Must See Video of the Day

I want to see the blue and red gears try to mesh.

Seriously what's so cool about this? Have them spin at the same speed, and lined up correctly at the beginning... (and nothing else gets changed during the demo)... it should be fine.
 
the video fulfilled my nerd requirements for the day... thank you ^_^
 
A chinook does basically the same thing. Rotors are mechanically timed and spins opposite directions so they don't hit each other.
.
 
LurkerLito said:
Come on it's from Japan, it must be for Giant Transforming Robots :)
Click to expand...

Closer.... so much closer!

the-one1 said:
A chinook does basically the same thing. Rotors are mechanically timed and spins opposite directions so they don't hit each other.
.
Click to expand...

Chinnok blades = approx 225 rpm at hover. The demo is up to 30 times as fast.

colinstu said:
...Seriously what's so cool about this? Have them spin at the same speed, and lined up correctly at the beginning... (and nothing else gets changed during the demo)... it should be fine.
Click to expand...

*sigh*
 
But what happens when you drop a bucket of ping pong balls in there?
 
colinstu said:
Seriously what's so cool about this? Have them spin at the same speed, and lined up correctly at the beginning... (and nothing else gets changed during the demo)... it should be fine.
Click to expand...

You assume every gear has the same exact ramp up/ramp down speed, the server motors all have identical timings, etc. Even a single ns worth of delay (which is typical of even high performance parts) would have these gears tear eachother apart, so you have to continually make minute adjustments to ensure proper timing.
 
Basically servo motors with encoders for syncing speed. You see things like this in industrial applications. Nothing all that special.
 
colinstu said:
I want to see the blue and red gears try to mesh.

Seriously what's so cool about this? Have them spin at the same speed, and lined up correctly at the beginning... (and nothing else gets changed during the demo)... it should be fine.
Click to expand...
wow you're right.. it's so easy

hold my beer
 
I wish whoever created that was around in the 70's. I could have used him to tune/sync the Tri-Power carbs in my GTO.
 
While cool, I think I'd rather watch the synchronized swimming team. :)
 
This is child's play with a servo with an encoder, especially with those gears with angled edges. Ah, I see someone already pointed that out, oh well.
 
not impressed, try shifting gears (standard car) by ear without pressing the clutch.
My dad used to do that.
 
Uvaman said:
not impressed, try shifting gears (standard car) by ear without pressing the clutch.
My dad used to do that.
Click to expand...

And nowadays you'll ruin the syncro's in a new cars trans.

Regardless, upshifting without the clutch is easy, the real magic is downshifting.
 
gamerk2 said:
You assume every gear has the same exact ramp up/ramp down speed, the server motors all have identical timings, etc. Even a single ns worth of delay (which is typical of even high performance parts) would have these gears tear eachother apart, so you have to continually make minute adjustments to ensure proper timing.
Click to expand...

Modern servo systems have been doing this for years. This is nothing new. I do work on machine tools and stuff like this is common.
 
That's actually not that hard to do. All that demonstrates is precision motor control under a single set of parameters replicated three times.

Mesh three sets of gears, at high speed, with each driving a different total mass, and all at a different drive ratios and you have something useful. This just shows me they figured out how to make precise throttle controls and clone it three times.
 
At no point does it give you a full shot of a gear stopped -- dollars to donuts though they also are running gears with an odd number of teeth.
 
colinstu said:
I want to see the blue and red gears try to mesh.

Seriously what's so cool about this? Have them spin at the same speed, and lined up correctly at the beginning... (and nothing else gets changed during the demo)... it should be fine.
Click to expand...

You know how I know you don't do anything technical?
 
There's nothing magical, it's just impressive they got them to match up so well at those speeds.
evilsofa said:
Not as impressive as a set of 3 gears that rotate in the same direction while in contact (engineers may find this video disturbing):

http://www.youtube.com/watch?&v=-1Gfc1Iq0GY
Click to expand...
If you are an engineer and find that disturbing, you should go to the college where you got your degree and ask for a refund :p
 
Uvaman said:
not impressed, try shifting gears (standard car) by ear without pressing the clutch.
My dad used to do that.
Click to expand...

That's not really impressive as it's a very different mechanism. The gears in a manual transmission are always meshed with each other. When you change speeds, you aren't unmeshing and remeshing the gears from each other, you are simply changing which gear set is engaged. If you actually tried to pull the gears in a car apart from each other and back together like this video with them spinning at any speed, things would go bang and you'd leave lots of chunks of metal on the ground.
 
No way to say this without sounding like I'm crapping on something that I, myself, don't have the capabilities to do...but is this actually impressive from a mechatronic perspective? Is this a difficult feat? I mean, you have motors with encoders that give very good and feedback, with accuracy down to sub-micron scales. So long as your circuitry/processing platform can communicate with and respond to the motors fast enough, and that you trust the encoders, all you're doing is running the components within their intended specifications. Yes, you need to be very accurate and precise with positions, movement characteristics (start/stop/acceleration/deceleration/velocity) for translating the independent gears. I guess what I'm trying to say, and no doubt it's going to sound dumb/coarse, but with the state of technology today, it's not all that impressive--from a technological standpoint. Am I able to do this? No. But maybe if I saw this 15 years ago, I'd be more impressed.
 
teh_chem said:
No way to say this without sounding like I'm crapping on something that I, myself, don't have the capabilities to do...but is this actually impressive from a mechatronic perspective? Is this a difficult feat? I mean, you have motors with encoders that give very good and feedback, with accuracy down to sub-micron scales. So long as your circuitry/processing platform can communicate with and respond to the motors fast enough, and that you trust the encoders, all you're doing is running the components within their intended specifications. Yes, you need to be very accurate and precise with positions, movement characteristics (start/stop/acceleration/deceleration/velocity) for translating the independent gears. I guess what I'm trying to say, and no doubt it's going to sound dumb/coarse, but with the state of technology today, it's not all that impressive--from a technological standpoint. Am I able to do this? No. But maybe if I saw this 15 years ago, I'd be more impressed.
Click to expand...

Yup, it doesn't become truly impressive until you tell me the price. They do this and it's relatively dirt cheap, it's impressive. They do this and it cost five times what they other guy who could do this cost you to do it and it's an embarrassment.
 
You must log in or register to reply here.
Back
Top