Motion Platform Possibilities

[TheFearow]

Hi,

I'm in the design and planning stages of a cockpit, and I'm currently considering an offer I had for motion.
I originally planned for no motion, due to the complexity and difficulty involved, but I've had an offer from a friend of mine who works with motion simulator systems for entertainment uses (e.g. theme park stuff). He's offered to give me copies of designs for a few of the systems he works with, if I'd like to use them. I'm just not sure what system would be best, and what would be the most useful for flight simulation.

All the systems have Roll, Pitch, and Yaw, all nearly unlimited rotation (limited only by the amount of turns before the thing wears through, so tens of thousands of rotations). I know of course the limitations and that things like inversion won't feel like the real thing in most ways, but I'm sure flying upside down would be a new challenge (especially for sim construction, I'm sure most aren't designed to be suspended upside down).

What my main question is at this point is whether more than 3DOF is worth it. I can optionally add a rotationally-adjusted surge (e.g. moving forward/backward in the direction the pit is facing) with between 0.5m and 1m travel and/or a real-orientation heave (up and down, in line with the earth not the sim) with 1m travel. There are limitations, mainly that both those extra movements would be limited depending on pitch, due to design and height limits. At horizontal they have full movement, vertical is 0.1m (or 0.2, I can't recall) movement, and just beside vertical is no movement, with it then becoming full movement again just after that. So for about 80 degrees on each side it's full movement (if that makes any sense).

I'm both not sure whether these would add anything, and especially for the heave, whether it's remotely feasible to interface them. It also adds complexity to the design, and construction. With the surge axis I then require extra high-strength actuators, which would add more cost. The heave axis is not much more cost, and makes entry/exit easier, although it may not end up being used for anything except lowering the sim for entry/exit. I'm not sure whether it'd be feasible to interface a real-world oriented axis with the sim, if that's the only translational axis I'm implementing.

The benefits of this system are no hydraulics, entirely powered by electric motors, and a lot of freedom of movement. It also requires a small area (5.5m x 5.5m, and my sim design is 4.5m long). It's biggest downside compared to a stewart platform or such is that it requires (precisely) 6.25m vertically. Which isn't too hard, I plan on spreading it across two stories (standard floor height is 2.5m including dividers here, so just sink it down slightly).

For now, my skills aren't really an issue, I'm a computer network engineer with a lot of background experience in electrical engineering (mostly microprocessor-based systems) and some mechanical engineering, so the building side shouldn't be that bad. And I've got help :)

Anyway, sorry for the long post, I thought I'd try and cover all of what I wanted to say. Any advice/stories/ridicule is welcome!

Thanks,
Matt

[kermit]

I think besides contruction limitions of your house/building also eletricwise,
You have to know what kind of cockpit are you building, what are the parameters of the plane,
a f 15 figther will have very different parameters then lets say a 737 or Cessna ,
I can`t find pictures of 737 do a 360 degree roll with holidaymakers on board:)
If you building a 737 or other airliners a 3 dof is all your need but if you have 6 dof within reach why not.
Also if your are planning on 360 degree rolls or high speed pitch/bank, you have to calculate the strengh of the material used.
Things can get very wrong when parts starts snapping off during a high speed S-turn.
Also you have to know your budget and planning, will you make motionparts yourself or are you planning on buying shelfstuff.
Before you start planning , its advisable to search the internet and get things clear in your head before starting this major event.
But when you start have a nice time and think of the end result
greetz
please make pictures during building , so we can follow you progres

[TheFearow]

I think besides contruction limitions of your house/building also eletricwise,
You have to know what kind of cockpit are you building, what are the parameters of the plane,
a f 15 figther will have very different parameters then lets say a 737 or Cessna ,
I can`t find pictures of 737 do a 360 degree roll with holidaymakers on board:)

I've seen one do it without passengers, but would make an interesting holiday if you were on board :). The cockpit is a semi-custom long-range trijet, low capacity (business-jet kind of seating).


If you building a 737 or other airliners a 3 dof is all your need but if you have 6 dof within reach why not.
Also if your are planning on 360 degree rolls or high speed pitch/bank, you have to calculate the strengh of the material used.
Things can get very wrong when parts starts snapping off during a high speed S-turn.

