Firefly Primer - Intro to GH/Firefly for Arduino, Pachube and more

The new Firefly Primer has been added to http://www.grasshopper3d.com/profile/firefly. The Firefly Primer is a 35 page introduction to using GH/Firefly with Arduino, Pachube and beyond. It covers basic installation, tools, examples, tutorials, basic electronics with many helpful links to get users started. We will be introducing many more tools in the next few months to allow GH/Firefly to connect with a wide range of additional micro-controllers, peripherals, actuators, sensors, web resources and more. Firefly has been beta-tested by over 60 users (architects, engineers, artists) and has proved to be a stable and versatile tool. We look forward to hearing what you think as we continue to develop Firefly!

Tags: Arduino, Firefly, Jason, Johnson, Pachube

Next Post >

Comment

You need to be a member of Grasshopper to add comments!

Join Grasshopper

Comment by Michael McCune on July 28, 2010 at 10:57am: Awesome Jason, that looks like it will work as long as you can control how long it goes in that direction. By that i mean not always running clockwise until it hits the max or min. Not to say what we were doing was remotely correct, but I think the second to last post by orfeus in that web post is close to what we were doing. However, we had to loop through a pulse. We then varied the ubound of that loop in order to control how long it would travel in that direction. The one thing that would have been great to implement from this code was the max/min cutoff so that it would not eat the fabric!

Comment by Jason Kelly Johnson on July 28, 2010 at 10:41am: Hey guys - This is a cut and paste from a web post that worked for me in the past. It should be pretty simple to incorporate this. I can do some tests when I get back to SF. I will try it with both a hacked servo and commercial model. I can add the hack info into the next version of the Primer.

he servo wants intervals of 20ms and pulse widths of 1.5ms if it's going to do nothing. Widths greater than 1.5ms turn the servo counterclockwise, widths lower than 1.5ms turn the servo clockwise. When the servo is receiving 1.5ms width pulses, it's possible to adjust the set point with a tiny Phillips screw driver.

Comment by Michael McCune on July 28, 2010 at 10:13am: Yeah, so maybe instead of an input being # or revolutions, it could just be milli. There was a group that bought a "real" continuous servo and they were having the same issues and value feedback. Your right though, depending on what size servo you have may change how fast it turns at that pwm, regardless it would be awesome to be able to input just those values without having to make the pulse pattern in gh!

Comment by Andy Payne on July 28, 2010 at 10:04am: If I remember correctly... the big issue was changing the PWM value changed the speed of the motor (which would correspond to less revolutions). So, in order for a component to be able to pulse correctly, you would need to know the speed of the motor at a given PWM value (which could be difficult to determine). I also think that since you were using the "hacked" servo, that it would behave differently than other servos. Regardless, I think as long as you knew the speed that the motor was operating at a given PWM value, you should be able to control the degree of rotation given a specified "pulse". I'll give this some more thought and try to put something together in the next week or two. I believe this same issue would apply for stepper motors.

Comment by Michael McCune on July 28, 2010 at 9:58am: Hey Andy,

I will get Luis to send you the arduino code that allowed us to "control" it, it still needs work but I imagine that some inputs may be(if i remember this right), input-rpm (i.e. the pulse width, i think 0=stop, <70 = left, >70 = right, and the closer you got to 70 the slower the motor turned., input-#of revolutions.

Comment by Andy Payne on July 28, 2010 at 9:33am: Hi Michael,
I can probably put something together that would do this... but can you tell me a little more about how this component might work? What inputs would it need? Wouldn't you need to know the speed of the motor at a given PWM value in order to know how long to send the pulse (which would ultimately correspond to the degree of rotation)? Let me know some of your thoughts on this, and I'll take a crack at it.

Comment by Michael McCune on July 27, 2010 at 9:48pm: Jason, a component to send out pulses to control a continuous rotation servo would be awesome ;)