Jump to content
Please support this forum by joining the SH Patreon ×
Sturgeon's House

Recommended Posts

So I got a request recently from {NAME REDACTED} as to whether we have a how-to guide or something for competitions. After a few moments of bitter, bitter laughter at the decade-plus of my life that I've spent cobbling together things that can maybe, sort-of, squint-your-eyes produce a facsimile of a realistic vehicle, I thought I'd share my process:

 

Quote

Hi {NAME REDACTED},

 

I'm going to post this on the forum, but thought I'd sent it to you first.

 

So I don't think there's a comprehensive how-to guide for anything at the moment, and all of us seem to prefer different approaches to modelling things. There are plenty of smaller tools out there for, for instance, calculating bullet velocities given a certain barrel length, bore diameter, shell weight and powder charge.

 

Generally the way I'd do something like a tank is to decide on the general layout and armament first, get some rough sketches done, and then start modelling the results. Normally I'll start with the fun bit: the main gun. For this you can either use an existing gun or else one of the afore-mentioned calculators + DeMarre and/or Longrods depending on the era in question. Once you know what size and configuration will work to hit parameters, you can then model that component individually and add it to the vehicle model. I do this for all the major components, using things like bulk densities for transmissions (to get the correct weight) and volumetric power outputs for given engine configurations (to get the right power output for a given size of engine). For some components, such as suspension and tracks, I either model and weigh the parts directly or else use available information to make good approximations for component weight. I work out the shape, configuration and volume of the rounds, and where they will be placed. I also use mannequins to model the crew compartments, although that's an area where I still make rookie mistakes (not having spent any real time in a turret and thus not knowing what might be important for a gunner, for instance, to have at hand). I use lots of reference material to sanity-check all my designs - mainly things like cutaways of real world vehicles and YouTube videos where people crawl around inside museum tanks.  

 

I use Sketchup a lot because I know it well and there are lots of parts publicly available if you want to, for instance, plonk a good-looking model of an M2 HMG on the turret roof. The best approach, however, would be to use something like Solidworks, because that allows you a fine degree of precision when it comes to estimating part masses and the like.

 

Anyway, once my initial model is done I tot up all the weights and start working on things like range, armour fraction, power-to-weight ratio etc. Often this is where whatever concept I had for my initial design breaks down completely and I have to go back to the drawing board. Generally I'm only happy after a few goes at the process, which having a quick workflow makes possible (a guy who spent two weeks making a good Solidworks model is not going to want to scrap it, whereas if I only take a few hours to lash something together I'm more amenable to starting over).

 

Some stuff I have to learn from scratch use things in ways they were never designed for - for a recent competition I decided that my tank absolutely had to have an ATGM, so I had to read up on control methods and laws and use hobby rocket modelling software to mock up the missiles and get an idea for their performance. For other stuff I've kludged together things like excel spreadsheets to help design shaped charges and composite armour packages.

 

I've been doing competitions like this for close on a decade now - ever since the World of Tanks days where the people that eventually started this forum would just sit and shitpost on the forums there. So I've gotten into a pretty comfortable groove ITO designing and I've got tools and approaches that work for me.

 

For planes, I'm also lucky enough that my dad has always been very into scale modelling and remote-control gliders, so I learned a couple of basic concepts at my dad's knee. So I'm decent at slapping together conventional aircraft designs with relatively benign flight characteristics. We're also lucky that there are a lot of available tools, rough simulation options and information out there which can be used to help the design process.

 

Ships I struggle a bit more with (not too much water where I live, so there have been fewer opportunities to experiment and learn directly, although I used to sail so at least I know the basics of ship handling) but thankfully there's also a wealth of information out there on how to design hulls and the like.

 

Anyway, I hope this helps.

 

All the best,

 

Toxn

 

Note: I was half-right - we definitely have supplementary info for aspiring pretend tank designers pinned to this very board.

 

Finally, I'm inviting our forum grognards and past winners to share their process for folk that haven't been here since before the last ice age, so that all can benefit.

Link to comment
Share on other sites

  • 3 months later...

