Phase II Testing - 6dof Sim Model...

[sagrada737]

Hello Folks,

I have started with Phase II of my testing for a potential 6dof Motion Platform for my 737-800 Sim Project.  The following video shows the Sim model, which is 1/8th scale, sitting on top of the 1/8th scale Stewart Motion Platform.  Within the video are descriptions of flight activity showing how the Platform responses to various "motion cues".

Keep in mind that I am still in the process of determining the correct balance in setting up the motion cue parameters to best fit with the 737 Flight Model (Sim-Avionics) operating with FSX.  So...  Some of the movements might be a surprise to you.  It turns out that FSX does some strange things with respect to supplying FSUIPC data that the motion software uses to command the 6dof motion software.

Also, in this test, I was using ActiveSky Next for realtime weather, which happened to be clear and winds calm.  As a result, the motion cuing is subtle.  In turbulence, things get exciting.  As you watch the video, imagine yourself in the cockpit "sensing" the flight movements.

In parallel with further testing the motion models, I will refine the mechanical design for the actual Flight Deck Base that supports the FDS Nose Section and all the other Sim systems - including the three 70" LCD Monitors supported on a "flying bridge" type of structure.  An interesting design challenge for sure 

Also, I have settled on a 220VAC 1.5 HP Motor/Gearbox (one of six required for the Stewart Platform) that I will conduct actual load/torque testing under FSX/6dof motion software/motion hardware.  This should yield some interesting data and information to confirm viability for such a drive setup.

Anyway, here is the video showing the Sim and Motion Platform models doing their thing.  Enjoy the ride 

Mike

https://www.youtube.com/watch?v=AUc9bTsAQNw&feature=youtu.be

[fsaviator]

Mike,
truly amazing stuff!  I am in awe of the process and the attention to detail.

I think I can speak for all of us in that we are all watching your work on this closely.

[727737Nut]

Very nice scale model!!   What software are you using?  I used BFF on my 3 axis motion sim.  Worked really well and really added to realism.   
Rob

[sagrada737]

Thanks for the comments guys.  It's good to have the encouragement - especially with a big project like this 6dof Motion Platform.

The 6dof software I am using is from BFF  I also am using their Shaker software for cockpit vibrations.  Both of these read FSUIPC from FSX to accomplish what is required - and both offer a high degree of customization for output parameters.

I am now in Phase III of my testing of the motion platform - observing what the Sim does on top of the motion platform in high turbulence.  I'm flying in fairly high turbulence over the Denver area, and itis making for an exciting ride.  Maybe I'll post a short video clip of this later, as it really gives you an idea of how the Stewart Motion Platform can move about in response to cue commands.

Mike

[meatbomber]

Hi Mike,

Great project me and my friend have been researching ways of adding motion to our cockpit and we are also looking at possible making it a 6DoF system.

I,m wondering have you ever put your sim on the scales?
We haven't dared yet but ours is a steel frame construction 2 seater with a large projection screen. So I'm thinking it will tip the scales loaded at about 1000kg and I'm wondering if the same spec drive system as yours can be used or how much more power will be required.

Could you post which motors/gear heads/inverters/controllers you are planning on using?

Thanks
Phillip

[KyleH]

Hi Mike, nice looking mock-up.
One thing I noticed or I guess really didn't notice was the simulation of acceleration and deceleration g forces. The platform should be tipping backwards at a noticeable angle when you accelerate at takeoff. Same with braking on landing.

[sagrada737]

HI Phillip, Kyle, Rob,

Thanks for the comments.  Regarding some of the motions seen in the motion platform 1/8th scale model...  I agree that in some modes, it does not seem to be moving enough.  However, I am in the beginning stages of "bracketing" the motion cue parameters to find the combination to yield the best result with my 737 flight model (FSX/Sim-Avionics).  With the mini-servo motion platform model, I may be able to get close to what the settings should be.  In reality, the mini-servos acting on the model will not yield exactly the same behavior as the real AC Motor/Gearboxes that will be used for a full-scale Sim.  In this regard, the mini-servo model serves a limited function in testing, but it has been useful thus far.  No doubt there will be a lot more tweaking of these motion cue parameters once a full-scale 6dof Motion Platform is constructed and functioning with my 737 Sim sitting on top and me inside.  I suspect that many of the motion cue parameters will change - perhaps some significantly.

