Motorized Joyrider

[kjg71]

Thanks for the reply Thanos!

Well I checked the velleman... You have to connect the HB-25 on a digital port and pulse it within the 1ms ~ 2ms range, with 1ms center that stops the motor rotating either direction. I suppose you can do this in programming. Don't use its analog ports cause they only output voltage, not frequency.

hmmm ....currently I am limited to just sending a 1 or 0 to the lines. I am not quite sure how to send something other or then that or what middle piece I can use to to convert it into a servo signal? Any suggestions? I suppose with two Digital lines I could control direction and on off to HB-25, but speed might be out.

I am almost thinking that if I want to work with HB-25 I will need to use a different controller (servo controller) that works with my programming?

K8055 and ezio is certainly not that (unless there is a good middle piece of electronic that converts pwm votage to pwm frequency (mS).

You would almost think that something like this is out there?

well .....I think that your method is really better then what I am doing with just voltage. If you are controlling speed, position, direction (and even ramping) through 1 line to a low-cost gear motor, then you have really got something. Speed & direction is great but position and ramping is even better. I would love to eventually be able to add those. (It must feel so realistic!)

Whats very interesting is that your not using physical encoders for position, but rather math to help guess (or know) the current position based on the angle/speed of the motor. At least that is the way I am understanding it

Sounds almost like a "virtual" encoder

If that is the case then ...Wow...that is really" innovative.

great stuff!!

regards, Kyle

[TronicGr]

Hi Kyle,

Yes, a servo controller is needed. Thats why I used Basic Stamp2. It can drive servos easily and in the same time be able to receive from the computer's serial port the data needed to drive them, input the motors position feedback thru simple potentiometers and calculate all the math that is required to produce a smooth motion.

But the Basic Stamp2 isn't designed to do more than 2 or 3 axis motion simultaneous. Its a slow microcontroller compared to other in its family. There are other that run in twice of its speed and can handle even more axis. Why don't you try Propeller? Its inside microprocessor consists of 8 BS2 that each one can run its own set of commands in the same time like working in parallel !!! Check Parallax's site for more info: www.parallax.com

*Update: in a quick search in parallax I came up with this Spin Stamp Microcontroller:
http://www.parallax.com/detail.asp?product_id=SS1-IC
Parallax's description: "The Propeller is a multi-processing chip with eight 32-bit processors (cogs) and shared memory with 32KB of RAM and 32 KB of ROM holding a font, math tables, and Spin interpreter"

I AM using simple potentiometers to encode its position but its not connected on the motors shaft as you may think but in the frames rotation joint to actually measure angles!! I use my proportional code to move the motor to the direction of the desired angle and i ramping it down a few degrees before it reached the setpoint to avoid overshoot and bouncing.

As you may already figured, I play with serial data streams and servo data. I don't use on-off logic at all! My motion cockpit doesn't even has limit switches since the code is limiting the max travel automaticaly!

Hope it all cames clear now.

Regards, Thanos

[TronicGr]

Hi Kyle, I was re-reading all your posts and here are some notes:


>>The whole project was to be motion movie tracks and timeline based triggered motion, so I am using lingo inside Macromedia director as it allows you to control scripts with game controllers, sound triggers and inputs from external sensors. The game controller efforts was taking it that extra experiemental mile.<<

So you must using K8055 and ezio because they have ready communication DLLs with Macromedia Director right? Thats to avoid getting more in the microcontroller communication details.



>>I am almost thinking that if I want to work with HB-25 I will need to use a different controller (servo controller) that works with my programming?K8055 and ezio is certainly not that (unless there is a good middle piece of electronic that converts pwm voltage to pwm frequency (mS).You would almost think that something like this is out there?<<


I have an idea. Since you have to use your boards "K8055 and ezio" to output analog voltages to a motor you could use a little trick: You can connect the K8055 analog voltage output an external microcontrollers ADC inputs that its programmed to translate voltage to values. The microcontroller can translate really easy the voltage values into servo commands. I already do that by reading the voltage that the position feedback potetiometers returns to the ADC circuits of my basic stamp! So you can easily have an external voltage-driven servo motor! Thats sounds like a nice little project! hehe... And you know what? I was already implemeted such voltage driven servo to tune the motion of my cockpit with voltage measured on a potentiometer before I had it connected to read actual flight data from the computer!!! If you gonna use basic stamp2 you can use my analog input HB-25 servo code. I can send you a copy to try it. Just a note: To filter the PWM voltage noice you may need to add a small capacitor as a voltage depot and make it steady enouph for the ADC circuits to measure.

