Apollo 11 simpit, need 64x64 encoder, or something else...

Thread Starter

baloo6969

Joined Dec 21, 2011
3
Hey all, I need to build an encoder/decoder for a Apollo 11 simpit. There are ~600 switches in the CM, and ~200 in the LM, with most having 3 positions, so I need an encoder into my PC that will read +3,000 different inputs.

A matrix encoder 64X64 would be perfect…is this even possible?

Does anyone have plans for one of these?

If this isnt possible, does anyone have any other suggestions?
 

SgtWookie

Joined Jul 17, 2007
22,230
You really wouldn't want to build such a monstrosity. How would you debug it?

Think more along the lines of using local microcontrollers that poll the switches in their "neighborhood", and if anything changes, send a message via a protocol like I2C to a master uC that has the capability to communicate to the PC via USB; which then tells the PC about the change.

The master uC would sit there and poll all of the slave uC's to see if they are "awake" or at least alive, and if they have any changes. The master uC could report a fault if a uC didn't respond in a timely manner. Meanwhile, if you leave a bit of time between the polls, a slave could grab the bus to report in if a switch changed.

You wouldn't want the uC constantly sending the status of each and every switch; that would slow down the PC running the simulation. You would want it to go through a 'switch position inventory' on start-up, and report the status of all of the switches.

You wouldn't want to try to build a system like that using discrete 74x or 4000 series IC's - it would take forever and cost a fortune. You could do with 8-bit midrange uC's that had plenty of I/O pins and I2C or other protocol available.

If you designed a generic "slave board" with connectors for the switches, power, communications and onboard programming, you could simply make as many slave boards as you needed, plus a few extra for spares/troubleshooting. It would be much easier to build/debug something like that.
 

monster_catfish

Joined Mar 17, 2011
116
Holy Toledo ! Now there is one humdinger of a project.

I have seen youtube footage of many impressive airliner sim pits built by home hobbyists, but have never heard of an undertaking anywhere near as complex as the one you are contemplating, assuming this is not a joke.

My I ask why on earth an ancient space rocket would be a worthwhile project for a DIY sim pit ? There will be NO external scenery available for adaptation from the current sim market, the flight dynamics will have to be cobbled together pretty much on the basis of conjecture, and, if I am not mistaken, the original Apollo 11 vehicle was one in which the occupants were basically along for the ride, and with no real flight controls other than trajectory modification thrusters at their disposal.

If your assignment is to construct a flight sim of your choosing, may I suggest a common airliner like the Boeing 727, 737 or 767, all of which are already tried and tested workhorses of the flight sim world, and replicated in excruciating functional detail by various software vendors.
 

thatoneguy

Joined Feb 19, 2009
6,359
If you haven't actually started building the Apollo sim yet, I'd suggest building a full fighter simulator as mentioned above.

Make an F-4 or similar. Get a genuine seat and panel from surplus if possible. "Windows" would be LCD Displays showing terrain from the flight sim. Realistic sounds of all sorts, along with realistic control stick are supoorted.

It's still a big investment, but the payoff is worth it, I've only seen one mockup of an F-111 ( I think, some mid-70's fighter), and once inside, you'd swear you were actually flying!

That way you don't need to write the software too.
 

Thread Starter

baloo6969

Joined Dec 21, 2011
3
No this is no joke.

There is a open source flight sim, "Orbitor"...with an addon "NASSP - Project Apollo"

You perform most of the events during the mission. Transposition, Docking, decent/assent both moon and earth, and staging.

It's an excellent simulator, that is an evolving and new functionality is added quite regularly.

Project Apollo, is fairly flexible...and translating the signals from a listener to the application will require some work, but certainly not as much work as the hardware.

There are several things i'm working on in parallel, Input into PC(this thread), Output(via a controller board, and multiple monitors for windows and gauges), and Software changes/development required to support these changes.


Thank you for your ideas; my problem is that the suggestions have raised more questions.
You could do with 8-bit midrange uC's that had plenty of I/O pins and I2C or other protocol available.

If you designed a generic "slave board" with connectors for the switches, power, communications and onboard programming, you could simply make as many slave boards as you needed, plus a few extra for spares/troubleshooting. It would be much easier to build/debug something like that.
Any suggestions where i can find more information or other threads that i can piggy back on? I think is my best option.

