Once the above initial design work was done, I spent a lot of time staring at it with a smoke in my hand. For the past week I have been revising every feature of the design to towards ease of build goals as well as generating a more cohesive whole. The electronics bay is something that I have been uncertain off. The amp modules are quite large, and I want to include more components like DSP and such. So this is in v2 now and the work completed so far can be built as is as v1
Version two is minus the internal electronics bay, has increased internal volume and also widened by 40mm. The electronics bay will now extend down from the curved base and form a stand for the cab. This adds the flexibility of building this as a passive or active sub. I am redesigning the heat pipe to draw air through the new electronics bay and out through the back of the cab. V2 model also features the skin and core layers of the sandwich. More to come
Version two is minus the internal electronics bay, has increased internal volume and also widened by 40mm. The electronics bay will now extend down from the curved base and form a stand for the cab. This adds the flexibility of building this as a passive or active sub. I am redesigning the heat pipe to draw air through the new electronics bay and out through the back of the cab. V2 model also features the skin and core layers of the sandwich. More to come
Not scaled as that would proportionally increase wall thicknesses to undesirable levels in various areas. Instead, just the sheet size needs increasing and that needs resizing of some elements in Cad
The design as drawn is rather aimed at being able to be stood up and stacked to use in taller cabs but only the internals get stacked. One piece skins and core sheet is used to span and scarf
Keep in mind that the sandwich schedule being used here is massively over engineered for the 6.5” driver and I would conservatively rate the structure to 30kw
For a 15” driver, just make the box wider in all dimensions and only scale the fillets. Wall thickness of 8mm with skins on both sides is plenty
The design as drawn is rather aimed at being able to be stood up and stacked to use in taller cabs but only the internals get stacked. One piece skins and core sheet is used to span and scarf
Keep in mind that the sandwich schedule being used here is massively over engineered for the 6.5” driver and I would conservatively rate the structure to 30kw
For a 15” driver, just make the box wider in all dimensions and only scale the fillets. Wall thickness of 8mm with skins on both sides is plenty
The project cab can be modelled in a program called Carlson Hulls in moments, and cutting files produced for each panel and made over a few days. That is the normal way of doing things where sandwich construction of monocoque structures are concerned. The modelling takes moments, but the construction process is very work intensive
The Cub Sandwich project is my attempt to evolve that and develop a better workflow. V1 is the result of building up the SOR or statement of requirements items one by one. As designed, it moves the constructions process to the software and automation and the actual handiwork remaining becomes just assembly. The electronics bay in this is designed to take 840wrms of amp modules as well as a DSP board. The stereo speakers are likely to exceed what the amps can supply, so I will be redesigning this part to take only the 600w sub amp and leave the stereo amps and other electronics outboard. The V1 design will be just an active sub with DSP and balanced preouts for stereo
Now with V2, the aim is to reduce the time it takes to create each part in automation as well as a cab with more internal space. The active driver is going to see a clean 600wrms. V2 will explore increased SPL by unfettering the driver from space restrictions. As well, the new electronics bay will allow larger amps for the stereo channels as well as space for an optional tube stage between the DSP and sub to experiment with sweet over driving
Take a close look at the left info panel in the pic. This is v2 reaching the state of completing the curved sides only in the sandwich layers. There is a 8mm core layer with 1mm skins either side. The schedule is single layer of 200gsm biax or dual 0.5mm alloy sheet each side of that 8mm PVC boatbuilding foam core material. Panel shows 3 sketches and the number of elements it takes to get to that stage. Most of the funny named elements are not even structure features. These are elements to help create another element or a feature. All those elements are just for the curved sides and setting overall design dimensions
The trick to keep FreeCAD stable while working is to reduce the number of sketches needed and instead exploit some padding and pocketing features that the program doesn't bring up as available in the task list
Below is a closeup of a rear corner, showing the skins and core. That back panel showing and sticking out is not a cab feature. That is there to keep everything even like a concreting formwork. That element will be removed once the flat sides are on and a proper forward transom will be created there as a structural item
The boxing formwork is applied to the main baffle area. As can be deduced from the info panel, this needs to be set wider with a new sketch and then trimmed back. There are reasons for doing it this way to avoid FreeCAD limitations, including cleaning up aberrations created during building up the sandwich, as well as those that will appear in more advanced stages. These two elements in place of the proper baffle and back panel absorbs 5mm of base body at front and back and one removed, leaves two beautiful, curved sandwich panels with the raw edges in the correct planes. Now FreeCAD will have its panties in a knot it those two panels didn't have something connecting them together, so another use from the boxing formwork
One bad workflow that I caught while ironing v1 was that I had created the proper side baffles too early. This limited my access to modelling interior elements. So this becomes a demonstration for the learning about iterations while developing something. For v2, I'll do all the inside work first before creating the side baffles for the PR units
So next up will be the revised heat pipe/ heat radiator system for the active driver and the new electronics bay. V1 had dual 80mm fans for the electronics bay and a passive heat pipe for the driver, with optional 30mm fans to make the heat pipe active. In v2, one 80mm fan will draw air into the electronics bay and a second 80mm fan will draw it out through the heat pipe and out of the back of the cab. This solves another worry about spills, all it takes is some rum to ruin the night! Lost a 500wrms per channel PA amp during a dancehall reggae event in a suburban hall in 1992 and weary since :O
Its actually going to be fun modelling the air induction system, a supercharger and a turbo
This got pushed back a bit due to a more urgent project. Design of a daybed incorporating bass exciters for late night HT. That design is now complete, so back to the Cub Sandwich development. I discovered one more FreeCAD limitation while drawing the bed frame. This is related to drawing a number of panels in one sketch. If the panel edges are connected as in the bed frame, FreeCAD makes some of the panels hollow tubes. A bit of WTF moment while trying to figure out what's going on there. Including a number of redraws and checking files for corruption before clicking on that it's a program thing
The floor is open to anyone trying to learn FreeCAD and draw similar elements, we can figure out the steps here
The floor is open to anyone trying to learn FreeCAD and draw similar elements, we can figure out the steps here
Didn't think to yet. FreeCAD is great for DIY'ing as is and I think it's me who is asking the program to do things that might be beyond what the developers can invest in a freeware. I can see a SolidWorks lic. in my future due to the nature of some upcoming projects like the custom coaxial driver
A couple of posts back I did all the steps to get to the pair of curved side panels. Here is a closeup of the section. 9mm PVC foam core and 1mm skin on either side. The design supports aluminium skins with the option of glass skins for increased durability. Now I have to create a rebate on these end edges and create the end panels which will sit in this step
This is where the model is now. I have removed the helper parts, the fore and rear formwork. It's just two curved sandwich panels supported by a temp fill. The edges provided by this fill will let me build the baffle and rear panel without creating new sketches and then carve out the internal structure