Results 1 to 10 of 10
  1. #1
    mtbr member
    Reputation:
    Join Date
    Aug 2008
    Posts
    7

    Idea! shiningbeam p7 driver - 2 LEDs with 14.8V?

    Hi,

    I am going to build my 2nd DIY bike light using 14.8V battery pack, 2x SSC P7 and 2x 3-mode drivers by ShiningBeam:
    3-Mode Regulated Circuit Board for XML, MC-E, SST-50 and SSC P7 (2.8A Max.)

    Taskled bFlex which I use in the first light allows me to connect 14.8V battery directly but max voltage input for the ShiningBeam driver is 6V

    Is it possible to make a proper wiring of this set while keeping the battery at 14.8V (so I can charge and balance it easily)?

    Thank you and sorry for this noob topic (I am going to learn more about electronics soon)!

  2. #2
    mtbr member
    Reputation:
    Join Date
    Oct 2008
    Posts
    1,297
    Not really. It is possible to use a poormans circuit to use higher voltage batteries with these drivers, but that circuit would require four leds.

  3. #3
    mtbr member
    Reputation:
    Join Date
    Aug 2008
    Posts
    7
    So I have to convert it to 2x 1S2P packs and feed it individually? That would be quite complicated to charge and balance :-/
    EDIT: OK, I have found out there is no need to balance parallely connected cells, but can I connect 2 leds to one driver? Or how to deal with it?

    Thanks!
    Last edited by niverin; 03-21-2013 at 04:06 PM.

  4. #4
    mtbr member
    Reputation:
    Join Date
    Oct 2008
    Posts
    1,297

  5. #5
    mtbr member
    Reputation:
    Join Date
    Aug 2008
    Posts
    7
    Oh, thanks! So is it ok to connect 2 x 4.2V (when charged) to the driver even there is 6V max input? Or am I missing something? LEDs will drop the voltage on regulator?

  6. #6
    mtbr member
    Reputation:
    Join Date
    Oct 2008
    Posts
    1,297
    putting one led in driver circuit and other in feed circuit drops voltage the 7135 chip sees, so all will be good with the battery voltage coming in at 8.4V.

  7. #7
    mtbr member
    Reputation:
    Join Date
    Dec 2012
    Posts
    16
    Hi there,
    I understand from your dilemma that you need a little bit of help understanding how these drivers work. Most drivers are inductive drivers, that use an inductor as a storing element for energy.
    1) Buck drivers. It lowers the voltage from input to output. This type of driver uses an inductor as a storing element of excess energy in the active phase (ON) of energy transfer (when current flows from the battery to the load). This inductor takes the extra voltage from battery to the load, storing energy L*(I^2)/2 that is recycled and transferred to the load during the so called OFF phase, when the main switch is OFF and there is no flow of current from the battery to the load. These drivers have high efficiency, in the range of 85-95%.
    2) Boost, buck-boost topology. This uses an inductor as a storing element for the energy in the active phase, then it delivers this energy to the load by self-induction phenomenon at the required voltage/current to the load. It can increase or decrease the voltage from the input to output. These work best in the case where the battery has a voltage range above and then bellow the load voltage during a discharge cycle.
    3) Simple linear regulators that work in current limitation mode. These are conducting all the current full time, taking as drop all the excess voltage from the battery to the load.

    That driver you intend to use is a pure linear driver. It behaves as a linear regulator in current limitation, or a current source. There is no inductive element to help regulate the energy transfer by recycling energy with minimum loss, the efficiency is low. All the voltage above the LEDs drop (3 to 3.4V for one led) is dissipated on the resistance of the current controlled mosfets in the circuit. Worst case calculation shows that from 8.4V (two elements fully charged) the driver sees in this connection recommended by ODTEXAS only 8.4-3=5.4V max. It's ok, but it has to take as drop the rest of the voltage down to the other led, that is 5.4-3=2.4V max. So the driver will take maximum 2.4V drop on its mosfets, and the power is voltage multiplied by current. Once the current increases, the drop on the leds increases and the drop on the driver decreases, but the power dissipation on the driver has a max somewhere. Make a simple calculation: for 6.8V on the leds at max current, power dissipation on the driver is (8.4-6.8)V*2.8A=1.6V*2.8A=4.48W, while total power drawn from battery is 8.4V*2.8A=23.52W. So, usefull power is 6.8*2.8=19W on the leds, while total power is 23.52W. This is 81% efficiency, while a switching inductive regulator can reach easily 86-90% efficiency.

