Pneumatics/hydraulics in comparance to other systems

Hello fellow simulator builders,

today I would like to write down something about motion simulators with pneumatics, and pneumatics in general comparing to other systems, as I have received some support requests concerning RnR and custom built simulators with pneumatics, and I think that there may be a lack of information regarding pneumatic systems. Be aware that this post is quite emotional and subjective.

Content:
1. Pneumatic basics (motion simulator related)
2. Hydraulic basics (motion simulator related)
1.1 digital valves
1.2 proportional/servo valves
2. RnR
3. My opinion

1. Pneumatics basis (motion simulator related)
Air is a compressable medium! That brings up a problem to deal with, as the system (cylinder, airflow) has to be really oversized, to achieve instant simulator movement. If either the airflow, or the cylinder diameter is undersized, you will get the following described effect:
The valve opens, and the simulator rests as long in position, until the pressure is sufficient enough to move the construction.
And as bad as that effect is as well, that when the pressure is high enough to push-start, the simulator starts to accellerate continously then.
In other words: when the valve opens, it takes a second or longer until the simulator starts moving actually, and when it does, it accellerates continously like a rocket. No instant and linear movement at all.
For example this one (have a look a playtime 0:20, try to figure out how long it takes to see cylinder movemet, after the valve opens)
http://www.youtube.com/watch?v=-IKq4ychtP0
It is obvious that the cylinder' diameter and airflow rate is insufficient in comparance to the simulator weight.
To compensate this effect, you have to oversize your pneumatic system that way, that your simulator is able to operate with very low pressure, and the airflow into the cylinders is sufficient, to fill the cylinders instantly.
The same issue is with the braking procedure:
The valve closes, but the mass inertia of the simulator keeps it in movement, until the counterpressure is high enough to brake-stop the simulator.

Pros:
-Cheap
-Save for home users (low pressure)
Cons:
-Air compresses, no linear and instant movement/braking possible
-Pretty oversized system is needed
-Very loud compressors are required (the silent ones do not provide sufficient airflow), and you may get thermal shutdowns often, as cheap compressors are not made for 100% duty cycle.


2. Hydraulics
Hydraulics are almost non-compressable. This is the most important advantage in comparance to pneumatics, as the hydraulic system operates instantly, when the valve opens.

Pros:
-Instant, linear movement
-As well very heavy simulators can be moved savely
Cons:
-Very expensive
-Dangerous (If an oil hose bursts or the system has been assembled by an unqualified person, compressed oil with several hundred bars could leak and cut your fingers or harm you in other ways)

1.1 Digital valves
Digital valves are bistabil (open or close). They offer no flowrate control, and therefor you can NOT control cylinderspeed dynamically.
To get somehow at least some different speed stages with I/O valves, they can be set in parallel configuration, operating them either pulsed, or with different air pressures.
The more valves are used in parallel, the more speed stages are possible.
Pros:
-Cheap
Cons:
-no direct nor stageless flowrate-control

1.2 Proportional valves
Proportional valve configurations are mostly digital valves combined with a PWM operated flowrate regulator, and need therefor an interface which outputs PWM signals, as well as an I/O signals for the digital valves.
Pros:
-Dynamic, stageless airflow control, which offers stageless motion simulator speed and accelleration control.
Cons:
-Very expensive

2. RnR
The RocknRide simulator uses digital valves without pulse mode ex works, and provides just single speed movements with original manufacturers hardware.
By using the X-Sim software and improved hardware like the DIY RnR interface, or the Velleman K8055, the RnR can be operated in pulse-mode or multi-pressure mode (with four 5/3 valves), which alters the single speed setup to a 3 speed setup.

3. My humble opinion

Though the X-Sim software was an incredible improvement for the RnR to the standard single speed mode those days, and though I am full of respect for the work Sirnoanme has done to improve the motion quality for digital valves by developing the port-output-matrix and the RnR USB pulsmode interface, I have to say that IMO digital pneumatics are stone-age technology today. The time for the digital pneumatics is over, since we are able to control simulators dynamically with motors or proportional valves now, with less effort than for a pneumatic setup even.

Comparing digital valve technology with analogue/proportional systems like PWM driven motors or servo-valve hydraulics, is like comparing steam powered cars...

http://www.youtube.com/watch?v=tzWZ71pP73M&feature=related

...with top-fuel dragsters...

http://www.youtube.com/watch?v=5mkRfok_oxY&feature=PlayList&p=EB9B01FC6D481AE7&playnext=1&index=2

So far I have not seen at least one pneumatic simulator with digital valves, either single speed or triple speed, which would come close to a wipermotor-, SCN5-, or servo-hydraulic-, setup in terms of motion quality.
I do have a RnR simulator at home as well. I have tried ages to get an sufficient motion quality to use it with racing simulations. Without success. Though the X-Sim pulsmode is way better than single speed, it is simply not good enough for professional simulations.
Its a funny toy to play arcade racing or flying games, but there is no way to use it for racing sims!

And for all of you that plan to use pneumatics for future projects, please have a really close look on all the existing projects with pneumatics, and try to rate the movements! Rate them, and compare them to the motor powered or hydraulic simulators. I do NOT recommend them anymore!

