Tweet
Excellent progress! Love the underwater photo!
-
If you can cut, drill, saw, hit things and swear a lot, you're well on the way to building a working model sub. -
Hi all,
Construction of the ballast roof has begun with an exacting measurement, in three dimensions, of the interior hull where the roof will be installed. Since this hull’s shape is asymmetrical in the extreme, I needed to create a template across the entire area. I couldn’t just measure one side and assume the opposite side would be a mirror image. I had to do it the hard way.
The other immediate challenge was building it inside the hull in a cramped space, then be able to remove it through the opening smaller than the jig itself. So with a little ship-in-a-bottle inspiration, I included hinges into the build, allowing the two sides to fold inwards. The “butterfly jig” arrangement worked perfectly. Once outside the hull I opened the two sides again and locked them into place.
The roof will be acrylic like the battery encasement, which will be thermoformed using a ‘trailer-park’ technique that I’m not really proud of. But this is a complicated piece and I’ve gotta ‘up’ my thermoforming game.
With the measurements derived from the butterfly jig, I’m assembling a very manly thermoform press out of ¾” birch plywood. The plan is to clamp the softened acrylic sheet into the plywood press and use my air compressor to slightly balloon one side up, rather than have a vacuum pump suck it down. I’ll probably have to sit on the damn thing for a while just to get it to conform to shape.
The sub’s design has evolved a lot over the years, and my old ballast roof design (which kinda sucked anyway) didn’t allow enough access to access components from the ballast area, which is obviously critical considering how much crap is in there now. So besides the quick’n’easy screw-top access, I’m including a secondary, “rarely-used” access hatch for major work. The “rarely-used hatch” will encompass the screw-top, one surrounding the other. It’s weird but why not. Normally I would avoid this arrangement, but this is the ballast tank roof. It will be bearing buoyancy forces and keeping the gas within the ballast tank, but remaining more or less freely-pressured across either side of the gasket.
The aluminum plates you see there are going to be the two sealing faces of the “rarely used” hatch area. I’ll be jigsawing them into two rings, drilled & countersunk. I hate jigsawing aluminum. Shouldn’t this be fun?
Anyway, as soon as the thermoforming task is done I’ll post more pics.Comment
-
I'm not sure if this would help but. I have had better luck when cutting thick aluminum 1/4" or more using coarse wood blades than fine tooth "metal" blades. They don't clog like the fine tooth ones do. It might be worth a test on some scrapComment
-
Also, if your saw is a variable speed one, slow it down. At higher speeds the heat generated at the teeth will gall the chips into the tips of the teeth. A little oil helps as well. Its still messy.Originally posted by greg w View PostComment
-
Thanks greg, thanks crueby!
I think you're both absolutely right. I have jigsaw-cut aluminum plate before, quite a bit thicker than this too, and found that slower RPMs on the jigsaw while using a coarser-tooth blade works much, much better. (for aluminum anyway)
My DeWalt higher-end reciprocating jigsaw is kickass. The T118B blade is still technically meant for metal but it looks just like a wood-cutting blade. It works perfectly. I use cutting oil as well just to be safe. Keeps the blade heat & alum dust to a minimum.Comment
-
Hi all,
The end of the year is nigh, and so I thought to submit a final, “screw you, 2020” posting in order to close this wretched year on an oblique positive note.
My best efforts were to get this sub/thing/monster/money pit completed and in the water by the end of this year. Unfortunately it didn’t work out due to a couple of disasters which tossed the timeline out the window. But I was able to pull it together & get the final part, the ballast tank roof, nearly completed. A longer post containing the construction process is forthcoming on this thread, but here’s an image of the roof as it presently exists.
The two lower pieces (presently clamped together with little black clamps) will be inserted, fitted & adhered into place inside the hull, then acid-cemented together to form a single part. Then the “rarely used” access hatch will get bolted into place, (that’s the aluminum ring interface) followed by the simple screw-cap with the release valve on top.
The thermoforming worked ….well, basically perfectly. This image shows what the acrylic copies looked like, fresh out of the press and ready to get cut into separate pieces. I’m blessed to have an acrylics expert within walking distance of my apartment (how weird is that?) to get critical advice on adhesives & techniques to use. So it’s simply a matter of time. I have only to patiently install these parts into the sub and then give it a try in the apartment’s submarine testing tank.
Here’s hoping everyone else is making progress on their hulls, and that we’ll have a new batch of subs cutting the water in 2021. Cheers, everyone.