I can imagine, and that would especially be a problem inverted (don't want my pedestal falling onto my overhead). The motion system provides a metal framed box, that's about the right size for the cockpit, so most of it was going to be built off that and I was going to use either metal or some strong wood for the internal cockpit structure.
Also a note, it would be an approximate either 4DOF or partial 5DOF system, since the heave isn't quite "normal" heave.


Also you have to know your budget and planning, will you make motionparts yourself or are you planning on buying shelfstuff.
Before you start planning , its advisable to search the internet and get things clear in your head before starting this major event.

I've done a reasonable amount of searching, and I'll certainly be doing more before construction, as the area which I plan to build it in isn't ready for a couple months. Most will be made myself, which is partially why I like this design as easy to build (relatively).


But when you start have a nice time and think of the end result
greetz
please make pictures during building , so we can follow you progres
I'll be certain to set up a blog and such with progress once I get going, probably even before then with plans and models. And I'll make sure to take lots of photos.

-Matt

[kermit]

Hi,
I wouldn`t advise you , not to use wood , if you are intending to do 360 degree rolls,
as it is a natural component and the platform structure endures a lot of forces in all directions,
you are better off using aluminium ( you could use wood for light lining offcourse).
If you have extra dof features you could rebuild a line to get heave action,
do you have a picture of the platform you are going to use?
greetz

[TheFearow]

I was mainly thinking of using wood for some internal stuff, e.g. instrument frames (MIP etc). Do you think even for that aluminium is a better choice?

I've thrown together a quick diagram of approximately how it works, although the details of the mechanics are left out, it's just an overview. I've attempted to label it for a better explanation.

Just as a note, I was looking at the wrong height in my original post, the actual specs for size would be:

Total Height: 6.5m
Maximum Width (either direction): 4.5m

Note that this doesn't account for the method of getting in/out, which would be separate (I'm guessing a fold-down set of basic steps would be best).

-Matt

[kermit]

Hi Matt, Nice picture
what are you going to use as main current?
I don`t want to put you down, but looking at your answer regarding internal structures and the use of wood,
I`m really wonder if you know anything about quality characteristics of materials
and how to calculate forces and other mechanical characteristics .
I wouldn`t be in your cockpit when a heavy Mip shears off and starts coming down on you while sitting up side down.
Do you know the maximum weight the platform can lift and the speeds,
and I would advise you to use aluminium(popped and glued if you can`t weld alluminium) instead of wood but steel would be better.
But only if you really want to go upside down, if you are going for a 3 dof you can use wood provided you use quality. just my 2 cents

Do you already have all the components for the flightsim itself like seats, mip, screens ,overhead and so one?
The design itself looks great, you can add heave iff you rebuild the colums .
In that case you have to use limit switches or program your software ,
as it will cut off banking at a certain height or you will slam into the concrete.
What will you be using for controlls like PLC /programmes
all in all I`m very interested in the end result
greetz

[TheFearow]

Hi,

Not sure what you mean by "Main Current" there.

Regarding materials, maybe I'm being unclear. I don't mean I'm using wood for anything strongly weight-(or load-)bearing, just surface and instrument frames etc. The main body, and the joints to attach everything to the body, would be metal (aluminium or steel, not sure which). Certainly wouldn't want a MIP falling on me, so I'd make sure everything was held in place well.

The weight the platform can lift I'm not sure exactly, from the original specs I'm basing off it's pretty reasonable, as they are made to lift simulator rooms with 6-8 people in them, including seats, railings, projectors, and effects systems. From what I've seen of them they'd be about the weight of a simulator. Speeds are an issue, and would depend on the motors and gearings I use. I'd probably experiment on what worked best, and I've seen sims based on the design I'm basing mine off go pretty fast. They'd be able to handle most aircraft activities fine.

The column plans I have come in two designs, one with heave (where the pitch pivots are mounted on track that moves up/down), and one without. I'm not sure whether the added complexity is worth it though. Software limits wouldn't be that bad, and I'd include some hardware ones as well. Certainly wouldn't want to be flung into concrete (or through the roof).

As for cockpit parts, not yet, as I don't have too much space right now. This setup is going to be in a dedicated space in a new building, along with my new workshop areas etc. So I don't have much constructed yet (although I have a lot of plans lying around).

For controls, I'd plan to run the whole thing off a modified PC, running a real time OS most likely. The motors themselves would be controlled through an I/O bus carried through the slip rings, with custom microcontrollers and motor controllers at each axis. I've got plenty of experience with those sorts of setups, and should be the best option. Not sure about running digital I/O through slip rings, does anyone know if I'd have a problem with interference/losses/wear/etc? Rotary transformers as used in VCRs etc might be another option, although I've never really used them before.

