LED Audio Spectrum Analyzer help

Discussion in 'The Projects Forum' started by dta333, Dec 21, 2011.

  1. dta333

    Thread Starter New Member

    Mar 6, 2011
    13
    0
    Hey all I was wondering if maybe some people could help me with a project I am working on.

    So as a project for my Winter break I want to make a bigger display for my audio spectrum analyzer that I purchased in the summer.

    [​IMG]

    The spectrum analyzer runs a small 20 x 20 led display right now.

    [​IMG]

    My plan is to make a bigger 20x20 display using 5mm LED's. I have all the LED's already and am in the process of making the display.

    This is how the LED's are connected

    [​IMG]

    The question I have is the power from the board is not strong enough to power the bigger display and I need to make a LED driver for it.
    A friend of mine suggested using mic2981 on the anode side of the display.
    This chip should work I think. Since there is more power going into the display will I also need something on the cathode side so I don't over load the board?

    Any help would be much appreciated

    Thanks alot
     
  2. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Well, here's where you got the thing to begin with:
    http://www.youritronics.com/forum/audio-spectrum-analyzer-v2-1-description-t169.html

    And here's an expansion board that is available for it for $140:
    http://www.youritronics.com/shop/expansion-board-p-11.html

    The expansion board has twenty N-channel MOSFETs, and twenty P-channel MOSFETs.
    Why not simply purchase the expansion board?
    Here is a topic about the board in their forum:
    http://www.youritronics.com/forum/expansion-board-v1-0-for-the-audio-spectrum-analyzer-t133.html

    I guess you must have found that already, as your schematic is precisely what is in the "example led matrix for expansion board.pdf" that is linked to in the 2nd paragraph of that topic.

    So, what is so mysterious about the driver? It's more-or-less laid out what you need to do; just find suitable MOSFETs, lay out & build a board.

    Or are you in over your head?
     
  3. dta333

    Thread Starter New Member

    Mar 6, 2011
    13
    0
    Well the reason as to why I'm not 100% on buying the board is because I would like to make the driver myself as more of a learning experience. At work I have access to an LPKF pcb fabricator so I would like to use this time to familiarize my self with making and designing pcbs and the machine. I also wanted to know if there was a different way then to just using mosfets. Because I see a lot of LED driver ICs so i was wondering if a smaller form factor is possible
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Hmm, must be nice to have a PCB fabricator available.

    Anyway, the Micrel MIC2981 is a Darlington, which is a big step down from a MOSFET.

    The Vce on those IC's exceed can hit 2v even with a 100mA load. That would result in 200mW power dissipation.

    I don't know what kind of current you're thinking of using on your LEDs, nor what duty cycle. However, the controller board is already using ULN2003's, which is similar to the NPN equivalent of those Micrel ICs; and they are using discrete PNP transistors for the source drivers.

    The high-power interface uses individual MOSFETs, because integrated Darlingtons don't cut the mustard; too much heat dissipation. Also, MOSFETs that come packaged with multiple channels are much more expensive than individual MOSFETs - and you have to de-rate them more due to the small footprint. It's not that they generate lots of heat, but the will generate SOME heat, and that heat has to go somewhere.

    I don't know where you are going to find a transistor as small as those MOSFETs used on their interface board, that are rated for the current necessary. There is just too much power lost in a regular transistor.

    Anyway, if you're going to design this thing, might as well build the interface board into the LED panel.

    You're going to need to do most of the work on this. If you get stuck, we'll try to help. However, at this point I don't even know if the displays on the original panel are common cathode or common anode.
     
  5. dta333

    Thread Starter New Member

    Mar 6, 2011
    13
    0
    Thanks a lot SgtWookie you have cleared up the confusion I had about why individual MOSFETS were used. When I first saw the already made board it didn't make sense as to why individual MOSFETS were used.

    So I guess I need to go on a hunt for good MOSFETS, the expansion board says it uses 6.3ma MOSFETS. Any suggestions?

    Also looking at the the original small display are common cathode or common anode.
     
  6. sheldons

    Active Member

    Oct 26, 2011
    616
    101
    if you post the make and model of your unit i shall see if i can source a service manual for it which will be a big help with the design of your board....
     
  7. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    MOSFETs are really pretty amazing. You can get MOSFETs that can sink a heck of a lot of current in surprisingly small packages nowadays. The International Rectifier IRLD and IRLU line are interesting; they are both in TO-251 packages, the first with leads formed for SMT, the 2nd with standard long leads for thru-hole mounting. You can get these with low Qg (total gate charge) and surprisingly high Id capacity for such a small package - if you keep the Vdss rating down. Vdss costs you a lot as far as gate charge goes; the higher the Vdss rating, the higher the Qg is.

    MOSFETs that are rated for at least 3x the current you want to sink/source, and that are rated for a Vdss of 20% to 50% higher than is expected. After that, look for a low Qg.

    I don't know, you tell me?
    You can find out using a multimeter set to "Diode Test".

    Sheldon:
    He bought the spectrum analyzer here: http://www.youritronics.com/
    Just like it says on the board.
    They don't have a service manual listed.
     
  8. dta333

    Thread Starter New Member

    Mar 6, 2011
    13
    0
    Ok I will look into those MOSFETS

    I had a typo in the common cathode or common anode statement

    I meant to say that I did check them and they are arranged in common cathode not common anode. Sorry about that
     
Loading...