I'd like to know whether a motor could be used to fasten and unfasten a joint that is otherwise moving. (If you're interested, I'm still further developing this invention / idea - for a mechanical manipulator).
I've found that there are a lot of camera "ball heads" some of which support hundreds of pounds -- so I think it might be appropriate to use one and motorize its dial, and also do the same thing at the other joints. I've found simple circuits like this.
But I'm not sure how it could be used to fasten or unfasten a joint! I mean, if it is able to tighten it tight enough (and then doesn't the motor get damaged once it can't move anymore) then is it strong enough to untighten it?
So my question is: what is the the generally accepted way to motorize a friction fastener (such as wingnut and bolt or any other fastener)
Is this possible to do with electronics? Or can only a person do it?
of course it is possible, and probably in common use.
look at the release mechanism of locking pliers.
as for the motor, a current limited motor driver is not that hard to find.
but, what you are saying does not compute.
'otherwise moving' ? a lug-nut on a car tire is moving from the view point of another car, but from the view point of the wheel stud, they are firmly in a fixed position in relation to each other.
but, if you are thinking of an artidulating arm, and want the elbo to move freely some times, then lock it at other times. the locking motor is still firmly fixted in relation to the locking mechansim that locks the shaft that is moving.
this is also done in all geared transmissiong. each gear that is not engages, watches thing rotate, but the one gear that is engaged, is in a fixed position in relation to the driving shafts.
if you have a knuckle joint, you can use a clutch mechanism to engage or release.
Thanks for your quick answer! Some of the vocabulary is new to me.
release mechanism of locking pliers.
Perhaps I wasn't clear, I meant in an electronic way, so that the fastening and unfastening is controlled from a switch that is away from the location. i.e. only wires go to where the joint is loosened or locked, then an electrical signal is given to loosen or lock it.
but, what you are saying does not compute. 'otherwise moving' ?
Yes, as you've expressed, the articulating arms should be fastened sometimes, and free to move other times, preferably without adding a lot of weight.
These are the points that need to be fixed.
ball joint and 4) moving pivot points. like this. 1) is a ball joint, the other 4 are simple joints/pivots. Not sure what the technical term is.
if you have a knuckle joint, you can use a clutch mechanism to engage or release.
is there such a locking mechanism that is electronic? i.e. can I run wires to a knuckle joint, and if I give one signal it locks, if I give another signal it moves freely?
It seems that what you've said above is mechanical, but I'm interested in the electronics of it. Thanks for your help.
it is difficult to discuss the electronics without the mechanical side, since the mechanics of it play a large part in what sort of electronics are needed. You could just physically attach a motor or servo to a nut, or you could have a motor/servo that (dis)engages a mechanism that in turn causes the joint to become fixed. It all depends on your needs and how much force you are dealing with.
On the electronic side, one method of preventing the motor from damaging itself or other parts is to detect current levels. A motor will draw more current as its load increases. You can use this data to decide when the motor should shut off.
In the world of electronics, if you can imagine it, more than likely it's possible!
XORduino:
it is difficult to discuss the electronics without the mechanical side, since the mechanics of it play a large part in what sort of electronics are needed. You could just physically attach a motor or servo to a nut, or you could have a motor/servo that (dis)engages a mechanism that in turn causes the joint to become fixed. It all depends on your needs and how much force you are dealing with.
On the electronic side, one method of preventing the motor from damaging itself or other parts is to detect current levels. A motor will draw more current as its load increases. You can use this data to decide when the motor should shut off.
In the world of electronics, if you can imagine it, more than likely it's possible!
I don't want to do it if I'd be the first to control a nut with a motor and measure the current level to see when it's tight. There has to be a better way. Plus, if it's tight enough to hold it rigidly, how can the motor ever hope to untighten it.
it seems like there should be some "standard" way to do this. I mean, it's simple: here's a moving joint, lock or unlock it with lightweight electronics - it can even take a few seconds.
surely this has been a need someone has had... I can't be the first
curious48:
I don't want to do it if I'd be the first to control a nut with a motor and measure the current level to see when it's tight. There has to be a better way. Plus, if it's tight enough to hold it rigidly, how can the motor ever hope to untighten it.
it seems like there should be some "standard" way to do this. I mean, it's simple: here's a moving joint, lock or unlock it with lightweight electronics - it can even take a few seconds.
surely this has been a need someone has had... I can't be the first
dave-in-nj:
you would have to have a time machine and go back a couple decades for that.
I was doing that in the 1980's
And I was taught by people many years my senior.
Are you talking about a mechanical solution only, or one that is remotely actuated with electronics? I really don't understand the exact mechanism you've proposed. I want to flip a switch and have a clamp hold a joint down, without that clamp weighing a lot. What's the best way of doing that? You've seen my picture above for the kind of application I want to use it in. Thanks.
It seems to me like this question is being asked the wrong way round.
Mechanical locking devices with remote control have been around for yonks - long before computers were even a gleam in Babbage's eye.
I think you need to describe the mechanical lock that you want to use and then ask what sort of remote control would be appropriate.
It should be easy to use a screw system which stops when a certain torque (amps in the motor) is reached. Just make sure the motor can exert more torque (more amps) for the unlocking process.