Thanks for the help/advice/discussion kermit, it's helping me very much :)

Thanks,
Matt

[kermit]

Hi,
I sorry some times I translate one on one with dutch sayings,
I mean Electric power
Maybe we can use both expertise, I`m working in the winches and capstansfield, and have a mechanical degree.
I`m know how to build things and know some things about pulling heave and stuff, and I can get every motor and gearbox I need
On the other hand I`m a complete noob about software and implenting software/plc to drive motion platforms.
still looking for motorcontrollers and plc to drive my motion platform 2 dof,
I want to use 220v 1.5 kw 1500 rpm 1 ph motor with a gearbox I=35,6 ratio ,
it will drive my platform easily at a rate of 36m/min both directions(as its only weighing about 300 kilo or 700lbs in total.
But can`t find a suitable plc and motorcontrollers for the setup
greetz

[TheFearow]

Hi,

For electric power, I'm not quite sure. DC is easier and cheaper to get through the slip rings and to control (I have experience with DC motor controllers) and I have a cheap source of DC motors, although AC motors seem more common. Not really my area of expertise. My house is already wired up for DC supply (very little AC buses), so that's a better option from an efficiency perspective. I have about 200A of 12VDC planned to be available in the new simulator room, and I may run a bar from the 24VDC bus into that rooom as well, which would give about 100A max. I have no idea if this is enough, but it should be if used properly. I'll also probably have 20A of 230VAC as well, although I could extend that and remove most of the DC buses.

It'd be great to share experience, software and networking are my main fields and I'm sadly lacking in the engineering and construction areas, although I do have some experience.
First of all I'm assuming by the fact you say 1 phase and 220v that you mean AC, not DC.
Assuming by motor controller you want a variable-speed controller, you're probably better off with a three-phase motor if you can, as far as I know. Almost all the controllers I have seen are for three-phase, and I remember reading somewhere that it's a lot harder to electrically control the speed of a single-phase motor. If you're not looking for variable speed, or would accomplish that non-electrically, then contactors would do.
For the software side, I'd recommend that you either go for a more powerful microcontroller (e.g. one of the higher-end PICs) or some sort of high-speed control from a PC. You want to reduce latency (lag) on the motion as much as possible. Do you plan to calculate all motion values etc in the sim software then send them to the microcontroller, or simply send the desired angles and have the microcontroller work out the values? (by values I mean actual sim orientation, taking into account the various tricks to fool your brain).

If you're looking for specs and/or good quality automation components (e.g. motor controllers, motors themselves, control circuits) I can recommend RS Components as a good place to start looking, although they are quite expensive.

Just out of curiosity, what is the standard AC frequency where you are? It would affect what controllers you can use.

Hope any of that helps, I tend to ramble on and get distracted sometimes, and it's late. If I've misunderstood what you mean, blame the time I wrote this :).

-Matt

[Roland]

Just my experience: Heave is a great addition to pitch and roll. It is more difficult to implement, but definitely worth it, both for GA and Airliner flying. For airpocket "stomach" feeling, you'll need at least 0.3m travel, and it needs to be fast (0.6m/sec actuator speed). for the small turbulence, (smooth) touch-down and ground bumps, small 5 ~ 10cm excursions are sufficient, but still need to be fast.

Measured total electrical power levels for fast heave step with 160kg payload: ~ 1.6kW peaks. Average power (in balanced system): ~ 120W. Power delivered by 3x12V batteries, driving PM DC motors in a 3-point setup.

[TheFearow]

Ahh, I can imagine Heave would be great especially for those. If I follow the heave addon for the original design, it simply moves the pitch pivots and mechanism up and down on a chain lift, which can go reasonably fast, although I'm not sure on exact specs. It has about 1m travel (each way, so total of 2m distance) at 0 pitch, and about 0.15-0.2m travel at 90 pitch. Somewhere around 80 pitch there's a zone with no ability for it, but that should be fine.
Would you recommend sticking with that system, or going an actuator-driven approach? I'd rather stick with what I'm familar with, although I can imagine speed etc might be a problem.

