Current Question

Discussion in 'General Electronics Chat' started by mbohuntr, Dec 29, 2010.

  1. mbohuntr

    Thread Starter Active Member

    Apr 6, 2009
    413
    32
    I need to bypass the 5 volt regulator on this... http://www.sparkfun.com/datasheets/Sensors/Proximity/SE-10.pdf

    so it will run on 4 AA's. (4.8v). As you can see, the datasheet is useless unless I missed something. One commentor posted it pulls 1.6mA @ 3.3v, so using my SWAG :confused: method, I was going to limit the input current to 20 mA bypassing the regulator with a resistor from VCC to the output of the regulator, and popping off the input leg of the regulator. Any simple and better solutions?? Does the circuit even need a current limiting resistor because the datasheet for a standard 7805 says it will limit its output current to 1A ???
     
  2. thatoneguy

    AAC Fanatic!

    Feb 19, 2009
    6,357
    718
    4 AA batteries give a 6 volt output. Most logic level circuits will work at 5.4V, so many times the supply is 4 AA Batteries with a diode inline from postive to drop the voltage to 5.4V without limiting current.

    It is better to have regulated power, such as using a boost converter for 5VDC regulated from 2 AA batteries (2.5-3V), though those circuits get a little complicate to retrofit in.

    Doing the above will void any warranty, so be sure it works before modification and you know your way around with surface mount components soldering technique.

    The 7805 will only regulate voltages over 7.5V, it then drops out. A LDO (Low Dropout Regulator) wil work to 5.5V input, but can only carry ~100mA-500mA. Neither one regulates/limits the current, too much current and they overheat and fail. The circuit should be designed to not draw much power at 5V to start with. A 500mA fuse in the power line will help against short circuits.
     
  3. marshallf3

    Well-Known Member

    Jul 26, 2010
    2,358
    201
    4 NiCd or NiMH cells will only put out 4.8V and will probably be totally fine - also rechargeable.

    I don't see any current requirement for the board but if you want to use regular alkaline cells remember that they often put out a bit over 1.6V when new = 6.4V = too much by itself and not enough to go through most normal regulators. A simple buck regulator could be built but I think you'd be better off with NiMH batteries which are now capable of putting out nearly 4 AH per AA cell, easily recharged and readily available. The ones you'll find at common stores will vary, shop wisely for good ones.
     
  4. mbohuntr

    Thread Starter Active Member

    Apr 6, 2009
    413
    32
    Thanks for the quick answers, I should have mentioned they will be NiMh batteries. I can't seem to grasp when to use limiting resistors at the beginning of a circuit and when it is not required. I get their use for LED's , but when else are they needed at the beginning of a circuit??
    I will clip 4.8 volts onto the output of the regulator, and test the result before I perform surgery. I don't suppose I can jumper the VCC and outputs together could I ???
     
  5. thatoneguy

    AAC Fanatic!

    Feb 19, 2009
    6,357
    718
    What you want to do is add voltage at the output of the regulator, not any further into the circuit, as you aren't aware of the needs.

    You may want to double check that no devices/traces/parts lead out from the +12V other than directly into the regulator. There should be a capacitor from +12V to ground, but +12V shouldn't feed anything else for your idea to work.

    Some circuits will use both +12 and +5V, such as motor drivers, but that doesn't appear to be the case here, but it is always a good idea to double check.
     
    mbohuntr likes this.
  6. marshallf3

    Well-Known Member

    Jul 26, 2010
    2,358
    201
    Not a good idea to do that until you've removed the regulator from the circuit or at the least put a bypass diode across it such that current won't be tempted to flow backwards through it to the input side cap(s)
     
  7. mbohuntr

    Thread Starter Active Member

    Apr 6, 2009
    413
    32
    Ok, tomorrow I will try to get a look at the backside of the board to try to trace it out more. Thanks.
     
  8. mbohuntr

    Thread Starter Active Member

    Apr 6, 2009
    413
    32
    I got a better look at the board , and it appears VCC only goes to the regulator, the only spot I can't see is the center pin. Is it possible to show me the locations for the bypass diode connections?? I was thinking of lifting the VCC wire, and scratching off a clean spot between the regulator output and the bypass cap. Is this the best way for SMT ?? I don't think my club-like fingers and neanderthal-like reflexes are up to connecting to the regulator output. :D
     
    Last edited: Dec 30, 2010
  9. marshallf3

    Well-Known Member

    Jul 26, 2010
    2,358
    201
    It's shown on most data sheets as a suggested option, basically the cathode goes to the input in and the anode to the output pin.

    http://www.innoengr.com/examples/reverse_regulator_rectifier.jpg

    We sometimes use them as a safety feature. Consider that there might be a short on one of the input caps, (might have said something else earlier) or there's just a big one on the input, this will prevent the current you inject at the +5V point from trying to go backwards through the regulator IC. I drew the schematic up backwards from a conventional layout because I thought it may make more sense in the application you're dealing with.

