I thought some people might get a kick out of seeing how the sand casting process that we used to make our blocks and crankcases works. It combines a bunch of different manufacturing techniques from rapidprototyping (commonly called “printing”), casting and machining. A mold into which the molten metal will be poured to make the engine block must be made. It has outer walls that define the shape of the outside of the block, and it has “cores” that define the shape of the internal cavities of the part. These are made out of sand that is held together with a binder material. The way these are made these days is to use a laser-sintering process. A computer-controlled laser hardens a thin layer of sand in the shape that is required for the core or mold part. Then a fresh layer of sand and binder is deposited onto the just-hardened layer and the laser makes another pass, building up the part layer-by-layer.
Sand cores for the MkII Clarke-Brayton engine
Above you can see a variety of the sand cores and molds that were used for our engine block/head. Basically, you need to have sand filling up all the spaces where you do not want metal. Building the mold is kinda like building a “negative” of the engine. All the cores are assembled together.
Special coatings are sprayed on to improve the surface finish of the metal and ensure that sand does not stick to it.
The cores you see assembled here will be passages within the engine block. Some of these passages will be to allow the air to flow through the engine as needed. Some will be for coolant to flow through making sure the engine does not get too hot.
There is a last piece that goes on top of the assembly you see above but unfortunately I do not have a picture of the whole mold ready for pouring. In designing the mold, special care needs to be taken that the molten metal will be able to fill up all the gaps completely, allowing air to escape through vents as more metal is poured in.
After the metal cools and hardens, all the sand gets broken up and cleared away, leaving just the metal part!
That part is closely inspected using 3D scanners to ensure dimensional tolerances were maintained and x-ray scanners to make sure there are no internal cracks or other problems that cannot be seen from the outside.
Next the part is put in to computer-controlled machining centers that will cut away excess material and provide all of the tightly controlled dimensions, surface textures and other features that are required to make an engine work property!
Here is a video that shows a similar process being done for a different engine block. This video was not made at the foundry that is casting our parts.
Our first block has been poured at the foundry – ACTech in Freiberg, Germany. I was tempted to title this post as, “It LIVES!” We have been working on computer models of this complicated part for so long, to finally see a photograph of a real block made of compacted graphite iron instead of 1’s and 0’s is very very exciting.
MkII Clarke-Brayton Engine block just after being poured at the foundry
The block is now being prepared for machining, after which it will look much sleeker! The bulk of our smaller parts are already out for quote and we anticipate beginning testing on August 1st. Maybe I will save the, “It LIVES!” post title for when the engine is actually breathing.
Happy Memorial Day, everybody!
I have been woefully absent from these pages the last several months since we started the design effort on the new engine. I am excited to finally be able to share what the team has been working on so dilligently. The MkII Clarke-Brayton Engine is the next step in dramatically reducing fuel consumption in trucks, automobiles and generators without increasing costs. It is also the next step in developing a highly efficient compression-ignition 100% natural gas engine that can meet the up-coming greenhouse gas emissions regulations. This is a boxer configuration split-cycle engine implementing what we have come to call the Clarke-Brayton cycle. The thermodynamics of this engine are virtually identical to our previous “CCI” design but are implemented in a much more conventional way. Everything that we published in our SAE paper at the 2013 World Congress holds true for this engine, but many of the difficulties related to the old engine are resolved.
MkII Clarke-Brayton Engine
Section of the MkII engine
As with the previous design, in the MkII air moves sequentially through three cylinders, starting at the mid-sized cylinders at the top of the section image. This architecture allows us to achieve a 56:1 compression ratio leading to a 30MPa peak pressure. It has far less surface area for heat loss than a comparable conventional diesel due to the very small bore of the combustion chamber. The small combustion piston area leads to lower forces on the crank than a conventional engine would have if it were able to reach similar pressures, reducing rod bearing friction compared to conventional architectures. A lack of net forces on the main bearings due to the opposing forces of the piston pairs reduces main bearing friction compared to conventional engines. It expands exhaust gasses all the way to ambient pressure before the exhaust stroke. Gas transfer from one cylinder to the next is begun at equal pressures on either side of the valve, which keeps velocities low, minimizing pumping losses and eliminates blow-down. The power is produced in almost a 50-50 split between the combustion (central) and exhaust (largest) pistons. There is a power stroke every revolution. All valves are actuated by overhead cams. Piston ring sealing is completely conventional, eliminating the dynamic effects of the old design and greatly reducing the reciprocating mass.
The major components of the MkII Clarke-Brayton Engine have already been released to the foundry for casting and everything else should be released for fabrication within a couple of weeks. We will test this summer at a globally renowned engine development lab and I hope to have results to share shortly after that.
A team of just three people designed the MkII from a back-of-the-napkin idea to a fully developed test engine in 7 months. Azra Horowitz and John Clarke have both put in herculean efforts to get this done in time despite a couple unexpected thorny technical challenges along the way. I could not be more proud to be working with them.