The BFF 6dof Motion Software has the ability to "Capture" the flight dynamics to help in setting up these motion cues.  I am now working with that feature in hopes of refining settings.  However, there will be a point where further refinements of the servo motion platform model will not yield any new data, and I will then have to take things to a new level that involves building up the real platform.  That may come after further testing of the model and one of the VFD/motor/gearbox units to determine how the real hardware behaves under real-world loads.

Phillip - Regarding sourcing of these systems...   Here are some links that might be helpful for you:

BFF 6dof Motion Software
http://bffsimulation.com/6DOF-Motion-Software.php

AMC1280USB Motion Controller Hardware
http://motionsim.blogspot.com/2013/04/6dof-platform-in-action-amc1280usb-ians.html

Hatachi Variable Frequency Drive (VFD)
http://www.galco.com/buy/Hitachi-America/WJ200-015SF#bottom

1.5 HP Inverty Duty Motor 220 VAC 3-Phase
http://www.automationdirect.com/adc/Shopping/Catalog/Motors/AC_Motors_-_General_Purpose_and_Inverter_Duty_(0.25_-_300HP)/AC_Motors_-_Inverter_Duty,_Marathon_(0.25_-_100HP)/Marathon_Black_Max_(0.25HP_to_30HP)/Y538

NORD Helical-Bevel Gearboxes
http://www5.nord.com/cms/us/product_catalogue/geared_motors/helical_bevel_geared_motors/pdp_helical_bevel_geared_motors_1517.jsp

Another consideration is Weight and Balance.   If you want to build up a Motion Platform, you will need to calculate the CG of your Simulator (mass).  In my case, I established the Datum from the back of the FDS Nose Section.  Since my Sim Project is still not yet fully integrated, I was able to weigh many of the Assemblies, eg. MIP, Seats, Overheads, Pedestal, etc.  This allowed me to establish the Arm and determine the Moment for each of the Assemblies that make up the Sim.  For my Sim, I came up with a total weight of 1820 lbs. for the Sim, its Flight Deck Base, and the Displays, which yielded a CG of 44.8 inches from the established Datum.  There are other issues with respect to CG that consider moments of inertia that impact torque and acceleration limits for the motion platform drive system/hardware.   In this regard, establishing CG location and limits are probably the most important for stable performance in a motion platform, within the capabilities and capacity of the motion control drive systems.  In this regard, I believe it is better to "over-build" the capabilities of the Motion Platform to ensure it can move the mass without straining or overheating of systems/components.

One design tool you might find useful is offered by Fly Elise-ng.  It is called "Motion Platform Designer" and has a nice set of functions to help you visualize system parameters for various motion platforms - including a 6dof system.  It also allow you to do simulations to see motion behavior with various inputs and resultant response and force applied to the system.  Here is the link:
http://www.fly.elise-ng.net/motionplatformdesigner

I hope this information is helpful to you.

Mike

[bernard S]

I have a full level d motion platform complete  and I can assure you.. that what you are undertaking is no simple project not to rain on your parade....but good luck with it all the same.

some further considerations knuckles to mount the jacks too they "are they going to be  single castings"  how are you going to mate the jacks to the floor plates..how are you going to support the floor area .... dont forget the throws... just the bolts alone to support these things are a fortune....  your inverter is too small mine is 480/200amps... how are you going to reset this beast?.... are you making provisions for maintenance ( to operate from cockpit is not wise ) are you going to have stand alone control cabinet?  and did I say money and more money its going to eat it

[meatbomber]

Bernd my understanding is that each motor will hav it's own inverter so in total 6.
I also think that a "hobby grade" platform and sim can be built quite a but less massive than a level D sim that has a daily utilization of 20hours for 6days a week all year round

I'm wondering how much the cab on your sim weighs?
Gee I really have to put ours on the scales (the advantage when you have it in an aircraft hangar)

I'm also wondering just how much travel one really needs if you don't want to simulate the edges of the flight envelope. I think an aborted takeoff in someone's basement can be quite impressive with half the g load that the level D makes.

