Page 4 of 7 FirstFirst 1234567 LastLast
Results 31 to 40 of 63

Thread: Next steps with a PhantomX AX Hexapod Mark II & BBB

  1. #31
    Join Date
    Jul 2014
    Location
    New Zealand
    Posts
    70
    Images
    3
    Rep Power
    19

    Re: Next steps with a PhantomX AX Hexapod Mark II & BBB

    Quote Originally Posted by CreedyNZ View Post
    Thanks, that's very helpful and might have hit on the problem. At the moment my power setup is like this:
    Attachment 6105
    which I suspect is causing too great a voltage drop. I'll add another powered hub like this and see if things improve.

    Attachment 6106
    Ouch! Might need plan B

    Click image for larger version. 

Name:	Hub.jpg 
Views:	189 
Size:	32.1 KB 
ID:	6107

  2. #32
    Join Date
    Sep 2010
    Location
    ಠ_ಠ
    Posts
    2,300
    Images
    27
    Rep Power
    278

    Re: Next steps with a PhantomX AX Hexapod Mark II & BBB

    Quote Originally Posted by CreedyNZ View Post
    Ouch! Might need plan B
    Ouch indeed. Of course, I had meant use one 'hub' per leg with each servo in the leg having at least one direct cable connection to the hub, so six in all for a hexapod... The barrel connectors are usually rated for only ~5A, the smaller screw connectors are rated for only 10~15A per pin, and the Molex SPOX connectors on the dynamixels are rated ~2A per pin. The two SPOX connectors of each dynamixel are internally connected together with 22~24AWG wires soldered directly to the pins on the back of the PCB, so there is also a nice voltage drop across each servo during heavy loading.

    Cleaner solution I had in mind is a custom hub with 3~8 dynamixel/SPOX connectors in a single line with exposed traces on the back of the PCB reinforced with solder, then using at least 16AWG for the GND and VCC lines to connect each limb hub to the battery and provide comms through a singe SPOX connector to the central hub (disconnected VCC pin to prevent accidental power distribution through weaker comms hub). A bit like below, but with mounting holes and back protected by insulating material after solder reinforcement (or just double sided foam tape).
    Click image for larger version. 

Name:	dxl_power_hub_v0.0.0 (filled).jpg 
Views:	162 
Size:	73.7 KB 
ID:	6108Click image for larger version. 

Name:	dxl_power_hub_v0.0.0 (ripped).jpg 
Views:	167 
Size:	107.1 KB 
ID:	6109


    Cheap/dangerous option is to splice heavier gauge wires into the GND and VCC lines of at least one of the existing 3-pin cables in each leg and collect those heavier wires to a central power connection to the battery to leave the powered hub only/primarily for comms.
    Please pardon the pedantry... and the profanity... and the convoluted speech pattern...
    "You have failed me, Brain!"
    bleh
    more bleh

  3. #33

    Re: Next steps with a PhantomX AX Hexapod Mark II & BBB

    exposed traces on the back of the PCB reinforced with solder
    You can run 85 mil wide traces on top and bottom, which will be enough. No exposed/solder needed.
    Especially if you design the hubs as a star, rather than as a linear strip.
    Although even with linear, 170 mil effective trace width is enough for > 10A continuous, which you won't actually draw.
    You can even use a GND pour on the bottom, and VCC pour on the top, to have the distribution be very wide, leaving the loss to only the solder connection to the SPOX, which is a very short distance.

    The 2A rating for the SPOX is based on long-term sustained current heating parameters. For walking, where your duty cycle is 50% or less, they aren't that bad in practice. The thin gauge wire (including the thin gauge wire inside the servos) is more of a problem.

  4. #34
    Join Date
    Sep 2010
    Location
    ಠ_ಠ
    Posts
    2,300
    Images
    27
    Rep Power
    278

    Re: Next steps with a PhantomX AX Hexapod Mark II & BBB

    Quote Originally Posted by jwatte View Post
    You can run 85 mil wide traces on top and bottom, which will be enough. No exposed/solder needed.
    Especially if you design the hubs as a star, rather than as a linear strip.
    Although even with linear, 170 mil effective trace width is enough for > 10A continuous, which you won't actually draw.
    You can even use a GND pour on the bottom, and VCC pour on the top, to have the distribution be very wide, leaving the loss to only the solder connection to the SPOX, which is a very short distance.
    I still prefer my fusible elements to be discrete and easily replaceable components on a PCB or in a holder/housing, and not the traces of the PCB. 'sploded traces are not fun. It may be overkill, but still cheaper to replace an intentionally self-destructive and disposable component than a PCB and all its mounted components.
    Please pardon the pedantry... and the profanity... and the convoluted speech pattern...
    "You have failed me, Brain!"
    bleh
    more bleh

  5. #35

    Re: Next steps with a PhantomX AX Hexapod Mark II & BBB

    Double 85 mil traces will not "fuse" unless you have no main fuse in your system.
    Even a slow-blow 20A fuse will blow way before the traces or connectors take any damage in the case of a short circuit.

