Shop OBEX P1 Docs P2 Docs Learn Events
24//7 backup battery? — Parallax Forums

24//7 backup battery?

LoopyBytelooseLoopyByteloose Posts: 12,537
edited 2010-08-01 06:16 in Propeller 1
Hi all,
I've got a Propeller Proto Board here, and I also have a tiny 4.8volt, 700maH Ni-Mh battery. This are handy little packs that were created as replacements for cordless telephones.

I am thinking of just inserting the battery (fully charged) in between the 5 volt and 3.3 volt regulators as a 24/7 backup battery. I have done some reading, but I can't clearly find out if this is wise or not. Of course there are issues.
1. Can a back flow of voltage in the 5 volt regulator damage it. And if so, would putting a by-pass diode on it resolve my concerns.
2. Will the battery over-charge in a stand-by status at 5 Volts. I've read that Ni-Cd (not Ni-Mh will continuously trickle charge on up to 6.4V when current is limited to 0.095ma). I suppose I could insert a current limiting resistor into the scheme and or a blocking diode.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?

aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan

Comments

  • smbakersmbaker Posts: 164
    edited 2010-07-31 17:01
    Loopy Byteloose said...

    1. Can a back flow of voltage in the 5 volt regulator damage it. And if so, would putting a by-pass diode on it resolve my concerns.

    From the LM1086 datasheet:
    datasheet said...
    When an output capacitor is connected to a regulator and
    the input is shorted, the output capacitor will discharge into
    the output of the regulator. The discharge current depends
    on the value of the capacitor, the output voltage of the
    regulator, and rate of decrease of VIN. In the LM1086 regulator,
    the internal diode between the output and input pins
    can withstand microsecond surge currents of 10A to 20A.
    With an extremely large output capacitor (≥1000 μf), and
    with input instantaneously shorted to ground, the regulator
    could be damaged. In this case, an external diode is recommended
    between the output and input pins to protect the
    regulator, shown in Figure 5

    Figure 5 shows a bypass diode in the expected place, from output back to input. It recommends an IN4002.

    One should note that this paragraph starts with "if the input is shorted...". If that were to happen for any period of time, then your diode would not last long. If we assume that you don't plan on shorting the input any time soon, do we need to worry about this? I'd guess not, but others may know for sure.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2010-07-31 17:14
    Well, it would be easy to just solder a by-pass diode between the Vin and Vout on the 5 Volt - LM1086. I guess 1amp would be enough.
    I am not sure how to compare 700mah battery to a 1000uf cap. I suspect the cap would create more intense momentary current, but the battary would not let up.

    And I am talking about an OFF condition, not a short. The 78xx series had similar issues with output capacitors over 10uf.

    And that is just Item #1.

    I still have item #2, overcharging and the control there of or the failure to get a successful backup.

    I would just do a bench test, but I doubt that I can get some LM1086 regulars to try the with. And I'd rather not damage the ones on board.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Ain't gadetry a wonderful thing?

    aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
  • Mike GreenMike Green Posts: 23,101
    edited 2010-07-31 17:17
    You've got the right idea, but the details will get you ...

    The 4.8V is a nominal output voltage. The voltage will rise as the battery charges and can reach over 5V. Since the regulator produces 5V, you won't be able to charge the battery fully. You need more voltage, something like 6.5V to 7V. With that, you will be able to use an LM317 regulator as a charge current controller.

    The charge current has to be limited to 70mA. Most NiMH batteries will do fine with a 1/10C (capacity) charging current indefinitely.

    Perhaps what you can do is use an LM317 as a current regulator between Vin and the battery, then a Shottky diode between the battery and the 3.3V regulator input. You'll need another Shottky diode between the 5V regulated voltage bus and the 3.3V regulator input (and battery output Shottky) to prevent damage to 5V devices if the battery voltage rises too high as it's charged.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2010-07-31 17:39
    Thanks Mike,
    This is much of what I expected. Essentially it seems that a backup battery needs to be placed 'over the top' of all the regulators the provide power to critical components because the charge voltage is more than the micro-controller can handle (or a mouse and keyboard in this case). The Ni-Cd wants 6.4v, but I do see that the Ni-Mh may want only 5.6v. So it appears at best I am left with an undercharged battery pack.