Lots of locking systems (e.g. the brakes on a car) rely on large leverage. What about a lever operated by a servo via a spring to limit the force?
Adequate force = torque = reduction gear
Time = Speed = motor rpm
The ball joint you refer to in one of the photos would be simple to tighten to the point of breaking the bolt... or a typical ball & socket joint, could also be addressed in several ways... tighten the socket around the ball etc.
if there is exposed 'volume' around the joint, you may be able to ditch the tightening motor, and apply a magnetic solution, or rotary solenoid...
One simple way could be with PWM-driven dc motor. Use 50% PWM when you tighten the nut and 100% when you unscrew it. The unscrewing torque should be higher.
You need a way to find out when the motor can't move anymore, to avoid overheating. Since sensing the current is an easy way to achieve that (when stalled, current increases abruptly), you can stop powering the motor when it reaches eg 50% of its maximum stall current.
Or use a power resistor and two diodes to limit the max current when motor turns one way and have no current/resistor limit when it's polarity is reversed.
blimpyway:
One simple way could be with PWM-driven dc motor. Use 50% PWM when you tighten the nut and 100% when you unscrew it. The unscrewing torque should be higher.
You need a way to find out when the motor can't move anymore, to avoid overheating. Since sensing the current is an easy way to achieve that (when stalled, current increases abruptly), you can stop powering the motor when it reaches eg 50% of its maximum stall current.
Or use a power resistor and two diodes to limit the max current when motor turns one way and have no current/resistor limit when it's polarity is reversed.
Is this done in practice?? It doesn't strike me intuitively as something that would work. You can't always unscrew something with exactly twice the torque you used to screw it in!
I would like to know if anyone has actually done this. It seems like a fairly simple requirement. I mean, you're locking or unlocking a moving joint. It's easy to describe.
But when I google locking joint with motor arduino unfortunately this thread (that I'm writing in) is already the top result and there are no other relevant results.
I would really like a link to some project somewhere that has done this. I don't want to experiment with something that isn't really proven... unless you vouch for a particular solution and can suggest it in detail given my requirements
There is virtually an infinite number of ways of solving this issue. As Robin2 pointed out, I think you need to choose a specific method you'd like to use, and then ask for details on how to go about it.
On the specific topic of a motor/servo being able to undo it's own tightening:
The force required to loosen the joint will generally be equal to the force used to tighten it.
The use of current limiting/detection to stop the motor can also be used to allow additional current for loosening the joint (as also proposed by blimpyway). This way, you're guaranteed to have sufficient torque for loosening the servo's own tightening.
AWOL:
Surely "arduino" is irrelevant as a search term?
I find it very relevant because it assumes beginner-level electronics background, it really breaks things down from what you're trying to do to a way to do it. If any project with any arduino has ever needed to electronically tighten and loosen a nut or similar fastener, finding it would immediately give me invaluable advice.
XORduino:
On the specific topic of a motor/servo being able to undo it's own tightening:
The force required to loosen the joint will generally be equal to the force used to tighten it.
The use of current limiting/detection to stop the motor can also be used to allow additional current for loosening the joint (as also proposed by blimpyway). This way, you're guaranteed to have sufficient torque for loosening the servo's own tightening.
Your point number 1 is extremely interesting!! I had no idea this was the case. In fact, I would expect it NOT to be the case. For example, consider screwing a wingnut in with a piece of fragile uncooked spaghetti. After screwing it in as tight as you can (so that the spaghetti bends and is about to break), would you expect to be able in all cases to unscrew it by reversing the motion? Personally, I would not: I would expect the spaghetti to break in some cases. This proves that the force required to loosen the nut could be greater than than the force used to tighten it.
Actually I'm curious if you agree with my thought experiment, or if you think I give the wrong results were I to do it. What do you think?
Since you seem to be worried by the possibility that a screw won't be un-screwable why not look for some other solution - such as a friction brake. Or design a system that does not require much screw torque to lock it.
One thing you need to decide (and have not already mentioned) is how firmly the lock must be applied. In other words, after the lock is applied how much torque must the joint be able to resist before it moves. That is a (or the) key element of the design.
In addition, the ability of the joint to resist torque depends very much on the design of the joint and the lock combined - including the materials each of them is made from.
Figure out the mechanics first. Make a working test piece. Then come to the Forum for assistance with the control system.
Is this done in practice?? It doesn't strike me intuitively as something that would work. You can't always unscrew something with exactly twice the torque you used to screw it in!
It isn't about exactly twice the torque, it's just making sure the unscrewing torque is significantly bigger than screwing torque. Maybe it needs three-four-five times the torque. It doesn't matter. With PWM you can turn a motor with 1/256 max static torque in one direction than the other.
How big exactly it's easy to make and test what basically is an electric screwdriver.
I would like to know if anyone has actually done this. It seems like a fairly simple requirement. I mean, you're locking or unlocking a moving joint. It's easy to describe.
I don't know about locking joints but if it is a nut and a screw involved, any electric screwdriver that is not junk should be able to unscrew a bolt that itself previously screwed. I see no such big dilemma.
Edit: this is a mechanical engineering problem, you-re more likely to get more relevant responses on a mechanical engineering forum. People here can better tell you how to control motor movement with an arduino, than how to make that motor lock a joint or grind cereals or pump concrete at 10th floor.