All that said, man how did you gets Level D sim

[sagrada737]

Hi Guys,

I certainly respect Bernard's comments and significant set of cautions concerning the "what ifs" for all the details of owning, operating, and maintaining a complex 6dof Motion Platform/Simulator.  Indeed Bernard...  It looks like you have your hands full, and I would guess that you are living all those concerns each and every day.

As for my Sim Project - it is strictly a hobby.  If it all goes in the dumpster tomorrow, I will simply chalk it up to a learning experience, and move on with other interesting things in life.

As far as considering building my own 6dof Motion Platform for my 737 Sim, this too falls into the hobby category, and my efforts thus far are clearly exploratory in nature, as I have some fun working on design concepts for a "hobby grade" 6dof motion platform that would support and manage the mass associated with my "hobby grade" FDS Sim.  It is what it is...  Thankfully, it doesn't have to be as massive as the Level D Sim shown in the photo Bernard provided.

By the way Bernard...   that photo was very impressive.  Wow!!!  That's a nice piece of commercial Sim equipment for sure!  I'm guessing that the linear actuators alone for that Level D simulator weigh almost as much as my entire FDS Sim Project (less motion platform).  Equally impressive is the total mass of that beautiful Level D Simulator, which no doubt requires "everything" associated with it to be equally massive and powerful - hence the 200 Amp Drives. etc.

One thing I take to heart from these recent comments, is the fact that such a hobby grade 6dof motion Platform is still a serious project, with serious considerations - not the least of which is safety.  Right now, I'm in a "design safe zone", where I have made a minimum investment to explore (really playing around) how such a 6dof motion platform functions, and how it might be possible to build my own full-scale 6d0f motion platform for my existing "hobby grade" 737 Sim. 

In all this, I am exploring how it might be possible to create a viable design concept, for a relatively simple 6dof motion platform system, that will support and move the 1,800 lbs of mass that represents my 737 Sim.  This is why I chose to begin with a simple but efficient approach that uses Variable Frequency Drives to power/control a 220 VAC Motor, using a fairly low cost Helical-Bevel Gearbox, which ultimately moves a Lever that articulates a Push Rod, which moves the Upper Platform according to commanded "cues".  It is my opinion at this point in time, that the BFF 6dof motion software combined with the AMC1280 Controller system allows for a low-cost solution for a complex problem.

So at this point, the motion software/hardware I am currently testing seem to be working nicely.  Of course, I'm only working with this 1/8th scale model, but it has been a big help to better understand and visualize how such a 6dof motion system functions, and how it might be possible to build up such a system for my current 737 Sim.  Even if I were to stop here, this has still been a great deal of fun.

My next step however is to test one segment of the 6dof motion platform legs (one of six) with the actual VFD/Motor/Gearbox as interfaced to FSX/6dof motion software/hardware, in order to determine just how well this combination will move its share of mass under the conditions demanded by the 6dof motion platform in actual operation.  This is where I will make a go/no-go decision on proceeding with this project.  My initial calculations show that it is possible for the VFD/motor/Gearbox to adequately perform the task, considering the mass I want to move around - including frequency response times, duty cycles in this rigorous start/stop/fwd/rev/accelerate/decelerate environment.  I will make this determination in the next few weeks of testing.  If it doesn't measure up...  Then I will shut down this idea and move on to finish up my Sim Project as a "static system" - not the end of the world 

The facilities consideration that Bernard mentioned is indeed a concern.  Fortunately, my Sim Room will accommodate such an installation, and the concrete floor will withstand the loads involved as anchored to the floor.  Keep in mind that my design concept for this 6dof motion platform is a hybrid system, that incorporates visual/motion cues, along with vibration/sound "cues" - all of which will help to minimize the demands on the motion platform.  As such, I only need to create a "hint" of motion sensation to accomplish what I have in mind.  In this regard, the success of such a low-end motion system will still need to be refined to eliminate (as much as possible) elements of "simulator sickness" due to asynchronous motion cues.  I'm thinking that this will be the more significant challenge.

