1. Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
    Dismiss Notice

Help with LM 3915 VU Meter

Discussion in 'General Electronics Chat' started by haxor12, Aug 12, 2009.

  1. BigE4u

    BigE4u New Member

    Joined:
    Sep 2, 2006
    Messages:
    61
    Likes:
    0
    The circuit shown in my past post, for portability purposes, i made use of a 9V battery. I also made this same circuit first on a breadboard, but i used +12v and used the same resistor. I'll make sure to get ahold of a 68Ω 1w resistor next time i build this permanetly for my PC.
     
  2. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,450
    Likes:
    937
    Location:
    Canada, of course!
    Simply measure the voltage at the input of the LM3915. It should be 0V without an input signal. Then all LEDs should be off. The input should connect to the music source with shielded audio cable.

    13V is too high since if you use 1.8V red LEDs then 11.2V is wasted by making heat in the LM3915 if all LEDs are lighted. An added resistor can share the heat.
     
  3. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,450
    Likes:
    937
    Location:
    Canada, of course!
    Think about the 8.16V across a 68 ohm resistor then wonder where the additional 1.8V for the LED plus 1.5V for the LM3915 will come from. 68 ohms is too high even if the supply is 12V.
    With a 9V supply the LM3915 will dissipate only 864mW which is fine and the resistor is not needed.
     
  4. dave

    Dave New Member

    Joined:
    Jan 12, 1997
    Messages:
    -
    Likes:
    0


     
  5. BigE4u

    BigE4u New Member

    Joined:
    Sep 2, 2006
    Messages:
    61
    Likes:
    0

    Weird problem myself using the transitorized peak detector.

    1st pic- original layout and 7 of 10 LEDs stay lit

    2nd pic- transistor reversed and 7 of 10 LEDs stay lit

    Perhaps due to the transistor being 20yrs old from RS but they worked perfectly with homemade testor and with DMM readings.
     

    Attached Files:

  6. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,450
    Likes:
    937
    Location:
    Canada, of course!
    The voltage shows that the transistor has an extremely low current gain probably because its collector and emitter pins are connected backwards.

    You know what? You are correct. This circuit has an input of 1.25V to light all the LEDs but the similar circuit in the datasheet has 10V to light all the LEDs. Therefore your LEDs light when the input is only 0.418V.

    In my VU meter I use an opamp (like shown in the datasheet) to make the peak detector that does not have this problem.
     
  7. BigE4u

    BigE4u New Member

    Joined:
    Sep 2, 2006
    Messages:
    61
    Likes:
    0
    So, the transistorized schematic is out and the op-amp design is in. thanks

    Ps- I was triple checking everything because i though i was doing something wrong too.
     
  8. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,450
    Likes:
    937
    Location:
    Canada, of course!
    I think the transistor peak detector will work with a full-scale input voltage of 1.25V if the diode is replaced by a 2N3904 transistor and the 10k resistor is increased to 100k or more. I'll try it sometime.
     
  9. cly_tech

    cly_tech New Member

    Joined:
    Jun 16, 2010
    Messages:
    10
    Likes:
    0
    hello, need some advice, what if i try to make the input is from electret mic+lowpass filter+LM3915 VU Meter ? i mean i wanna make a VU Meter for the low frequencies only(100Hz and below may be the voice input) and the scheme in my head is like this :

    [​IMG]

    correct me if i'm wrong, coz i'm in a very need to make a project with this scheme, wanna make the max output(if all LED or only the tenth LED is on = 30dB, and that is gonna be the triggering input for a microcontroller)
     
  10. cly_tech

    cly_tech New Member

    Joined:
    Jun 16, 2010
    Messages:
    10
    Likes:
    0
    sorry, forgot to put a 2,2k Ohm value at the resistor at the positive leg V+ at the microphone.. and sorry for bad english.. :D
     
  11. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,450
    Likes:
    937
    Location:
    Canada, of course!
    The electret mic has an impedance of about 3.3k ohms so it should be powered from a resistor with a higher resistance like 10k ohms. The 10k resistor should have a filtered voltage supply.
    Then the mic and the 10k resistor have a combined impedance of 2.5k ohms so the input impedance of the preamp should be 12.5k ohms or more.

