selecting power source for mobile robot

Discussion in 'The Projects Forum' started by HaPPy_BoY, Dec 5, 2010.

  1. HaPPy_BoY

    Thread Starter Member

    Dec 5, 2010
    36
    0
    Hi. I'm currently doing a mobile robot project. I'm planning to use switching regulator between the power source and the motor load. I like to know whether, if sharing/using same power source for both the load and also to provide gate signal to the regulator, will be inefficient ?
     
  2. tyblu

    Member

    Nov 29, 2010
    199
    16
    Can you show a schematic? I don't get it.
     
  3. wayneh

    Expert

    Sep 9, 2010
    12,126
    3,048
    It's a good idea to isolate a high current load from any logic or sensitive circuits such as comparators, op-amps, etc. It can be really hard to maintain a good ground for the sensitive stuff if there's a current flowing through the same conductor. Ground won't be ground anymore.
     
  4. HaPPy_BoY

    Thread Starter Member

    Dec 5, 2010
    36
    0
    Here's a schematic of the regulator. At the mosfet gate, I was thinking of making own oscillation for the signal, using crystal and comparator.

    So, my question is that whether it will be inefficient if that I connect the input voltage supply of the crystal and comparator to the power supply (11.1 V) of the circuit ?
     
    Last edited: Dec 5, 2010
  5. HaPPy_BoY

    Thread Starter Member

    Dec 5, 2010
    36
    0
    I see. Thanks for the advice. Any power supply recommend other than lipo to make the oscillation works ?

    Also, do i need to separate the grounds? One ground for the load and one ground for the logic/sensitive circuits ?
     
    Last edited: Dec 5, 2010
  6. thatoneguy

    AAC Fanatic!

    Feb 19, 2009
    6,357
    718
    I keep the two supplies fully isolated without issues.

    The logic/sensor battery gets very little drain, whereas I replace the batteries for the motors quite often with another rechargeable pack.

    Use a lot of decoupling caps on both power supplies, as some electric motors emit RF, and the wiring around the robot acts like antennas.
     
  7. wayneh

    Expert

    Sep 9, 2010
    12,126
    3,048
    I can't comment on the PSU but I wholeheartedly endorse separate ground WIRES. You can have a shared ground, but make sure any connections to it are over conductors that are more than adequate for the current. Don't pass any substantial current over the same conductors you use for logic ground. It's easy to think that "ground" appears at the end of any wire connected to true ground, but that's just not true.
     
  8. tyblu

    Member

    Nov 29, 2010
    199
    16
    Separate ground planes are usually not required unless you are drawing a large amount of current. Just good decoupling.

    http://goo.gl/Mhl7Z
     
  9. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    I don't know exactly what you're trying to accomplish with that incomplete "regulator circuit", but it won't be very efficient.

    R5 is simply wasting power.
    L4 and D2 are on the wrong side of MOSFET Q2.
    R6 will simply limit the effectiveness of C7; they both might as well be omitted.
    Q2 has no control for Vgs.
    D5 and C4 will simply waste power.
    R3 and L1 serve no useful purpose that I can discern besides being expensive and wasting power.
    You have four diodes (D5, D1, D4 and D3) in series between the battery and the output, which will all have voltage drops and waste power.
    R4, L3, Q1 serve no useful purpose that I can determine.
    R2 is so large that you will hardly get any current output.
    There is no feedback or voltage reference to compare a feedback to.

    Why don't you start off by telling us about your motor(s), and its' current and voltage requirements? It would be a lot easier to design a chopper driver using a current sense resistor and a comparator or an IC than trying to correct what you have so far.
     
  10. HaPPy_BoY

    Thread Starter Member

    Dec 5, 2010
    36
    0
    Hi SgtWookie. Thanks for your replies. At the output, I have a servo controller with 10 servo motors. 5 servo motors needed to be moved simultaneously. The servo controller needs 6V. The total current will be 5A for 5 servo motors to move at the same time, right? For the power supply, I have 11.1V 2200 mAH Lipo batt.
     
  11. HaPPy_BoY

    Thread Starter Member

    Dec 5, 2010
    36
    0
    Is it recommended that if the regulator is to be made as boost converter so that the output is to be a total of 18V to accomodate 6V servo controllers with 10 servos and two 12V 1.1Watt dc geared motor?
     
