This is an archive of a previous version of Sodarace.

Please visit http://sodarace.net for the latest version.

forum  |  »» sodaplay
 
»» forums 

subject: ===Dedicated Help Thread=== *new constructors please read p.1 first!

5096 replies on 340 pages. most recent reply: Tues, Aug 28 3:17 AM by OBstakel

»» back to topic list  

This topic has 5096 replies on 340 pages [ « | 1 ... 9 10 11 12 13 14 15 16 17 ... 340  | » ]
»» previous topic   »» next topic  


axilux

»» models

Re: ===Dedicated Help Thread=== *new constructors please read p.1 first!   posted: 06-Dec-02 14:22   »» 
no, it wasn't that one, Lectvay made it, I'm pretty sure of that
/ax

axilux

»» models

Re: ===Dedicated Help Thread=== *new constructors please read p.1 first!   posted: 06-Dec-02 15:39   »» 
The model was: TriangleLaticeInertiaBoat by Spideylegs, sorry for making such a mess.
/ax

warptera

»» models
»» homepage

Re: ===Dedicated Help Thread=== *new constructors please read p.1 first!   posted: 06-Dec-02 15:43   »» 
Okay,

Here is an exercise for this “rainy day” (meaning we can’t go outside and sodaplay) of constructing. You can’t save your work, but saving isn’t as important as knowing how to perform construction tasks again and again so maybe it is for the best.

I bet people would like to use some of the complex spinning motors they see in the zoo. In fact, the zoo would be a better place if everybody knew how to properly anchor and implement these motors.

I can’t save a custom example to my “example” account so use this pre-existing counter rotor:

http://sodaplay.com/constructor/player.htm?&getmodel=warptera+4pt_Angle_Counter_Rotor

Like I said a few minutes ago in the “sodazoo” thread, this motor looks funky, but it works exactly like the old counter rotors in the sense that there are two circles of masses rotating along the same path in different directions.

By anchoring one circle (or “side” of the motor) to a frame, the other side can spin freely while supplied with the necessary leverage to push a larger mechanism.

Begin by making wave speed zero. Now manually advance the wave to make sure that all eight power points of the motor are easily differentiable and don’t overlap.

Now go into “construct” mode and create a perfectly horizontal fixed spring stretching from left to right, above the motor. Now create five vertical fixed springs that all intersect the horizontal line at roughly even intervals. Don’t worry about their placement, they will be moved shortly.

Now go into simulate mode again. The wave should be at max amplitude but zero speed, g = zero, k = ~80%, while f=~60%.

[warning: this step can be frustrating at first] now examine the motor and Identify two masses in the power ring of the motor that are directly across from each other. It doesn’t matter which pair, just make sure you choose an opposable pair that are sufficiently clear of the linear motor parts. Move the motor up to the guideline (toward the left side) and position the two masses right on the horizontal guideline. Sometimes it is tough because the opposite mass will tend to move when you shift the other. Just be persistent.

Now place two of the vertical fixed lines so that they intersect the horizontal line exactly where the masses are sitting. Now that your guideline is fixed remove the motor. If done correctly you have now created a single guideline with the exact length that exists between the two masses at all times during the motion of the motor.

Now go back into “construct” mode and place two free masses on the intersections that constitute your new guideline. Once they are perfectly placed, connect them. Now you have used your guideline to create a spring that has the exact length between the masses in the motor.

Now go back into “simulate” mode. Use the spring you just created to make divisions on the horizontal guideline equal to the original length. You will have to be just as patient as when you placed the motor on the line. Use higher friction if it vibrates too much.

When completed you should have a single guideline with five vertical divisions all equidistant to each other. This has created a distance exactly equal to 4 times the distance between the masses in the motor that you originally measured.

Now discard the short measuring spring or delete it preserving the masses because you now need a longer spring. Using the large guideline, make a spring that is equal to 4 times the original length. This length will be the long length of the tension spring.

Now you must attach each of the end points of your long spring to one of the two original masses in the motor with springs of zero length (place them exactly over each other and click until it creates a “zip” spring). Once that is done, zip the two masses from the motor together as well.

If you have done everything correctly (and if I’ve explained it correctly), you should go into “simulate” mode and see the motor just as it was, excepting the large “spike” going through it. This tension spring can be used to attach the motor to a frame, while still allowing the other half of the motor to spin unimpeded.

There are many examples of this technique in the zoo. There are many possibly configurations for the tension spring such as placement, # of masses etc. Please enjoy your new skill and enjoy the new possibilities it provides.

Please post if any of this is unclear or if you have further questions.

