the same time, the most powerful VGs (horsepower wise) are pretty conservative on their torque numbers. That further proves that the biggest enemy of these blocks is flex and barely anything is being done to properly address that issue.
So you are now probably sitting there thinking to yourself, "Ok, so why does VG flex more and other blocks flex less?". To answer that question, please refer to the three images below.
Here we have three engine blocks that powered iconic Japanese cars from the 90s. Left to right: 2JZ, RB26, VG30DE. Please look carefully at these three pictures and tell me why you think the first two are less likely to flex.
Exactly! The crankshaft centerline in JZ and RB blocks is located much deeper in the casting, where as in VG, crank centerline is pretty much at the bottom of the casting. All the extra material around the crank from extended oil pan rails is forming a rigid structure that is much harder to flex. I am not saying these blocks do not flex at all. Everything flexes at some point, but these blocks are capable of well north of 1500whp (where billet caps become necessary). Even then, the weak point is that caps are not connected to frame rails by factory design in both applications and aftermarket ladder braces go hand in hand with billet main caps to tie everything together and form a complete, rigid structure around the crankshaft.
As you can clearly see now, billet main caps will do nothing to address inherent issues that cause block failures in VGs. Until you extend those pan rails way below the crankshaft centerline and tie main girdle or caps together with those rails, the issue will remain. You need to create a rigid structure around the crankshaft. Only then everything will remain aligned keeping block materials and bearings happy.
I only brought up two of the most popular cylinder block examples for the sake of this post. You are welcome to do your own research and look at other block designs that live happily in 4-digit land. They will all have a rigid structure all around the crankshaft that ties all parts of cylinder block casting together, spreading the load evenly over large sections of material.
So does this mean that VG is hopeless and you should swap it out? Not exactly... As I mentioned at first, even when you look at VG in its factory configuration, it is still a very capable engine. It also inherently produces more torque than the other two engines, has more displacement over RB and is more compact. On top of that, VG is already proven to delivery 900whp in a fairly reliable fashion and that is not a bad number even by today's standards. In my opinion, there is actually not a lot that needs to be done to make this engine block superior to the other two, and many others. Well, actually quite a bit, but it is nothing compared to putting people on the Moon. In part 1, I mentioned that my work was inspired heavily by cylinder block development that Electramotive did for Nissan in the 90s. They did build an all aluminum block that was capable of 1200hp. Keep in mind that it was 1200hp in endurance racing where engines are being pushed to their limits for hours and hours at a time. Their block was developed for a GT racecar and was designed (for starters) to be mounted in a tubular chassis. They did not have attachment points on that block for engine mounts (instead a block plate was attached from the front), no mounting points for accessories and many other differences that made that block unusable in a factory Z32. With my project, I aim to significantly improve on what was done back in the 90s, make it affordable to a slightly-above-average Z32 enthusiast and make the block a drop in part that would work using most parts that you already have.
Lets start bringing this block into existence using some visual aids.
The whole point of this project is to address block flex. Lighter weight, better cooling efficiency, cylinder sleeves and world domination are just welcome side effects. Taking into account everything that was mentioned above, we will begin by extending oil pan rails downward by 50mm (this is modified version 1 of the block that I call PG30V1 for time being and things will likely change as we go on with development). We are also getting rid of factory cast iron girdle in favor of billet main caps (material will be determined later since I really want to run high grade aluminum, but it may not work based on expansion rates so FEA testing will determine caps' fate). We also do not want to concentrate all that clamping force in one area, so VR38 main studs will be used. There are 4 studs per cap attaching it from the bottom and also 2 more bolts will be going into the cap (1 from each side). Do you see how we are already manifesting the aforementioned "rigid structure around the crankshaft"? And this is only the front view. I will design main caps a bit later, but based on Electramotive cap design, I think I have a good understanding of how I want things to look.
Let's take a look from the side.
As previously mentioned, the bottom part of the block is extended downward by 50mm. On that extended part there are wide ridges that are located in same places as main journals. They are also extended upward into the original block model. You can also see four bolts that will be holding each cap from the side and mating it to sides of the block. All features are rounded off and fillets are used whenever possible. Sharp corners are stress points and should be avoided at all costs.
There isn't much space around the crank, but I am not abandoning my goal to run 6-bolt main caps. As previously mentioned, main and head studs will be borrowed from VR38. A nice touch for Nissan purists out there and a part you can source from ARP in three different grades, including CA625.
I have to give credit where it is due and thank one of the real MVPs in the community for sending me scans of Z32 engine bay and factory Z32 engine mounts. You are the best!!!
For engine mounts, I quickly built rough solid models that we could “mate” to the block model. Engine mounts came out very accurate. You can see how well the mount stud aligns with mounting hole on subframe in assembly renders below.
Engine bay point cloud was converted to surfaces really quickly just to give us a general understanding of body panel shapes. Now having the engine bay AND accurate engine mount models, we are able to position the block exactly (and I mean EXACTLY) where it would be located in engine bay and eliminate all guess work/speculation about bottom end fitment.
As you can see in second picture, even with extended block (by 50mm in V1) there is still a ton of space for us to work with. I will be working on dry sump bottom plate first and this clearance gives me a lot of freedom to create exactly what I want as opposed to having to cram things in and making compromises.
Once again, this post was meant to help you visualize the direction we are taking with this project. It is still very early in development stage, so naturally things will change and move around. I won’t lie, this excites me a bit more than it should.
To be continued...