Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Running switching regulators in parallel

  1. Jun 7, 2013 #1
    For a high altitude balloon I need to power a radio that draws 1A but briefly peaks at 1.4A for a fraction of a second. <SEE EDIT>

    I was thinking of using two of these in parallel:

    Or one of the above item and one of these in parallel:

    But I am not sure if you can run these types of regulators in parallel. Some places have said you can, others have said you need to use oring diodes, so I am not sure what to do.

    This is for a High Altitude Balloon, so this does need to be reliable.



    Ok, I am an idiot. I made incorrect measurements. The peak draw is actually ~1.3A to ~1.5 continuous. (I.E. not for a fraction of a second, for all the time I am keying the radio)
    Last edited: Jun 8, 2013
  2. jcsd
  3. Jun 8, 2013 #2


    User Avatar
    Gold Member

    As for paralleling those regulators, why not ask the manufacturer if it can be done?

    The R-78B6.5-1.0 data sheet shows output overcurrent set to 2A (with automatic recovery). The performance graphs show typical output ripple voltages at output currents up to 1.2A. Accordingly, it seems one device would power your radio, even if it draws 1.4A for a fraction of a second. You could first try it on the ground easily.
  4. Jun 8, 2013 #3
    Ok, I am an idiot. I made incorrect measurements. The peak draw is actually ~1.3A to ~1.5 continuous. (I.E. not for a fraction of a second, for all the time I am keying the radio)

    I am very sorry about that mistake.

    So it looks like I will have to ask the manufacturer.

    Thanks for your response.

    If anyone already knows about paralleling these regs please post - it may take a while for the manufacturer to get back to me.
  5. Jun 8, 2013 #4


    User Avatar
    Gold Member

    OK. Why can you not find one regulator that outputs say, 2 Amps?

    You have not told us your input voltage or your output voltage. Those would be useful to us who are trying to analyze your design. Thank you.
  6. Jun 8, 2013 #5
    That is another option. The problem is that I need a black box solution (no external parts), that I can get access to easily & cheaply in Australia.

    I could say buy this from digikey, but it is almost $40 shipping:

    From RS I can get free shipping but they only stock the 1A model.

    This is a student project & we are on a very tight budget as you can imagine :)
  7. Jun 8, 2013 #6


    User Avatar
    Gold Member

    That Digi-key 945-1719-5-ND is only rated at 1.5 Amps (maximum). That does not seem to be a "robust" choice. Sorry your have such a limited budget, but you need reliability, as you mentioned. You need a system that you purchase, assemble, and test thoroughly on the ground before you fly it.
  8. Jun 8, 2013 #7
    I am sorry - I totally agree. I do want to choose the best option, and if that means paying a substantial amount for it I will. I just find it a bit hard to justify $40 shipping for a single component :)

    Ok, so that component doesn't seem to be a good match anyway. Looks like I need a 2A reg at 6.5V to be safe.

    Our input voltage is 10-13.3V

    I will keep searching, if anyone has any recommendations they would be welcome :)
  9. Jun 8, 2013 #8
  10. Jun 8, 2013 #9


    User Avatar
    Gold Member

    Many manufacturers specify their DC to DC converters by their power output. Seems you need (6.5V X 2A = ~15W)

    Using Google search for 15 Watt DC to DC converters I found several. Here’s one that may work for your project: PST-DC292 Cost US$ 18.95

  11. Jun 8, 2013 #10
    That could work but it is super heavy . This is a high altitude balloon project so weight needs to be low as possible :)
  12. Jun 8, 2013 #11

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    the way I read their datasheet there's a minimum 10 ma load.
    Can you stand one diode drop or do you need precisely 6.5 volts?

    A 10 ma LED at each output would assure minimum load, then diodes to the radio would let you parallel them - and give you an "on" light for each.

    They make a 1.5 amp medical version, 6.5 would be a special. Might inquire if they've made that one for somebody already.
  13. Jun 10, 2013 #12
    IN addition to the Vin / V out -- what do you expect the duty cycle to be. i.e. How long ON ( 1.3-1.5 A) and how long off?

    What is the Voltage Supply range allowed by the radio?

    Also - when keying ON the radio will you see a surge? If you are close to the specs for one converter - perhaps adding some bulk capacitance to the output will help.
  14. Jun 10, 2013 #13
    Re Jim:

    Thanks for your response. Ok, I have read about using oring diodes to run switching regulators in parallel. Have you had success with doing this?

    In relation to the medical grade 1.5A version. I have seen it, however I do not have a easy way to acquire it, the shipping from digikey is $45. So If I can make the 1A version work running in parallel then that would be considerably more cost efficient.

