Capacitor selection for multiple power modules

I'm trying to design a board that needs multiple voltages, and I've settled on some of TI's PTx series, since it seems that they're going to be pretty straightforward to implement.  My current design has three supplies:

PTH08000W:  3.3V, with two MSP430s, GPS, magnetometer, gyro, and bluetooth module (<200mA)
PTH08000W: 5V, powering two remote encoders, one digital, and one analog (<50mA)
PTR08060W: 5.5V, powering two servos and an R/C receiver, drawing up to 3A at stall.

All of these will be pulled from the same 7.4V lithium battery, which brings me to my question.  The datasheets for each of these supplies are very specific about what the capacitances should be on input and output.  I'd love a little advice on what to use for the input capacitors, since there will be three supplies in parallel.  Should I use a single larger capacitor (~220uF?) or put 100uFs as close to the input pins as possible?

Any advice is welcome, up to and including recommendations for the capacitors themselves- or even advice on the output capacitors for the specific loads mentioned.

Thanks,

Ben Stiles
NCSU grad student
Advanced Diagnosis and Controls Lab

  • The recommended input capacitors for:

    PTh08000W  100uF capacitor  is required with Rrms ripple current rating of 650mA. The suggested input capacitors are located on page 9-10 in the product  specification Each capacitor  should be located < 1cm from  the input power pins.

    Output capacitance is optional,. However, a 100uf polarized capacitor or 100uF of ceramic capacitors is suggested.

    PTR08060W 100uF capacitor  is required with Rrms ripple current rating of 650mA. The suggested input capacitors are located on page 9-10 in the product  specification The  capacitor  should be located < 1cm from  the input power pins.

    The ouput capacitor is required for this  module.The output capacitors are also located on page 9-10.

    PTH08000W:  3.3V, with two MSP430s, GPS, magnetometer, gyro, and bluetooth module (<200mA)
    PTH08000W: 5V, powering two remote encoders, one digital, and one analog (<50mA)
    PTR08060W: 5.5V, powering two servos and an R/C receiver, drawing up to 3A at stall.

    How are you planning to prevent current when not in use? Are you controlling each  inhibit pins? Each module requries its separately control open drain FET for each inhibit pin.

    Tom

  • In reply to Tom Guerin:

    Thanks for the quick response, Tom.

    When not in use, I'm just planning on disconnecting the power.  The 5V will have a FET controlling the Inhibit pin for safety reasons- This circuit is being build for an autonomous sailboat, and the 5V powers the wind sensors mounted at the top of the mast.  In case the boat turns over, I'd like to have a way of shutting off power to those very exposed sensors.  According to the data sheet, it doesn't appear necessary to us the inhibit pin unless needed.

    I'm still unclear about my question though.  Since each of these power modules requires 100uF of capacitance, do I simply need a single 100uF capacitor mounted within 1cm of the PTR08060W?  In theory, this would provide appropriate capacitance for all three modules.  In practice, though, would it be better to put a 100uF capacitor near each supply?  This would give a total of 300uF capacitance on the input net- is that acceptable?

    Thanks again,

    Ben.

  • In reply to Ben Stiles:

    Ben:

    The input capacitor at 300UF with 100uf dedicated for  each module is the preferred method for input filtering.

    I do have a concern about the PTR08060W and mechanical stability when assembly. Shock and vibration unless the  complete completely encapsulated  will can separate form the module.

    Regards

    tom

  • In reply to Tom Guerin:

    Thanks for that.

    I'll plan on using 6 100uF capacitors total then- one each for the input and output of each module.

    As to the mechanical stability, is there any part in particular that I should be concerned about?  I was actually planning on removing the right angle header pins and mounting the device horizontally- mostly for space reasons.  With the device in hand, I see also that there are two "extra" holes at the top two corners of the device.  I could easily design the PCB with matching holes and use unconnected pins soldered to the device and the PCB for extra stability- giving the entire structure 7 pins for mechanical support.  I should mention as well that this is more for a proof of concept project on a 3' sailboat to be sailed on a lake, not an ocean going vessel.

    Ben.

  • In reply to Ben Stiles:

    PTR08060 top two corner pins are internally connected to power pads

    Pin 2 Vin pin = left top corner  blank pin

    pin3 ground= right side corner blank pin.

    The removal of the present pins could strip the copper vias interface.

    Tom

  • In reply to Tom Guerin:

    Additional data Tom Guerin wrote the following post at Thu, Jun 23 2011 9:24 AM:

    PTR08060 top two corner pins are internally connected copper very small traces to each power pads. These small inner layer traces will not support  the power and curren t. the inner copper traces  will open up if used as the substitute power connections rather than the power pins.

    Pin 2 Vin pin = connected via small copper trace  to left top corner  blank pin

    pin3 ground= connected via  a small copper  trace to upper right top  corner blank pin.

    The removal of the present pins could strip the copper vias interface.

    Tom