Here is a sneak peek of our 6 DOF arm prototype. It uses the Low Profile Axis for the forearm rotate and wrist rotate. It uses the same heavy duty bearing for the counterbalance that the Low Profile Axis uses. The arm is not powered up in this image, yet the forearm is able to hold position due to the counterweight. Springs are ok for balancing arms, but the counterweight works better.
This arm has a very nice “Industrial” look to it. It looks like it could lift a tank!
This brings up a question I had about the low-profile axis. Namely what is the point of it? It seems to me that in any of the contexts that you have shown this low-profile rotational axis (e.g. this arm and the biped, both of which are very nice by the way) you could have simply mounted a load bearing bracket directly onto the servo horn (in much the same way that the bracket is directly attached to the rotational shoulder servo in the bipeds). Although I suppose it would look somewhat clumsy in the case of the arm.
Is the reason you want to mount the bracket onto a gear and then use the servo to rotate that gear that you don’t want the servo horn to be directly bearing weight in the axis perpendicular to its rotational axis (i.e. pushing or pulling on the servo on the axis that points directly up through the middle of the horn and into the servo gears)? If that is a concern then this does seem to be a very clever means of eliminating that. I’d have to think a bit about how it could be applied to the shoulder of the biped arms though…
What’s the point of it?.. I am very excited as to the assemblies this part will make possible. It simply adds rotation on the same axis without requiring the length of the servo to be added to the assembly. Perhaps a few image will better illustrate this point.
This image shows the Low Profile Axis without anything attached to it.
This is the part used for a Gripper Rotate. Doing this without the LPA would require the assembly to be 1.25" longer, and whatever is connected to the round servo horn will be supported by only the servo horn. There are many assemblies that do not allow the use of a “C” bracket to reinforce it.
This image shows the Low Profile Axis used as a forearm rotate. This enables a true 6DOF arm to be built. The closest thing to a real industrial arm configuration available. It should be apparent that this is not possible with the brackets alone. The ball bearing assembly can withstand a great deal of load to be applied.
I was going to build and shoot the 3rd axis for a humanoid biped hip but ran out of time. This part will allow a 5DOF leg to be made into a 6DOF leg and only add .375" to the length. Hope this helps…
Very cool Jim,
I definitely see how the assembly allows some designs that otherwise wouldn’t be possible. Thanks for the illustrations.
So is it true that the ball bearing assembly can bear more weight than something being attached directly to the servo horn?
I have a question of my own.
One of those pictures shows that the low profile-axis gear can be placed on the top OR bottom.
Could it be placed on the top AND bottom, as well?
If so, I could give my biped a tail that rotates ~180 degrees back and forth.
Besides the nice addition of another whacking limb, I’d also be able to use the tail to balance the biped, instead of having to interupt my gait to use one of those joints.
Could I then stack another waist rotate upside-down on top of the other, so that the biped could rotate at the waist, as well?
perhaps the solution is putting a smaller gear on the bearing and larger on the servo so that the rotation is lets say double. should save on brackets and servos
nick
It’s already in process. Just waiting for some sample units to test the fit. 
what about doing something like this so that the servo could support the weight when you dont have room for all those gears
x4.putfile.com/1/2317411522-thumb.jpg
i could really use one of these
nick
Great idea and design Nick. I don’t think you need, or want, the curves on the ends of that metal add-on to attach it to the servo though; rather the bends down should occur prior to the attachment point making the thing like like a square wave in effect. That’s the way I was thinking about it anyway. That seems like it should even be something that Hitec makes to increase the versatility of the servos for weight bearing…
That was kind of what I was thinking about in questioning the rotate mechanism (e.g. for the biped). It just seemed that you could mount the servo in such a way as to directly have the servo horn perform the rotation as long as supporting the weight weren’t an issue. It would be a more efficient use of space and weight as well… There are a lot of case (like with the arm) that it other set up is absolutely necessary though.
