New video using smoothing routine(linear) and my bicycle speedometer reed switch. Using the smoothing routine only on the downstroke. Much smoother operation without the bounce at the bottom of the stroke and it can go faster now without using the homemade twangy switch.
Guess next version will be using multiple reed switches so I can change the angle on the fly.
(original video text follows)
Here is video of version two where I just added a potentiometer as planned and suggested by others, to control delay time of the cycle to find most efficient (fastest) cycle. Please disregard my blathering on the video as I got just about everything I said wrong.In fact the first time it reversed direction I didn't even notice as I was looking at my computer screen. This is a good example of how not to do science.
But I do believe from the one trial (I will do more now), I see that counter clockwise rotation is the most efficient and at about half second of delay. Kind of makes sense as an overhand throw is probably more torquey than an underhand one (or getting a gravity assist?). Would be interested in what anyone thinks about it or how better I can measure it. After running some more trials I will think about modifying the program to find the best delay time itself. That should be version 3. Thanks for watching.
Yeah, it is kind of jerky. Not sure how to fix that. Mounts are probably too flexible, linkage is too flexible, counterweight is off. And it bounces terribly at the bottom of the stroke. Thats what I get for using fishing poles and hot glue.
I am sure ,there is somewhere a post in LMR related to the smooth servo movement. I tested it myself,and it really is very smooch. Please search for that. It will work just like a charm.
I can’t quite work out what that twangy thing attached to the angle bracket is doing. Is it a limit switch of some sort?
Surely, that’s what you need, either mechanical or an optical interruptor type to detect when the wheel is at it’s low point and then kick off a new cycle (possibly with a programmed delay) so the servo operation would be self regulating.
Edit
I’ve just noticed the croc clip on the frame. So the wire frame hitting the spring completes a circuit. Is that doing anything at the moment?
Yes, my original plan, and still is, was to use a reed switch mounted on the gyro frame in place of the twangy switch (I like that). But I am at a new school in China and dont have any students yet who can order me electronic stuff from taobao.com, and I couldnt wait so came up with the twangy switch inthe interim. I am using a programmed delay, but now controlled with the pot.
My future version will use many reed switches mounted on a circular pad attached to one side of the gyro frame so that whenever the magnetic wheel axle rolls around it will activate the switches.
I finally figured out how to calculate the curve. It is a simple matter of looping through the proper set of numbers to get the required curves. If you add a duration, you can stretch out the curves. You may want to consider adding a step value to your loop too. Figuring out the degree numbers you need to loop through is just a matter of looking at the degree numbers of a sine wave.
you have a very nice toy there and the sound and operation of it reminds me of an indigenous foot operated husking/grinding wooden apparatus called ‘dhiki’ from the foot of the Himalayas in Nepal-minus the gyro wheel of course…haha.
I wonder if you could use solar to make it go on for ever....
I will get around to this eventually - I have to do lesson plans - crap.
I wonder two things:
1. If I can get by by using a simple smoothing routine without the curve. I have used that before to smooth out vibrations so I could use a tilt sensor.
2. Will this slow the servo down so much that it wont have any momentum to throw the wheel or will it just be so slow that the top speed of the wheel wont be much.
Just a note about something that happened in the video, it was going as fast as it could in one direction then when delay was increased again, it stopped and went in other direction even faster. Anyone find that interesting or know why that is happening? I don’t anything about physics.
One way it is throwing action. The other way it is receiving . The gyro momentum seems to be adding to the servo movement on throwing while subtracting on the other side. Slower side, it is pointing towards the axis of rotation. Faster way, it is pointing away from it probably bringing in centrifugal forces. Of course all these forces are adding up or subtracting on the servo movement force… in my observation…edited …
The speed difference is why I mentioned the jerky movement
I would imagine you could get a better handle on what might be happening if your throws weren’t so wibblely-wobblely timey-wimey. (Bonus points if you know where that phrase came from. :D)
