Shop OBEX P1 Docs P2 Docs Learn Events
Is this a viable option for a higher current regulated switching supply? — Parallax Forums

Is this a viable option for a higher current regulated switching supply?

rwgast_logicdesignrwgast_logicdesign Posts: 1,464
edited 2013-02-14 18:36 in General Discussion
Lets say I had a 22v power supply and I wanted to regulate down to 10v or so, but be able to supply 10amps of current. Could one take a chip such as a 555 (I know a 555 is only good to 16v so its supply would have to be regulated too), adjust its duty cycle until it was outputting 10v, then use a mosfet like this one

http://www.ebay.com/itm/IRF530-IR-Power-MOSFET-N-Channel-17A-100V-/180827883096?pt=LH_DefaultDomain_0&hash=item2a1a2e8658

to supply the current demands.

If this would work, does anyone know if there is a chip out there I could use to generate a duty cycle directly from a 22v supply?

Comments

  • Mike GreenMike Green Posts: 23,101
    edited 2013-02-14 10:59
    You're talking about a lot of power to be switched. You're also talking about an open loop regulator (no feedback) and it wouldn't take much of a mis-step, say a brief power surge or drop in power needed, to blow up the whole thing (and whatever you're running from this). Either buy something designed for these power levels or look for a tested design on the Internet or spend the time with a variety of datasheets and Application Notes from switcher chip manufacturers to learn how to "roll your own".

    That MOSFET is a relatively old one. It may be rated at 17A and 100V, but the on-resistance is 90mOhms. At 10A, that means a drop of about 1V and a power dissipation of 10W. You'll need a good heatsink for that with the TO220 package and that on-resistance is probably best case (to look good in advertising). Without looking at the datasheet carefully, I don't know what a really typical on-resistance would be and under what conditions. Commonly, you need a decent drive voltage for the gate and plenty of drive current to charge up that gate capacitance. There are much better MOSFETs available.
  • rwgast_logicdesignrwgast_logicdesign Posts: 1,464
    edited 2013-02-14 11:30
    Wow ok, thanks for that, didn't realize that this would be that easy to blow up. Its very hard to find a 10 amp switching regulator thats the only reason I was looking in to this, I know ST has one but the chip is fairly expensive and I would have to design a PCB as theres no readily available modules using it.

    More and more lately projects I would like to do require mosfets, in the 10 to 20 amp range. Do you have a suggestion on a good mosfet that can handel 30 watts or so without a heatsink? I know there are plenty out there, just not sure which ones are quality I guess...
  • Peter KG6LSEPeter KG6LSE Posts: 1,383
    edited 2013-02-14 15:15
    Attachment not found.

    the 34063 is a VERY usefull chip . and with a External Fet and a proper rated Inductor it can handle the current you need ,...

    and has Feedback!!!!!!

    and Yes it can do Buck, or Boost .,
    Peter
  • localrogerlocalroger Posts: 3,451
    edited 2013-02-14 15:37
    Such a homemade "switching" regulator wouldn't give you any advantage over a linear regulator. Actual switchers are more efficient than linears becuase they use an inductor to store and release energy, which does not involve the resistive losses you will see in either a linear or "switched" scheme. You'll find your "switcher" dissipates just as much power as the linear regulator as it charges the capacitor you need to maintain power when the switch is off.
  • rwgast_logicdesignrwgast_logicdesign Posts: 1,464
    edited 2013-02-14 16:14
    I have a few adjustable switching regulators around, ones I know how to use and feel comfortable with, I guess just adding a fet on there output would have been the simplest solution...
  • Mark_TMark_T Posts: 1,981
    edited 2013-02-14 16:48
    One issue with switching regulators is the choice of inductor - this is always a compromise. Unless you choose to go for an air-cored inductor you'll
    have to deal with saturation and non-linearity and iron-losses. With an air-cored inductor you'll have a large and cumbersome device with a lot of
    copper! However it might fair better in open-loop control (and its inherent resistance and power-dissipation ability might help with overloads).

    Consider switching at 100kHz, with 0.2V and 2A ripple, using synchronous switching buck circuit (2 MOSFETs), you would want 25uH into 250uF,
    an air cored 25uH inductor capable of handling 10A with low resistive losses is a large beast, perhaps a toroid taking up a
    cubic decimetre or so. More normally seen in long-wave radio transmitters!
  • Hal AlbachHal Albach Posts: 747
    edited 2013-02-14 18:36
    " I wanted to regulate down to 10v or so, but be able to supply 10amps of current...".

    Is about 12 Volts close enough?

    First thing that came to my mind was to use a PC APX Power supply, the 12 volt rails have lots of muscle. They even have a seperate 12 V cable, 4 or 6 pin, Brown and Black, for plugging into power ravenous graphics cards. Plenty of sites on the web that show how to modify them so that they can be turned on without a PC MOBO. Lots of folks get elaborate and make 'em into bench supplies.

    Hal
Sign In or Register to comment.