The amount of power may be a bit annoying, that's a lot of amperage going to one motor. I would probably want a bigger supply then, for Heave at least. The other axis aren't really doing much lifting (ideally), so they should be fine (except maybe Pitch, if I don't balance that well). Out of curiosity, how long were the peaks in your experience? For the duration of sharp movements or mainly rush current at the start of movements? I'm thinking of putting some large supercaps on the motors to cushion some of those spikes from hitting my supply, although that won't be helpful if they keep pulling for a while.

[Roland]

A chain lift construction can be build strong, and makes it easy to add a weight balance system. (lots of bungee)
But the pivots need to be super strong, and need to run in some kind of rail system to avoid wiggling. Building a structure with sufficient stiffness can be quite tricky.

But keep in mind that heave acceleration needs to be fast to be able to feel anything. With higher payload, the forces will increase according F=m*a. same for power.
For DC motors, using higher voltage motors (i.e. 50V in stead of 12V types) will help reducing current and reduce conduction losses considerably.
Note that all high power DC motor control are based on switch-mode PWM . They rely on motor inductance, so adding supercaps across the motor will result in smoke.

Acceleration duration is normally short, around 0.2sec. Of course longer travel makes longer acceleration possible.
Some scope plots on current during heave step :http://www.simprojects.nl/platform_drive_calculations_iii.htm

[TheFearow]

The original I'm basing this off had the pivots mounted on four metal bars running vertically as rails, which were very sturdy, and the lift itself had brakes and locks for when there was no heave needed. I can imagine the stiffness would be hard, especially since the columns themselves are turning on the yaw. It may help that the system is well secured at both the top and bottom, although I'd still need solid supports.

As for motors, I'll look into options for higher voltage - my main reason for wanting 12V is that my house is primarily 12VDC supply, although there are 24VDC buses. For payload, would counterbalancing the heave system (similar to how it's done with elevators) be helpful? It might be possible, and I can imagine it would help reduce some load, although the benefit might make it not worth it.

I meant supercaps on the motor supply, so for motor controllers on the voltage input of the controller, not the motor itself. Although now you mention it, the other way might work well as a cheap fog machine replacement, no fluid needed :).

Good to know about normal duration, and with longer travel (I'm looking at 1m at straight and level) basic turbulence and dips and things could probably be reconstructed truly, which would be nice.
That's great to read, and a lot of interesting information. I've had a bit of a read around, and a very nice site :).

Thanks for all the help/suggestions/feedback!

-Matt

[Roland]

For payload, would counterbalancing the heave system (similar to how it's done with elevators) be helpful? It might be possible, and I can imagine it would help reduce some load, although the benefit might make it not worth it.

Counter balancing with counter weight like in elevators will double the mass that needs to be moved. Using bungee is light weight, and won't add extra mass.
Using air-pistons and expansion vessel is another low mass solution, and more flexible for tuning the force. Succesfully applied by HI6sim http://www.xs4all.nl/%7Ejippes/hi6sim/index.html

[hca]

Hi Matt,

Based on a scan of the topic I suspect you are heading toward building a sim similar to what I have been working on nearly every day for the last 4 months.

If so, i can relate a bit as to where you are headed, not to frighten you off, but to assist you if you are deviating from the normal plus minus 30 degree type motion platform. I will describe what i have been doing and mention the additional s that crop up.

Briefly my sim 2 stated out as a rotating motion assembly into which i would transplant plant my single seat generic sim 1 now in the lounge room which is built with a steel frame with a view to motion

My sim 2 consists of a rotating platform 3 meters in diameter, on top of which sits a gimbal frame 1.7 metre above the rotating table (as in gyro), in this gimbal frame (just under 3*2 mtrs), sits the cockpit it self. The original scope was for a full Boeing size cockpit but for now to take the sim1 single pilot cockpit.

So we have unlimited pitch, roll and yaw for 360 degrees. Yaw in the rotating base, roll in the gimbal on the base then pitch with the cockpit inside the gimbal. Power flows and data through sliprings on the 3 bearing axis's.
I considered heave and surge but the shed is just not big enough and nor is the mains supply.

For a start without the cockpit and the access stairway and retracting walkway, there is 400 kg of 50mm*100mm steel box, and quite a few 6 mtr lengths of assorted size steel bar. 10 1 inch flange bearings, 5 1 inch custom machined shafts, 5 custom machined 1 inch boss/flanges plus the sliprings and 8 steel roller wheels.