  6. #36
    Join Date
    Jul 2014
    Location
    New Zealand
    Posts
    70
    Images
    3
    Rep Power
    19

    Re: Next steps with a PhantomX AX Hexapod Mark II & BBB

    Thanks for the guidance. I've found a way to reduce the shipping cost of the powered hubs, so plan to move initially from the current situation of 1 hub powering all six legs to three separately powered hubs, on for each pair of legs. At the same time I'll also upgrade the master switch and fuse housing, as I suspect these may also be too light and adding the potential for voltage dropout. If that doesn't work I'll try the more elaborate solutions above.

    At the same time I like the renumbering idea from r3n33, so might try that too to minimise the downtime if any servos still revert to 1.

    Will let you know if increasing the number of hubs improves the situation

  7. #37
    Join Date
    Jul 2014
    Location
    New Zealand
    Posts
    70
    Images
    3
    Rep Power
    19

    Re: Next steps with a PhantomX AX Hexapod Mark II & BBB

    Hi

    Quick update on how the move to three independently powered hubs is going:-

    So far, after about 2 weeks, I'm pleased to report that I've experienced no instances of an AX12 resetting itself to 1.

    Early days yet, but so far so good.

  8. #38
    Join Date
    Jul 2014
    Location
    New Zealand
    Posts
    70
    Images
    3
    Rep Power
    19

    Re: Next steps with a PhantomX AX Hexapod Mark II & BBB

    Quote Originally Posted by CreedyNZ View Post
    Hi

    Quick update on how the move to three independently powered hubs is going:-

    So far, after about 2 weeks, I'm pleased to report that I've experienced no instances of an AX12 resetting itself to 1.

    Early days yet, but so far so good.

    Spoke too soon! Just had a single servo revert to 1!

  9. Re: Next steps with a PhantomX AX Hexapod Mark II & BBB

    Quote Originally Posted by CreedyNZ View Post
    Spoke too soon! Just had a single servo revert to 1!
    Hi Creedy, thanks for all the code examples in your Github account...I'm trying to do the same thing, but then use WebIOPi and a camera to control the hexapod over the internet.

    Currently I have the raspberry2 connected via a usb-ftdi cable that came with the Arbotix-M, but like you I had a problem of having to reset the board everytime python runs.

    I read in your code that you are using a GPIO pin to reset the Arbotix-M using a pull-up resistor, but where on the Arbotix-M did you connect it to? The reset jumper? The schematic has me a bit confused

    Your hexapod looks amazing!, did you make a Youtube vid yet?

    Thx!

  10. #40
    Join Date
    Jul 2014
    Location
    New Zealand
    Posts
    70
    Images
    3
    Rep Power
    19

    Re: Next steps with a PhantomX AX Hexapod Mark II & BBB

    Glad to hear the you're finding the code useful, though all I've done is modify the code from those that came before me.

    I simply connected the GPIO to the reset jumper (pin closest to the CPU) and it works well for resetting the Hexapod on startup. No pull up required, the RPi2 brings the pin low enough to reset and back up high enough again to run.

    Jim Bob was getting too fat, so I've made a few mods since the last photos to lighten the load and to add a mast for the compass to eliminate the remaining mag interference. I'll post some update photos and vid when I get a break in the work schedule next week.

    Look forward to hearing how you go

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. Question(s) PhantomX AX Hexapod Mark II Kit to Quadruped?
    By siempre.aprendiendo in forum Humanoids, Walkers & Crawlers
    Replies: 3
    Last Post: 04-23-2014, 11:34 AM
  2. News/Announcement PhantomX AX Hexapod Mark II Kit
    By etwx in forum Garage Sale
    Replies: 2
    Last Post: 04-08-2014, 08:31 PM
  3. Wanted: PhantomX Hexapod Mark I
    By fostac in forum Garage Sale
    Replies: 2
    Last Post: 01-03-2014, 12:21 AM
  4. Question(s) Quadruped Mark II PhantomX Robot Turret
    By hwan we in forum Humanoids, Walkers & Crawlers
    Replies: 1
    Last Post: 03-16-2013, 07:35 PM
  5. Question(s) PhantomX AX Quadruped Mark II PhantomX Robot Turret Equipped??
    By hwan we in forum Robotics General Discussion
    Replies: 1
    Last Post: 03-11-2013, 02:09 PM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •