The Fenix-Store
Results 1 to 13 of 13

Thread: serial vs. parallel wiring for LED's

  1. #1

    Default serial vs. parallel wiring for LED\'s

    I have been wanting to hook up 2 LED's in my Mag Lite Solitaire (and eventually 3 LED's in a Mini Mag) and was wondering...

    I know there are 2 different ways of doing the wiring: serial & parallel.

    Now, if you hook up 2 LED's in serial, won't each of the LED's be individually dimmer than a single LED?

    If you took the same config and hooked them up in parallel, wouldn't each LED draw the maximum power from the battery thus making each LED brighter (at the expense of a shorter battery life)? Isn't this a better/brighter method?

    Newbie answers/photos/wiring diagrams/explanations are greatly appreciated! [img]images/icons/grin.gif[/img]

    Question 2:
    How do you figure what resistance is needed in each (serial & parallel) wiring method?

    Example #1: Let's assume I have a MN21/23 12v battery and 2 orange (2.2v) LED's in serial.

    Example #2: Let's assume I have 1 AAA and 2 N batteries for 4.5v and 2 violet (3.6v) LED's in parallel.

    (Other examples would be appreciated...) [img]images/icons/grin.gif[/img]

  2. #2
    Mostly MIA Administrator

    Join Date
    Jun 2002
    Location
    Utah
    Posts
    7,448

    Default Re: serial vs. parallel wiring for LED\'s

    I'll try to work my way through this but my advice will undoubtedly be slightly off and will need to be corrected as I just bearly learned all this stuff myself.

    I am not sure HOW to get these numbers but you have to go...and this has been converted for easy calculator entry in an algebraic, RPN users can use it too though [img]images/icons/grin.gif[/img]

    Rs=(V-Vled)/(Iled)

    Rs = Resistance Needed
    V = Supply Voltage
    Vled = LED Volatage
    Iled = Led current

    So...on to the examples...

    If I am correct (I am probably not 100% correct) you should be able to drive 1 LED or 100 with only the voltage required...in example one that would be 2.2V but I do now know the LED current.

    What I need is how many mAh you want to run it at. You have to determine that. Lets say you want to run one orange led (2.2V) from a 3 Volt supplyat 30 mAh... you go

    Rs=(3-2.2)/(0.03) to get

    26.67 or aprox 27 Ohms


    Am I even close to right? I need to figure out how to do this all too so...

  3. #3

    Default Re: serial vs. parallel wiring for LED\'s

    <BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by ChrisM:
    I have been wanting to hook up 2 LED's in my Mag Lite Solitaire (and eventually 3 LED's in a Mini Mag) and was wondering...

    I know there are 2 different ways of doing the wiring: serial & parallel.

    Now, if you hook up 2 LED's in serial, won't each of the LED's be individually dimmer than a single LED?

    If you took the same config and hooked them up in parallel, wouldn't each LED draw the maximum power from the battery thus making each LED brighter (at the expense of a shorter battery life)? Isn't this a better/brighter method?
    <HR></BLOCKQUOTE>

    In general, if you have the choice and means to do serial instead of parallel, that's what you should do. I've heard a number of reasons for doing this, but the one that's stuck with me is that different LEDs have different forward voltages -- and putting two in parallel will invariably cause the the LED with the lower forward voltage to draw more current than the other. LEDs vary quite a bit from unit to unit, so this happens quite often (this is why manufacturers have been known to "hand-pick" LEDs with known forward voltages for their specific requirements) So you end up with less-predictable power consumption and brightness.

    Putting the LEDs in series equalizes the current draw, but requires you to provide sufficient voltage for both. If you're working with ordinary 1.5V alkaline cells, this may be different to balance against size/weight requirements. That's why DC-DC boost converter circuits are used -- to cut down on the required voltage from your batteries. Some boost converters are "regulated", meaning they attempt to provide either a constant voltage or current to the LED regardless of battery voltage. Each has its own advantages and drawbacks to be taken into consideration.

    I'm a bit of a newbie as well, so someone feel free to correct me if I left something out. [img]images/icons/confused.gif[/img]

    -XCN-

  4. #4
    Mostly MIA Administrator

    Join Date
    Jun 2002
    Location
    Utah
    Posts
    7,448

    Default Re: serial vs. parallel wiring for LED\'s

    So if you do them in series and you have 2 3V leds (This is hypothetical people) then you need 6V to drive the 2 LEDs? That doesn't seem right but maybe it is
    [img]images/icons/confused.gif[/img]

  5. #5

    Default Re: serial vs. parallel wiring for LED\'s

    [QUOTE]Originally posted by Saaby:
    I'll try to work my way through this but my advice will undoubtedly be slightly off and will need to be corrected as I just bearly learned all this stuff myself.[QUOTE]

    I think you're pretty much correct, except for your use of "mAh", which is short for "milliamp hour". That's a unit used to describe battery capacity, in terms of the number of total milliamps you can draw over a specific period of time (given fixed voltage, as you theoretically have in a battery).