    Re Windadct:
    V in - ~13V down to 10V
    V out - 6.5V
    Current when listening - ~100ma
    Current when transmitting - ~1.5A

    The radio will transmit once per minute for ~8 seconds MAX each time (probably more like 5 seconds, but 8 to be safe)

    I don't think the radio surges. I think I made a bad reading as RF was getting into the multimeter wires.
  15. Jun 10, 2013 #14


    User Avatar

    Staff: Mentor

    I just caught up in this thread. One *very* important tip when working with radios is to NOT use a switching power supply to power them directly. You need to put a linear regulator between the DC-DC converter and your radio receiver circuitry, or you will de-sensitize the RX circuitry significantly.

    It looks like it can still work, but you may want to use a SEPIC topology DC-DC conterter, so the output voltage can be held at a steady 12V, even after the battery starts to droop. That would let you use a standard-dropout linear regulator for your 6.5V radio supply voltage. Alternately, if you could get a standard DC-DC buck converter to provide you with a steady 7.5V, then you could use a low-dropout (LDO) linear voltage regulator.

    The layout of your PCBA (PC board assembly) and/or wiring of the modules also needs to be done correctly, to keep digital circuit noise from getting into the RX radio circuitry. It took us a while to learn all of the tricks here where I work. If you could post what your assembly and wiring will look like, we could try to give you some tips...
  16. Jun 10, 2013 #15

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    I've done it before but not with this particular device. It was a different DC-DC converter with isolation.
    The LED indicator meets minimum load for you.
    The regulator that's set highest will hog the load until resistance of the diodes lets them share.

    Thought I had a solution for you with this regulator:
    http://www.ti.com/lit/ds/symlink/78sr174.pdf [Broken]
    but can't find a place for you to buy them.

    Original National datasheet showed how to make LM317 into a 3 amp switcher.
    http://www.hep.upenn.edu/SNO/daq/parts/lm317.pdf page 10 of 20
    But I realize you want a 'drop in' part.

    Sorry , but I don't have a better answer than diodes to make them share load..
    I though about plain old linear LM317 but it wastes quite a bit of power during transmit period..

    Try a bench check using the diodes with 10 ma LED to keep each supply happy about minimum load. I think it's a reasonable and practical solution.

    Beware of noise. Like Berkeman I have chased switcher noise all over analog circuits. I have not much radio experience and am sure glad he rang in .

    old jim
    Last edited by a moderator: May 6, 2017
  17. Jun 11, 2013 #16
    Is this only an issue on RX? As we are not really using RX, only TX.

    Thanks for your tips, if this is going to be an issue of TX then we will have to go for the linear solution you suggest.
  18. Jun 11, 2013 #17


    User Avatar

    Staff: Mentor

    It is mainly an RX issue. The desensitization of the RX stage can be on the order of 30-40dB, which makes the receiver close to useless. If you will only use the RX up close (like before you launch the balloon), then it will probably work.
  19. Jun 13, 2013 #18

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Another thread came up with this little switcher
    looks right flavor, 3 amps, has two more pins for output adjustment..
    see if they're available in your neck of the woods ?

    http://www.tracopower.com/fileadmin/medien/dokumente/pdf/datasheets/tsr3.pdf [Broken]

    Here's the distributor "Down Under" :
    Last edited by a moderator: May 6, 2017
  20. Jun 16, 2013 #19
    Ok. I am thinking I will use the LED, diode, switching solution Jim suggested. I can just use a small receiver on a linear regulator for RX.

    Ok, so I am thinking of getting this:

    I am just not sure what kind of LED to use or how to wire it, do I need to use a resistor? Also, I think that it would be good if I could find a diode with a lower voltage drop, but I can't see any on RS.
  21. Jun 16, 2013 #20

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    The diode you linked is a Schottky, and low forward voltage drop is characteristic of Schottkys. It looks like a fine choice.

    Most any LED will do.
    Here's one that your supplier says he can deliver in ten days with no shipping charge:
    Looks like they come 100 to a pack so you'll get a lifetime supply.
    or two in one package
    It'll need a resistor to set current

    to size resistor for ten milliamps
    subtract from your 6.5 volts the forward drop across LED;
    6.5 - ~2.4. = 4.1
    That's voltage across resistor
    R = E/I , I = .01 amp so you want ~ 410 ohms
    you might aim at 12 ma to leave tolerance ,~ 341 ohms
    360 and 390 are standard values .

    That current rating on LED's is for full intensity. They'll be quite visible at lower current.

    Were his catalog easier to search i'd have looked for 10ma LED's with built-in resistors.
    a 12volt 25 ma might draw about 10ma at 6.5 volts, if avoiding the resistor is worth an experiment

    Good luck !

    old jim
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: Running switching regulators in parallel
  1. 5V Switching Regulator (Replies: 13)

  2. Regulators in parallel (Replies: 3)

  3. Switching Regulator (Replies: 4)