Dropping down one Volt

Discussion in 'Power Electronics' started by AgentSmithers, Aug 19, 2016.

  1. AgentSmithers

    Thread Starter Active Member

    Jan 14, 2011
    52
    0
    Hi Everyone!!
    Alright quick question. I have a 6V battery and a uC that can only handle 5.5 volts max AKA ATTiny2313.
    Is the best way to drop the volt down is by using a diode to eat the 0.7V and bring it to a save value or just use a resistor. The diode would at least provide for reverse terminal protection if connected backwards..
    What is more efficient?

    http://www.mouser.com/ds/2/389/CD00000444-249828.pdf
    I was going to use this because I had one on my work bench but I saw the Dropout was set to 2 volts bringing it down to 4V and I need 5 for my LCD screen :-\.
    Thanks!!
    Sorry for the lame question :p I would use a regulator if it was any more :p
     
  2. Bordodynov

    Active Member

    May 20, 2015
    637
    188
    Use the stabilizer on 5 volts. There is much miscellaneouses stabilizer with small fall of the voltage (LowDrop).
     
  3. hp1729

    Well-Known Member

    Nov 23, 2015
    1,935
    218
    How much current are you taking about? Only mA? Then a diode works. More current, less than 100 mA? LM78L05. Up to 1 Amp LM7805.
    Don't fret that "5.5 V" limit so much. The chips will run okay at 6 Volts. You just can't rely on the same specs for performance. Play with them. If you run them on 6 Volts, run them all on the same 6 Volts. What you want to avoid is driving from a 6 V system to a 5 V system.

    I have ran TTL chips at 12 V briefly (by accident) with no harm, but I wouldn't recommend it.
     
  4. DickCappels

    Moderator

    Aug 21, 2008
    2,646
    631
    Two diodes might be ok. In general, it is not a good idea to run semiconductors at the maximum ratings. Be careful of charge current into bypass capacitors damaging the diodes.

    If this is going to power an ATTIN2313, then you are looking at a load of less than 10 ma, so efficiency is probably not a concern.
     
  5. #12

    Expert

    Nov 30, 2010
    16,247
    6,739
    This datasheet shows that a 1N4000 series diode will use up 0.6 volts at 10 ma., less if the load is less. That's why you can get away with using 2 diodes to get rid of 1.2 volts like DickCappels said to do. You will get 4.8 volts at 10 ma and more voltage if the load is less than 10 ma. If the load is higher, just use one diode. They can survive a 30 amp start surge, so that seems safe to me. Measure the actual voltage when you hook this up to be sure 10 ma is the right range of current for your ATTiny.

    Those regulators you named use up about 1.3 volts minimum at 10 ma, so a 5 volt regulator won't be regulating at 5 volts. They will knock 6 volts down to 4.7 with a fresh battery, and the won't be regulating, they will be starving. That's why Bordodynov was talking about low drop out regulators.

    If you want to be precise, go with Bordodynov. If you can get in the right range with diodes and the fact that battery voltages decline as they get used up, the diodes might work.

    Here are some LDO regulators:
    http://www.mouser.com/Semiconductor...0srsdZ1z0w7feZ1z0wa4oZ1yx5k7vZ1z0shyuZ1yoii43
     
  6. RichardO

    Well-Known Member

    May 4, 2013
    1,228
    382
    Keep in mind that there is no such thing as a 6 volt battery. :(

    As soon as you connect a load and use the battery for a while its voltage will drop. How quickly the voltage drops depends a huge amount on the battery and the load.
     
  7. ian field

    Distinguished Member

    Oct 27, 2012
    4,412
    782
    The de-luxe solution is to use a buck-boost converter. It drops the voltage down when the battery is fresh and switches to boost mode as the battery comes down. The design isn't trivial, but the likes of Maxim & LT etc do off the shelf chips that do most of the hard bit.

    If its an ultra low power micro, you can get away with a diode Vf - but you have to watch out for "nominal" voltage ratings, if you charge a 6V lead-acid battery in situ the terminal voltage can go up to 7.5V. It'll be nearly 7V fresh off charge.

