PDA

View Full Version : Head Tilt Range



drachels
10-08-2015, 04:29 PM
Since getting and assembling my HR-OS1, I've been reading all the previous posts. One that really caught my eye (http://forums.trossenrobotics.com/showthread.php?7620-Camera-Vision-Processing-%28within-the-HR-OS1-Framework%29), started by r3n33, made reference to the limited tilt angle that made it impossible for the robot to see it's own feet. Has anyone measured the angle their robot is able to attain? I used to make my own golf clubs and remembered I had a tool for measuring angles, so I dug it out and manged to take some rough measurements.

With the stock build, I am able to look up 22 degrees and down 32 degrees. The up angle would need to be limited even further in actual operation as that is pushing against the cables.

619561946196

In a reply at the bottom of page 1 of that thread, tician hinted at using a C-bracket and a redesigned P1. I've made a mock up and boy what a difference! Using a Biloid F1 bracket with the angled arms going up and forward, I can get 90 degrees up and 110 degrees down! Does require flipping servo 19 to horn up instead of horn down.

I just finished printing r3n33's Chappie head, (thanks for the files) and with it on the same mockup I can still get 90 degrees up, but only 47 degrees down. The extra thing under Chappie's chin limits downward travel. Using the Chappie head would require both servos to be on top of the P1 plate. Without it, the pan servo could go on the bottom side of the P1 plate.

6197

Gotta print a red antenna as the orange one just doesn't look right.

I'm about 80% done on designing both types of P1 one plates I need, then 3D print and try them out. Gotta see what I can come up with for a thrust bearing.

tician
10-08-2015, 04:41 PM
The thrust bearing (http://www.mcmaster.com/#5909k34) I have is serious overkill for an HROS1 unless you are planning to do frequent headstands, so you should be fine if you just add a hard/slippery plastic bushing/washer on top of the plate, or augment the plate print to extend to the face of the horn, to push against the bracket attached to it.

drachels
10-08-2015, 05:37 PM
or augment the plate print to extend to the face of the horn, to push against the bracket attached to it. I had just about decided this would probably be my best way to go. But, ... since I had things apart to do a little measuring, I realized there is another, simpler possibility I need to investigate...tomorrow. Fresh, hot, salted, boiled peanuts are just to messy to eat while playing around with the hardware. :happy:

drachels
10-09-2015, 12:04 PM
The simpler possibility thought that I had last night was just to relocate the right hand side Head Tilt Swing Arm from the inside to the outside of the right hand Head Tilt Cam. This eliminates the first mechanical stop that limits forward tilt to 32 degrees. A little wordy and possibly hard to understand, so I added a picture. This results in the same max of 22 degrees looking up, but adds an extra 10 degrees looking down for a down angle of 42 degrees. I would be interested to know if this would be "enough" to see a regulation size soccer ball. I haven't got video working on my robot yet, so I don't know if this is enough, or not.
6198

r3n33
10-09-2015, 04:00 PM
6197



Holy cow! That is SO cool :) Thank you for sharing the print! I hope you didn't have too much trouble with supports. I've been using the latest version of Slic3r with pillar supports for a little while now and it hasn't been too bad.

Did you print the head that holds the camera forward or angled downward? Did the camera fit well and do you think there was there enough thread space to securely install the camera?

drachels
10-09-2015, 04:15 PM
I haven't had a chance to actually mount the camera in it yet. Been trying to get an stl file ready to print some hands. The models I have right now from Trossen are REALLY slowing down 123D Design. Every time I make a minor adjustment it seems to take 30 seconds or more to give me back control. Did you print yours fingers up, fingers down, or some other orientation?

I also hadn't planned on mounting it until my son has a chance to do some detailed painting to make it look more like the movie version. He does that kind of stuff professionally and of course, doesn't have time right now for family. LOL Maybe this weekend, perhaps.

r3n33
10-09-2015, 05:15 PM
I printed the hands fingers up so far.. That left just a little bit to support in the center area. I've even tried to modify the model a bit myself but these STLs seem to have a very high polygon count and it slows down my favorite 3D software too O_o

I can't remember the exact steps I took but I think I had to use 3DS or Inventor to export the STL again but at a lower resolution. There may even be some software out there for just this sort of polygon reduction task.

drachels
10-09-2015, 06:11 PM
Trying to redo the file for those hands is on my "to do" list, but I have a few other things ahead of them. I think there is possibly something wrong with them as a file for AX-12 servos is much more responsive than the hands, but the servo has quite a bit more detail to it than the hands.

It's funny, I am retired, but there are still not enough hours in the day for having fun. I have REALLY got to make room so I can move my two new 3D printers into my computer room so that I can have all three going at once. (My wife really, really, really wants her dining room table back.