    So you think that the battery should just go in front of the +5 regulator and trust the 'low-drop out' feature to use what the 4.8v battery has to offer. (or did I miss something in all those Scottky diodes?) I was thinking I might just seek out a different high voltage battery pack.

    I wonder if limiting current by a mere resistor will work as well. I know that a LM317 would be more precise, but I suspect that insertion of it causes a 0.7 voltage drop. I think I can bench test what the battery wants with a 7805 and see what happens. A by-pass diode on the LM1086 seems to be absolutely wise as the board has a 1000uf cap on it I have to look more closely at the schematic.

    I need to think more about exactly what I am hoping and what I want to do.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Ain't gadetry a wonderful thing?

    aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan

    Post Edited (Loopy Byteloose) : 7/31/2010 6:02:02 PM GMT
  • Peter KG6LSEPeter KG6LSE Posts: 1,383
    edited 2010-07-31 18:02
    I was on the phone with Saft about some Batts I orderd for a project last week .( NiMh pack for a Mobile Ham Radio.)

    We were talking shop about batt chargers for simple apps like this . I wanted to use a COTS wall wart .. the tech was telling me that C/20 is the MAX float charge rate for NiMh cells Under 4AH . .And C/10 rule of thumb is for NiCd only ..

    Loopy these little honeys are VERY good at doing NDO work .
    http://www.dimensionengineering.com/AnyVoltMicro.htm

    """The output voltage is set with a small screw potentiometer on the side of the AnyVolt Micro. Once the output voltage is set, it does not matter whether input voltage is higher, lower, or the same as the desired output i.e. there is no dropout voltage."""""


    Peter KG6LSE

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "Carpe Ducktum" "seize the tape!!"
    peterthethinker.com/tesla/Venom/Venom.html
    Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway. —Tanenbaum, Andrew S.
    LOL
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2010-07-31 18:24
    Well, I hadn't properly done all my reading.

    First, I now understand that the 1000uf cap is not usually involved as it is there for powering servo motors and only becomes an issue if the servos are powered at +5 regulated. In that case, a fly-back diode is very wise.

    Second - and most importantly - the LM1086 documentation on page 12 actually has a circuit for a backup battery to charge at 6.5 Volts. It requires a second LM1086-5 to operate in parallel, some modification to the original to elevate its output to +5.2 (still in spec for the +5.5 limit) by inserting a 50 ohm resistor.

    In sum, this is definitely doable with the little 4.8V 700mah Ni-Hd that I have charging at 6.5 volts. (It is more about keeping the size to a minimum than about using what I have. I might even just use 4 AAA Ni-Md in a conventional carrier.)

    And Peter KG6LSE, yes I agree C/20 is better than C/10 charge rate. C/10 is generally considered a 'slow charge rate' whereas C/20 is more accepted as a 'float charge rate'. And thank you to all the HAM radio websites, where they take battery recharging very seriously.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Ain't gadetry a wonderful thing?

    aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
  • Peter KG6LSEPeter KG6LSE Posts: 1,383
    edited 2010-07-31 18:57
    You can Lift a 7805 with a few diodes in series with the Ground lead on the Reg . and unlike a resistor the diode is a tad more stable .

    I used a 7805 lifted with 5 diodes to charge the 7.2V pack on DORTHY robot and it works well .

    Peter KG6LSE

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "Carpe Ducktum" "seize the tape!!"
    peterthethinker.com/tesla/Venom/Venom.html
    Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway. —Tanenbaum, Andrew S.
    LOL
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2010-08-01 06:16
    Thanks, I was looking at the schematic and in occurred to me that I could leave the Propeller Proto Board intact if I made the other regular the elevated voltage one and then connected its output to the +5 side of the existing regulator. The backup battery would connect through the existing barrel plug and go through the switch circuitry.

    That led me to wondering whether I needed to match the existing regulator or whether I could use a 7805. You seem to say the 7805 works fine in this context. After all, the back up battery needs to rely more on the low dropout feature than a wall wart. But I do see that Parallax chose a 1.5amp regulator as the combined demands of a a mouses, a keyboard, and the Propeller may exceed 1amp as a steady state demand. So I am pondering what I care to do about choosing the right regulator.

    I will consider using diodes instead - even it I use merely one Scottky to go up to 5.3 volts. Why so?, I want to be able to safely power a keyboard and mouse which accept +5.5 volts max.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Ain't gadetry a wonderful thing?

    aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Sign In or Register to comment.