coding for roller cams
22 Oct 2010 14:08 - 22 Oct 2010 14:12 #4795
by grey1beard
Replied by grey1beard on topic Re:coding for roller cams
OK, here's the initial build.
The project is to provide an apparent random eye movement for a chameleon model that my daughter is building. The c. is sitting on the top of an arm chair and it replicates the fabric of the chair which she is re-covering.
It's one of those "Daddy, would it be possible......." projects.
Chameleons move their eyes indepentantly, and having them reposition at a rate of about once every 6 secs, each in a variety of directions was the challenge. The width of the head level with the eyes is about 4cms, and each eye moves with a radius of about 7cms over 90 degrees.
I've started by building a small timer cct, based on a 555 to pulse a solenoid. The plunger pulls a pawl that you can see at the top of the photo, which pulls both ratchets.
In the middle of the photo, the two halves of the cam mechanism are laid out, so imagine that the right hand one is turned over and drops on top of the left one, with the lower transparent layer, plus a couple of paper washers, separating them.
That separator carries the non-return indents, two short piano wire lengths.
The solenoid/pawl is mounted to align with the ratchets through the gap at the left of the separator.
The sneaky part of the design is that one ratchet has 24 teeth, the other 25, so that the stepping pattern takes 600 pulses to repeat, about an hour, and who's going to watch it that long !
Each ratchet gear is cemented to a cam with 24 and 25 lobes on it(some are the same height, so no movement at that point).
The cam follower levers will pull piano wire connectors to the eye moving levers, not yet built.
The solenoid/pawl/ratches worked until I put some small load on the levers, and immediately the friction between the steel levers and the steel cams, all 1mm thick, caused it to refuse to budge.
My pathetic attempts to avoid this by polishing the edges of cams and levers before I built it (I did see that it might be a problem)was nothing like a solution.
What I'm currently doing is to replace the cams with 3mm nylon, and fit the levers with rollers, then see how it goes.
John
EDIT photo too big, will retry.
The project is to provide an apparent random eye movement for a chameleon model that my daughter is building. The c. is sitting on the top of an arm chair and it replicates the fabric of the chair which she is re-covering.
It's one of those "Daddy, would it be possible......." projects.
Chameleons move their eyes indepentantly, and having them reposition at a rate of about once every 6 secs, each in a variety of directions was the challenge. The width of the head level with the eyes is about 4cms, and each eye moves with a radius of about 7cms over 90 degrees.
I've started by building a small timer cct, based on a 555 to pulse a solenoid. The plunger pulls a pawl that you can see at the top of the photo, which pulls both ratchets.
In the middle of the photo, the two halves of the cam mechanism are laid out, so imagine that the right hand one is turned over and drops on top of the left one, with the lower transparent layer, plus a couple of paper washers, separating them.
That separator carries the non-return indents, two short piano wire lengths.
The solenoid/pawl is mounted to align with the ratchets through the gap at the left of the separator.
The sneaky part of the design is that one ratchet has 24 teeth, the other 25, so that the stepping pattern takes 600 pulses to repeat, about an hour, and who's going to watch it that long !
Each ratchet gear is cemented to a cam with 24 and 25 lobes on it(some are the same height, so no movement at that point).
The cam follower levers will pull piano wire connectors to the eye moving levers, not yet built.
The solenoid/pawl/ratches worked until I put some small load on the levers, and immediately the friction between the steel levers and the steel cams, all 1mm thick, caused it to refuse to budge.
My pathetic attempts to avoid this by polishing the edges of cams and levers before I built it (I did see that it might be a problem)was nothing like a solution.
What I'm currently doing is to replace the cams with 3mm nylon, and fit the levers with rollers, then see how it goes.
John
EDIT photo too big, will retry.
Please Log in or Create an account to join the conversation.
23 Oct 2010 13:24 #4812
by grey1beard
Replied by grey1beard on topic Re:coding for roller cams
andypugh wrote:
Better still, you could use a real PC and EMC2 with a cellphone camera in each eye to track movement for maximum freak-out value.
Once you have played with an Arduino for a while, they start to look like an almost universal solution to this class of problem.[/quote]
Andy, don't tempt me
The whole family are quite capable of wilder flights of fancy than I've ever admitted to here.
Hence this particular mad project.
John
Better still, you could use a real PC and EMC2 with a cellphone camera in each eye to track movement for maximum freak-out value.
Once you have played with an Arduino for a while, they start to look like an almost universal solution to this class of problem.[/quote]
Andy, don't tempt me
The whole family are quite capable of wilder flights of fancy than I've ever admitted to here.
Hence this particular mad project.
John
Please Log in or Create an account to join the conversation.
Time to create page: 0.082 seconds