Hello everyone,
I need to build a robotic arm that’s at least 12" from elbow to wrist and 12" from elbor to shoulder. I was hoping i can use the servos from lynxmotion.com, but I’m worried about the low torques will slow it down, if it ever lifts the arm at all 
. The arm would also have to lift 2.5 lbs at the wrist. Any suggestions on how/where to get the servos i need for this mega-project? ANY HELP WOULD BE GREATLY APPRECIATED
THANKS! 
Your design requirements are somewhat vague. You need to specify any limitations on the “arm”, as there are other ways besides servos to possibly get the job done. How much $$$ do you have to spend on the project?
a 24 inch long arm?!?!? 
you gonna need some beastly servos, even if u do get the most powerful servo, idk if they would be able to lift 2.5 lbs along with that ton of stress they are under already
Thanks for the reply zoomkat
the arm has to perform a series of repetitive movements, (more specifically moving a fencing sword side to side, up and down etc. )I just chose servos, because of the readily available servo controller and program that lynxmotion.com offers. I thought this is the easiest way to go - but if you have another idea/suggestion to accomplish the mentioned tasks, please feel free to post. I admit, I’m a total newbie 
the $$$ limit is supposed to be minimal since it’s school project, i’m not sure yet but maybe $800?
so am I looking into industrial servos? are they as easy to control as the hobby servos found here?
well, as far as i know, even the beastliest servos would not be able to do that, i wish i could tell you looks around for equation to find how much servos can hold
but it is quite a bit of weight
Hi,
here, try this thread :
lynxmotion.net/phpbb/viewtop … 1&start=60
in one of the posts from captain hes gives some links to linear actuartors
here is one of the sites : firgelliauto.com/default.php?cPath=86
and ultramotion.com/products/bug.php
there pretty expensive, but maybe you could search for others a bit less expensive.
only an Idea… but that would have the strenght your looking for…and more (150 lbs to 500 lbs!!!)
Sam
yea, i was going to suggest that, but i couldn’t find it, but yea, linear acutuators would be a better choice
much more powerful than servo
Well, one thing to shoot for is keeping the drive motors off of the arm and back in the trunk if possible. If you look at the below arm you see that the gripper servo is actually at the rear of the arm. It uses a cable in a sleave to get the power to the gripper. These cable setups are some what like cable brakes on a bike. You can make crude linear actuators using nuts on threaded rods turned by something like a small drill motor. You would probably need to make a gimbled joint for the shoulder.
Or… go the reverse-engineer route and make a much lighter sword.
If you want something that’s going to be quick and strong enough to bother fenicing with as is, you’ll need powerful motors/servos or linear actuators.
This pretty much leaves you all alone for learning how to control them.
Small hydraulics might also be answer.
What about balsa wood?
thanks Sam for leading me to the spider-bot thread! Also, thanks to all those who replied in such a short period of time. so I think I’m gonna go for the linear actuators for my project…
Captain, how are you controlling your actuators? Are you writing your own progam?
I plan on writing my own program to control the acutators. I’m pretty sure I’m gonna need a H-Bridge for each actuator. I really won’t know until I finish designing the robot, and begin the prototyping phase of the project.
Currently I’ve completed about 10% of the design phase.
cool! can you give us info when you start??
Sam
Sure, I’ll be glad to help once I start. Isn’t that what forums where invented for? 
Hi all,
I’m curious as to how this project is progressing; I happen to be researching possibilities for a fencing arm myself.
I started thinking along similar lines (that I might be able to find a suitable programable arm) but the weight/speed issues seem to present a great number of difficulties.
I’m now reducing the problem to the needed start/end states/positions of the 1lb ‘blade’ and wondering if an ‘arm’ is actually the best way to manipulate the ‘blade’ through these states.
The purpose of my ‘arm’ (as is the original poster’s I believe) is to simulate the basic attack thrusts of foil fencing at a medium or variable speed to allow for solo training in the parry of each thrust.
To fuction in this way the device must be capable of moving the "hilt’ of the foil to the four points of a square about 2’x2’, adjusting the angle of the ‘tip’ about 45deg. inward from each point, and then advancing the assembly about 2’ forward and back.
As a proper parry is made by blocking the top third of the attacker’s blade with the bottom third of one’s own blade the “foil” manipulated by the machine could also be shortened by 2/3’rds creating just over ten inches of ‘blade’ to manipulate and a weight of an ounce.
This certainly might de-anthropamorphize the ‘arm’ one practises against, but might make the manipulations needed quite a bit more achievable with what we have to work with.
I’d love to know what people think about other ways to potentially solve this problem!
Kind regards.
holbrook
The other problem that I see (perhaps I just misunderstood) is that the 2 lbs weight of the fencing sword isn’t a dead weight 2lbs at the end of the arm, right? It’s a 2 lb pole extending out another several feet…
2lbs of solid mass at the arm is one thing, a 2lb pole that is 4 foot long is going to put SIGNIFICANTLY more strain on the arm.
yeah but you should be able to find the center of mass relative to where you hold the foil and use that as your moment arm… until you like actually hit something and then it all goes right out the window.
If you used gas shocks (tailgate) to take some of the weight off the motor or servo.
It would also absorb impact loads.
