Hi,
I am new to the Lynx 6 so would really appreciate it if you could provide me help on:
thanks 
When you say that you need it to rotate 360 degrees, are you saying that you just need to be able to swing through a full rotation, with a bit of overlap, and back again, or that you need for it to be able to rotate continuously through several rotations, without worrying about twisting or binding wires?
If you just need one rotation plus overlap, I’ve seen some products out there (you’ll have to google for them, I’m afraid) that put a regular servo through a gearing stage and external pot, to achieve multiple rotations. There is also a sail winch servo made by Hitec (and probably several others), which gives three rotations, but I know very little else about it, and have never worked with it myself.
If you’re looking for continuous rotation, absolute positioning of the arm within that rotation is going to be an engineering challenge, as you’ll need some way to tell where it is in said rotation. This can be achieved in various ways, but the ones that leap most readily to mind are:]Absolute encoder/:m]
]Relative encoder with a “zero position” index mark/:m]
]Stepper motor with “zero position” index sensing/:m]
]Directional sensing/:m]
]Narrow-beam beacons to tell the arm when its facing in the desired direction/:m]Admittedly, I have never implemented these, but I’ve read various articles about them, and seen examples of each, which should also google.
Thanks Seamus!
I would ideally want continuous rotation. let me see what i can google.
Hi, essentially my project requires the base of the lynx 6 to rotate continuously . I don’t the require shoulder movement, it stays upright at all times. I also can do without knowing the NUMBER of rotations, I just need to know when a rotation has been completed.
here are the potential problems i see, I hope you can help me out:
thanks.
For continous rotation, you could most any motor, although you would need to have some setup to reverse and control the motor if needed. To keep from twisting wires, you could use one of the slip ring devices used on telephone cords to keep them from twisting. This would provide for four wires (tx, rx, v+, and ground).
if he actually needs to get significant power up to the arm for electronics and servos then you will need something beefier than a telephone slip ring. if you “just” needed power, which would require all the electronics and a wireless connection of some sort, you could use something like a mercotac 205.
I wonder if the new lynxmotion rotary base could be modified with a round PCB and some spring loaded ball connectors to make a slip ring connection… hmm. 
If the servos needed to be powered, one could put a battery pack on the rotating base and run power from a charger to the batterys thru the slip ring to keep them somewhat topped up. I think the coninously rotating base would present more problems than necessary, especially determing its position. I looked at the phone slip ring some time ago, and it might be possible to stack two, but it would be small tedious work.
There’s also a question of driving the base. A stacked slip-ring type of setup (as used for phones, etc) is fine for a platform with an open lazy-susan type of bearing, or one with a central axle that you could fit the wiring through and mount the ring rotors to the bottom. In a Lynxmotion-type base though, where the base rotation is provided by a motor mounted directly to the rotating base and bolted to the static chassis, you may end up needing a concentric ring-and-contact design to allow for a central motor mounting.
Edit to add: Also consider that one should always consider moving or sliding contacts to be intermittent in nature, and design accordingly, with at least one redundant set of contacts. This is true for power contacts, and especially true for any set of contacts that will have data traveling through them. the faster your data stream is, the more sensitive it becomes to minor glitches and inconsistencies.
If your data path is one-way to the arm, and you don’t expect to read any data back from it, another option might be to provide power via slip rings or some similar method, and send the commands to the onboard controller via an optical path. Photons don’t twist.
You can buy through-bore slip rings.
You can also use ball bearings as slip rings if they are mounted on plastic or shielded. They work well for small applications.
hi, thanks again.
The continuous rotation servos dont have position feedback. So you will not be able to do any automated movements as you will not know which way the arm is facing.
Also, the servo will not try to hold its position if there is any force on it, so it could be facing anywhere.
If you need multiple rotations knowing the position of the servo, you would have to use either a hitec HS-785HB (this is a 1/4 scale servo though), or a Dynamixel AX-12+ Actuator (this uses a different kind of communication though). I don’t know of any standard size servos that can do multiple rotations with position feedback.
Hope this helps.
2 more q’s:
In a typical 3 DOF arm setup (shoulder, elbow, wrist, gripper) you only need 6 connections.
-4 signal connections (one for each servo)
-1 common +Voltage
-1 common -ground
Each servo doesn’t need it own power. They can be tied together but their grounds must go back to the receiver/ssc-32. You can knock one more connection off and mount your servo power on the rotating base and then you won’t need a + Voltage going back. So now only 5 connections.
As for the height, you can buy thin and wide slip rings to keep your rotational surface low. It all depends on what you use. A lot of slip rings are bearing based and can be used as the rotational bearing themselves since you have a small load.
Is there any way I could knock off the control signal wire too? i was thinking a wireless connection between the ssc 32 and the motor by using some additional components… does that make sense?
If you have a battery powering the arm itself already then using any number of wireless serial adapters will allow you to link the ssc-32 to the PC. Probably a bluetooth based solution would be more than adequate for this task and it is one of the cheaper options too.
The absolute minimum number of connections would be zero, but that would involve placing a battery, an independent servo controller, and some sort of wireless receiver on the upper (rotating) portion of the arm, with another servo or continuous motor of some sort (plus its power and control electronics) attached to the base for rotation.
At this point, you’re probably looking at designing a new baseplate for the arm anyway, so as long as you’re doing all of that, why not just flip the base rotation motor upside-down, attach it to the arm’s baseplate, and have a passive base that supplies nothing but physical support for the arm?
This final setup would have a battery-powered arm assembly, with onboard control electronics, fed by a wireless data link, and nothing but a physical connection to the base, so it could spin around in circles all day (or at least until the batteries died). The downside is that this is a lot of mass to be swinging around, so you’d probably want something fairly beefy in the way of a shoulder rotation unit, which would probably draw a lot of power, which calls for bigger batteries, which are heavy, which requires a beefier shoulder rotation unit…etc. Or, you could use something that’s geared way down for rotation, and sacrifice a bit of speed for accuracy and lower current draw. You probably don’t need it to sling around fast enough to pitch a ball anyway.
u r a genius!
lemme see how that works out
hi, so i modified the servo and now it moves 360 degrees (it was pretty easy… my earlier q look stupid now ) … but how do i control it ? i’m using matlab… do i need basic stamp dedicated to it?? confused
thanks
You probably need to have matlab send the appropriate control string to your servo controller. If the servo is modified for continous rotation, adjust the pot for a mid position in the servo control band such that the servo doesn’t rotate. Then send the servo controller a higher or lower position value to make it rotate one way or the other.
i see that after i adjust the pot so that there is no rotation at 1500, the servo behaves in this manner:
this is very different from the normal servo behavior. i wanted to know the electronics behind it…
thanks.
