Russian Arcturus Concept Submarine Build
Collapse
X
-
Alrighty. We got this licked (provided the props actually push the boat with decent thrust).
First off, let's talk linkages. The Arcturus, as I engineered these files, has a modified x-tail configuration. The control surfaces are not aligned, and the horns for each all need to point towards the centerline of the boat thanks to the limited room back there, meaning that some creative engineering needs to take place. Typically, the planes that are across from one another are in line, meaning you can connect them with standard yokes and run a single rod to the horn. Not gonna work in this case as the rotation needs to be reversed thanks to the placement of the horns. All control horns need to point toward the centerline of the boat. If you can figure out a way to get one pointing in and another out, you will be able to simplify the linkages and do away with the assembly I'm about to show you.
First off, here is the configuration with a highly technical and ultra-precise rendering of the setup:
And here is the photo of the linkage assembly needed to convert the linear force in the applicable directions to each control surface:
In this setup, control surfaces 1 and 4 need to work in tandem, and 2 and 3 need to work in tandem. 1 is connected to the forward assembly under the shaft, 4 is connected to the same shaft, but above it. Move the shaft in one direction, and rudders 1 and 4 move in the same direction. Without the assembly up front, they'd move in opposite directions. Same goes for 2 and 3.
Okay, now onto propulsion. The stern is tight, especially with all the linkages coming forward from that area. That makes making room for tubing for pump propulsion very problematic. A simple 1/8" drive shaft slips in there nicely, however. All we needed to do was engineer a prop back there to push the boat.
Here is the solution I came up with:
Parts needed: 20mm diameter brass props, 1/8" stainless shafts, 1/8" universal joints, 3D printed mounting bulkhead, 5/32" brass tube (for the thrust/reverse thrust bushing). It may be that we need to block the opening beside the prop shroud to stop water cycling around it, but we can test that first to see if it's needed. If we do, we might just get away with some plumbers epoxy pressed into the opening to make a nice, smooth filled pocket there. Heck, you could even sculpt in some cool venty-looking thingies, too.
The 3D printed inserts were done on my resin printer. I tried on my filament printer, but the fitment was poor. I find that filament printers (at least mine) make parts that are about 0.2mm larger than the file dictates. This necessitates a lot of sanding for perfect fit, and these parts are pretty thin. I just used the resin printer and it fits perfectly as you see it here.
This boat will be the test bed for our upcoming 250 Series SubDriver, which I hope to have parts for within a week. Once we have the prototype cylinder assembled, we'll drop it in this boat and see what it does in the water!
BobComment
-
Alrighty. We got this licked (provided the props actually push the boat with decent thrust).
First off, let's talk linkages. The Arcturus, as I engineered these files, has a modified x-tail configuration. The control surfaces are not aligned, and the horns for each all need to point towards the centerline of the boat thanks to the limited room back there, meaning that some creative engineering needs to take place. Typically, the planes that are across from one another are in line, meaning you can connect them with standard yokes and run a single rod to the horn. Not gonna work in this case as the rotation needs to be reversed thanks to the placement of the horns. All control horns need to point toward the centerline of the boat. If you can figure out a way to get one pointing in and another out, you will be able to simplify the linkages and do away with the assembly I'm about to show you.
First off, here is the configuration with a highly technical and ultra-precise rendering of the setup:
And here is the photo of the linkage assembly needed to convert the linear force in the applicable directions to each control surface:
In this setup, control surfaces 1 and 4 need to work in tandem, and 2 and 3 need to work in tandem. 1 is connected to the forward assembly under the shaft, 4 is connected to the same shaft, but above it. Move the shaft in one direction, and rudders 1 and 4 move in the same direction. Without the assembly up front, they'd move in opposite directions. Same goes for 2 and 3.
Okay, now onto propulsion. The stern is tight, especially with all the linkages coming forward from that area. That makes making room for tubing for pump propulsion very problematic. A simple 1/8" drive shaft slips in there nicely, however. All we needed to do was engineer a prop back there to push the boat.
Here is the solution I came up with:
Parts needed: 20mm diameter brass props, 1/8" stainless shafts, 1/8" universal joints, 3D printed mounting bulkhead, 5/32" brass tube (for the thrust/reverse thrust bushing). It may be that we need to block the opening beside the prop shroud to stop water cycling around it, but we can test that first to see if it's needed. If we do, we might just get away with some plumbers epoxy pressed into the opening to make a nice, smooth filled pocket there. Heck, you could even sculpt in some cool venty-looking thingies, too.
The 3D printed inserts were done on my resin printer. I tried on my filament printer, but the fitment was poor. I find that filament printers (at least mine) make parts that are about 0.2mm larger than the file dictates. This necessitates a lot of sanding for perfect fit, and these parts are pretty thin. I just used the resin printer and it fits perfectly as you see it here.
This boat will be the test bed for our upcoming 250 Series SubDriver, which I hope to have parts for within a week. Once we have the prototype cylinder assembled, we'll drop it in this boat and see what it does in the water!
Bob
Well Bob,
Nice job on the engineering! Somewhat complicated, but certainly doable. Thank you very much for taking the time to post up your designs. The function of a V tail configuration is still BIT HAZY in my mind, but ounce I have put it all together I am sure it will make sense!
Great job Bob, thanks again!
Rob
"Firemen can stand the heat"Comment
-
X tails are more complicated to engineer, but offer 1.4x the effectiveness of standard cruciform planes/rudders. In the video link below, I go over how they work. It starts at 19:53 in the video:
Comment
-
That does help in explaining how to design the linkages and also how it functions. Very unique design for sure!
Rob
"Firemen can stand the heat"👍 1Comment
-
Comment