I hope no one takes this as my being defensive - I'm just having fun in this season of my life, trying to keep my mind active doing interesting and complex things.

Best Regards,
Mike

[sagrada737]

Hi Guys,

I decided to "crank up" the turbulence settings within Active Sky Next, and see how the 6dof Motion Platform model would handle the "harsh movements".  I found a thunderstorm on approach to Heathrow Airport, which made for an exciting ride!  By the way...  I purposefully have the model base unsecured to better "excite" the mass on the upper platform - a basic way to see resonance in the system.

On a serious note...   The motion cues induced into the Sim model in this test video are fairly abrupt.  It makes me realize that such active shock would be way too excessive for my "hobby grade" Sim to withstand.  In any case, this testing gave me an idea of the moments of inertia, and how such an elevated mass might behave with these kind of motion cues. 

Also, I have been experimenting with the various motion cue parameters, for improved performance, eg. Braking and Take-off.  Lots for me to learn in all this.  The scale model has helped a great deal to visualize how changes in cue parameter affect the resultant movement of the upper platform.  No doubt these settings will change with a full-scale motion platform - should I go down that road.  Anyway, enjoy the test video.

Regards,
Mike

https://www.youtube.com/watch?v=V39c7f2rReQ&feature=youtu.be

[jfuenmayor]

Hi

I sent the link of your video to a friend who is a full time instructor for Airbus in their Level D simulators.
From what I understand, it can be made pretty rough in those, to the point that the instructors do not like to go into severe turbulence in them. I ll send you the feedback once I get it.

JF

[sagrada737]

Hi Joaquin,

Thanks for the post.   It will be interesting to see what your friend has to say about the turbulence.  I have a friend that is a 777 Senior Captain and he hates simulators - perhaps for that very reason.  Certainly that kind of "violent" turbulence is not pleasant - either in a Sim or a real aircraft - big or small.

Mike

[Nat Crea]

Mike I love that model in the latest video..its insane!
I could watch that all day...
Even IF you don't get to full scale stage...well done mate.

Nat

[fsaviator]

You keep driving on, Mike.

I appreciate you have the time to tackle the issues I wish I could, so I am living vicariously through you!

[bernard S]

@ Sagara737, to give you an idea of weights each jack on mine weights 750lbs each and you will need a spare jack and servo...we had to dig down 5 ft to support each leg on a 3 inch steel plate . the cab unit weights about 2200 with out people at 250 lbs each x5 the base alone to support the cab prolly weights in at an easy 3000lbs with control loading   the pump is about 10,000 lbs the knuckles lower weight about 1200 lbs and the smaller knuckles about 750 .... about 300ft of hosing at 3000 psi... and the system is rated for 35,000 lbs per corner   to give you an idea of overall size its a monster   about 19ft 6 at widest point some 28ft long without the collimated visuals attached the deck as a stand alone is 19ft long and its is 14ft high without the full extension oi the jacks at another 8 ft    but as we say in TX go big or stay home :-)

the instructor station is separate to the cab unit for comfort

[Sam Llorca]

It looks good Mike!
  On take off and braking I think you need to increase the pitch limit in order to feel it inside the cockpit and tone the turbulence down a bit on the software, my opinion. When I first start it to test my platform some of the settings were set too high and at one time it ripped a couple of cement anchors right out of the ground cracking the floor surface, I tone it down quite a bit thru the software and now I can enjoy it a bit, I say a bit because after you have a few scares on it, it will make your hands and arm pits sweat every time even tho it works very smooth now. I think is a great, huge, complicated, task for a flight simmer but not for Mike, GOOD JOB MIKE.