I calced out wheelchair motors would move every thing fast enough via a suitable reduction drive from 24volts. only problem was switch mode supplies i could find were not really up to providing the required peak 40 plus amps. Add 2 truck batteries to the weight on the rotating platform. I figured one motor on each axis to be enough as long as the sim was balanced but worked on using 2. So far one motor rotates the gimbal and platform 360 degrees in just under 2 seconds(no cockpit installed as yet). Here in Bali there is no source of suitable reduction drives, but motor bikes are dirt cheap. it turns out they run 15 to about 40 teeth to drive the back wheel, fabricate a boss for the motor and weld on a 15, then an idler with a 15 and a 42 tooth, then another custom made 1inch boss with 42 tooth and 2 chains. and its more than mobile enough for a flightsim, probably even a fighter but thats not the aircraft dynamics I plan to simulate. Dont forget you might want to stop and reverse this moving mass quickly.

Ok the cost of those machined parts here in Bali probably ran to 300 plus USD , but we are adding kilos all the time, and these are moving kilos that the motors must be able to control safely while the weight is moving fast and has a lot of leverage. ( ie upto 2 meters from the rotation axis). Consider flex and torsion in you plans, also metal fatigue and stress points. I think you will ditch wood and ally if you are building some thing similar to me.

Based on the results achieved with 1 wheelchair motor for the rotating platform and 1 for roll on the gimbal which are both mounted on the platform, I figured i would go with 1 on the gimabal frame for the cockpit roll. Thus 2 meters of 50mm*100mm steel box filled with scrap steel and concrete (28KG)is added diagonally opposite the pitch motor to restore the balance on the gimbal frame in lei of the second motor and drive.

Balance is important and difficult to perfect in a spheroidal machine with out huge motors, buy an electronic scale!

If you don't have a hoist in the roof, i recommend you plan on building some kind of removable lifting system in the towers that hold the gimbal to lift it and later the cockpit into the gimbal. 6mm steel rope, steel gate rollers and a 1 ton chain hoist worked for me.

Next consider how the sim will be kept stable on both axis when getting in and out. add more electric actators and steel locking pins complete with the mechanisms to move them and the (dueled for safety) sensors to control them.

Also how to get into and out of if? I ended up with a 2 metre high staircase and retracting walkway, add the other half of the double garage and another electric actuator plus 30 metres of 1 inch steel box and a few more sheets of 15mm plywood . Hand rails? add another 30 metres of steel rod and a bit more box.

All of this stuff needs to be made strong and safe, over engineer the steel work and lots of interlocks, you plan to ride in this thing, and probably take your kids, it need to be safe. This kind of machinery can take your fingers off, break your arms, or crush your body, or dice your kids while you fly. Big weights, powerful motors and lots of mass moving fast. Its a no go zone in the immediate vicinity

Obviously this kind of kit is not entrusted to Microsoft. every thing here is controlled by microchip pics and linux.

Also a balance adjust system is required unless you use huge motors, add another 100kg of steel and concrete a dozen more steel roller wheels and a few more actuators.

For linear actuators I have used a lot of 18 inch satellite dish positioning actuators, i anticipate a total of 10 complemented with 7 12v wiper motors and a handfull of electric car door lock motors. Why satellite actuators,? here they are very cheap complete with the 220v power supply/controller (50USD) which is controllable via the antenna socket with DISQ? which is easily generated on the pin of a pic. The same pic can monitor the sensor in the actuator to confirm commands are performed. Plus they can be cut down from 18 inches and are better than anything I can fabricate locally to do the same job. Possibly a bit slow but if you want fast up goes the weight, power requirements, and cost.

So now it stands in the double garage which also had to be modified to give height clearance, in all its glory waiting to have the sim 1 modified and dropped in. along the wat i realised sim 1 was only built with1.5 inch anle was not up to the grade. So now i am building a new sim base and frame to renovate the old sim 1 up to strength and dual pilot. ie, full rebuild. The last 4 weeks, and over 40 bearings have gone into the dual rudder pedals with force feedback, centering units and auto pilot control. currently working on the dual yokes which will be a similar setup, but thats another strory


As for the cockpit, aircraft style adjustable seats subwoffers, and modified genuine instruments add up in Kilos very quickly, I think the ally TQ is over 15 kilos, then add a couple of fat guys and a case of beer!

Fortunately all of the pics and half a dozen PC motherboards are pretty nominal in the weight game. Ah plan to run your PC code and OS of thumb sticks in ram, HDs dont like this kind of motion!

