While I do not care for the Crust Crawler hexapods at all, they did do their pan/tilt assembly right. I really like their S3 Pan/Tilt base, since it can look a full 180 degrees up/down and left/right.
Has anyone mounted one of these on an SES compatible robot? I believe the S3 could be mounted to an ASB-06 on the front of a robot. There would be at least two holes for mounting screws, and possible three or four depending on the vertical spacing of the slots.
8-Dale
I dont like the design personally. I don’t like the industrial tan color or all the slots, and the SES brackets can be made to do the same thing and achive 180 deg pan tilt. All that you have to use is the asb-05 or the asb-10
SES FTW 8)
Absolutly! SES is “For The Wise!” 8)
I don’t know about the “right” part - but they did do it differently.
While their pan/tilt uses a turntable assembly similar to the structure of the Lynx molded shoulder base, to provide more support to the bits being panned and tilted than just a servo shaft will allow, I suspect that it is necessary to provide the extra support, because the aluminum structure and mounting plate hang so far out over the edge. The design looks like it could be fairly easily duplicated in Lynxmotion SES parts, using a BR-KT base rotate kit and some brackets, but you’ll still be front-loading the assembly, and forcing the mechanism to work to keep your sensors pointing anywhere other than at the ground.
Both their design and the Lynx pan-tilt kit on the “assemblies” page do this, placing the mounting plate well away from the axis of rotation. Both of these assemblies may be fine for a lightweight sensor package where the structure and sensors don’t drag the servo down too much, but without counterbalancing or load-sharing springs, the tendency will always be there for the nose to dip downwards and resist pointing up. A “quick and dirty” solution is to rotate the terminal bracket from a horizontal position, with the mounting plate on a vertical plane, to a verical one, with the mounting plate horizontal. This would, in effect, “balance” the sensor payload on top of it, and lessen the down-tilting tendency. Many (most?) common robotic sensors are designed for vertical mounting from the rear, rather than horizontal mounting from the bottom though, so for a lot of things, like sonar sensors, the Sharp rangers, etc., you’d need to work out a 90-degree bracket of some sort. It would also tend to add to the height above the ground, which is where most of the obstacles you’re trying to detect live. The difference may only be an inch or two, but depending on your sensor and the mounting geometry, it may make the difference between seeing or not seeing something, significantly affect the detection range, or prevent it from seeing an area immediately in front of the robot, where you’re often trying to detect things in the first place.
This whole front-to-back pan/tilt thing was discussed a while back when someone was looking at mounting a pellet gun on a turret. With a heavy load mounted to the front of a pan/tilt, the servos will always be working against gravity, and it will quickly “wilt” when power is removed. If a system is devised where the axis of rotation passes through or near the CG of the gun/camera/sensor-suite/badger/whatever, it will minimize the amount of work the servos must do, and minimize the amount of power required to move them. It also has the upshot of reducing wear and tear on the mechanics and electronics of the pan/tilt servos.
Again, for small, lightweight things, the load won’t be very high even if they’re hanging out there on a big aluminum bracket, but everything adds up. I suspect that all of those slots are there on the crustcrawler design not just to allow for flexible mounting, but also for weight-reduction purposes.
I am not intending to prop or bash anyone’s product here - just making a few observations.
I actually like the pan/tilt setup I have on W.A.L.T.E.R. now. It gets +/- 60 degrees pan and +90/-30 degree tilt. If I made a small cut out in the deck, I could get a fill +/- 90 degree pan. The 30 degree down tilt is plenty to see anything on the floor except something directly underneath the robot. I’m using an ASB-18 attached to the pan servo, which connects to the tilt. An ASB-10 attached to the tilt servo is what the sensor(s) attach to. I have a 3D model of it which would be easier to show, which I’ll post later.
8-Dale
A couple of very simple 180 deg pan tilt web cams I made. The small black one was in service for ~5 years until the servos wore out. The larger white one was in service ~3 years until the computer died. Rubberbands need to be changed every ~6 months. 
The key to easy 180 deg pan/tilt operation is to use small cams.
http://www.geocities.com/zoomkat/pix/picolo-pt.jpg
http://www.geocities.com/zoomkat/pix/ptcam.jpg
http://www.geocities.com/zoomkat/pix/ptcam1.jpg
http://www.geocities.com/zoomkat/pix/ptcam2.jpg
Very High Tech!
Gettign back to the SP3 Pan/Tilt, I think thats an outrageous amount of money for the P/T, but they deffinately did do it right! the servo is @ an angle so you get more tilt in more desirable degrees up or down.
Still not worth the 100 though, even with our high Canadian currency 
It’s alright looking but the price is insane. I can’t believe someone would pay &80 for a a couple aluminum brackets. You could build 8 SES pan/tilts for that price… and it wouldn’t be that awful grey color. 
We think alike! The price is just to high for a single function design. When you get bored with this thing, you cant reuse the brackets and then it becomes a $90 dust collector.
Grey color? I thought that is had a hint of tan. Oh well, old age I guess…
I wouldn’t pay that price for an S3 when I can design something using all SES parts that works the way I want it to work. What I designed is actually a Tilt/Pan versus a Pan/Tilt, in order to get it to sit where I need it to be on WALTER’s new body. I’ve got three different front turret designs I am playing with now. I did get some ideas from the S3 design though.
8-Dale
I can’t see anybody buying a pan/tilt anything for small cams. Way too easy to just make one (assuming one has some basic construction skills). The lost art of DIY. 
This is true - my comments were in keeping with the initial discussion of the crustcrawler pan/tilt assembly. If all you’re swinging around is a sugar-cube camera and a couple of sensors, there’s rarely any need for anything as elaborate as a huge platform like we’ve been discussing.
It probably only really becomes necessary when you’re aiming something with sufficient mass to waggle around from its own inertia, that causes a servo to exert more than a trivial amount of force to move it, and/or causes the mechanism to “wilt” when power is removed.