So, I made some minor modifications on my BS2 servo code so it can position the motor just by feeding it with analog voltages (0 volts to +5 volts) to be able to use it with your existing velleman or ezio controllers. Don't expect huge resolution as it is just an 8-bit microcontroller but its very robust in what it can do. With this, all you get is a 2 analog input and 2 servo outputs. As I said before you could always try other microcontrollers Spin Stamp Microcontroller (Propeller) that have ten times the performance of the simple Basic Stamp2. But then again, a different programming is needed which I don't have the knowledge to do.

If you like the idea and want to try it, PM me your email to send you the BS2 code and the schematic with the connections.


Regards, Thanos

[kjg71]

Thanos,
Sorry about the delayed response....had a relatives funeral, a wedding and preparation for a new job interview coming up ....so this weekend was so swamped!!
yes I am using those microcontrollers for their high level ease of use with the director environment /DLL utilizion.

little trick: You can connect the K8055 analog voltage output an external microcontrollers ADC inputs that its programmed to translate voltage to values.

I'll try to explain how I was (eventually) going to do positional encoding and then realy back how I understand your method so I am clear (I actually had to read your post a couple times)

currently plan was to use linear pots that will move with the cables / mechanics and attach them to the ADC (Analog to Digital Convertor) inputs. The software would read those and store the value as reference and querys the current position until it reaches that (as to when the motors stops). This logic of course is done in the same code that run the motor oporation. This of course requires that I have physical encoders vs that nice virtual encoder method you are using. My I/O boards will not generate servo commands though.

your method almost sounds like you did use physical encoders at fist? see quote below:

voltage measured on a potentiometer before I had it connected to read actual flight data from the computer!!!

...so sounds like you got it to work....that's good

Now, on to your method......

It sounds like I would only use the k8055 or ezio to send PWM then BS2 (with a small capacitor before entering)....this final run-through would convert it to servo code.
As you are stating, speed and direction would be no longer set in pwm voltage, but rather by the distance traveld to get from one position to the next (with response set in BS2 so motor and supply is not stressed to much) I actually don't set my pwm speed full either or it would rip everything apart.

In this method, it sound like BS2 and HB-25 could be the missing link I was refering to as the "PWM voltage to servo signal convertor".

......Lots of things to consider here. I am wondering if I shouldn't just finish up the mechanics, get all to work well with limiters and pwm speed and direction change and then to the "switch" over to the "pwm voltage to servo " way

I can see my current setup being pretty easy but I will not have the accuracy of the servo way if I don't gear up for that. At some point I think this servo way will adapt itself better to recieve game telemetry as well (it's hard to say on that right now).
One major drawback is these boards only have 2 pwm ports each. I would like to see if I could make a 6 port pwm K8055 and have the whole thing on 1 board.

I love the idea you have, but might hold off until I can round out the system as a whole. I think this would be a major upgrade / replacement for any physical encoder idea I would have.....to some extent my current positional code would be streamline too.

I will PM you to get that code and schematic so I have it on record. and for everyone else reading, I will post a link to my site end of the today or tomorrow morning at the latest so you can see all this craziness!

regards, Kyle

[TronicGr]

Hi Kyle,

As you are stating, speed and direction would be no longer set in pwm voltage, but rather by the distance traveld to get from one position to the next (with response set in BS2 so motor and supply is not stressed to much) I actually don't set my pwm speed full either or it would rip everything apart.

Actually the voltage will provide the final position you would like have the motors to be placed. So for example a 2.5 volt will center the motor position, an 0 volt value will position the motor to its far left end, and a 5 volt will position the motor to the far right end. Any other intermediate voltage values will place the motor to various positions. You will not have to worry about speed anymore. As long you sending smooth voltage changes to my BS2 servo controller you will have a smooth positioning.

If you can, use the two pure analog output ports of Velleman controller to avoid any noise that the PWM may produce, as its designed to drive motors.

I will send you the BS2 firmware along with all the connection diagrams.
Regards, Thanos

[kjg71]

Here you go Thanos, plus everyone else for that matter

...here is my site after cleaning things up a bit.

FINALLY!!

The "Feelfx" project was to be the exploration of movie tracks that would add tactile and motion effects to your seat / chair.

While I was pleased with the initial test, I was quickly moved into trying it with gaming, as such to provide interactive feedback from various game controllers to the motors.

The Final idea is to run 5 DOF. The pictures are 6 months old as I have been working on add-ons new motions (only yaw and pitch is shown) and improvements in the software and seperate controller houseing with quick connects.

The site has a somewhat official feel and a "first look" video that seems like a marketing video, but know that the video was kind of a fun thing that was put together and serves more as a example of a product demo as I do small side videography work and wanted to do other type of video. The mention of patented stuff and mentioning "we" and "corporation presents" is not real it's just added to make it sound more official. ....Again, just ficticious fun here.

It may help to download the 13 Mb "first look" video and run local for best performance. The other videos should not have an issue. From a technical standpoint, everything in the video is currently running or close to being done, except counsel platform and rudder pedal support. Some of the graphics in the video could be improved, but as mentioned, it's just a little sample thing.

Even though I gave the site a somewhat business persona......know that the entire project is a "one man" show (me) ...and that The site is there to offer as much information and software relating to the project as I have collected Downloads are also free (when the downloads are ready).

I don't think I would be where I am if it wasn't for the sharing of knowledge that is out here on these forums. Contrary to what the project might look like, know that I have no immediate plans to sell electronic controls, plans or build anything for sale. Instead I am wanting to get the site out so it may provide enough information for others to get started on their own designs. As I have said before, you can really make motion work without spending alot. The $2000-$6000 solutions are really out of reach of most. Their are other ways.


Maybe later down the road I may do something more with it if things turn out well with it, but I think for now it's best to just share the site as an idea resource as so many have before me.

You'll notice this is very much different then a cockpit design that may be found on this forum. I tried to think universal in that it is geared to driving, space craft, flight and movies. However I have been working on the attachments for wheels, yokes, pedals etc. to have the option to be less "chair" and more "cockpit" Modularity and universal was really my thinking here.

After the initial link below, choose the large "motion system" buttons to enter the site

enjoy......feedback always welcome

http://www.feelfx.com/

[TronicGr]

After major tuning my servo code and rearrangement of the side motor mount, I took another couple of video of it moving fast like ****. I need to install a new racing bucket seat with six-point belts before I'll attempt to ride it again with the new tuning. Now it works with 50% more speed than before and the reaction time is also reduced down by 50%.

I integrated some g-forces and enviromental effects into the motion so now it just follows the swing of the plane during turbulence. And as if that wasn't enouph I did some crash tests to see how hitting the ground in various angles affects the cockpit's motion.

You may noticed on the 54second position of the video on the cessna, I impulsively extend my hand to catch the rail because my mind was really believed that I was going to fall off the side even if the logic part knows that the cockpit is restricted from flipping over. You can't think of the immersion it creates... With an enclosure canopy will be really scary!!!


Full motion sim on Cessna v2:
[ame="http://www.youtube.com/watch?v=e44LdBINYWw"]YouTube - Full motion sim on Cessna v2[/ame]


Full motion sim on F16 v2:
[ame="http://www.youtube.com/watch?v=1SX_oBSGoKk"]YouTube - Full motion sim on F16 v2[/ame]


Regards, Thanos

[kjg71]

Thanos,

Wow, that's so smooth. I suppose the 6-point belt seat would add to the experience. Seems like you would get a little G-force pressure there with the belts that may add to make it seem even better.

I know you mention. "enclosure canopy " ...but what is it like in the dark?...surely the experience changes?

If you have seen my site you'll notice that I use cable/chain, but I have wondered if I can do better with cable/pulley...Your setup does look insanely smooth.

regards Kyle

[TronicGr]

Hello to all,

I made some new video with dogfights on Combat Flight Simulator2:

I mounted the camera inside the cockpit to take the first one!!



Full Motion Flight Sim on CFS2 Corsair Cockpit View:

[ame="http://www.youtube.com/watch?v=dO6i7YCOvJk"]YouTube - Full Motion Flight Sim on CFS2 Corsair Cockpit View[/ame]



Full Motion Flight Sim on CFS2 Corsair:

[ame="http://www.youtube.com/watch?v=dYOmygb5XtQ"]YouTube - Full Motion Flight Sim on CFS2 Corsair[/ame]


Regards, Thanos

[tour93]

I was researching ideas for a car racing sim and found this forum.
I would guess car racing sim would be easier to manage than plane sim.
All your videos are really interesting and congratulation on your simulator, looks very good.
But I am kind of annoyed with what I see.
Your simulator and everyone I have seen match whatever your joystick is doing:
bank 10° left, simulator banks 10° left
bank 30° right, simulator banks 30° right or whatever maximum it is capable.

So if you want your sim to act like the plane (attitude) it's fine, an outside viewer can compare your sim attitude with what he sees on the screen and that will match.
But if you want your sim to act an airplane what you fly it needs to act to match what you should feel in a given attitude:
High speed bank left, sim attitude should be up and banking a little bit on the right to match the feeling of being pushed down your seat.(centrifugal force), and so on.
Remember "Back to the future" ride at Disney.

For a car it would be easier:
turn left, sim bank right
turn right, sim bank left
accelerate, sim pitch up
decelerate, sim pitch down.

Any thoughts?

Alain