    But the 2.2k resistor and the extremely low 150 ohms input impedance of your preamp shorts the output of the mic to almost nothing.

    C2 shorts the output of the opamp. Do you want a peak detector circuit instead?

    In my Sound Level Indicator project I used a dual MC33172 opamp with the first opamp as a preamp with a high input impedance and a gain of 101, and the second opamp as an inverting peak detector circuit with a transistor. You can add a lowpass filter 0.015uf film type capacitor across R5.
     

    Attached Files:

  12. cly_tech

    cly_tech New Member

    Joined:
    Jun 16, 2010
    Messages:
    10
    Likes:
    0
    oh i see.. thanks, this really give lamp in my head.. i'll try, so, the 0.015uF will be put at R5 output line(going to pin 1) to other side (going to Pin 2) line right ? will make the cutoff frequenciess = 100Hz ?
    and can i see how you put the dual op amp, or the single 358 layout ? it kinda hard for me to see how the line will be in the PCB from the scheme(although i really understand how it work now)
     
    Last edited: Jun 16, 2010
  13. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,450
    Likes:
    937
    Location:
    Canada, of course!
    If the resistors and 0.015uF capacitor have a tolerance of 5% then the -3dB cutoff frequency is 107Hz. The slope from the simple filter is gradual so that the output is 1/100th (-40dB) at 12kHz. Frequencies above 12kHz will not be reduced more than -40dB.

    I used stripboard to make my circuit. It looks like this:
     

    Attached Files:

    • Like Like x 1
  14. cly_tech

    cly_tech New Member

    Joined:
    Jun 16, 2010
    Messages:
    10
    Likes:
    0
    ahhh i see, i'm gonna make it with real PCB, i'll try put the board layout image as soon as i draw it in the next post, and i want u to check it... many thanks uncle scrooge.. :D
     
  15. cly_tech

    cly_tech New Member

    Joined:
    Jun 16, 2010
    Messages:
    10
    Likes:
    0
    uhm.. double check on the preamp.. did the R12 = 100 Ohm ??
     
  16. cly_tech

    cly_tech New Member

    Joined:
    Jun 16, 2010
    Messages:
    10
    Likes:
    0
    done.. here it is the image

    [​IMG]

    heheheh, still with the same question, is the R12 = 100 Ohm ? oh yes, and is the R10 = 1000 Ohm(the one to the V+) ?

    if this is alright... than it means my low frequencies sound detector is ready for the next step.. heheheh :D
     
    Last edited: Jun 16, 2010
  17. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,450
    Likes:
    937
    Location:
    Canada, of course!
    You have R4 connected to ground by mistake instead of to R2 and R3.
    You have round electrolytic capacitors shown as rectangles that are too long.
    The output voltage at pin 10 of the LM3915 is high enough to destroy the input of the micro-controller.
     
  18. cly_tech

    cly_tech New Member

    Joined:
    Jun 16, 2010
    Messages:
    10
    Likes:
    0
    ahhh yes, the R4 will be corrected, and as for the electrolytic capacitor.. well, it's just for the lay out in the pcb, i just put it up as long rectangle figure(5mm) so i can draw the line between any capacitor legs if it needed... hmm, any suggestion/advice about "from where" can i get the 30dB LED ouput(to become the input for the microcontroler) ? or how should i modify the out put pin ? so it will be harmless as an input..
     
  19. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,450
    Likes:
    937
    Location:
    Canada, of course!
    Two resistors will make an voltage-divider to reduce the max output from the LM3915 to +5.0V.
     
  20. cly_tech

    cly_tech New Member

    Joined:
    Jun 16, 2010
    Messages:
    10
    Likes:
    0
    hm.. what if i put a 10k Ohm trimm resistor ?
    (hehe, plss correct my value if i'm mistaken :D)
     
  21. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,450
    Likes:
    937
    Location:
    Canada, of course!
    You can use two resistors to make a fixed attenuator or use a trimmer resistor to make an adjustable attenuator.
     

Share This Page