Also, as mentioned, the possibility of utilizing different gear ratios with the low profile axis rotation is the biggest advantage I can see. You could even get a full 360 degree swing out of it, with minimal cost in terms of accuracy, with the right gears… That would be VERY cool.
sannyasin, normally i woud agree with you on the square wave shape, but the bends are too sharp to get a suitable mounting point at the servo. also during fabrication it is usually easier (ie cheaper) to do successive bend in the same direction.
also with proper gears you can get the full 360 degrees, but you lose approx half your torque and half your accuracy.
even so i have a need for about 4 of these with a 360 rotation and another 4 of my own design
for my design i had intended to use a thrust bearing, but im hoping that these heavy duty bearings will be sold in the near future.
nick
Here is a bracket that looks like it may work for you. It’s the Hitec 55807
We stock them.
lynxmotion.com/Product.aspx?productID=335&CategoryID=87
But they are a bit pricey. It comes with a few teflon rings that go in between the horn and the bracket to reduce friction. Just an idea…
lol, I told you Hitec should make something like that (apparently they already do)… I really didn’t know that was available but it is exactly what I was describing. 
I have always been thinking of worm gears for robot arms. I feel that the self locking mechanism of such a gear should be profitable because the arm would keep his position even if you switch the power of.
that arm is pretty dang shnazzy 8)
Hello, danger! danger! newbie alert! I have recently had a robotic epiphany! WOW, these sort of things really do exist outside a lucas production! I have been doing alot of research into bipedal robots and have noticed few utilize gripping hands. Why? Are there not many good ones available or exist? Seems like they would be much more fun to play with if you could grapple an opponet or just pick things up. Any suggestions, direction, help locating such a thing? I’m really new at this and could use a hand, (heh, heh).
Lynxmotion has a biped you can build that has gripping hands. It does not have individual fingers if that is what you are after, but it still has the ability to grab objects.
Here is a pic of it:
you know, I think I must be operating on 8megs some days! after I posted this message I found the hand you provided, thanks by the way for not making me feel like a complete dolt. these are VERY cool and exactly what I was looking for! any reason we dont see these used more often in robo-one?
I can think of a few.
First and foremost, they’re relatively new.
Secondly, they don’t come from Khondo, which was the company that everyone had to make do with before Hitec and Lynxmotion (::squish::), so fitting them onto the slightly different servo horn might be daunting?
Now, the reason that I wouldn’t use them in Robo One, is that they’re rather heavy and rather short, compared to the other possible hands.
Granted, it could be beneficial to be able to grab your opponent in a fight, but there’s the whole problem of stopping the servo before it overstrains itself.
Since the size of what it’s grasping would be difficult to pre-determine, it would necessitate the use of pressure sensors, or the like.
However, there are more advantages, like being able to use the added weight to shift the biped’s COG, and to pack a more powerful punch (although it may be an overbalanced one).
There’s also the other events to think about (which, I appologise, as I always seem to forget about them).
While I’m not sure that they’d work well with the throwing or door-opening event, I believe that they’d suit the eagle event very well.
In the end, it’s up in the air as to what they’d work best with.
Personally, I won’t be using either of the lynxmotion hands, but have instead cut one long sword (two blunted dowels) “hand” out of aluminum stock, and have made one short battering ram “hand” out of four of Lynxmotion’s steel hex standoffs.
When I get the Lexxan sheets in my next order, I’ll be cutting my own brackets to fit them.
That’s the beauty of the Lynxmotion parts, though; they’re very modular, and also very tolerant of being cannibalized.
I’d say, if you like the look of the gripper hands, go for it!
Cool idea of the sword. Sort of puts a gladiatorial spin on things. How about saddle up a hexapod or quadrapod and fit the biped with a lance! 
Let the games begin!
In all seriousness, good reasons for not seeing alot of hands in use,but you have to admit there is a certain joy in pulling the cats tail for effect!
thanks, James