5v regulator getting hot.. also mystery problem

Discussion in 'The Projects Forum' started by Mnfwctbrb, Jul 18, 2012.

  1. Mnfwctbrb

    Thread Starter New Member

    Jul 18, 2012
    1
    0
    Forgive me but I am a bit of a novice when it comes to building electronics. I am making a usb charger that will take 6-9 v via batteries. Possibly more. I bought a 5v regulator and rigged it up. I have two issues.

    First, the regulator gets way hot. I actually found this forum by googling my problem, the solution there with a greater input was to add some resistor or something. I was wondering what exactly it is I need to add based on my power source being 6-9 v. If at all possible I would like to have the ability to use a larger power source, and I can not have a heat sink as this is basically just a cable.

    Problem two is a bit harder to find answers to.. as is, my cable is able to take 6-9 v and take it down to 5, and successfully power every usb device I have except for one- my phone. It even powers my old phones.. but not my Motorola DROID x2. What might the reason for this be?
     
  2. R!f@@

    AAC Fanatic!

    Apr 2, 2009
    8,740
    759
    First we need to know ur 5V reg number and what kind of devices u are connecting to it, meaning, we need the device spec too
     
  3. Pestatio

    New Member

    Jul 16, 2012
    3
    0
    I can only speculate, 5V regulator is a 7805? Mostly the regulator overheats when to much current is drawn. Like when it short circuits? Basically just a cable? Meaning there is no circuit board? Did you attach some capacitors from the input to ground and output to ground? This will help if the input voltage varies a lot. This might be why your motorola does not want to charge... All of this is only speculating according to the info you have given us.
     
  4. Markd77

    Senior Member

    Sep 7, 2009
    2,803
    594
    The thing is, a well behaved USB device will draw up to 500mA.
    If you are using 9V then the regulator has to drop 4V at 500mA which is 2W. It's going to need a heatsink. This could be as simple as putting it in an aluminium box, with the regulator (in TO-220 package) bolted to the box.
     
  5. wayneh

    Expert

    Sep 9, 2010
    12,087
    3,027
    I screwed mine to the sidewall of an Altoids tin containing the rest of the circuit and connectors. It gets hot, but it's lasted for years.

    [update]Whoa, didn't catch that your supply voltage could be 6v, or less as the batteries fade. Won't work. The comments about current draw and sense voltages are spot on. Your cable will work fine for low current devices and fail for big smart ones. Smart devices often look for specific voltages on the USB data pins. You need to understand what your particular device is looking for. Lots of data on iPhones. Not sure about Droids.
     
    Last edited: Jul 18, 2012
  6. mcasale

    Member

    Jul 18, 2011
    210
    12
    I see another problem, if you're using an LM7805 regulator. It is specified to have a 2 volt (typical) dropout voltage. So, if yours is typical, the output will be at 5 volts only when your batteries are above 7 volts.

    Also, assuming you are dissipating 2 Watts in the regulator, the junction temperature rise is about 130 deg C !!! You just cooked the 7805.
     
  7. nomurphy

    AAC Fanatic!

    Aug 8, 2005
    567
    12
    First of all, 6V is probably to low of a voltage to feed the regulator, there won't be enough headroom to maintain proper regulation. Let's assume that you can supply a constant 9V to the regulator input. Then, you need to determine the maximum load on the regulator output; let's assume it is 500mA.

    9Vin - 5Vout = 4V
    4V / 500mA = 2W (quite a bit for smaller regulators)

    Let's add an input resistor to try and help out the regulator, where you probably don't want less than a 7V input to the regulator in order to maintain headroom.

    9V - 7V = 2V
    2V / 500mA = 4 ohms
    2V * 500mA = 1W

    This removes 1W of dissipation from the regulator and moves it over to the input resistor, and indicates that at the maximum load of 500mA the resistor (and hence the regulator) will drop 2V and 1W. The input resistor should be a 2W to 5W device. This obviously makes it easier on the regulator, but can the regulator handle 1W of dissipation without heatsinking??? --read the datasheet to determine what can be done, or upgrade to a "better" regulator.
     
    Last edited: Jul 18, 2012
  8. wayneh

    Expert

    Sep 9, 2010
    12,087
    3,027
    Will a "dumb" USB device ever draw 500mA? I thought it was max 100mA unless a handshake was established to move up to a higher current.

    I use a 7805 with a 12V (auto) supply, and it doesn't blow up when charging a range of devices including my iPod touch. I assumed that's because it never draws more than 0.1A.
     
  9. gerty

    AAC Fanatic!

    Aug 30, 2007
    1,153
    304
    And if it's a 78L05 it's only good for 100 ma if it has a heat sink.
     
  10. BMorse

    Senior Member

    Sep 26, 2009
    2,675
    234
    You are using a stable supply that does not deplete over use, so the output of your regulator is more stable and can handle charging most devices.... he is using batteries, and a 9 volt at that..... would be better off using NimH AA's, can handle more current draw than normal AA's.....

    but since the op mentioned he has it connected to a cable, he probably has not attached any decoupling or smoothing caps on the input or output of the regulator which is probably causing it to oscillate and not have a stable output which is causing some issues with the charging circuits for some of his devices....

    I have actually built a device to charge my kids' phones and ipod's off of 2 AA's using the MAX757 IC.... so they can charge the "charging" batteries before going anywhere, and then can use it in case of emergencies to quick charge their devices......


    AND, the L7805CV put out by ST Micro can handle 1.5 amps of current draw, just an FYI....
     
    Last edited: Jul 18, 2012
    R!f@@ likes this.
  11. R!f@@

    AAC Fanatic!

    Apr 2, 2009
    8,740
    759
    Well put B.
     
    BMorse likes this.
  12. wayneh

    Expert

    Sep 9, 2010
    12,087
    3,027
    Could be. I followed the datasheet and also added big electrolytic caps in mine.

    But I guess my point was that I don't think I EVER pull 500mA through mine since that might make it hotter than the Altoid tin could handle, and maybe blow the 7805. Since it hasn't blown in several years, I speculate that 500mA mode was never activated by any attached device. Or maybe the 7805 has been able to protect itself against overcurrent, overheat.
     
  13. nomurphy

    AAC Fanatic!

    Aug 8, 2005
    567
    12
    The Junction to Ambient for the ST L7805CV is in the 50C/W to 65C/W range (similar to TO-220 LM7805), which means at 2W it will run 100C above ambient at best (too high a temp). Note that at 1W and 25C ambient, it will still run at 75C which is pretty hot to the touch (or 65C + 25C = 90C for other pkg types). The saving grace is that all have some kind of internal current-limiting and thermal shut-down.

    If the OP could provide an actual output current, then accurate assesment could be made without ASSumption.
     
  14. wayneh

    Expert

    Sep 9, 2010
    12,087
    3,027
    I could easily believe mine gets that hot, even with the connection to the tin box. Not 100°C, though. So I guess my charger is self-limited. Not the best design to shoot for, I suppose.
     
Loading...