Page 13 of 13 FirstFirst ... 3111213
Results 121 to 127 of 127

Thread: DIY rotary DJ mixer

  1. #121
    I decided to rebuild the smaller 2ch mixer which was built from leftover parts and then disassembled to build the xover.

    First I built a 19" enclosure.. the bottom panel is hardboard, sides are 18mm thick pine and the top panel is plywood. The depth at the back is 60mm and 40mm at the front so it's angled, the panel height is 300mm which is 6-7RU.

    The transformer/power supply is from an old tape deck and has 630mA fuses, it provides 30VDC to the preamp and 9VDC to the power on LED. The AC input operating voltage is 230V/50Hz.

    The channel configuration consists of a 6,3mm mono input channel and two RCA stereo channels which do not provide any amplification, ie. they're passive. They're summed into a bus that is preamplified and further fed to the master outputs.

    The summing bus I built from a 12-pin screw terminal that has 4 left, 4 right and 4 ground parallel into L/R/gnd terminals so in other words the ground signals are summed.

    The mono channel is split into L/R via 1kOhm resistors to provide stereo separation between channels. The channel strips use 100kOhm potentiometers for volume control (the mono channel is a single gang, the stereo channels are dual log)

    It also has 2-track in/out, with the input connected to the summing bus and the output taken from the master preamplifier prior the master vol. The master outs are 6,3mm unbalanced via a 100kOhm stereo log, the preamp has trimmers and is wired to the summing bus using a PCB/bus connector (so it's easier to service)

    It's missing knobs atm, I'm also looking to add a headphone monitor section with PFL/master cue (which requires trim pots, a 2-pole 4-pos rotary switch, level pot and an output jack, possibly a headphone amp too), per-channel HPF/EQ and -20dB pad switch as well as balance controls.

    Here's a pic of the internals with the PSU and summing bus/master section visible :



    ..and a pic of the front panel in place :


  2. #122
    Modified the 3RU mixer so it now has a L/R input trim.. basically 1kOhm trimmers in the back panel with a through-chassis hole so they can be adjusted with a flat-head screwdriver. The trimmers are wired between the RCA inputs of the line channels and the corresponding channel preamplifiers.

    This is for adjusting the input level of the source (ie. tape deck/CD player) to prevent the channel preamplifier from clipping/distortion in case the output level of the source is too high as I noticed they may vary greatly between different devices.

    Additionally it allows adjusting the L/R balance if the output levels of the connected device are not equal for one reason or another (output circuitry, cabling etc)

    Another modification I did to the smaller 3ch mixer in my earlier post is a Ch1 mute switch, eg. if it's used as a mic channel since not all microphones have an on/off switch (edit : also added rubber feet for tabletop operation)

    EDIT : here's a pic of the input trim mod from the inside:



    and a front panel pic of the tabletop mixer build:


  3. #123
    Here's a more recent build pic :



    It now has knobs, I also added an internal supply voltage trim to the preamp as well as a 3,5mm headphone output with volume control.

  4. #124
    Added handles too for looks, racking, cable strain relief etc..



    Next I'm looking to add preamps (mainly the mono channel), channel PFL and EQs but I think the main functionality is there.

    Probably ok for stuff like karaoke (wireless mic receivers usually have some sort of built-in output preamps) with background music from external source(s) and in general if there's no need for beatmatching but I haven't tested it yet (it's essentially the same small 2ch device I built a few years ago but it has a mono channel with mute, master cue as well as 2 track in/out and uses a built-in linear PSU)

  5. #125
    Also added a 2-band master EQ, cleaned the wiring scheme a bit and sketched a functional circuit/block diagram of the signal flow in the front panel.



    The EQ sounds surprisingly good through headphones when I tapped into it post-preamp with a smartphone (it may sound different when driven with a preamp as the cutoff points tend to drift with the input level)

    It has theoretical cutoff frequencies of -12dB for the LF at 30Hz and -18dB for HF at 40Hz although I have no scientific data to back this up because I don't have an oscilloscope, instead I used a filter formulae to compute the values (I assume the filters are -6dB/oct since they're 1st order passive RC circuits)

    The filter configuration is 220ohm to 100uF/50V in the LF and 100nF/63v to 4k7ohm in the HF, I based my earlier calculations on the HF capacitor being 1uF but it's actually 100nF, by accident I misread the value.. with 1uF the corner frequency would've been 33Hz but it sounds quite nice nevertheless.

    The LF has resistor in series and cap to ground, the HF has cap in series and resistor to ground.. they're in parallel with 10k output volume pot in the HF and 1k in the LF.

    This particular filter configuration sort of scoops the low-mids resulting in a nicely warm sounding EQ in my opinion but it's a matter of taste really (I had to make a lot of compromises as I put the unit together from junk and spare parts I had left over from other projects), the downside is there's a voltage drop across the EQ (the design is basically a 1st order version of a 2-way crossover I built a few years ago)

  6. #126
    Here's a quick graph of the filter frequency response using an online graph calculator and center frequency values derived from a filter formulae (I haven't used a function calculator in over ten years, last time in high school so it's a bit crude) :



    It uses a linear equation to calculate the values. The dots represent the "ideal" 1st order -6dB/oct LF slope (blue line), the hollow ones are the HF (purple line). X-axis is the frequency (in Hz) and Y-axis is the loudness (in dB).

    The equation works by multiplying the center frequency by 2 and decreasing 6 from the dB scale in LF, and in HF by dividing the center frequency by 2 and decreasing 6 from the dB scale, so in other words :

    HF : ln(x2)^2-√a=(y-6)-b
    LF : -ln(x2)^2-√a=(y-6)-b

    where a is the center frequency (in Hz) and b is the loudness (in dB). It's somewhat inaccurate but gives "better" results if the center frequency a is first divided by 2 or multiplied by π.

    The 0dBu (red line) is set at 100 as a margin. The green line is the Y-axis (dB) value where the passbands intersect.

  7. #127
    I also put together a prototyping board/platform :



    It has a 2,1mm DC power jack, 3xAAA battery holder and a +/- screw terminal block for configurable power connectivity, as well as a 400 point solderless breadboard and a screw/crimp terminal GPIO for interfacing with external modules, input sources and output devices etc.

    It's assembled on MDF using screws, double-sided tape and hot glue and stands on rubber pads with a 210x245mm footprint. It also has a small array as a reminder for keeping track of the wiring scheme and a functional schematic/block diagram.

Posting Permissions

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