• Help matching impedances

    10/15/2019 at 22:51 3 comments

    Hey there!

    I'm looking for some help in understanding impedance matching.  I'm hoping one or more of you technically savvy electronics folks will jump in and give me a simple circuit recommendation.

    Specifically I want to try to build a passive "pad" that goes between the output of a Fishman Prefix Plus built-in guitar pre-amp and the input of an IRig HD2 guitar-audio-to-usb converter.  My issue is that, as it stands, I have to turn the volume control of the pre-amp as close as I can to it's minimum value (like 0.01, which is iffy), and set the gain of the Irig input hard to it's minimum value to get any kind of decent sound.  Anything else results in lots of distortion.  I'm trying to get a clean guitar sound.



    I get the feeling that by using the extreme lowest setting on the pre-amp I am not in the sweet spot of the pre-amp design.  I feel like it makes the sound tinny ... I'm always lacking bass when I use the device (compared to just plugging my guitar into the Yamaha Stagepas PA).  It's just my noob idea that the amp was probably designed to be used in the middle, rather than at the extremes of its gain settings.

    So, I'm thinking about making a passive "pad" that basically consists of two resistors. 

    input    O------- R2 --------+----------------O output
    ground  O------------------- +----------------O ground

    My basic, very limited, understanding is that such a passive pad consists of a voltage divider created by resistors R1 and R2.   I really have no idea what values to use, although I sort of think their ratios should be about 9:1  where the value of R2 is about 9 times greater than the value of the "shunt" resistor "R1".   Rather than just guess, I thought I would ask the community.

    The specs for the devices are copied from the user manuals  Here is what I see:

    FISHMAN Guitar Pre-amp

    • Nominal Input Level: -20dBV
    • Input Impedances: 20MOhm
    • Output Impedance: "Less than 3.5kOhm"
    • Nominal Output Level: -12dbV

    IRIG HD2 audio-to-usb converter

    • Maximum Level: from 307mvpp to 8.36Vpp
    • Gain Control Range: 28.7 dB
    • Input Impedance (guess): 380 kOhms

    I cannot find the actual input impedance specification for the IRIG HD2.  There is a reference in the user manual that it is a "high-Z input".  I am guessing it is simllar to the input impedance for the IRIG2 which IS specified as  380kOhms.


    So, given those specs, what should be the values of R1 and R2?

    Any other thoughts on what I'm trying to do here?   I'd sure like to get a good sound and a feeling of control instead of a tinny sound and a pre-amp that is turned down to 0.01 ....

    Thanks in advance for any replies.

    - Patrick

  • Why doesn't someone make a WS2812b Switch Array?

    09/07/2019 at 06:17 9 comments

    Two things:

    • I can’t believe this worked (almost the very first time) !
    • I’m surprised nobody else seems to have done this before !

    OK.  I have been looking for months on the net for somebody, anybody, who has combined WS2812b addressable LED’s with switches to create a switch array.  At the end of this I explain why I would want such a thing, but for now just please understand that I could not find any other examples of anyone selling, or even trying, something, like this. 

    Perhaps no-one needs it, or perhaps it exists, but is just not on the web, but perhaps no one has thought of it before.   In any case, it seemed to me that it would be fairly straightforward to combine a switch with each ws2812 in an array (or a strip), and be able to tell which button is pressed, if any, based on the timing of the signals being sent to the LED’s.

    Now please understand that this is very crude hardware and code.  I didn’t even expect it to work.  The LEDs are supposed to have capacitors from power to ground, most schematics recommend a resistor from the output pin of the Teensy to the input pin of the LED, and I’m pretty sure that no electrical engineer out there would recommend hooking the DIN (data in) pin of each LED through only a switch directly to a bus line to an input pin on the Teensy.   Something has to fail, right?  Oh, and always use a separate power supply for the ws2812’s lol 😊

    So, please let me say that I will appreciate any comments, or contributors, that can design a proper circuit, help me to understand how it should work, and/or really want to dig in to write good code for this purpose.   


    I have a couple of ws2812 arrays sitting around (an 8x32 and a 4x4), but I don’t have any strips.  That’s a shame because the strips look like you can easily cut one off and work with it.  You probably have a strip you can just cut up!  I don’t want to cut up the arrays, but I do happen to have a bunch bare ws2812 LEDs that I bought experiment with.  So I rummaged through my stuff and found a little circuit board I could solder them to.  That being about my third try to air-solder SMD’s, I glued them on with superglue, a little too much in one place, and then glommed some paste solder on em with a toothpick, and blew hot air on em.  Even though it was only 215C, I basically tortured the LEDs as I had to burn through the superglue on one of the contacts.  Nonetheless, they conduct electricity, and there are no shorts. 