    Drop the hour to get "mA", and you have the correct units for those equations. [img]images/icons/smile.gif[/img]

    I've heard varying things about how much total voltage to supply to an LED circuit with a shunt resistor (something along the lines that LEDs should consume no more than around 80% of total supply), but I've never heard a straight explanation for why one should bother. (Obviously, the shunt resistor serves other functions, such as preventing overvoltage/overcurrent to the LED.) If anyone has a clear, straightforward explanation for that figure, please post it here.

    Links referring to the 80% figure:
    http://misty.com/people/don/ledd.html http://www.plasma-ireland.com/lp/applications.html

    -XCN-

  6. #6

    Default Re: serial vs. parallel wiring for LED\'s

    <BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Saaby:
    So if you do them in series and you have 2 3V leds (This is hypothetical people) then you need 6V to drive the 2 LEDs? That doesn't seem right but maybe it is
    [img]images/icons/confused.gif[/img]
    <HR></BLOCKQUOTE>

    Yes. This is a function of Kirchhoff's Voltage and Current Laws (KVL and KCL, respectively):

    KCL (or the 1st Law): for a given junction or node in a circuit, the sum of the currents entering equals the sum of the currents leaving.

    KVL (or the 2nd Law): around any closed loop in a circuit, the sum of the potential differences across all elements is zero.

    Link: http://theory.uwinnipeg.ca/physics/curr/node8.html

    Those laws form the basis of circuit analysis. If you're not familiar with any of this, you should pick up a basic electronics book and work through it.

    In a nutshell, KVL tells us that 2 LEDs placed in series will require a voltage source equal to the sum of their individual voltage requirements. KCL tells us that both LEDs will draw the same amount of current, since we're just working with a single closed loop.

    If the LEDs are in parallel and have the same forward drop, then the voltage will be the same in each branch -- but so will the current consumption. KCL tells us that, when the branches come back together to return to the battery, the currents in each branch add together, so you end up drawing more current in total (twice as much in this example).

    -XCN-

  7. #7
    *Flashaholic* James S's Avatar
    Join Date
    Aug 2002
    Location
    on an island surrounded by reality
    Posts
    5,081

    Default Re: serial vs. parallel wiring for LED\'s

    This sums up exactly what I was looking for, I think. If I'm running 2 20ma 3.6v LED's in series i should give 20ma at 7.2 volts. And in parallel I should give them 40ma at 3.6v correct?

    I wired up 5 of them in parallel on my breadboard and thought to run them off 3 AAA NiMh, which at full charge should give me exactly the correct voltage. When measuring the current through my meter though I'm only seeing about 55ma, shouldn't I be seeing 100ma? Do these batteries just not provide that much power? This measurement was taken without any current limiting resister or other componant involved. I think I will switch to a NiMH 9volt so that I can run pairs in series, but I'm just trying to understand what is supposed to be happening before I solder anything together.

    Thanks,
    James

  8. #8

    Default Re: serial vs. parallel wiring for LED\'s

    Yes, you have distilled the answer correctly.

    However, LED's are special. When you put two of them in parallel, and measure 40ma going through them at 3.6v, what may be happening is that 35 ma is going through one of them, and 5 ma is going through the other one - leading to one very bright, and one very dim LED. If yours appear about the same brightness, then you have gotten lucky and have reasonably well matched LEDs. This is why LED's are seldom hooked up in parallel.

    Now, you said you were using NiMH cells. At full charge, they will be about 1.4V apiece, very rapidly dropping down to about 1.2V. 3 of these gives you 3.6V, when no current is being drawn (when you hook them up, the current draw will drop the voltage a little bit). Depending on the characteristics of the particular LEDs you have, this will drive anywhere from 10 to 50 ma through each LED. I think this is what you're seeing.

    For your amusement, here are some numbers I measured for LED Current versus Voltage on a bright-white 5mm LED I have in my possession:

    2.9V 0ma
    3.0V 3ma
    3.2V 8ma
    3.4V 17ma
    3.6V 30ma
    3.8V 48ma
    3.9V 60ma

    Notice that this a lower-voltage than normal LED - white LEDs are often shown with typical voltages of 3.65V at 20 ma. For this LED, 3.5V is probably enough to drive 20 ma through it. Five of these in your circuit would probably draw much higher current than the five that you have.

    You can see that a very small change in voltage (from, say, 3.4V to 3.6V) can nearly double the current flow. LEDs from different manufacturing lots will have similar kinds of current changes with voltage; an LED from a different lot may not draw 30 ma until 3.7V, or may do it at 3.5V.

    Good luck with your experiments.

    /frank

  9. #9
    Flashaholic*
    Join Date
    Aug 2002
    Location
    Albany, NY
    Posts
    769

    Default Re: serial vs. parallel wiring for LED\'s

    Nice posts... (no sarcasm!)