Cheers,
Sam

[jfuenmayor]

Hi Mike.

You have generated quite an interest here, I commend you for a fine work.
My friend thinks that the movements do need softening. He also mentioned
that in the Airbus,  the autopilot dampens them to avoid over stress of the structure.
I am certain you can accomplish that.

Keep up the good work.

JF

[sagrada737]

Hi Guys,

Joaquin - Thanks for the feedback from your friend.  It's good to hear his recommendations about "softening" some of the motion cues.  It turns out to be quite a challenge in re-configuring the 6dof motion software, as when you change one parameter, it tends to affect other parameters - so there is a balancing consideration that has to also fit with the flight model - 737 in my case.

Regarding over-stressing the structure...   I'm reminded of a helicopter model I once built.  Once I had it "tuned" and flying properly, it seems to hold up fine - that is, until I would lose control and crash.  Years later, I got my helicopter rating.  I learned in a Eurocopter AStar and subsequently flew a Robinson R-22.  The AStar had a fully articulating rotor system that was subject to "ground resonance" if the helicopter was not controlled properly in close proximity to the ground.  It only took one bump mishap to induce this blade resonance, which would literally shake the helicopter to pieces in a matter of seconds if the rotor system was not "loaded" by lifting the aircraft off the ground.  The R-22 had a different rotor system and was not subject to this kind of ground resonance, although it too had is resonance vibration issues in going to idle to full rotor speed.  It is truly amazing how these kind of structure can resonate - sometimes to the point of component/systems failure to the point of destruction. 

Shortly after getting my helicopter rating, I decided to build a two-place experimental helicopter called the RotorWay 162F.  This was a kit helicopter and a great project that took about a year to built.  For a lightweight sport helicopter, it actually felt a lot like flying the AStar, which had hydraulic assist for its flight controls.  However, one thing was common to all of these helicopters - the had significant vibration in virtually all flight modes.  I noticed this mainly when doing major systems inspections on the RotorWay 162F.  Things that were otherwise suppose to be secure, would tend to loosen and develop wear patterns.  It really gave me an appreciation for how mechanical systems are subject to shock and resonant frequency vibrations. 

With respect to designing and building a 6dof Motion Platform, I consider resonant vibrations no different, as evident from my testing with the 6dof motion platform Sim scale model.  I certainly concur with your friend's statement about "softening or subduing" the motion cues to prevent structural damage.  Fortunately, the BFF 6dof motion software allows for "adjusting" these  motion cue parameters.

The challenge for me is to come up with a balance of "Sim movement" on the motion platform that allow the Pilot sitting in the Sim to trigger vestibular sensations that yield a sense of acceleration/directional movement in sync with the flight.  I think this can be done, but not necessarily determined from the observations gained from the little Sim model shown in these videos.  This is why I need to build up one of the VFD/Motor/Gearbox segments to ensure that there is adequate torque to handle the dynamic force requirements for a full-sacle 6dof Motion Platform with the 1,800 lb. Sim mass.  Assuming this can be accomplished, I have confidence that the "motion cues" can be modeled and fine-tuned to fit nicely with the 737 flight model.

Sam - Thanks for the comments and encouragement - and cautions.  Having a good structural foundation for mounting the Motion Platform is certainly important.  Not only does the foundation need to adequately support the static mass of the platform/Sim, but it needs to be able to stabilize the inertial forces that come about by dynamically moving the elevated Sim mass - especially in a Yaw, Sway, Pitch, and Roll mode - or combinations thereof - all with worse case start/stop/fwd/reverse considerations, with fast response times associated with these motion cues.