    The bFlex is an inductive switching regulator that only touches the power transferred from the battery pack to the load (LEDs). It's a buck, lowers the voltage delivered to the leds from the battery down to the level required by 1, 2 or even 3 leds in series.
    Last edited by uiflorin; 03-24-2013 at 08:09 AM. Reason: Forgot calculation

  8. #8
    mtbr member
    Reputation:
    Join Date
    Aug 2008
    Posts
    7
    Thanks to all for suggestions and explanations! I really appreciate that huge amount of help!
    Uiflorin, many thanks for your detailed post. But please what do you suggest as the best solution for driving 2x SSC P4 when have 2x those shiningbeam drivers and want to change mode for both LEDs by one switch?
    I have individual 18650s and holders with all variants so I can make any battery configuration.

    Thank you!

  9. #9
    mtbr member
    Reputation:
    Join Date
    Dec 2012
    Posts
    16
    Hi Niverin,
    I can give some general advice with the options that you have, so you can choose whatever suits you best, or whatever you can afford to spend. I worked in the past in this field of switching power supplies, this is "my turf" sort of speaking. It's a long reply, with all the details you need to understand the limitations of each solution. Maybe I will post this in a separated thread as a cook book for DIY enthusiasts.

    You want to drive both LEDs simultaneously with the same driver or separate drivers, but both on the same level of light set by one switch. Logic. OK, now that this is set, we have something like:

    Driving both LEDs simultaneously can be done connecting them in parallel or in series. The drivers can be buck, boost or buck-boost. Usually what you find as driver on DX.com is buck. Some of them, depending on the circuits used, can be modified to work as boost or buck-boost by simple changes on the board and maybe the coil replaced. Hard to do that if you don't know electrical design. What are the choices left? Use what can be found on the market as it is, with the best results.

    1) Parallel connection with buck driver. Requires a current capability equal to the sum of currents for each led. For maximum current you need 6A, this is a little high and the dissipation on the wires is high, so thick wires. This is something that needs a good driver. If you find one to deliver 6A, go ahead. The voltage needed from the battery is twice the forward voltage of a led, which is max Vf=3.4V, so you can use a 2S battery. No way to use a 4.2V battery (regardless of number of elements in parallel), not enough voltage. Another downside of parallel connection is the current balance between the leds. Two or more leds in parallel will share the current based on the resistance of the wires, matching of Vf/If characteristics, thermal resistance of each LED, etc. In the end is hard to make them have the same current, always one LED will take a little more, one less. Some low value resistor like 30-100 miliohm in series with each LED will help the sharing, or just the wires can play this role if they are trimmed to the same length and their resistance adds 100-300mV drop for equalization of Vf/If characteristics of the leds, but some power is wasted. It works, but do not go to max current, becomes dangerously for one led.

    Do not try to use two drivers with outputs in parallel , they do not work, they will pull currents from each other and eventually burn themselves up . The linear drivers do work in parallel, but this solution is out of question because of limitations and efficiency.

