GearHeads Forum

Bill:

Seems its default color index was -2.. a crash. I have added safeties to make sure that one doesnt happen.
It will be in next release.. couple of days likely...

Thx
Art

Bill:

Crashes her eon that one file as well.. Ill see whats up with it.

Art

Art

I've begun regression testing some of my previous projects and assets with version 2.0B of Ticker. I downloaded again your modified version of my first attempts at Newton's Cradle (Newtonv1.zip). I was hoping to see what settings you had used in the model, instead Ticker.exe crashed.

Ticker crashes while loading the project file. Ticker displays "loading project" and then crashes.

This is the first time that I've seen such a crash.

Last few lines of the Ticker log File:

ker\ticker\model.cpp(774): Mesh Create: Sky
ker\ticker\model.cpp(797): Mesh Stage2: Sky
ker\ticker\model.cpp(806): Mesh Stage 3: Sky
ker\ticker\model.cpp(814): Mesh Stage4: Sky
ker\ticker\model.cpp(774): Mesh Create: Sky
ker\ticker\model.cpp(816): end Mesh Create: Sky
icker\tickerview.cpp(1228): Completed Sky
icker\tickerview.cpp(1234): ShadowCameras
icker\tickerview.cpp(1240): Particle Engine....
icker\tickerview.cpp(1245): Gear Design Init.
er\ticker\ticker.cpp(293): Application Opens.
icker\tickerview.cpp(2078): Action...
icker\tickerview.cpp(2315): First Frame Completed. 60FPS begins
icker\tickerview.cpp(188): Remove Objects.
er\ticker\ticker.cpp(308): Exit App.
er\ticker\ticker.cpp(75): Exception code: 0xe06d7363 ((null))
er\ticker\ticker.cpp(75): Exception code: 0xe06d7363 ((null))
er\ticker\ticker.cpp(75): Exception code: 0xe06d7363 ((null))
er\ticker\ticker.cpp(75): Exception code: 0xe06d7363 ((null))

Hi Bill:

Yes, there is a disconnect between assets like bars and where COM ( center of mass) and center of rotation are. I likely will find a way
to rationalize this. If you attempt to rotate a bar, it will simply rotate on its center , as it may have many shafts it only knows its center so
it cannot be rotated in the normal way. If you have abar placed and wish to move it to a new orientation, set gravity to off, start a simulation and use the Drag/Snap feature on the sim page to allow you to simply drag it to a new orientation, when your happy, press the LOCK button, the simulation
will now start and be in that orientation at rest. A bar should rotate if hit by gravity and is connected to a shaft as the center will try to move and be limited by that connection, as a result the move will be proper, though the force is actually generated by gravity on the object center of mass.
You can drop a center of mass on the end, and its like attaching a weight where ever you put it. I am still working on this section of the forces
involved with assets due to their nature of being connected anywhere and everywhere at once. SO dont get too hung up if things work a bit weird,
I will be making changes as I find better ways of constraining them. In Auggie the COM setting actually moved the objects center, but here its like a small variable weight you can place anywhere on the object. Again, forgive weirdness in this beta, I am zeroing in on the final solutions.

Thanks
Hope your having fun,
Art

Art:

Thanks for the updated version of my attempted simulation of Newton's Cradle. I've loaded the Newtonv1.zip file that you sent to me. I haven't had much time to play more with the Newton's Cradle.

I've been exploring the creation/ use of Assets. I've tried using Vexx and Fusion 360 to create objects and STLs needed for the creation of Assets.

I'm thinking that Fusion 360 might be way easier and precise to use for the creation of parts I'll need to experiment with the Denison gravity escapement. I've noticed a possible difference of Z axis definition between Fusion 360 and Ticker.

The bars of my Newton's cradle were created via Vexx. A number of things surprised me. I thought that gravity should swing the bar when the bar is at an angle relative to the vertical... it didn't. I added a shaft to the lower end of the bar and gravity did swing the bar back to vertical. I then rotated the bar about it's shaft further and my added shaft did not move with the shaft. I noticed the same disconnect when I tried to designate Mass Center.

I'll continue to review your 9 Videos on Ticker to see other features I may have missed.

Bill Michael

Hi Bill:

( I got some time so I couldn't help playing..)

Here's a modified version of your newtons cradle attached. I did find if I removed "Lock Position" that the legs floated away.
The shafts on the legs were not locked to the table. I selected each top shaft and used the "Fuse" button on each that floated away. This locked
them in place so that "Lock Position" could be removed safely.With lock position on, as the legs are assets, they didn't collide ( but the
bottom shafts did..which is where what bounce you had came from.).
(This is a screwup here, when I allowed fusing a shaft to the workbench, the message I gave you was and is ambiguous, you cant
tell if it fused or unfused( which it will do if you fuse a fused shaft). I have modified for next release so that the message will
tell you if the shaft was just fused, or released from its fuse to the bench.

After that I set all masses to default. If you run it now, you will see if come down and hit, then bounce away. About 3 of the legs ( maybe 4) are affected, but the force is not transferred well across them enough to make it go "full newton".

I also set "Angular Damping" on each leg and shaft to 0 from .75. This stopped much of the force from being lost by damping.
I think Ill redo that code so that 0 is default on angular damping. Feel free to play with gravity itself, restitution( which may effect this
type of thing ) and any other variables to see if you can make it go "Full Newton." I suspect friction must be lowered and it may be impossible
due to my own internal assumptions on friction. I will investigate why all this is so.

I will release a new version this weekend with the fixes I just described, but your current version should show this newtons
cradle very close to actually working..

Art

Hi Bill:

Leave this one with me, I haven't built a newtons yet, but I had noticed that swinging links have an issue with
shafts at times. I am investigating this one and will let you know when I figure out and fix this type of arrangement.
Christmas stopped my debugging till now, and Ill work on this in the coming week. It looks good by the way, and should
be able to make work, but I see several problems in the way the system is seeing it.

The Wall gravity is simply a selector as to if gravity is applied as to a wall contraption, or one on a table. You have it
right, this malfunction is all me and I suspect I introduced it in one of the later versions as I repaired a certain malfunction
I had been searching for at that time.

Art

Art

Attached is a zip file that contains the project file of my first attempt at creating Newton's Cradle. It's similar to my long ago attempt at creating the cradle using Augie's Vector processor.

I've been experimenting with some of the Expert settings: mass, Angular damping.

I see what their effect is but I'm trying to grasp the implications of Wall Gravity vs Gravity settings in the Ticker simulator. I don't quite understand the workings of the Mass Center feature.



Bill