    The other thing to keep in mind, besides the good information about the LED's, is that a battery isn't a constant voltage source. There's an open circuit voltage, and an effective series resistance that causes the output voltage to be lower under load. Plus, alkaline batteries have a voltage drop that is significant as soon as you use the battery. Smaller batteries have higher series resistance. LED lights without regulators take the battery series resistance into consideration when figuring the LED current. (Hence a recent post about using a small 12volt battery to drive an LED. The high series resistance of the bulb was almost all that was needed to keep the LED current acceptable.)

  10. #10
    Flashaholic*
    Join Date
    Aug 2002
    Location
    Albany, NY
    Posts
    769

    Default Re: serial vs. parallel wiring for LED\'s

    Originally posted by Albany Tom:
    Nice posts... (no sarcasm!)

    The other thing to keep in mind, besides the good information about the LED's, is that a battery isn't a constant voltage source. There's an open circuit voltage, and an effective series resistance that causes the output voltage to be lower under load. Plus, alkaline batteries have a voltage drop that is significant as soon as you use the battery. Smaller batteries have higher series resistance. LED lights without regulators take the battery series resistance into consideration when figuring the LED current. (Hence a recent post about using a small 12volt battery to drive an LED. The high series resistance of the bulb was almost all that was needed to keep the LED current acceptable.)
    <font size="2" face="Verdana, Arial">....er "series resistance of the BATTERY"

    I shouldn't post at 2am. I really shouldn't.

  11. #11
    Flashaholic* MrAl's Avatar
    Join Date
    Sep 2001
    Location
    New Jersey
    Posts
    3,071

    Default Re: serial vs. parallel wiring for LED\'s

    Hello there,

    Frank S.:
    Nice data! That gives a very clear picture of
    the typical Nichia white LED.

    For the original question by ChrisM:
    Since you intend to use a Mini Mag, you have
    to decide what batteries you will be using.
    The simplest mod is to use three 'N' size cells
    in series and make up a conductive spacer to
    take up the extra room, usually inserted last
    after the three cells have been inserted.
    The conductive spacer has the resistor you intend
    to use inside of it.
    I did this exact mod on a Minimag but using only
    one LED. You can lightly file the LED leads
    to fit into the bulb socket in the head of the
    Minimag.
    If you intend to use more then one LED, then
    you will have to wire them in parallel because
    there wont be enough voltage to drive two or more
    in series unless you choose another type of
    battery. Problem is, runtime will be greatly
    affected by your choice of batterys. The longest
    run time seems to come from the N size cells,
    but there may be other ideas here?

    To make sure all three LEDs in parallel get
    the same current, you may check the brightness
    as someone else pointed out. If any one seems
    too dim, replace it with another and test again.

    You will have to modify the head piece a bit more
    then with only one LED though. For one LED,
    you only have to drill or file out the hole
    a little bigger, and everything works the same
    as the original Minimag. For three led's, i guess
    you will have to come up with your own mounting
    strategy, or use a tail cap switch and glue everything
    in place.

    Let us know how you make out with it.

    Good luck with it,
    Al

  12. #12
    Flashaholic
    Join Date
    Aug 2002
    Location
    WI
    Posts
    206

    Default Re: serial vs. parallel wiring for LED\'s

    Something I like to do when parallel connecting LEDs is to place a series resistor on each LED if possible. This will allow each LED to drop it's own forward voltage. It also allows for smaller wattage resistors to be used. The amount of room may not allow for this and than the before mentioned matching of LEDs is critical.

  13. #13
    Flashaholic*
    Join Date
    Dec 2001
    Location
    Portland, OR
    Posts
    565

    Default Re: serial vs. parallel wiring for LED\'s

    Originally posted by Xcandescent:


    I've heard varying things about how much total voltage to supply to an LED circuit with a shunt resistor (something along the lines that LEDs should consume no more than around 80% of total supply), but I've never heard a straight explanation for why one should bother. (Obviously, the shunt resistor serves other functions, such as preventing overvoltage/overcurrent to the LED.) If anyone has a clear, straightforward explanation for that figure, please post it here.

    -XCN-
    <font size="2" face="Verdana, Arial">The 80% figure comes from a desire to reduce the sensitivity of the circuit to small changes in component values or supply voltage. If you could absolutely guarantee that the supply voltage was exactly the LED voltage, then you wouldn't need any resistance at all. But in real systems there are variations in the supply voltage, variations in LED forward voltage, etc.

    The greater the fraction of the load which is resistance, the less the variation in load current as voltage, (or LED forward voltage), changes. However the greater the fraction which is resistance, the greater the losses and the less efficient. Thus the 80% compromise

    -Jon
    Since a small change in the voltage applied to an LED will lead to a large change in current, the greater the fraction of the circuit which is LEDs, the more sensitive the circuit is to sm

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •