A new guy with new project and silly old questions

[RufusDufus]

Correct, I agree with all your comments...

My idea to use servos is just because I have a load of them already, I was looking at alternate ways to determine the resistance that the person is applying and adjust the servo accordingly - may end up too complex to do in practice.

My thinking around mounting on a HANS was that you would also need to be using a seatbelt as with any HANS and that would secure everything in place - again just ideas at this stage. For anyone unfamiliar with a HANS refer Wikipedia: http://en.wikipedia.org/wiki/HANS_device or google. These are mandatory in most major motorsports these days.

Experimenting is half the fun with DIY projects!

[sirnoname]

If you only need realtime acceleration feelings you can connect your brain directly with this device:
http://www.youtube.com/watch?v=ZlNfBrXYYTc

I do not agree your acceleration thoughts because you are here on the good old earth. This means you have a earth acceleration known as earth gravity. If you make a pitch movement you have two accelerations, one of the starting movement and one of the increasing earth acceleration after you go out of the center. You must mix that two values everytime. If you take the car accelerations out of a game you will not get (and this is mathematical described in the plugin source code) a 3D virtual environment. The plugins are all limited to 2D calculations without earth gravity. (i.e. rfactor plugin source of the game homepage). Using a sensor which will take this values, ignoring the pitch earth acceleration etc. is not correct enough
Finnally you can say that the plugins have a general leak of accuracy for that idea. The X-Sim graphic driver does work in 3D space to make a alternative for the official game support. But it has for now no earth acceleration switch.

[Aloupol]

I'm not sure to understand. I suspect I didn't get what you mean about the 2D and 3D calculations and earth acceleration switch.
I don't care about the earth's gravity, since the mass force it induces to my head (aka my head's weight) is the same in my living room as it is in any race car. I don't have to simulate it in order to simulate car's ride since it's already present.
If the vertical component of the acceleration takes into account the earth's gravity, and so is 1 (I mean 1 g, or 9,8 ms-2) in the pits and 0 at Flugplatz (with a no-downforce car), then I have to subtract 1 to it in order to get only the acceleration due to ride.
If it is only the ride acceleration, which would be 0 in the pits and -1 at Flugplatz, then I have to take it as this.
When I'm in a race car and cornering at, say, 1,5 g to the right, the mass force applied to my head is 1,5 x g x M to the left, if M is my head's mass. If my head with helmet is 4 kg, the centrifugal force will thus be something like 60 N. Whatever I do, in the same time, the earth's gravity applies 40 N downwards.
At home I have to reproduce the 60 N of cornering force, not the 40 N weight, already there since I live on good old earth.
With my Scenic (......) if I go Flugplatz at 180 km/h, I will get a 0 g figure of total acceleration. Even if for a short time, and kill myself a few after. In that situation, the simulator has to lift my helmet with a 40 N force, in order to compensate the weight, and give me the illusion I'm flying.

[sirnoname]

First you must understand the gravity factor which equal to acceleration.
Lets talk about an example with daytona nascar. The oval is a good example that roll and side accelerations must be correct calculated to compensate them to zero at speed x. Because the plugins do not calculate in 3D this will not happen with the roll angle (g = 9,81 m/s^2 * angle) and lateral acceleration.
The compensation does not work like you sit in a car and do the same speed because of the wrong calculation. In case of Nascar the worst situation.

[Aloupol]

So, if I understand, I have to take into account roll and pitch (and yaw?) angles in global (track) xyz reference to calculate real acceleration components in a local (car) xyz reference, which reflects what does the driver experiment?

[Aloupol]

Hello all.
I'm still on this project but I'm stuck. I don't know how to extract the game acceleration values and bring them to my Arduino board.
Pictures here: gallery/album.php?album_id=165
The Arduino is connected via USB, the virtual Com link works, I can send data via the serial monitor to operate LEDs, servos, motors etc. with keyboard strokes, but I don't succeed to send the rFactor data.
I know it can sound like a silly question to you, because it's straight forward for most of you. Informatics noob inside...
Can anyone help me to sort it?
Thanks.
Daniel

[sirnoname]

Sounds strange.
The description does not tell me where you have a problem.
If you use the wizard all plugins etc. will be copied and installed.
In the end you have to verify the moving datas in the converter input setup.
Is this part OK?

[Aloupol]

Héhé yes !!! At this stage it works, since 10 minutes.
I did all the tuto things hudreds of times and it didn't work but today it did...
I'm trying the next step: send "something" to the USB virtual com port. I will see if the Arduino Rx LED flashes.

[DEADBEEF]

To lower the drag, especially when low torque is applied, I plan to use a force sensor (instead of position sensor in a servo) in a PID feedback loop.

You shouldn't need any force sensors or anything as your motor's current draw will be proportional to its torque output. So to work out how hard your motor has to pull just measure its current draw then multiply by some constant to get the force being applied.

So you could do something like...

if ( current * x > desired_torque ) voltage -= 1
if ( current * x < desired_torque ) voltage += 1