Best,

w

Spideylegs

»» models
»» homepage

Re: ===Dedicated Help Thread=== *new constructors please read p.1 first!   posted: 06-Dec-02 16:52   »» 
This is totally off topic, but I don't think this needs a new topic or anything. Is kevino gone? Has he gone the way of mono and brucemiller? He hasn't made any new models or posted or anything. I can't tell if he has any new models in his pub because of the glich. This was on my mind and I decided to post.

Whatever,
-$pidey

warptera

»» models
»» homepage

Re: ===Dedicated Help Thread=== *new constructors please read p.1 first!   posted: 06-Dec-02 16:59   »» 
Kevino has been around forever.

he has told me that he has taken as long as a month off. I'm sure he'll find inspiration and renewed vigor sometime and return to us with new fantastic models to delight and inspire.

-w

roey_schurr

»» models

Re: ===Dedicated Help Thread=== *new constructors please read p.1 first!   posted: 06-Dec-02 17:04   »» 
warptera, I tried understanding your explanation, but it took me some time... I used massandwiches insted of guidelines. It is always easier for me to use. I think that in this case, it might easier for everyone to use. That is, ofcourse, if I underastood what you said and built it correctly. I hope we could save soon and you could see what I have done.
I am going now, but will be back later. I will explain what I've done. If you could, please follow my steps and tell me if this is what you meant.
Thank you, have a wonderul day...

lonelyswedish

»» models
»» homepage

Re: ===Dedicated Help Thread=== *new constructors please read p.1 first!   posted: 06-Dec-02 17:26   »» 
i was doing a bit of experimentation with that motor/method you posted warp, and came up with something interesting. alas, i accidentally deleted a spring/mass for which i had no guideline, so the model was unable to be completed. i'm going to go back again later and try again once the problem with the accounts is fixed.

anyways, here's what i found out: creating a chassis with the method you have described (indeed, any other method would also have this result) not only yields a 4-point rotary motor, but also 4 individual 2-point rotary motors which all rotate in the same direction. these motors are created by the points of the two 4-point linear motors which you combined with tension bars to get the original counterrotarty structure.

while the points in each 'motor' are 1/2 out of phase with each other, each pair of rotating points in this structure, as you go around the motor, is 1/4 out of phase with the ones before/after it. this means that you can achieve leg motion by standard rotary motor methods (such as creating a 3-point tension bar using the motor point as the end and attaching the middle point to a chassis) as well as by methods used in standard linear motors by translating the rotational motion into linear motion.

a note on rotary-linear translation: this just means somehow attaching a point or set of points to a mass on a rotary motor such that one or more of the new points moves in a linear fashion. probably the easiest way to do this is to attach the rotary point to a mass, then attach that mass to a chassis point. the sum of the lengths of the springs used should be greater than the maximum distance between the rotary point and the chassis point used. this allows the rotary point to push/pull the extra mass back and forth and, since the distance of the mass from the chassis point is fixed, it will move linearly back and forth about the chassis point.

hope that made sense - i think this will make some cool new models once the problem with the accounts is solved.

warptera

»» models
»» homepage

Re: ===Dedicated Help Thread=== *new constructors please read p.1 first!   posted: 06-Dec-02 17:58   »» 
Roey,

the reason I prefer guidelines is that I too often need to measure distances within the center of a larger structure. Since the mass sandwich traditionally uses the corners and sides of the boards it makes this impossible.

I assume you made a replica spring and then transferred it to the corner to make your guide. For me this would be an undesirable extra step but for those who find the fixed mass guide easier I suppose it is worth it.

Also, I prefer guidelines because of their ability to create measurements in two dimensions. This includes hypotenuse guidelines as well as the very complex 2D guidelines i've seen jeckyll use.

Anyway, I hope all this conversation becomes moot with the advent of the upgrade.

LS,

I believe I understand what you say about rotating linear motors yeilding multiple rotor possibilities. I have noticed some of these during the construction of my "Anaximander" and other models but no new model has sprung out of my imaginination yet. I am eager to see what you developed.

To all,

Regarding what I said to Roey about hypotenuse guidelines and what LS said about his deleted spring length (kind of):

I have noticed during construction that many lengths which I believed demanded to be determined experimentally, actually turned out to have concrete geometric relationships to the rest of the structure.

In the beginning of my career here, I used to make tension springs by "winging it". An example of which is in Z_Axis. The legs masts are all equal of course, but the lengths were just designed according to what I thought I needed.

Since then I have slowly realized the interrelations that exist. It is nothing mystical, just geometry. Almost every time I need to determine a particular length it turns out to be equal to... say 1/2, 1/4, 2 or 4 times the length of another existing length, or perhaps the hypotenuse of a right triangle constructed out of that length.