    It's also a nice safety feature should anything short out your input supply, the residual voltage remaining on the other side of the regulator IC can easily bypass it. Not really that necessary for most modern regulators but you never know. 1N4000 rectifiers are cheap, I keep a variety of them around but rarely use them as if I need a single simple rectifier diode I'm more tendent to use the 1N5400 series since the price isn't that much different. I suppose even a common 1N914 of 1N4148 would be fine unless there's a large amount of charge on the regulated side that wants to feed back into a huge input cap or a short.

    $3.80 will get you 100 of these:
    http://www.mouser.com/ProductDetail...GAEpiMZZMtEwUVCuofpuFOn%2bHIvCcyB6OioqsTJAaU=
    $4.00 will get you a bit better:
    http://www.mouser.com/ProductDetail...=sGAEpiMZZMtvcUztdGSumAHvEdnpxAfbT6y1qQFz/ng=
    or $11.00 will get you 100 of these:
    http://www.mouser.com/ProductDetail...GAEpiMZZMtEwUVCuofpuFOn%2bHIvCcyB6OioqsTJAaU=

    Even better pricing can be found if you look around, sometimes you can find the older 1000PIV 2.5A rectifiers (or something close) for a few pennies each on a surplus website. I generally like to buy the 400V ones as I often use them as snubbers (which do not need to be as high speed as many people think) because the back EMF when the current is cut off from an inductor doesn't occur as quickly as you might believe, just depends on what you've got driving it and unless it's an almost 0 time cutoff the reverse EMF doesn't happen in nanoseconds. Even good MOSFETs take a period of time for them to cut off due to the gate charge that's inherent to most of them.
     
    Last edited: Dec 30, 2010
    mbohuntr likes this.
  10. marshallf3

    Well-Known Member

    Jul 26, 2010
    2,358
    201
    I was thinking of lifting the VCC wire, and scratching off a clean spot between the regulator output and the bypass cap. Is this the best way for SMT ?

    That will work fine, or you can just order a free sample of http://www.chipquikinc.com/ and easily remove the regulator from the circuit in a matter of seconds if you follow the directions. That will leave you with a nice place to solder your new input power wire to the board and then you won't even need the reverse rectifier since you've determined that the 12V isn't necessary.

    Just a warning, once you've tried that ChipQuik solution you'll wonder how you ever did without it. The free sample arrives within days and has plenty to do several mid-sized ICs or a ton of smaller devices. The way it works is the solder part of it is an extremely low melting point alloy that bonds with the existing solder and thus remains liquid for plenty of time in order to lift an entire chip just from the latent heat left in the component. Follow up with a bit of solder wick and some isopropyl alcohol to clean off the flux.

    A note about buying Isopropyl alcohol: Always pay the extra 50 cents to get the true 91% stuff as opposed to the 70% which is already mostly saturated with water. The only thing is that you can't use the 91% on most clear plastics as it will lightly fog it, stick to the 70% if cleaning something you want to see through again.

    Common chemicals you should collect over time should be another thread in itself I think I may start sometime today. I've got an arsenal of stuff but about half of it is not for most people to keep around such as 70% Nitric Acid just to name one.
     
    mbohuntr likes this.
  11. mbohuntr

    Thread Starter Active Member

    Apr 6, 2009
    413
    32
    Surgery was successful. I read a couple of articles on SMT's, and opted to heat up the pads enough to lift the regulator. I should have done this BEFORE two cups of coffee...:rolleyes: Thanks to both of you!!!! The unit operates as it should at 4.5v. I certainly learned a lot here!!! Marshallf3, I couldn't order the soldering solution as I wasn't sure what to call my business;)
     
  12. marshallf3

    Well-Known Member

    Jul 26, 2010
    2,358
    201
    Glad it worked, I think we all had a feeling that it would.

    Call your business anything you like so long as it's different from anything common. Mine is called Innovative Engineering - has been for decades - but there's another firm out there that took the name not long ago. When I got my Federal & State tax ID stuff I had to register it as Innovative Engineering Concepts

    Anything halfway creative, i.e. not something like "Mike's Garage" will work, it also helps to have an e-mail address from one of the less than obvious free ones like @yahoo or @hotmail
     
Loading...