Having very recently finished a design cycle, I can now offer insight into how I do things.
The first thing I do is look at the requirements, and the available options (whether limited by name or simply similar in performance), and try to figure out a first order approximation of what it is I'm hoping to get done. It's usually at this point that I sketch out the design in pencil and make a list of design features I intend to include in the design - this typically helps solidify the concept in my mind, as well as making sure I don't miss anything major along the way. I set myself certain design goals to guide myself along the way at this point, as they shape to an extent how I want the design to end up.
It's also at this stage that I tend to obsessively google ballistic charts, gun blueprints, and internal pics of relevant vehicles, for cribbing the designs off later.

This approach is very "front heavy", in that it requires you to have a very good image of what you want to get done in your mind before you ever really touch the CAD software, but it does mean that other than minor tweaks, you only CAD once. The Norman, Fox, .224 Rapier, and now BT-5-76/43 were all substantially first shots. Some details were ironed out as the design progressed (typically, precise dimensions are initially guessed and then adjusted as needed), but the concept, down to, say, the location of the extra periscope on the roof, was sketched out on day 0. This approach also means modelling is easier, as you are focused on carrying out the decisions of the design committee (you) and not making the decisions themselves.

As the design progresses, the day 0 first order approximations may turn out to be incorrect. The BT-5-76/43 was initially supposed to only have thick side-skirts along the fighting compartment, with thinner ones to the rear, but then I realized I had a sufficient mass budget to allow me to go wild, so I did.

As you go, it's important to keep whatever goals you set for yourself in mind. With a sufficiently blank slate, it's easy to get lost and start designing spaceships with All The Features TM, when you should be focusing on core capabilities. For the Norman, the core was a tank on par with late Centurions; for the BT-5-76/43, the core was "I do not want to touch the driveline or the turret ring in the process of improving this vehicle". Some features (like vision cupolas) are basically free improvements if you remember to include them; others (like fuckoff big guns) require substantial tradeoffs which must be viewed in the context of their effect on the bottom line of the design goal.

I tend to start from the turret, then the hull, then the armament, then systems, and finally suspensions and the like. The wonders of parametric modelling mean you can make a rough hull shape, design components to fit, and then rejigger the hull as required by the systems.

There's no good replacement for pencil sketches and barely legible scribbles on paper, in terms of focusing one's design intent into practice. CAD is just a way of translating that scribble into something others can understand too.

Link to comment
Share on other sites

12 minutes ago, N-L-M said:

Having very recently finished a design cycle, I can now offer insight into how I do things.
The first thing I do is look at the requirements, and the available options (whether limited by name or simply similar in performance), and try to figure out a first order approximation of what it is I'm hoping to get done. It's usually at this point that I sketch out the design in pencil and make a list of design features I intend to include in the design - this typically helps solidify the concept in my mind, as well as making sure I don't miss anything major along the way. I set myself certain design goals to guide myself along the way at this point, as they shape to an extent how I want the design to end up.
It's also at this stage that I tend to obsessively google ballistic charts, gun blueprints, and internal pics of relevant vehicles, for cribbing the designs off later.

This approach is very "front heavy", in that it requires you to have a very good image of what you want to get done in your mind before you ever really touch the CAD software, but it does mean that other than minor tweaks, you only CAD once. The Norman, Fox, .224 Rapier, and now BT-5-76/43 were all substantially first shots. Some details were ironed out as the design progressed (typically, precise dimensions are initially guessed and then adjusted as needed), but the concept, down to, say, the location of the extra periscope on the roof, was sketched out on day 0. This approach also means modelling is easier, as you are focused on carrying out the decisions of the design committee (you) and not making the decisions themselves.

As the design progresses, the day 0 first order approximations may turn out to be incorrect. The BT-5-76/43 was initially supposed to only have thick side-skirts along the fighting compartment, with thinner ones to the rear, but then I realized I had a sufficient mass budget to allow me to go wild, so I did.

As you go, it's important to keep whatever goals you set for yourself in mind. With a sufficiently blank slate, it's easy to get lost and start designing spaceships with All The Features TM, when you should be focusing on core capabilities. For the Norman, the core was a tank on par with late Centurions; for the BT-5-76/43, the core was "I do not want to touch the driveline or the turret ring in the process of improving this vehicle". Some features (like vision cupolas) are basically free improvements if you remember to include them; others (like fuckoff big guns) require substantial tradeoffs which must be viewed in the context of their effect on the bottom line of the design goal.