So in short, I advise you to look out for these fairly complex geomtric moving relationships (like the length of the tension springs in relation to the resulting length of the linear motors used above in that counter RAM). They will help you attain exactly the length you need, releiving you from "winging it", possibly resulting in some unwanted flex or torsion in your models/motors.

-w

the_duck

»» models

Re: ===Dedicated Help Thread=== *new constructors please read p.1 first!   posted: 08-Dec-02 15:49   »» 
Hi, I'm new. Here's my first working model, using warptera's motors:
http://sodaplay.com/constructor/player.htm?&getmodel=the_duck+octopede

How do you build those rotary motors so perfectly, like in this:

http://sodaplay.com/constructor/player.htm?&getmodel=display+Triangular_lattice_walker_2

warptera

»» models
»» homepage

Re: ===Dedicated Help Thread=== *new constructors please read p.1 first!   posted: 08-Dec-02 17:41   »» 
duck,

very cool. that's a great job. here's a tip. if you anchor the motors together like this, it'll be a little more efficient:

http://sodaplay.com/constructor/player.htm?&getmodel=example+the_duck1

also, since those motors are 6pt ones, two more legs on each side would complete the elliptical motion.

that second model was made by axilux I believe. he made it precise using a grid or guideline. I imagine you used such a method to make the legs on your fine walker so even.

best,

w

axilux

»» models

Re: ===Dedicated Help Thread=== *new constructors please read p.1 first!   posted: 08-Dec-02 17:44   »» 
In this case, a bit of simple geometry is handy.
I just created an equilateral triangle, and then used guidelines (read some posts and I think you will understand the concept of guidelines) to put the springs inside.
How did I create the rotating complex inside the triangle?
I first made a guideline where I marked the middle point, so that it consisted of two equally long lengths beside each other. From this you can construct the rotors.
In a perfect hexagon, the sides of the form (from one corner to another) is equal to the length from a corner-point to the exact middle.
In a triangle, the length from corner to corner is twice as long as the length from corner to center.
The rotating points is center points, at rest they lay precisely in the middle of the triangle.
I used the guideline I metioned to measure the lengths.
Tell me if you don't understand me, because I am not very good in explaining.
/ax

Post Scriptum:I like your octopede :-)

axilux

»» models

Re: ===Dedicated Help Thread=== *new constructors please read p.1 first!   posted: 09-Dec-02 15:05   »» 
Flex structures are really nice, but I can't get them functioning. I don't have a model to post, I'm sorry, but is there any principle on how to get the muscle systems to work?
I may come up with some examples later.
/ax

roey_schurr

»» models

Re: ===Dedicated Help Thread=== *new constructors please read p.1 first!   posted: 09-Dec-02 15:10   »» 
axilux, flex structures DO have some rules. I think warptera is working on an article that explains it all just great (he explains wonderfully, he is the one who taught me how to use flex chain). Anyway, until it is done I can send you the email that he has sent me, explaining it all. Or I could just teach you...
I would love to do so, but I have to run now. I will be back later tonight, so just tell me what you prefer.
have a nice day all.

warptera

»» models
»» homepage

Re: ===Dedicated Help Thread=== *new constructors please read p.1 first!   posted: 09-Dec-02 15:24   »» 
Roey & axilux,

The series of emails Roey and I exchanged were specifically geared toward making a replica of Kevino's "TriangleQ". It involves one possibility of programming flex chains that I refer to as compression waves (due to its similarity to the way sound waves move through air, as opposed to the way a sine wave oscillates perpendicular to the direction of the wave).

It is quite detailed but it may contain some material you are interested in. If this would be of interest, Roey can forward it to you. Otherwise you can wait for my tutorial or we could cover some flex basics in this thread.

Best,

w

axilux

»» models

Re: ===Dedicated Help Thread=== *new constructors please read p.1 first!   posted: 09-Dec-02 15:30   »» 
If roey could e-mail me that, it would be great as it may be of my interrest.
I was just out for some simple muscle placement, but all this sounds good. I can wait.
I have to run now, but I will write again.
/ax



This topic has 5096 replies on 340 pages [ « | 1 ... 9 10 11 12 13 14 15 16 17 ... 340  | » ]


»» previous topic   »» next topic  

»» back to topic list  »» top of the page  

PLEASE READ THE FORUM GUIDELINES AND ALWAYS PREVIEW TO CHECK MESSAGES BEFORE POSTING...
...Help keep our forums creative and constructive. Thank you.


»» forum home