  12. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    6v at what current?

    You haven't provided any information about the motors until the beginning of your reply. You say "servo motors", but no specifications other than you think you'll need 5A to move them.

    Do you have a manufacturer and part number for these servo motors? Otherwise, it will be very difficult to figure anything out.

    That's a draw of 110mA (0.11A) current over 20 hours. With larger current flow, your battery life will be far less.
     
  13. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    You don't normally operate them in series.

    Apparently, you need a 6v supply for the servo controller.

    You may need that same or a different voltage for the servos themselves; I don't know because you have not provided that information. I have no clue how much power the servos will require.

    You now say you need 12v for two 1.1 Watt DC geared motors; each motor will then require ~91.7mA.
     
  14. HaPPy_BoY

    Thread Starter Member

    Dec 5, 2010
    36
    0
    I'm sorry for the lack of information. The servo motor distributor/manufacturer is Cytron. I'm using five C55R (max 7V) and five C40R (max 6V) servo motors according to their part number. The link for the information can be found at http://cytron.com.my/viewProduct.php?pid=Eyw4GioFFAszLxQkEzUzEQfMWm8DBPPkBZlGiX8DeH8=and http://cytron.com.my/viewProduct.php?pid=OAIKOQ4lDSwhIxUwJi88MdT2W4Jwruq4N/28lEPo8Sk=.
    From their website, they do not provide any clue to the current except some small hint on the enquiry/forum that the servo motor can draw up to 1A in case there is obstacles to prevent them from rotating. To be on safe side, i want to build a circuit which can provide 1A for each of the servo motor in case they really need it.
    It is a 16 channel servo controller.The servo controller has two separate supply input; 5V for the MCU on board and another input is for the servo motor connected to the controller. That input is based on servo spec and it will be max 6V due to limit of C40R servo motors. The info from the website is http://cytron.com.my/viewProduct.php?pid=FBoZEA45IQU4JDAJCRIHEvcPVfKFOasbfqAXc/4jaKA=. No current specification can be found on it as well. Currently the 5V supply to the controller can be provided, except only the input supply for servo motor which I could not guess the current required.

    My team currently trying to build a mobile 6-segment snake-like robot and the servo motors is to provide the repetitive snake locomotion. Each segment will have two servo motors to provide 2-axis motion; vertical and horizontal. The two dc motors will be positioned front- and back-end to provide forward movement. The robot is supposed to be able to move in uneven surfaces like grassy landscapes, and climbing up stairs.
     
  15. HaPPy_BoY

    Thread Starter Member

    Dec 5, 2010
    36
    0
    I like to know if connecting 5V 10MHz triangular waveform mxo45 crystal oscillator to LM311 comparator with 2-3V threshold voltage could make a logic signal to be used at the mosfet gate? Or will LM358 be adequate enough to make logic signal ?
     
  16. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    From the sounds of things, your selected battery will be inadequate for your robot; you'd be lucky to even get a couple of minutes of operation from it. The load will be so heavy, the battery may even overheat and rupture forcefully.

    You need a more capable battery or batteries.

    MOSFET gates are basically capacitive in nature. You charge the gate to turn them on, and discharge them to turn them off. The charge/discharge is relative to the source terminal. Rds(on) is specified with Vgs at a given value.

    Total gate charge is given in nC, which is nanoCoulombs. It can take a considerable amount of current to charge/discharge a MOSFET gate at higher frequencies. Even 100kHz is challenging for a novice.

    I think you would be well-served to start with a less challenging project, preferably a kit, in order to have something that you can assemble and get working immediately.
     
  17. HaPPy_BoY

    Thread Starter Member

    Dec 5, 2010
    36
    0
    About the battery, it worries me too. At one time, the plan is to drive only 5 motors, either moving left and right or up and down, and the 2 dc motors. Will it work if there is two 11.1V 2200mAH LiPo and one 7.4V 1300 mAH?
    On the mosfet, would it be easier if i just buy a mosfet gate driver like IR2110? Or perhaps maybe using Darlington pair or power transistor with lower frequency could be easier for me to handle ?
     
Loading...