Any way there a few out dated pics on my site in the sim 2 album if you are interested.
http://www.bali-gold.com/Avionics/thumbnails.php?album=8

Also, theres no drawings, except the acad of octagons for the motion base to cut the steel.

reg Harry

[TheFearow]

Hi Matt,

Based on a scan of the topic I suspect you are heading toward building a sim similar to what I have been working on nearly every day for the last 4 months.

If so, i can relate a bit as to where you are headed, not to frighten you off, but to assist you if you are deviating from the normal plus minus 30 degree type motion platform. I will describe what i have been doing and mention the additional s that crop up.

Briefly my sim 2 stated out as a rotating motion assembly into which i would transplant plant my single seat generic sim 1 now in the lounge room which is built with a steel frame with a view to motion

My sim 2 consists of a rotating platform 3 meters in diameter, on top of which sits a gimbal frame 1.7 metre above the rotating table (as in gyro), in this gimbal frame (just under 3*2 mtrs), sits the cockpit it self. The original scope was for a full Boeing size cockpit but for now to take the sim1 single pilot cockpit.

I'm still planning for a similar-to-Boeing cockpit, with quite a few customizations, to be a sort of custom-designed aircraft.


So we have unlimited pitch, roll and yaw for 360 degrees. Yaw in the rotating base, roll in the gimbal on the base then pitch with the cockpit inside the gimbal. Power flows and data through sliprings on the 3 bearing axis's.
I considered heave and surge but the shed is just not big enough and nor is the mains supply.

I'm thinking mostly the same, except the sliprings would only carry power and emergency stop signals, the motors themselves would be controlled wirelessly (although I may switch to a 2-pin bus for controlling them through the sliprings). At this point I think I am going to add Heave, but Surge is out, I don't want the complexity. And it's not one of the most important ones, at least not during flight. I'll use tightening seatbelts as I've seen on some designs to get a braking effect.


For a start without the cockpit and the access stairway and retracting walkway, there is 400 kg of 50mm*100mm steel box, and quite a few 6 mtr lengths of assorted size steel bar. 10 1 inch flange bearings, 5 1 inch custom machined shafts, 5 custom machined 1 inch boss/flanges plus the sliprings and 8 steel roller wheels.

There's a lot of stuff that goes into this. I'm not the best metalworker, is it worth studying into this more as well?


I calced out wheelchair motors would move every thing fast enough via a suitable reduction drive from 24volts. only problem was switch mode supplies i could find were not really up to providing the required peak 40 plus amps. Add 2 truck batteries to the weight on the rotating platform. I figured one motor on each axis to be enough as long as the sim was balanced but worked on using 2. So far one motor rotates the gimbal and platform 360 degrees in just under 2 seconds(no cockpit installed as yet). Here in Bali there is no source of suitable reduction drives, but motor bikes are dirt cheap. it turns out they run 15 to about 40 teeth to drive the back wheel, fabricate a boss for the motor and weld on a 15, then an idler with a 15 and a 42 tooth, then another custom made 1inch boss with 42 tooth and 2 chains. and its more than mobile enough for a flightsim, probably even a fighter but thats not the aircraft dynamics I plan to simulate. Dont forget you might want to stop and reverse this moving mass quickly.

I've got less of a problem on supply, with the house wiring being DC at 12/24v, and offering a lot of amperage. Reduction drives etc I can get through an engineer friend, nice and cheaply hopefully. Or rip them out of other things as you did :). And for stopping, then plan was to use brakes (the same ones used in the safety system) assist the motors in direction changes by slowing the mass in the direction it's going.


Ok the cost of those machined parts here in Bali probably ran to 300 plus USD , but we are adding kilos all the time, and these are moving kilos that the motors must be able to control safely while the weight is moving fast and has a lot of leverage. ( ie upto 2 meters from the rotation axis). Consider flex and torsion in you plans, also metal fatigue and stress points. I think you will ditch wood and ally if you are building some thing similar to me.

I think the strength of the design in my case is improved since it's attached to the solid building at both top and bottom, but I can see what you mean by having to use stronger materials.


Based on the results achieved with 1 wheelchair motor for the rotating platform and 1 for roll on the gimbal which are both mounted on the platform, I figured i would go with 1 on the gimabal frame for the cockpit roll. Thus 2 meters of 50mm*100mm steel box filled with scrap steel and concrete (28KG)is added diagonally opposite the pitch motor to restore the balance on the gimbal frame in lei of the second motor and drive.

