Shop OBEX P1 Docs P2 Docs Learn Events
Sneak-Peak at a New Power Pack for the Boe-bot — Parallax Forums

Sneak-Peak at a New Power Pack for the Boe-bot

Matt GillilandMatt Gilliland Posts: 1,406
edited 2011-04-10 20:33 in Robotics
Hi everybody!
Here's a sneak-peak at a prototype Li-ion power supply for the Boe-Bot. It's a double-sided board with components on both sides. It uses a similar circuit to the one discussed here.

All status indicators are visible through the rear horizontal slots in the frame. A standard wall charger is plugged into the board out the "backside" of the robot base, as shown below...
100_4402.jpg


The circuit has fewer terminal connections than the 3x4 version (they're not needed), but it does retain the same polarized molex connector to barrel plug cable, as well as an on-board replaceable fuse.

100_4403.jpg


The Cells are rechargable between 500 and 1000 cycles. Rough estimate right now is that a Boe-bot can travel a little over four miles between charges.
100_4405.jpg


At 500 charge cycles, this would replace 2000 "AA" alkaline cells. @ Sam's Club prices here in CA (.49 each) that's almost a thousand bucks. At a thousand charge cycles...hmmm...stuck on the math here, but it's higher I think.

-Matt
1024 x 768 - 153K
1024 x 768 - 133K
1024 x 768 - 130K

Comments

  • doggiedocdoggiedoc Posts: 2,241
    edited 2011-03-30 09:02
    Sweet! I'm going to need about 8 of those.


    :D
    Paul
  • RobotWorkshopRobotWorkshop Posts: 2,307
    edited 2011-03-30 09:53
    Do you have a diode on the charger input to protect against someone using the incorrect charger? The voltage drop of one diode shouldn't make that much of a difference on the input and is a cheap way to protect against reverse polarity.

    Robert
  • RobotWorkshopRobotWorkshop Posts: 2,307
    edited 2011-03-30 10:13
    Hello Matt,

    Actually if the design isn't finalized yet I would go one more step. Where the charger plug comes into the board add a small bridge rectifier and a cap on the output. That way you can use any DC charger (within the voltage range) and the polarity no longer matters. Also, with a cap, etc on the output you could probably use some AC adapters as well. The board would be much more versitle. Please consider it!

    Since this will probably be used in a classroom setting think of it this way. The students are probably going to plug in what ever charger is sitting around that fits in the plug. If you add the bridge and filtering above then your board will probably still just work. That's much better than turning into toast because someone didn't use the original charger......

    Also, I see status LED's on there. Are those signals brought out to a .100" header some where? If not can a couple pads be added right next to the status LED's?? It would be nice to be able to easily check if the batteries are low or currently charging. In particular this would be good on a Boe-Bot so it knows when it needs a recharge and can also decide NOT to move when plugged into a charger. If the signals aren't available then someone could easily read them if there are two pads next to the LED's on your board. On one of my latest robots I wired in an optow/resistor in parallel with a status LED and run the output of the opto to a Propeller pin to read the status.

    I can definitely use some of these boards and hope you think about these other features.

    Robert
  • Matt GillilandMatt Gilliland Posts: 1,406
    edited 2011-03-30 11:02
    Good idea Robert. In the case of the Boe-bot, at the very least a diode would be a good protective measure, because it will probably always be charged by a wall-wart- especially in a classroom environment.

    The 3x4 does not have a diode, so it allows for a lower working input voltage to charge the cells (such as how it's running on a couple of solar cells now). The 3x4 is well marked with the required polarity, which provides both clarity as well as versatility.

    In this case however, I think you're exactly right, since the Boe-bot *swallows up* the board entirely, and the jack (and markings) are nearly invisible when he's right-side-up, a protective input scheme (either diode or FWB) would be pretty good insurance indeed!

    On the status LED's: The LED's are off while he's not plugged into a charging source. The Blue LED's blink (when fully charged) and/or remain on (while being charged). Red LED's indicate either a broken cell or an out of temperature condition (in both cases resulting in the charger shutting off.

    Also, to prevent the boe-bot from driving off the table while charging, the circuit completely disconnects the Cell voltage output to the bot's Boe. While charging, the bot can't drive off a table. If someone wishes to program the boebot while the cells are charging, they can simply plug in a second wall-wart to the boe (which of course defeats the safety feature of "charging lockout") - but that's up to the user.

    -Matt
  • RobotWorkshopRobotWorkshop Posts: 2,307
    edited 2011-03-30 11:12
    Good idea Robert. In the case of the Boe-bot, at the very least a diode would be a good protective measure, because it will probably always be charged by a wall-wart- especially in a classroom environment.

    The 3x4 does not have a diode, so it allows for a lower working input voltage to charge the cells (such as how it's running on a couple of solar cells now). The 3x4 is well marked with the required polarity, which provides both clarity as well as versatility.

    In this case however, I think you're exactly right, since the Boe-bot *swallows up* the board entirely, and the jack (and markings) are nearly invisible when he's right-side-up, a protective input scheme (either diode or FWB) would be pretty good insurance indeed!

    If lower voltage is a concern on the charger input and you still want to offer protection then there is still an option. I'd add a 2-pin header in parallel with the diode. That way you can use a jumper to bypass the diode and for lower voltage charging. Jumper off = Protection diode enabled, Jumper on = Protection diode bypassed, lower voltage charging. All that is needed is a diode, 2-pin header, and a .100" jumper.

    You can't control what people are going to plug into this but you can make it more robust by protecting it from some of the things they shouldn't be using....

    Robert
  • WhitWhit Posts: 4,191
    edited 2011-03-30 13:03
    Matt,

    Neat looking product! Are the servos mounted with the shafts and cables toward the front because they have to be with this power supply? Looks like maybe so...
  • Matt GillilandMatt Gilliland Posts: 1,406
    edited 2011-03-31 08:00
    Whit wrote: »
    Matt,

    Neat looking product! Are the servos mounted with the shafts and cables toward the front because they have to be with this power supply? Looks like maybe so...

    Good eye Whit. Actually, it will fit either way. Sometimes I mount the servos this way for a longer wheelbase - in this case I also used the chassis for developing a new solar panel kit (coming soon). With the large "sail area" and mounted ping, the longer wheelbase gives greater stability.

    100_4407.jpg


    ooops. how'd that get in there?
    -Matt
    1024 x 768 - 79K
  • ercoerco Posts: 20,256
    edited 2011-03-31 08:35
    Great idea. Nice near net battery fit.

    4 miles between charges? WOW. Can you get 4 miles out of a servo before it falls apart?

    Non-standard connectors are a simple way to prevent reverse polarity charging. You could always use polarized left-handed metric Fahnestock clips!
  • ajwardajward Posts: 1,130
    edited 2011-04-10 16:40
    Good eye Whit. Actually, it will fit either way. Sometimes I mount the servos this way for a longer wheelbase - in this case I also used the chassis for developing a new solar panel kit (coming soon). With the large "sail area" and mounted ping, the longer wheelbase gives greater stability.

    Oh cool! I assembled Bob as laid out in the instructions and when I added the Ping kit he became a bit nose heavy. This evening's project will be to remount the servos. Thanks for the idea Matt!

    Amanda
  • RobotWorkshopRobotWorkshop Posts: 2,307
    edited 2011-04-10 17:38
    So from the picture above are we to believe that each one comes with donuts?
  • FranklinFranklin Posts: 4,747
    edited 2011-04-10 20:33
    So from the picture above are we to believe that each one comes with donuts?
    ONLY if you are signed up...
Sign In or Register to comment.