Venturing into the wood(s) – Literally
Up until this point in my career, my knowledge of woodworking consisted of what could be nailed, screwed and beaten together from 2×4 lumber. I suppose the project I am writing about tonight, to the experienced tree widdler, would not be classified as woodworking, but for lack of a better term, I will call it that.
Our friends over at Spage Sport asked us to help develop a universal front race car splitter. Starting with base dimensions and ideas from Spage we started working on some concepts and running through some CFD tests of the features. The original designs had a lot of inspiration from common frontal ground effects found in Motorsport which we already know function well. Being that this is to be a universal item we were limited on how radical of a design we could produce.
An example of one early revision is shown below. The splitter would feature a centre element and mirrored outer geometry. The centre element would bring in as much flow as possible, constrict, then diffuse. The idea here, being that we create as large of increase (Delta) in the velocity of the flow under the splitter as possible. The area of increased fluid velocity creates a low pressure zone.
Being that we did not have a full car model to work with for this splitter design, we had to create a mock system in which we would use for analyzing the splitter variations. This has both positive and negative effects on the design. Being that this is a universal item, it will work differently on every vehicle in which it is mounted so having its performance linked to a certain body shape is not necessarily ideal, but analyzing it without a full vehicle body behind it limits our analysis’s realism (you could say).
Our mock system included a very conservative, generic upper bumper shape as well as a road surface 3 inches below the lowest point of the splitter, to simulate a reasonable ride height.
Moving through a series of design changes and tweaks, running analysis at each stage, we arrived at a design we were satisfied with.
Our design was refined based on both raw data from our analysis as well as visual plots. The data we looked at was the flow velocity delta, as well as the static and dynamic pressure values across the lower surface of the splitter. We also payed very close attention to the visual pressure and velocity plots in our mock system. Of course – the images I have were taken of my computer screen, with my cellphone, brilliant.
Below is one of our fluid velocity plots. It is quite evident, the principle discussed earlier in this article. In the image, the velocity is identified by coloured streamlines ranging from blue to red. Blue is representative of the nominal velocity in the system (i.e. vehicle speed) and red is representative of the maximum velocity in the system with a range between them. Being that pressure is inversely related to fluid velocity, it is clear that with this splitter, a low pressure area (downforce) is being developed beneath it.
Our final design features a single central splitter throat and outer contoured diffuser elements. At the leading edge of the splitter, each of these features have slight inlets which we found promoted flow and ultimately allowed us to develop greater pressure differentials. Below, is a cross section image of the splitter. The leading edge is on the left and the diffuser exits are on the right. Note that this entire splitter is a single 3D contour.
In any design project, you arrive at a point where you have to leave the comfort of your desk chair and venture into the tangible stage of manufacturing. So this is the part of the article where we recount our first real venture into the woods, and when I say “the woods” I really mean the wood shop, by which I mean the CNC wood shop, which really in fact is another office and another desk chair.
It is by good fortune that we have access to a high production wood shop with various CNC machines intended for wood, opposed to the CNC machining centres we are used to. So, in order for Spage to build a seamless infused carbon fibre splitter from our 3D model, we would have to create a mold. Our goal was to produce a 3D plug from solid high end MDF which could be sealed and used for building the end product. Above, the model for the mold in which we would produce is shown. To grasp the scale, the rectangular base of the 3D plug is exactly 4′ by 8′ – in other words, a full sized sheet of drywall.
When building something of moderate size from wood, you can simply glue up some boards and go about running them through some hazardous wood working equipment. For this splitter mold however, we needed a starting material size of 4′ x 8′ x 4.5″ in thickness, which is not something you can go grab from your local lumber yard. To our good fortune, the wood shop supplier had material sizing which would allow us to produce our build-up in just three sheets.
When the material showed up, it was apparent we were not about to make a piece of furniture. The three sheets combined weighed in at a pleasant 528 pounds, so, the weight of a small Harley Davidson motorcycle. The first step was creating our work piece. Armed with 4 litres of moisture resistant wood glue, a paint roller and a full sized glue press, we were ready to create our work piece. We stacked each sheet into the glue press with no less than a litre and a half of glue between each sheet. The glue press we had access to only had a 2′ throat so the extremities of our work piece had to be clamped manually after applying initial pressure with the press. Below is an image of our glue-up.
With our raw material brick completed it was time to fire up the 3-axis CNC router table. This is where our model would come into play. Essentially, our Pal who runs this production wood shop, handed us the keys to their smaller of two production machines and said “don’t crash it”.
Fortunately, programming a 3-axis router table is the same as programming a 3-axis machining centre, so we spent some time in another office, came out and loaded our five hundred pound stick of MDF onto the machine.
Due to the height of the plug, we had to detach the dust collection shield on the spindle. This seemed like it would not be a problem, well let me be the first to say, this indeed was a problem! During the main roughing program, the machine and surrounding area resembled what I would describe as a moderate to severe prairie dust storm. Standing 15 feet away from the table at the machine control we were geared in respirators, hoodies and glasses. In this attire it was barely tolerable. I do not envy wood-workers…
After many hours, we neared completion of our roughing program, and as we continuously blew the wood chip piles off our work piece, our beautiful splitter shape began to emerge.
Leaving just a small amount of material for finishing we loaded up our finishing program. This program would run the full length of the work piece continuously with a very minute step-over and utilize a ball-nose end-mill. This process is identical to finishing a 3D billet part in a machining centre, the only difference being this part was eight feet long and four feet wide. Our finishing program ended up in the 2+ hour range.
After our finishing program had completed we had a finished splitter plug. Due to the very small step-over in our finishing tool-path, we achieved a great surface finish, well, for wood that is. We had removed close to half of the original five hundred pounds in material and thanks to our respirators breathed hopefully none of that in.
After inspecting our mill work, we loaded this slab onto our open car float (it would not fit in the back of a truck) and headed back to the Spage shop. The carbon lay-up for this splitter is now in process and we cannot wait to see the finished product.
Typically when it comes to parts of this scale, we are only involved in the design and CFD portion of the project. Getting a chance to produce the mold and learn a new trade is always exciting. We will post an update in the weeks to come with the finished product. Feel free to contact us with any custom Motorsport design project!
Check-out Spage.Sport on Instagram