Balance is important and difficult to perfect in a spheroidal machine with out huge motors, buy an electronic scale!

With what I'm looking at, the motors would be in pairs and evenly spread across, so balance of them isn't a problem. Balancing the sim will be, so I'll have to counterweight that somehow.


If you don't have a hoist in the roof, i recommend you plan on building some kind of removable lifting system in the towers that hold the gimbal to lift it and later the cockpit into the gimbal. 6mm steel rope, steel gate rollers and a 1 ton chain hoist worked for me.

It's going to be in a purpose-built room, so I will probably add a hoist in the center of the assembly on the roof, and probably two separate lesser ones on either side, to make moving the simulator and such into place a lot easier.


Next consider how the sim will be kept stable on both axis when getting in and out. add more electric actators and steel locking pins complete with the mechanisms to move them and the (dueled for safety) sensors to control them.

I had planned for locking pins as well as friction brakes on all the chains, which would be safety locked, with a wired AND on safety switches as well as intelligent control for during operation.


Also how to get into and out of if? I ended up with a 2 metre high staircase and retracting walkway, add the other half of the double garage and another electric actuator plus 30 metres of 1 inch steel box and a few more sheets of 15mm plywood . Hand rails? add another 30 metres of steel rod and a bit more box.

My idea was to use a simple staircase on rails, with a pulley system to move it towards/away from the sim. If heave is added, which at this point I want to do, then the sim's height off the ground wouldn't be that much during park (at maximum negative heave).


All of this stuff needs to be made strong and safe, over engineer the steel work and lots of interlocks, you plan to ride in this thing, and probably take your kids, it need to be safe. This kind of machinery can take your fingers off, break your arms, or crush your body, or dice your kids while you fly. Big weights, powerful motors and lots of mass moving fast. Its a no go zone in the immediate vicinity

Since it's planned for a dedicated room, I'll have it integrated into my house locks (all doors use electronic locks) to prevent access while operating, as well as a multitude of sensors and cutoff switches to prevent anyone getting near it. I'll also have a camera system so I can monitor the simulator room while flying, if need be.


Obviously this kind of kit is not entrusted to Microsoft. every thing here is controlled by microchip pics and linux.

For control, I plan on controlling it with dedicated hardware, a bit more than PICs, most likely a customized real-time server, and probably running a customized real-time OS, although I'd use one based on Linux. Software is my specialty, so I know the requirements and best options there.


Also a balance adjust system is required unless you use huge motors, add another 100kg of steel and concrete a dozen more steel roller wheels and a few more actuators.

What exactly do you mean by balance adjust system? Do you mean like moving counterweights?


For linear actuators I have used a lot of 18 inch satellite dish positioning actuators, i anticipate a total of 10 complemented with 7 12v wiper motors and a handfull of electric car door lock motors. Why satellite actuators,? here they are very cheap complete with the 220v power supply/controller (50USD) which is controllable via the antenna socket with DISQ? which is easily generated on the pin of a pic. The same pic can monitor the sensor in the actuator to confirm commands are performed. Plus they can be cut down from 18 inches and are better than anything I can fabricate locally to do the same job. Possibly a bit slow but if you want fast up goes the weight, power requirements, and cost.

I'll have a look at those - they could be useful for a few things in my design, not just the motion system. Most of the motion in the design I have so far is done by geared-down motors directly, not linear actuators though.


So now it stands in the double garage which also had to be modified to give height clearance, in all its glory waiting to have the sim 1 modified and dropped in. along the wat i realised sim 1 was only built with1.5 inch anle was not up to the grade. So now i am building a new sim base and frame to renovate the old sim 1 up to strength and dual pilot. ie, full rebuild. The last 4 weeks, and over 40 bearings have gone into the dual rudder pedals with force feedback, centering units and auto pilot control. currently working on the dual yokes which will be a similar setup, but thats another strory

I've got plenty of space, from a dedicated room, and my sim design is being coordinated with motion system design from the start, so shouldn't be many problems there.


As for the cockpit, aircraft style adjustable seats subwoffers, and modified genuine instruments add up in Kilos very quickly, I think the ally TQ is over 15 kilos, then add a couple of fat guys and a case of beer!

