DC Voltage Amplifier

Discussion in 'The Projects Forum' started by ducati996, Dec 13, 2007.

  1. ducati996

    Thread Starter New Member

    Dec 13, 2007
    5
    0
    I am working on an LED (RGBs to be exact) lighting project. I have an RGB controller that works well, but the RGB lights that came with it are crap, so I'm trying to adapt newer, better lights. The problem is that the RGB computer uses a 0-4VDC (+/- 0.5V) variable power source to the RGBs. My new RGB lights require 12VDC to burn at full intensity, so somehow I need to amplify the 0-4VDC signal to a 0-12VDC variable signal. I have tried building a noninverting amplifier using an op amp model 1458, with a 12VDC source and the 0-4VDC for the input. The problem is that I only gained about 0.5VDC and my attemps to build multiple amplifier circuits in series have failed - not to mention it would require a large moving box to house all the circuitry when completed. Any help would be appreciated. Thanks.
     
  2. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    It would be helpful if you could post the part number of your RGB LED so that we can track down a datasheet for it.

    hgmjr
     
  3. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    I suspect the your LED's need 12 volts at some current above the 10 ma or so the 1458 can output.
     
  4. ducati996

    Thread Starter New Member

    Dec 13, 2007
    5
    0
    If this works, I'll be using multiple 1-foot strips on RGBs. Here's a link to the data I found on the RGBs: http://superbrightleds.com/specs/LB_RGB.htm

    For now, I am testing the system with a 1 foot strip of Radiantz LEDs, part number 4002-21.

    The RGB controller I have runs each color on a separate circuit, so I'll have to build 3 separate amplifiers and each would have to be able to handle up to a 5Amp load.

    Thanks again for the help.
     
  5. Distort10n

    Active Member

    Dec 25, 2006
    429
    1
    A simple solution would be a power amp like the OPA549. Power it with +/- 15V to start with. It is not only the voltage that you need, but the current source capability. The RGB bar datasheet says that current consumption per 1 foot bar is 200 mA. Makes sense, LED's are power hungry.
     
  6. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    If I interpret the datasheet literally, it would seem that the output intensity would remain relatively constant at 20000 mc for the range of input voltage from +6V to +12V. If I am correct then that implies that the input voltage is regulated. It is difficult to predict how the regulator (if present) will behave for voltages below 6 volts. You may want to hook it up to a variable power supply with adequate current rating and see what happens as you vary the power going to the LED stick from 0 to 6 volts.

    hgmjr
     
  7. ducati996

    Thread Starter New Member

    Dec 13, 2007
    5
    0
    I checked out the data sheet on this from TI (http://focus.ti.com/lit/ds/symlink/opa549.pdf). I'm not an electrical wizard, but it looks like it may work. Would Fig. 10 be my best working model for a wiring diagram? It looks like terminal 4 would be my 0-4V signal from the controller and terminal 8 should be connected directly to terminal 6 for the max 10A current limit. Let me know if this is correct.

    As far as the regulator works, I'll check to see what happens below 6V. Hopefully it won't be a problem, but if it is, I have other options for the RGBs.

    Thanks again for the help.
     
  8. Distort10n

    Active Member

    Dec 25, 2006
    429
    1
    Yes, figure 10 is a great template to modify. You are correct your signal voltage will be at the non-inverting input (pin 4); however, I would not recommend that you connect Ilim to Vref for 10A.
    The OPA549 outputs a continuous 8A, so I would suggest using 1.87k to connect Ilim to common.
     
  9. ducati996

    Thread Starter New Member

    Dec 13, 2007
    5
    0
    Should I still use the 1k and 9k resistors on wiring from terminal 3 or should the resistance be changed? Would two 10k trimmers be suitable replacements to help "dial-in" the proper output voltage?
     
  10. Distort10n

    Active Member

    Dec 25, 2006
    429
    1
    One pot to replace the 1k resistor would be fine to change gains.
     
  11. ducati996

    Thread Starter New Member

    Dec 13, 2007
    5
    0
    Thanks for the help. As soon as I can find and order a few opa549, I'll give a project update. BTW, In diagram 10, it shows a 74HCT04 and a resistor running an LED that would light if the opa goes into thermal shutdown. I have a few CD74HCT04M96G4 coming. Would these be compatible, and if so, how would I connect one to work? Thanks again.
     
Loading...