I am sorry to disappoint some of you guys. But I am quite long a member here in this forum, and I supported and went along with decades of projects with pneumatic setups. And not one of the builders has been satisfied with their result. I have soldered and sent quite a couple of RnR USB interfaces to all over the world, and no one is in use now! No one of them, including me, was satisfied with 3 speeds!

Sure, you can add 16 valves or more, and make port-output matrix configurations with even more than 3 speeds, But think about all the valves, compressors, air-hose conncetores, interfaces, the loud compressor noise, water seperator, and all the additional mess arround your rig for a result, that would not come close to a cheap wiper-motor setup still. You get wipermotors for 10Euro. Why spending a couple of hundred or thousand dollars for a pneumatic setup, that offers alot worse motion quality?

Fortunatly Thanos came to our site someday, and brought our community his Basic Stamp Controller interface with dynamic speed, and later on his AMC1.5 interface. On top he arrived in good time even, as we had to deal with the backdraw of the failed stepper project. He gave me new motivation to keep supporting the X-Sim project, as I felt confident, that his controller and Sirnoname' X-Sim software are the future for the X-Sim project.
From this time, I dont care about digital pneumatics anymore. The day I've seen Thanos hardware working that brilliant with the X-Sim software and dynamic speeds and accelleration, the digital pneumatics died for me.

Though the SCN5 simulators in comparance, are a pretty awesome, but expensive pieces of hardware, there is nothing that meets the spirit of the X-Sim project closer than Thanos DIY AMC interface and DIY H-bridge, in conjunction with cheap windscreen wiper motors and the X-Sim software.
With hardware for ~150Euro (DIY controller, DIY h-bridge, and wiper motors), we have been able since then to build up pretty awesome working simulators that work much better than everything that I have seen that time.

So let me tell all of you that plan to use pneumatics or hydraulics. Save money and buy a proportional valve setup, which will give you a minimum amount of motion quality, or desist from pneumatics/hydraulics at all.
I'd be appreciated to see someone using proportional valves for pneumatics or hydraulics, and I would offer support as far as I can help. But dont ask me for digital pneumatics anymore please. I do not find it worth to spend my freetime for it anymore.

And to all of you that keep trying to build a digital valve setup. Good luck, and do not say that I havent warned you!

Regards
Christian

Excellent overview for those who have not been watching all along.

Can't say that I agree with the above quote though. I guess it depends upon how you define the spirit of X-Sim. I define it as a superior piece of software that provides superior motion quality developed by a group of people passionate about realistic simulation. To me, the whole idea is to build a community around this superior quality of motion, and help others who Frex users, etc... to see the light.

Others may define the spirit of X-Sim with some other definition that is more along the lines of making motion simulators cheaply available to everyone.

All a matter of perspective I guess.

Great overview though! and love the steam powered car vid.
:

Your definition is the spirit of X-Sim2 :

Edit: after reading it again, I really have forgotten to cater to the SCN5s as much as they deserve.

Sure, for all of you that either want to have the non-plus ultra hardware, or for all of you who dont want to solder DIY interfaces, the SCN5 are the best choice for personal use car racing simulators.
At least for racing simulators...

Initially I just wanted to point out the issues with pneumatics, but could not control myself to write down some additional, general sentences about X-Sim and alternative systems :

regards

Hi,

If you want my opinion too, yes I think digital pneumatics sucks.

Here is my emotional and very subjective story part:

You know, I entered into the motion simulator world many years ago. All started when Lurker (Keith Daniel) decided to share his work on an MS Flight Simulator software interface that can be connected to basic stamp microcontroller. He actually asked for help interfacing the basic stamp2 electronics with actuator that can be used to move a motion cockpit:
http://forums.parallax.com/forums/default.aspx?f=15&p=1&m=70615

Guess what, his intentions was to use pneumatics with digital relays in first. Instead I showed him that precise motion can be achieved using simple servos for modeling purposes. So in first his idea with pneumatics sounded strong enough for me too, and even started to look for pneumatics equipment in my neighborhood. I even devised a way to have precision placement of the pneumatics cylinder by having them follow the motion of a simple servo using this clever pulse mode!






But after some software simulation of the motion to find the proper size of cylinders to use, we found that they suffer from the greatest flaw ever... AIR is COMPRESSABLE!!! That means that it could work weightless but as soon you put a mass of the rider on it the system would fail !!!! Only small cylinders could be used but they had no power then!!

So after a year of doing nothing on the project, I noticed that Parallax had available for sale the HB-25 motor drivers, that can be used with fairly large motors. I purchased some and start looking around for a way to build the cockpit with motors this time. I fall onto joyrider design and you know the rest of the story as it continues here. I used x-sim software in less than a month I had my cockpit in working order with Keith's software, after I had to change a few things on the basic stamp electronics by adding an RS232 buffer to make it more compatible with profiler's USO module.

The projects around motion simulators seems never ending as more new features arise every day, thanks to Sirnoname and his brilliant piece of software!!!


Best Regards, Thanos