    I added some pins to the board, so then I had three ws2812’s I could play with on a breadboard.   If you want to build one of these, you probably can just cut one off a strip and use alligator clips or whatever it takes to hook them up.

    The first circuit I made used jumper wires.  Much to my surprise I hooked up the 5V and ground and got the Teensy example basic ws2812 test program working with my little 1x3 array of LEDS in a few minutes,.   

    So then I decided to try to write some code to make it work as a switch array.


    As you send serial data to the LEDS (actually, to the first LED’s DIN – data in - pin) using the ws2812 protocol, each LED takes 24 bits to display an RGB value.  OK, actually it’s GRB, but that’s not important!  

    After the first LED “eats” the first 24 bits, it starts passing the remaining bits onto the next LED. So the second LED does not see any bits (rising edges) until the first LED is done with those first 24 bits.  Then the second LED “eats” the second 24 bits before it starts passing the remaining bits to the third LED, and so on.

    The other thing worth knowing is that each bit in the ws2812 protocol, both ones and zeros, has a HIGH and a LOW part. ...

    Read more »

  • Everything looks like a (plastic) nail ..

    09/02/2019 at 19:27 0 comments

    To a man with a hammer (3d printer) ... everything looks like a (plastic) nail …

    OK.  Whew. 

    Turns out that 3D printing is addictive.   Who would have thought?   It’s been about a month now since I got my (first, only) Prusa MK3s printer.   I’ve pretty much been running it, or doing Fusion 360 design projects, 24/7 for the entire time.   My sleep patterns are out the window, watching the sun rise as I wait, sleepy eyed, for a 3D print to finish before crashing for a few hours to wake up at noon, 1, or 2 and start again.

    To say this has been a minor digression from the project would be an understatement.

    I got it, basically, because I know at some point I’ll want a custom box of some kind for the system components.  I also got it because I have an inkling of an idea for creating different kinds of floor controllers, ala a mouse-pad for your feet, or gyro-accelerometer gesture detection shoes, or something like that. 

    To whit I’ve also built quite a collection of little gyro-accell modules (ADXL345’s and MPU6050’s), a bundh of esp8266 modules, and have even got a collection of Velostat, conductive fabrics, threads, copper tapes, and the such with, once again, the general idea of creating a controller of some kind that is more portable, and expressive than four shiny metal push buttons on a metal box.

    But what I didn’t expect was how the 3D printer would take over my life.  

    I think I’m over the initial hump.  As I’m writing this I’m also learning how to post things to Thingiverse. So I posted this: https://www.thingiverse.com/thing:3837992 .. a system of brackets and fixtures for making desk, table, and wall mounted 12V LED lamps.

    I don’t want them to be part of my hackaday.io page, per se, but in this case I at least want to note why there has been no activity on the project for the last month.

    That’s because not only is it obligatory to print the basic examples, then design your own phone case, but it is almost compulsary to add lights, and an rPi, to the Prusa.   So while learning Octoprint, Fusion 360, and 3D printing in general, I ended up creating this whole darn fangled 10mm lighting system, not only for the Prusa, with a 12V dimmer circuit (ok, maybe this is a subproject), but also as a clamp on desk lamp, and a desktop magnifying lamp as well. 

    Here's the 12V dimmer I made for the Prusa:

    Sheesh, for some reason I ended up thinking about CNC, and want to have one of those little CNC machines, maybe hook my Dremel up to it, now.

    I guess the other thing I hadn’t realized was how appropriate a 3D printer is for life on a sailboat. Instead of keeping a bunch of spare parts, it is conceivable that many things can be 3D printed on demand.   I discovered that early in the first week … Had some toilet hinges that were broken … there was a darned set of STL files for the exact model of my Jabsco toilet on thingiverse!   I downloaded it, sliced it, printed em, and they’re better than the original hinges from the manufacturer (who tried to save plastic and they characteristically break in the same place).

    Anyways, it’s been almost a month and only now am I semi-ready to start going back to the electronics.  I had to get a new Audio Injector Octo board because the I2C stopped working on the one I had, so I had just diverged into the Zynthian project, got blown away by that, and was also messing with the 8266-gyro stuff.    I chatted with Flatmax, and got 4 CS42448 chips (like $10 each) and will try replacing the chip on the bad board (my first SMD soldering try, esp with 64 pins), so I think the first thing is going to be to go back to the AI Octo Rpi Bare Metal setup, test the new board, and then try to fix the old one.

    But I had to put something up about the stupid 3D printer.  I’m waiting for thingiverse to render an STL file, I think, and then I can upload the 30 or so...

    Read more »