Thanks again!
 

SgtWookie

Joined Jul 17, 2007
22,230
There are a LOT of microcontrollers out there, with lots of different features.

A couple of 8-bit PICS that don't require crystals with lots of I/O pins you might look at are:
1) the PIC16F88x series ($2-$2.50 in small qty); these are an updated version of the PIC16F87x series, less expensive, slightly different features, but mostly compatible.
http://www.microchip.com/wwwproducts/Devices.aspx?dDocName=en026561
Datasheet: http://ww1.microchip.com/downloads/en/DeviceDoc/41291D.pdf
PIC16F882/883/884/886/887; they have different features as far as memory, number of pins (28/40/44), packages (DIP or SMT), number of I/O pins (24 to 35), and various other options. They can operate with either an external clock/crystal/resonator, or the internal clock, which can operate up to 20MHz.

2) The PIC16F193x series ($2-$2.50 in small qty)
http://www.microchip.com/wwwproducts/Devices.aspx?dDocName=en538149
These are relatively new, about the same price as the PIC16F88x series, but have more features, more memory, and can clock at up to 32MHz. The downside is that being relatively new, 3rd party compilers may still be somewhat limited. The C and assembler that come with MPLAB are limited to 2K. You can find 3rd party support for it though. B.Knutsen has a shareware C compiler that supports it, and doesn't have a memory limit:
http://www.bknd.com/cc5x/

The PIC18F series is where the USB support starts. You may wish to stick with all one model of uC, so that you don't have to learn multiple architectures while you're trying to handle everything else.
Have a look at the PIC18F4550; ~$3.50 each :
http://www.microchip.com/wwwproducts/Devices.aspx?dDocName=en023498

Microchip.com has a ton of information on their website - so much so, it can be rather bewildering.
Here is a page that lists ALL of their uC's which support USB:
http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=2654

USB support info is here:
http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=1486&redirects=USB

But, before you can think about selecting a microcontroller, you need to sit down and make a detailed list of every feature that you require. That will help you narrow the list of options in a big hurry.

I don't know how much experience you have. This project will be very complex. However, if you build the project using a modular approach, keeping the individual modules simple and as identical as possible, with a standard method of interfacing with those modules, you will be able to leverage those relatively simple modules into a more complex system.

It will be far easier to troubleshoot as well. Things are going to break. You will want to make the fault-finding as easy and quick as reasonably possible.
 

Thread Starter

baloo6969

Joined Dec 21, 2011
3
Thanks Sgt,

I am writing up the requirements, and figuring out exactly what is needed.

Once this is completed, i will post the needs, and maybe you guys can give me some guidance as to which microcontrollers i should go with.

I am a computer programmer, with lots of experience in circuitry. I'm am in no way an electrical engineer, but I’m confident that with some guidance i can figure this out. I see this as an opportunity to learn a lot.

As for modularize this, yes. I've started a high level project plan, with each step being(or will be) broken into individual phases. Looking at this from 30,000 feet, it's a nightmare. But when you look at each step, it is challenging, but certainly a interesting project. I've also started a list of "Issues and Risks", i'm up to 120, just sitting here tonight...oh boy.

Also, on several other forums i'm getting push back "Why"? Well, it's more than just getting in a seat and 'flying', actually there is very little 'flying' in space. I've flown several simulators, and nothing is more challenging than the NASSP Apollo simulator. My final goal, is to fly the entire mission, locked in the LM/CM for ~12 days. With my only 'outside time, is a few hours to run around outside. Yes. 12 days. It's crazy, but how much fun would it be? Well probably not a lot of 'fun' but it will be quite an experience once i've spent all the time building it. Odds are, once i'm done with 12 days i will want to burn the simpit!
 

spinnaker

Joined Oct 29, 2009
7,830
My final goal, is to fly the entire mission, locked in the LM/CM for ~12 days. With my only 'outside time, is a few hours to run around outside. Yes. 12 days. It's crazy, but how much fun would it be? Well probably not a lot of 'fun' but it will be quite an experience once i've spent all the time building it. Odds are, once i'm done with 12 days i will want to burn the simpit!
How much fun spending 12 days in a simulator? I would think sticking pins in your eyes would be a million times more fun. :)
 
Top