As far as "having a few scares"...   There is no doubt that building up a 6dof Motion Platform system is serious business.   My goal is to take in consideration as many of the factors as possible to eliminate "scares" associated with this kind of project, in its operating environment - both static and dynamic.  Fortunately, this is not the first time a 6dof motion platform system has been designed or built, and its features, capabilities and limitations are well known.  The highly sophisticated BFF 6dof motion software seems to be very stable, incorporating safety consideration for human intervention, shut-down, broken communication links, etc.  But what is a bit more unknown is the mechanical/structural/form-factor requirements for a given design.  In this regard, the safe support and movement of the Platform mass is the main design consideration, with added consideration for "safety margins" within the design to remain within a safe and reliable operating envelop.  To a degree, such a 6dof Motion Platform is a "life support system" when you consider real people in the  cockpit, so identifying potential failure modes is part of the equation.  In this regard, basic engineering sense as-well-as common sense come into play.

Bernard - Wow!!!  That is quite a list of weights for the various parts of your Level D Simulator.  Massive for sure!  And I'm sure it needs to be that way for its commercial intent at the scale of its design and performance requirements.  It seems smart to have the Instructor Station outside of the "Sim Pod".  Those guys don't need to get motion sickness if they can help it 

The Hydraulic systems on your Sim have some serious working pressures.  People don't realize how critical it is to have reliable, high-quality hydraulic lines with those kind of pressures.  As pin-hole leak at those pressures will slice right through a person's hand.  Gotta be extra careful with hydraulics - especially with high frequency response to mechanical forces.

Another wow!   That Sim Pod is roughly 20' off the deck!  You better wear a parachute    Just kidding of course.  My 6dof motion platform upper platform will be level at 4' off the floor.  This bodes well for helping to manage the myriad vertical/lateral force vectors associated with the operating dynamics of the motion platform as transferred to floor mounting/foundation system. 

Perhaps I could come down your way someday to see your Level D Sim - if that's possible.

Anyway - thanks again guys for the interest, constructive comments, and encouragement for this project.  Much appreciated.

Touching base on another project...   I am just about finished with the design and fabrication of my J-Rails for my Flight Deck.  When they are completed, I'll post a thread to share the results.  I think they came out very nice, and will help to solve the problem of getting in and out of the 737 cockpit seats along side of the Pedestal.

Mike

[quid246]

Hi Mike, nice looking mock-up.
One thing I noticed or I guess really didn't notice was the simulation of acceleration and deceleration g forces. The platform should be tipping backwards at a noticeable angle when you accelerate at takeoff. Same with braking on landing.

This is correct... the pitch angle of the sim will be MUCH greater than the "visual pitch" of the actual aircraft take-off to simulate that "pushed against the seat" feeling at T/O, etc.

I was in an AC 777 sim not so long ago, asked my brother to hold my camera, so he put it down beside the jump seat.  After I pulled up at V1, the camera literally flew through the air and hit the middle of the back door.  So I assuming externally the sim was pitched well in excess of 50 degrees.

Another consideration might be to include some kind of "knock" actuator... as the Boeing/Airbus sims will thumb as you are on the ground to simulate the bump from rolling over expansions joints.  It's a small thing but I found really added to the realism.  Don't think FSX/P3D has an output for that, but perhaps just take your ground speed when on the ground and for every x distance traveled trigger the knock motor.

[sagrada737]

Hello Allan,

Thanks for the constructive comments...   I'm in the middle of machining an Actuator Lever for the motor/gearbox so I can complete my load testing.

Regarding the motion platform angular cues...   I have since increased the radius on the Levers to yield a max angle of 15 degrees for Roll and Pitch.  This will be nothing like what the commercial motion platforms yield, but I believe it will be adequate to provide the sensations necessary for the various motion cues.

Regarding the runway bumps...   I am using a separate vibration system to simulate this kind of activity directly to the Pilots.  This system will reside within the Flight Deck, but isolated from the motion platform so as not to induce shock into the sensitive Sim parts/systems.

Best Regards,
Mike

[blueskydriver]

It's been a while for me, but trying to catch up on posts...

-Bernard, do you have a YouTube video of your Level D Sim? I'd love to see it and you in action.

-Mike, can you post some updated photos please?

Reading all of this data is much easier with some videos and photos for reference.

John