    2) Independent buck drivers, one for each LED, completely separate. Voltage is the same as previous solution, 2S battery, just ease the strain on the driver, because you use two. Perfect balance of current through the leds, each has the current controlled, instead of shared. Because you buy drivers as individual items, you need to interface them, use a single switch to control both of them. Possible, but needs reverse engineering, identify the driver, find the datasheet, understand the schematic, understand what kind of mode control is there (voltage level, current, or PWM from a side circuit), do the modification.
    • Usually the power level of the leds are set by a adjusting the ON/(ON+OFF) ratio of the driver (called duty-cycle) with a low frequency of 200-400Hz. So, the driver always pushes max current through the LED, but it is ON only for a fraction of time, the output being averaged. It's called PWM, pulse width modulation, and the relation between duty-cycle and output current is linear. That means, if it's ON all the time it delivers max current, if it's ON 1/10 of the cycle the current is 1/10 of max. Being performed at high frequency for the human eye, the persistence of human vision makes the average output light to be equal to that obtained by driving exactly that 1/10 of the current continuously. The reason for doing this PWM is efficiency (driver always works with the same efficiency) and light color consistency (the color tint is given by the instantaneous current through the led, not by the average)


    Kind of complicated , easy for an engineer that knows electrical design, but the idea is: you don't really know if it's possible to synchronize them before you buy those drivers and reverse engineer them .
    Most dual XM-Ls lights have this solution, but it's all-in-one design, a single PCB with two power drivers and a central controller for both. MagicShine MJ-880 has this kind of driver, independent power driver for each led, but a microcontroller sets the modes of both of them synchronously by PWM. It is possible to set different current levels for each led, for example if one led has flood optics and the other spot, such that the power for each led can be adjusted smartly in programs (more flood, more spot, etc). This is advanced design, such thing will be heaven for experiments, too bad I don't work in this field, I design sensor ICs now .

    3) Series connection with buck driver. Requires only 3A for XM-L led. There are plenty of drivers around that drive 3A. The voltage required for this connection is higher than Vf=6.8V. Because of drop on the wires and on the main switch, two battery elements are not enough, you need 3S or 4S battery. Efficiency is higher by 2-5% than parallel connection, less current is switched. You get one control switch. Get a driver that can drive 2 leds in series (outputs at least 6.8V or specifies that can drive 2-3 leds) and accepts 3S-4S battery (14.8V). I will advice this, simplest and most robust solution, no tampering with the circuit, just solder, plug and play.

    4) Series connection with a boost or buck-boost driver. I don't know if you can find one, usually is custom design.

    • The boost driver has the ability to generate a voltage higher than battery, let's say from 2S battery you can power 3-4 leds in series (that require 10.2-13.6V). It is used as a custom design on some 3-4 XP-G designs.

    • The buck-boost is another type of driver that has the ability to generate voltages in a broader range, from bellow to well above the input. This can power 2-3-4 leds from 2S-3S-4S batteries in any combination, with some performance limitations of course. In your project can power 2 leds in series from any 2S-3S-4S battery. This will be the best choice in terms of battery compatibility, but it's a custom design, you need to build it.


    I hope this helps more than it confuses.
    Good luck!

  10. #10
    mtbr member
    Reputation:
    Join Date
    Aug 2008
    Posts
    7
    Hi,

    I have IP67 momentary switch but it is off-on-off type, is there any easy way how to use it to change the modes?
    Thank you!

Similar Threads

  1. Three xml LEDs one reflector
    By troutie-mtb in forum Lights DIY - Do It Yourself
    Replies: 47
    Last Post: 04-17-2013, 02:53 AM
  2. Red leds
    By troutie-mtb in forum Lights DIY - Do It Yourself
    Replies: 11
    Last Post: 01-27-2013, 06:55 PM
  3. Anything promising with these new LEDs? MLC and MLE
    By gticlay in forum Lights DIY - Do It Yourself
    Replies: 2
    Last Post: 03-21-2012, 02:03 PM
  4. ShiningBeam Drivers
    By BlueMoon100 in forum Lights DIY - Do It Yourself
    Replies: 2
    Last Post: 09-03-2011, 06:16 AM
  5. So thats where all the S2 leds went too
    By troutie-mtb in forum Lights DIY - Do It Yourself
    Replies: 17
    Last Post: 02-08-2011, 11:20 PM

Posting Permissions

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