BBComment
-
Looking good crispy creator.Comment
-
Ben,
That is well done! Stay safe and healthy! Look forward to meeting you in 2021!If you can cut, drill, saw, hit things and swear a lot, you're well on the way to building a working model sub.Comment
-
“Beautiful! Nice work! I have ….no idea what that is.”
It seems that, in all the time I’ve spent in the laboratory recently, I’ve been less than coherent as to what it is that I’m doing. So a recap:
If you’ve been following along all this time, you might remember that among my most important goals with this submarine is to give it the highest possible freeboard for its size. I think the term I used was “absurdly high freeboard.”
To do that, in some sense I designed much of the sub’s interior basically inside-out. I installed all the guts occupying the front section (comprising about two-thirds the total length of the sub) and then used the hull itself as the ballast tank. So mathematically-speaking: the ballast tank’s displacement value becomes “whatever’s left over.”
So far so good. The final task was to seal off the front section with a roof, made of clear acrylic so I can still see inside the tank when I want to.
Acrylic is clear, but is also nice because it can be thermoformed. If you need a shape, and it isn’t too crazy, you can heat it up and put it into a press then let it cool and trim off the edges.
The down side is that acrylic has some idiosyncrasies. It can be a pain in the butt to glue to other things. Acetone-based adhesives work well with some “clear plastics,” but if the chemistry isn’t compatible then it might not adhere at all, or worse, there will be a bad chemical reaction that causes micro-fracturing.
Acrylic also doesn’t like “open cuts” all that much. Over time, tiny cracks can propagate off the edges of a cut if left untreated. The solution here is to run a tiny torch flame (crème brulee torch is fine) along the edge to re-melt it. It's like cauterizing a wound.This relieves any stresses in the material and stabilizes it.
Oh, and never use blades. Always grind. Good life advice.
Goal:
This is a “gas” system, so I’m not constrained to a cylindrical shape for the sake of a piston drive. The shape of the roof had to span the entire area of the ballast front section. It had to rise in a spot where the drain valve will be, (gas rises) and it needed at least one access hatch in order to get to the guts housed inside. Ultimately I chose to have two.
There are two access hatches on the acrylic piece planned, one inside the other. The upper hatch is just a screw-cap meant for regular, basic access. The larger one encompasses the screw-top, and is a bolt-down cover meant for major work when I’d need to pull guts out and put them in. It involves a ton of tiny flush-head bolts clamping them together, so it’s only meant for overhauls, mostly.
Getting the Proper Measurements:
One day I’ll be able to scan this submarine hull with a phone app & print a brand new one with all my dumb mistakes fixed, like for example: better access to the hull. As it is, the only way to get to the guts is through a narrow top section running fore and aft. This opening is smaller than the planned ballast roof piece that will have to be placed inside. (!) I know, everyone has their issues with sub access. But working this way has been really tough.
Also, embarrassed to say, but the hull shape is not axially symmetrical. So to get accurate measurements of the hull interior I had to build a temporary jig that would define all four edges of the final ballast roof. AND, the jig had to be removable.
So I built a “butterfly jig” out of basswood, with hinges incorporated into the design so I could fold the sides in like butterfly wings and remove it out of the narrow hull opening. It’s a sort of “ship-in-a-bottle” approach.
Building the Thermoforming Press:
So I got the dimensions I needed from the butterfly jig. Next step was to make the press. The thermoforming press had to be stalwart! This thing had to survive my weight on it, and I knew the acrylic would put up a fight getting into just the right shape, so I had to overbuild it a lot. I used lots of thick plywood sandwiched into two halves, top and bottom, and aligned with four dowels. I believe the engineering term is it’s a “smushy thingy.”
Those aluminum rings you see in the pictures serve two purposes; one, as matching clamping surfaces for the “overhaul hatch,” and eventually as the final hatch surfaces themselves. So both parts being finished pieces, they had to be veeery carefully machined to be both as strong and as light as possible.
Thermoforming-proper:
I created a crude aluminum frame to hold the acrylic sheet when placing it into my kitchen oven. The edges are clamped around the perimeter with ordinary spring-steel clamps you can get from a stationery store. (once the acrylic sheet is clamped into the frame, I remove the little handles from each clamp because they just get in the way) The clamps might not inspire confidence, but as the acrylic softens the clamp’s ability to “bite” into the acrylic improves greatly.
The refined thermoforming “magic numbers” to soften the acrylic were determined to be 340-degrees for 4 minutes 10 seconds.
This value is critical because you need to strike a balance between getting it hot & soft enough to conform well, but not so hot that it begins to form bubbles in the acrylic.
Besides the temperature & duration to avoid bubbles, I’ve also found it critical to eliminate any strong sources of convection inside the oven. If there are gaps along the edges of the oven’s bottom tray, cover them over with aluminum foil. Otherwise, the heat blasting from below will cause bubbles to ‘bloom’ in that area.
(One aspirational plan was to use an air compressor to create a positive pressure under the soft acrylic plastic in order to have it “bulb” upwards a bit. I discovered that there was no practical way to seal the lower spaces, specifically around the corners. So I abandoned that plan. You may see a Schrader valve in some of the pictures, but it never ended up getting used.)
After some loud trial-and-error swearing and a smelly kitchen, I successfully thermoformed three perfect copies. Then I cut the copies into three pieces; one large piece for the main perimeter, one for the “overhaul hatch,” and one for the top screw-cap. The aluminum rings, which had been used to help clamp the acrylic to shape, were removed from the thermoforming press and installed onto the acrylic pieces as the two halves of the “overhaul hatch.”
For the thermoforming event I’ll let YouTube link speak for itself. And you get to see what my kitchen floor looks like.
Final Installation:
Construction of the ballast roof has concluded. Overall it required an enormous amount of work on the front-end in order to get reliable & accurate results at the back-end. I’m fairly happy with the process and the results. Since the roof consisted of two pieces with an overlap, I knew I’d need at least two identical thermoforming jobs from which to cut the final pieces. As mentioned, I pressed three. But using this method I could press a hundred if I wanted to.
Fitting the roof into the hull required cutting the larger, gorgeous, main perimeter roof piece into two parts (gasp!) I know, ridiculous, but it simply could not have gotten in there otherwise. Once the two parts got carefully ground down & sanded to fit, I re-glued them back into one piece & into the hull itself. Then I installed a cork gasket between the large section and the overhaul hatch to maintain a good seal.
Once I assembled everything together I realized that the drain valve was absurdly tall & wouldn’t clear the inside edge of the conning tower. There was never any need to make this thing as tall & skinny as it was, it was just an early design when (again) I figured I’d have all the clearance in the world for whatever design I could think of. So many design iterations later, it became obvious that it just wouldn’t fit. So that’s in the process of rebuilding, with a stockier design.
All for now....Comment
-
Hi all,
The Feed/Drain Assembly is finally installed in its little spot, tucked behind the battery box and in front of the old WTC section. You can see it there, bathed seductively in blue light though the access hatch.
There was time today to calibrate the two Viper Sub10 ESC's, which turned out to be laughably easy to do. So that's checked off the ol' list.
Just have to seal around the propel tank's schrader valve and I should be able to give the entire ballast system a formal check. Then to the testing tank!
A note of major major concern though; The rock under which I live apparently didn't have sufficient wifi strength to carry the news that Rust-o-Leum had bought Testors back in 2019, only to malevolently wipe it off the face of the Earth. While this is tragedy enough that the Model Master line's gone, (which now explains the price spike) it also means my only source of Ozone-safe propellant ("Aztek") is disappearing as well. Does anyone have any alternative source for this product? ("Piston peeps" kindly spare me, "gasbag Ben", the prostelyzation to go with a piston system. My design's already irrevocably committed to gas.) (fart joke ref# 2847912)
BBComment
-
If you can cut, drill, saw, hit things and swear a lot, you're well on the way to building a working model sub.Comment
-
Thank you, Tom!!!
Supastaaaah!!!Comment
-
All systems confirmed as functioning, on dry land anyway. Just waiting on a warm(ish) day to get the Crisp in the apartment complex's testing tank to test the buoyancy and fore-aft trim.
I did something lame, overstressing the overhaul hatch when I was tightening the bolts, & caused a couple of cracks in the acrylic. I've sealed it for now & it should work fine for the tests. But at some point I'll replace the acrylic for this part. Not a huge deal to thermoform another one, or two, or five.
Wednesday ought to be in the 70's. Will post a video if it happens
https://youtu.be/NDvEAj-c5h0 .Comment
-
Okay all, The first testing tank test tended toward tumultuous.
But it was a mixed bag of good & bad news, nothing unfixable or untunable, but shows I've got (ahem) more work to do.
I cut together a movie of the event & dropped it on YouTube, in case anyone's interested.
Comment
-
Glad you isolated the leak. Any reason you didn't just run the servo wet? Either by potting it or using corrosion x?Comment
Comment