I tend to start from the turret, then the hull, then the armament, then systems, and finally suspensions and the like. The wonders of parametric modelling mean you can make a rough hull shape, design components to fit, and then rejigger the hull as required by the systems.

There's no good replacement for pencil sketches and barely legible scribbles on paper, in terms of focusing one's design intent into practice. CAD is just a way of translating that scribble into something others can understand too.

So our processes are basically opposites of each other :lol:

Link to comment
Share on other sites

6 hours ago, N-L-M said:

Having very recently finished a design cycle, I can now offer insight into how I do things.
The first thing I do is look at the requirements, and the available options (whether limited by name or simply similar in performance), and try to figure out a first order approximation of what it is I'm hoping to get done. It's usually at this point that I sketch out the design in pencil and make a list of design features I intend to include in the design - this typically helps solidify the concept in my mind, as well as making sure I don't miss anything major along the way. I set myself certain design goals to guide myself along the way at this point, as they shape to an extent how I want the design to end up.
It's also at this stage that I tend to obsessively google ballistic charts, gun blueprints, and internal pics of relevant vehicles, for cribbing the designs off later.

This approach is very "front heavy", in that it requires you to have a very good image of what you want to get done in your mind before you ever really touch the CAD software, but it does mean that other than minor tweaks, you only CAD once. The Norman, Fox, .224 Rapier, and now BT-5-76/43 were all substantially first shots. Some details were ironed out as the design progressed (typically, precise dimensions are initially guessed and then adjusted as needed), but the concept, down to, say, the location of the extra periscope on the roof, was sketched out on day 0. This approach also means modelling is easier, as you are focused on carrying out the decisions of the design committee (you) and not making the decisions themselves.

As the design progresses, the day 0 first order approximations may turn out to be incorrect. The BT-5-76/43 was initially supposed to only have thick side-skirts along the fighting compartment, with thinner ones to the rear, but then I realized I had a sufficient mass budget to allow me to go wild, so I did.

As you go, it's important to keep whatever goals you set for yourself in mind. With a sufficiently blank slate, it's easy to get lost and start designing spaceships with All The Features TM, when you should be focusing on core capabilities. For the Norman, the core was a tank on par with late Centurions; for the BT-5-76/43, the core was "I do not want to touch the driveline or the turret ring in the process of improving this vehicle". Some features (like vision cupolas) are basically free improvements if you remember to include them; others (like fuckoff big guns) require substantial tradeoffs which must be viewed in the context of their effect on the bottom line of the design goal.

I tend to start from the turret, then the hull, then the armament, then systems, and finally suspensions and the like. The wonders of parametric modelling mean you can make a rough hull shape, design components to fit, and then rejigger the hull as required by the systems.

There's no good replacement for pencil sketches and barely legible scribbles on paper, in terms of focusing one's design intent into practice. CAD is just a way of translating that scribble into something others can understand too.

 

This is why you're a genius, and I'm not. I'm a very dedicated idiot.

Link to comment
Share on other sites

I’ll add more context later, but here’s what (I have figured out) NOT to do when designing your tank/ gun/ boat/ thing: 

 

1. Don’t hyper fixate on one aspect or feature of your creation; get the idea into the rough area that you’re going for, then move on. Trying to get all those details perfect as you put them down will only slow you down, and drive you crazy. 
 

2. Don’t make your ideas too small; make your creation 50-100mm longer/ wider/ taller (for vehicles) than you think it needs to be, I’ve found that it’s easier to make a thing smaller than it is to make them bigger. Also, it allows for growth if you find an aspect lacking. 
 

3. Don’t be too “experimental” with it; fancy technology is always tempting to use, especially weight and space saving technologies to meet requirements in a competition. But are those technologies actually used in real world, production vehicles? Were there teething problems when it was implemented? Know the technology/ features you’re applying to your creation before you try to utilize them. 

 

I might think up more things when I finally get home, but those are the biggest things that held me back in the cascadia and californium competitions. 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...