I can see how much the average cockpit would weigh - as well as the fact that the cockpit design itself that I'm using is complex, various effect devices and such that I'm using certainly won't help the weight. Although I can't imagine it would get too ridiculous.


Fortunately all of the pics and half a dozen PC motherboards are pretty nominal in the weight game. Ah plan to run your PC code and OS of thumb sticks in ram, HDs dont like this kind of motion!

I'm going to be using customized low-profile and low-partcount computers off of SSD's, so no problem with motion causing problems there. And fans for everything, because convection cooling isn't the best when you tip everything upside down.


Any way there a few out dated pics on my site in the sim 2 album if you are interested.
http://www.bali-gold.com/Avionics/thumbnails.php?album=8

Also, theres no drawings, except the acad of octagons for the motion base to cut the steel.

Very interesting, I've had a quick look and I'll be sure to read it in a bit more detail later. Looks good :).


reg Harry
Thanks for the help/advice/suggestions/etc. I'm starting to get a much better idea of what I need to do etc. It's not a small job ;). I've got about 6 months before work can fully start anyway, so plenty of planning time!

-Matt

[kermit]

There's a lot of stuff that goes into this. I'm not the best metalworker, is it worth studying into this more as well?

Hi Matt,
Again , what do you know about materials?
If you want to throw a few hunderd pound of weight in all kind of directions,
You have to know about them.
Putting weight between 2 collums with a lengh of about 6.5 meter you will encounter flex, putting a lot of strain of joints and welds.
If you don`t know much about metalwork or gearing, you have to read and learn.
How do you want to make construction drawings without knowing what your joints,profiles,matrials can or can`t do?
Also if you can calculate the materials you know,
how much certain constructions will weigh and what you must use to be safe.
People have died or got into accidents ,
because they didn`t know how to make good welds or know how to secure joints or constructions.
I have a lot of experience in enginering but I don`t know no much about electronics so I have to learn or stay away.

May`be you can start to learn how to make good welds or make simple constructions just to know what you will encounter.
It won`t cost much, but you gain experience

What budget do you have,
greetz

[TheFearow]

Hi Kermit,

I know the basics about engineering and materials, strengths, flex, etc. What I'm not the best at, at least yet, is actually building things. I'm planning on doing a lot of study in my period before main construction/proper designwork starts, which is about 6 months, and I'll be doing scale tests etc. I also know a couple engineers who've said I can run designs past them if need be, which I'll definately do. They also offered to help with design and to help me learn things I'm not sure about, which will be helpful.
I'm also going to work on a few smaller projects in the mean time, which will help a lot I'm guessing.

As for budget, I'm not one of the people that expects a full motion system for $100. I've worked on similar projects professionally (although my job was never in the mechanical side), so I know the sort of money it takes to make them. Factoring in the amount I'm doing myself, or that I'm scavenging, It's still not a cheap hobby. I'm not sure on exact figures, but I know what I'm in for.

-Matt

[kermit]

Hi Matt,
I will follow your progress in your project.
Maybe you can explain the software you will be using,
As I`m working on a 2 dof with a option for 3 dof , for now I `m in contact with a inverter specialist.
To drive my platform with 220 volt AC motors and gearboxes, also still thinking about hydraulics
greetz

[simcraft]

Matt, based on your designs your cockpit seems very similar to the center of mass architecture we use for our motion systems. There is an interesting article (http://www.simcraft.com/education/5) that address the benefits (or lack of rather) of additional DOF beyond 3 (either RPY or RPH).

Basically the abstract is that professional airline pilots reported in almost all cases, that 3DOF sims produced motion simulation quality comparable to that produced by the much more expensive and complex 6 DOF Steward platform. The study was conducted using nine combinations of architecture and motion algorithms designed to provide a study of a broad range of motion simulating low and high frequency accelerations in all degrees of freedom.

Good luck with your project, correctly adding motion to your cockpit is a very rewarding experience.

Michael Boardman (http://www.linkedin.com/in/mrboardman)
SimCraft (http://www.SimCraft.com) Information Officer

[kermit]

Hi Michael,
Did you overcome unbalanced platform,
and could you explain how without telling mfr secrets offcourse, also is it poss to buy your plc only please
greetz

[simcraft]

Did you overcome unbalanced platform.
Balance is as important to a successful motion platform as it is to flying an airplane successfully. Moving the center of rotation into line with the center of mass can be pretty easily accommodated by adjusting the height of the frames.