Shop OBEX P1 Docs P2 Docs Learn Events
Voltage Reg Problems? — Parallax Forums

Voltage Reg Problems?

mhamen3mhamen3 Posts: 69
edited 2012-11-30 06:56 in Propeller 1
I'm having a bit of trouble with a board I made.

I accidentally shorted battery terminals (at least I think that's what happened) while the board was hooked up to the battery. It' was a 3 cell Lipo hooked up to a 3.3v regulator. Soon after the voltage reg got very hot very quickly. So I changed out the voltage regulator and changed out the capacitors for it too but I'm experiencing the same symptoms. I can't see any obvious shorts on the board.

Anyone have any advice?

Comments

  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2012-11-29 07:06
    First of all, some regulators have thermal protection (most actually) and even short circuit and reverse polarity voltage protection. Others do not.

    So if you can look at the tiny lable on the regulator (this might require a magnifier) and identify what you have, the whole problem might just be solved by letting it cool down and starting over. The actual regulator may not have suffered any damage.

    Meanwhile you can download and read the PDF file that tells what protection your regulator provides.

    On the other hand, some of the older regulators - like the LM7805 have less protection. I believe that one only has thermal shut down. Cooling down may still solve the problem. But reverse polarity voltage might do serious harm and running current backwards through the LM7805 is also a disaster.

    ~~~~~~~~~
    You say you changed the voltage regulator and still have problems. That might be to a tiny solder bridge in your construction from the Vdd to the Vss. Look around your bypass capacitors and supply connections.

    I have learned to build board and power them up without any chips other than the voltage regulator installed. A lot of times, there are solder bridges I didn't notice. I keep a thumb on the regulator as I do this as early warning to unplug.
  • mhamen3mhamen3 Posts: 69
    edited 2012-11-29 08:20
    Thanks a lot, Loopy. I'll check this out when I get home.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2012-11-29 11:43
    BTW, I would worry more about damaging the Lithium battery with a short circuit than a voltage regulator. I've had some very costly short circuits with Lithium batteries.

    It might be better to use a cheaper battery or a AC converter when testing new boards.
  • Mark_TMark_T Posts: 1,981
    edited 2012-11-29 16:04
    I'd go further and suggest always using a bench power supply with settable current limit for initial testing - prevent and detect over-current issues safely. You can also bring
    the voltage up slowly whilst monitoring current - looking for suspicious behaviour!
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-11-29 16:16
    Like Loopy, I'm not sure how shorting the battery could harm the regulator or anything else downstream. OTOH, reversing battery polarity could cause significant damage, depending upon the regulator being used.

    BTW, a three-cell LiPo puts out 11.1V when charged. That's a very high voltage to be providing to a 3.3V linear regulator and could explain the overheating, notwithstanding the shorted battery.

    -Phil
  • Duane C. JohnsonDuane C. Johnson Posts: 955
    edited 2012-11-29 17:34
    There are many regulators that can't handle a shorted input.
    LM317 Figure 3
    LM340-n / LM78xx Figure 9

    The diode requirement is somewhat dependent on the value of the output capacitance
    However, I always add these capacitors for safety in my low volume designs.

    There are others.

    Duane J
  • jmgjmg Posts: 15,173
    edited 2012-11-29 18:18
    mhamen3 wrote: »
    I'm having a bit of trouble with a board I made.

    I accidentally shorted battery terminals (at least I think that's what happened) while the board was hooked up to the battery. It' was a 3 cell Lipo hooked up to a 3.3v regulator. Soon after the voltage reg got very hot very quickly. So I changed out the voltage regulator and changed out the capacitors for it too but I'm experiencing the same symptoms. I can't see any obvious shorts on the board.

    Anyone have any advice?

    Was this board ever working ? What is the voltage regulator ?
    A battery-side short can do some damage, when the short is released, depending on the inductance available.
    The voltage spikes could kill sensitive downstream parts.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-11-29 18:59
    Duane C. Joihnson wrote:
    There are many regulators that can't handle a shorted input.
    Right. I hadn't considered the effects of reverse capacitive discharge through unprotected regs.

    -Phil
  • Duane C. JohnsonDuane C. Johnson Posts: 955
    edited 2012-11-29 20:25
    Some of the newer ones are protected for this, but it makes me nervous.
    Especially for battery stuff where the short is easier to do.
    Besides a rectifier diode is pretty cheap for the peace of mind.

    Notice the second diode for the adjustable regulators!!!

    Duane J
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2012-11-29 21:15
    I want to say some more about solder bridges.
    I have had these show up as thin as a human hair and make the voltage regulator sizzle. It takes magnification and close inspection to see them.

    So I just take a utility knife and run the point between all the traces on the printed circuit board before I start looking. At times, this cuts the solder bridges that I can't easily see and no further looking is required.

    Yes, you can put in protective diodes, but a 1n00x diode has a voltage drop of 1 volt or more, not just 0.7v that most people guess. A lot of battery life can be used up with a protection diode. The linear regulator is already taking 30% for its share.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-11-29 21:27
    Loopy Byteloose wrote:
    Yes, you can put in protective diodes, but a 1n00x diode has a voltage drop of 1 volt or more, not just 0.7v that most people guess. A lot of battery life can be used up with a protection diode.
    The protection diode being discussed is connected from the reg's output back to its input -- not in series with the input. It's to protect against sudden capacitive voltage dumps from output to input, not reverse input polarity.

    -Phil
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2012-11-30 01:20
    Okay, I see now. As I recall, if you have over 10uf of capacitor on the output side of the LM7805 and you suffer an input short, the capacitor discharges through the regulator and destroys the internal circutry. In that case you must have a sort of fly-back diode to bypass the regular with the short.

    If you use a 10uf or smaller capacitor, you are likey to not have enough energy stored to do damage.

    Still, the LM7805 will die from reverse polarity as well.

    I prefer the LM2940-5 and the LM2940-3.3 But I am sure that newer and even better protected devices exist. Everything uses voltage regulators and it is a big portion of sales. So newer and better is constantly fighting for market share.
  • mhamen3mhamen3 Posts: 69
    edited 2012-11-30 06:15
    Wow, thanks for the tips guys.

    I'm using the LM2937. The board was working before, I probably ran it for a good 2hrs in 10 to 15 min bursts. The reg would get warm but never hot enough that I couldn't leave my finger on it. I'm thinking it is a solder bridge somewhere. I've already started milling out a new board, there were some trace sizes I wanted to change anyway.

    Thanks to all who helped. All this advice will certainly give me something think about when I need voltage regulation in the future.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2012-11-30 06:56
    LM2937 is better than the LM780xx seriies. Parallax has used a lot of them on various products because customers are happy, less boards are broken.
Sign In or Register to comment.