    Using a series pair of lithium cells for 7.2V would require only a simpler buck converter - or a 7805 if the current draw is so low that wasted power is irrelevant.

    The downside of lithium is the critical terminal voltage and ensuring cell balancing.
     
  8. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,386
    1,605
    I am no big fan of using a diode as the voltage it drops (and hence your supply voltage) can and will vary with load and thus is both unstable and noisy. Adding caps can help with the noise, but a low voltage dropout (LDO) regulator is a better choice.

    *Which* one I recommend depends on the total current draw, and if you can use SMD or just thru hole.
     
  9. ian field

    Distinguished Member

    Oct 27, 2012
    4,412
    782
    A diode shouldn't be too noisy for the decoupling caps the TS should be using anyway. Its always an option to search datasheets for diodes with the sharpest knee curve, that will at least limit the damage of Vf varying with current. In any event - a little "give" is beneficial. With such a small margin, you actually want the diode to drop more when the battery is fresh.

    A typical LDO regulator should work with input/output differential as low as 0.5V, so far; I haven't found any mention of battery chemistry - I'm thinking the flat discharge curve of nickel chemistry might be best as the TS seems to be cutting it a bit fine.

    A hurdle for the switchmode (buck/boost) route; is quiescent current draw. If the TS is doing the coding, a spare port pin can be used for power management. A series pass PNP (non regulating) transistor feeds Vdd and is activated by a pull down pushbutton, once started - the MCU pulls the port pin low which takes the place of the pushbutton. Inside the MCU; an activity monitor detects when nothing happened for a while and releases the port pin.
     
  10. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,386
    1,605
    A diode solution is at best a kludge which may work for a bit. It is in no sense a proven dependable solution worthy of a production unit. A cap can at best reduce the noise intrinsic in this scheme that is due to fast switched currents. If I was to use this on a breadboard it would only be there until better parts arrive.

    An LDO is a better part. For low currents you can find them with dropouts as low as 0.03 volts, they are quiet and dependable. And fairy inexpensive to boot. Plus many have an enable pin.

    If the battery starts to go a LDO will keep on outputting whatever voltage it can, riding down with the battery.

    Using a switcher to boost up a dead battery may extend your runtime for a short time, but once a battery starts to drop off it is no longer reliable.
     
    GopherT likes this.
  11. ian field

    Distinguished Member

    Oct 27, 2012
    4,412
    782
    I guess it all depends on what you call it - loads of manufacturers put series reverse polarity protection diodes in production units. At some point; the Vf has to be included in the design calculations. Maybe they moved something else around to accommodate the Vf - or maybe the Vf killed two birds with one stone.....................
     
  12. ScottWang

    Moderator

    Aug 23, 2012
    4,848
    767
    The Dropout = 2V that it was when the current = 1A, so normally we will treat the Dropout as 2.5V.

    ATtiny2313 datasheet --
    Operating Voltages
    – 2.7 – 5.5V (ATtiny2313)
    Speed Grades
    – ATtiny2313: 0 – 10 MHz @ 2.7 - 5.5V, 0 – 20 MHz @ 4.5 – 5.5
     
  13. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,386
    1,605
    Extraordinary claims require extraordinary evidence, or as they say on Wikipedia "citation needed."
     
  14. ian field

    Distinguished Member

    Oct 27, 2012
    4,412
    782
    "Extraordinary claim"?!!!!! its fairly common practice on items that people like you might stick the battery on the wrong way round.

    If you're [Moderator's note: mild insult removed -please play nice] to see what's out there on the market right now - I'm not going to bother leading you by the hand............
     
    Last edited by a moderator: Aug 24, 2016
  15. AgentSmithers

    Thread Starter Active Member

    Jan 14, 2011
    52
    0
    Thank you to everyone that chimed in. I appreciate it! :)
     
  16. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,386
    1,605
    Not a single solitary example. I thought so.

    (Pssst... next time try a FET for low voltage reverse bias protection. Wastes near zero power, an important consideration when using batteries.)
     
    Last edited by a moderator: Aug 24, 2016
  17. ian field

    Distinguished Member

    Oct 27, 2012
    4,412
    782
     
Loading...