Category Archives: Electronics

How to Use Kobiconn 174-R819 Banana Plugs

After a bit of searching for the best deal I could find, I eventually settled on the 174-R819 Kobiconn Banana Plugs for my new Synthesizer. My previous Synthesizer used 1/4″ plugs/jacks because they are what I have been previously comfortable with and I had a certain affinity for the coaxial-natured noise reduction associated with the connectors. However, after a good bit of research, I came to the conclusion that I would much rather use banana plugs/jacks. The biggest selling points for me were the simplicity of cable construction, the stackability of the plugs (goodbye, “multiples” modules, which allow the use of one output to drive several inputs!), and the decreased panel area that the jacks take up.
Anywho, here is a short pictographical explanation of how to use them most effectively.

Backlight Modification for Roland TR-707 Drum Machine

Back in 1985, I was born.

But that’s beside the point; that same year, the Roland Corporation also manufactured my TR-707 Drum Machine! I received it as a birthday present in 2004 (or thereabouts), and after playing with it a few months, me being the hacker that I like to think of myself as, I proceeded to modify it in many different ways. Among them, I wired a TON of toggle switches to the panel, forming a matrix of connections between various sound effect data lines.


These are fairly well-documented circuit bends, and most of the connections create some very unique sounds and effects that I can instantly recognize when I hear in a track (I say that like it happens a lot, but it doesn’t and all I can really think of right now is Last Step (aka Venetian Snares) – “Last Step”). I used it to make some videos recently (before the mod I made this post to discuss was performed):

I have yet to find any articles on the backlighting of a TR-707 or similar device (or perhaps this is just me being too lazy to figure out just what similar machines might exist and searching for “ backlight mod”), so this was something I had to figure out on my own. The LCD in this machine is based on Twisted Nematic technology, so simply putting a light behind the display did nothing for me; in fact, less than nothing. It prevented the LCD from showing anything at all! I later realized this is because this was completely diffused light source. The way the display normally works is there is a reflector with a polarizing film, so that ambient light from outside passes through an outer polarizing film, then the LCD, another polarizing film attached to a reflector, then back out again. Thus, the light you see is polarized in one direction throughout the layers; the segments show up as dark spots in the reflected light because the liquid crystal is twisting the light out of phase with the rest that is reflected behind the LCD, and the outer polarizer thus blocks it from getting through; the segments represent light that is blocked.



So I knew what I needed finally, which was another polarizing film like that found on the outside of the LCD. Long story short: I had an old broken TI-89 graphing calculator which is conveniently based on the same technology, and the film is just slightly smaller than the TR-707 LCD; in fact, it is the same size as the silk-screened grid on the thing, which is all that needs to be backlit anyway! Excellent!

Now for the important, non-ranting bits of information! My backlight consists of a 1/8″ thick piece of mirrored acrylic cut to fit behind the LCD.


I then took some white LEDs, and ground them down to fit flush against the edge of the acrylic; in total, two sets of three, powered by the -10V regulated supply I had to make with an LM317 because I burned out the transformer on board many years ago (that’s another story to tell altogether, but suffice to say that I didn’t know what was messed up for 3 years and only fixed it last year at some point, finally).

I also went ahead and put another transformer in from a wall-wart so that the whole thing can be powered from a normal computer-style power cord. There’s also a small fan to keep the whole thing cool.


The other edges of the acrylic are covered with foil tape to keep light inside, the surface was roughed up a bit to improve diffusion of the light, and the TI-89 polarizer is stuck there. The whole thing is then stuck behind the LCD and everything put back together. Fairly simple idea, but I’m the only one I know of who has documented this process. To do it to other machines, it may not require more broken TI-89s, because the polarizing film can be purchased online in a few places, such as this one.


It is wired with the backlight on all the time because the LCD is readable in both normal conditions and the dark so I didn’t care about setting up a separate switch for it. I find it works nicely, despite some swirls of distortion in various parts (due to the polarizing film not being meant to be removed and placed elsewhere). Finally, I used GIMP to mock up some new labels for the panel (that was a lot of fun, I swear), and printed out on white adhesive label sheets.




1:9 scale TARDIS Model with Lights and Sounds

Kim has been a fan of Doctor Who for as long as I can remember, but she helped get me into it a few years ago and now we’re to the point where we watched the last 3 seasons episode-by-episode as soon as they aired (thanks, p2p filesharing!). Well, for her birthday this year, I built her a TARDIS! Not like the last one, either.




This TARDIS is made from clear acrylic of various sizes and cuts and painted with “Phthalocyanine Blue” acrylic paint (which gives a decent approximation of a wood texture up close). There was lots of hand routing, drilling, cutting, and fast-paced handiwork involved! I mean, I originally put most of it together over the course of the 5 days preceding her birthday back at the beginning of June 2011, and then I had a few more days of apologetic finishing to do the following week. We ended up taking it to A-Kon 22 and carried it around the con, where we got lots of compliments; it was a lot of fun! This past week, I rebuilt the electronics to hide everything at the top and make some modifications to the way it all functioned.


On the details side, it was a lot of work. I even had to force myself to get into Arduino programming. Well, kinda “forced”. I’ve done some work with PIC microcontrollers in the past, but I had been in this sort of “analog phase” (well, obsession) for the past few years so I had an annoying anti-new-stuff attitude going on that I feel a need to apologize for. I saw the Arduino platform as too simple for me and I still do think that a lot of it has been wayyy over-simplified; people who don’t even know how to solder are able to put together complex controlling systems that others have programmed. I also saw it as kinda pricey until you realize what you are doing (I started out with a ~$19 Arduino Pro Mini and a ~$15 FTDI Basic Breakout Board), but now I know that all I really need for any project is a ~$4 ATMEGA 328P IC and some other cheap components to build a project around.


I built the sound system system with LadyAda’s Wave Shield design and code; it was pretty much throwing some components together and getting it all to work pretty much right away. I needed something quick, and I surprised myself by having a working perfboard soldered together the day after I received the components I ordered from Sparkfun and Mouser.


There is a reed switch and a magnet on the right side door that triggers an internal 1W RGB light (100% red, 50% green; kind of a gold-orange in person) and door opening/closing sounds appropriately. The lantern light is a 3W warm white LED, and the interior light is a second 1W RGB LED; when sounds are playing, an LM324 (best op-amp for single-supply low-voltage circuits) is setup with an active low-pass filter and envelope follower and 2n3904 driver for the interior light (it basically stays on the entire time something is playing) and an active high-pass and 4x parallel 2n3904 driver to handle the 900mA for the lantern LED! I have to wait until I have some more spending money to order some wide-angle white LEDs for the “police box” signs, but I have some blue LEDs that are working nicely for now.


There are 12 buttons hidden on the bottom twelve panels of the sides and back; each one triggers a different sound stored on an internal SD card. There is an external 16V input for charging the 5x 2400mAh NI-MH batteries powering the whole thing, a switch for turning the “police box” sign lights on and off, and another switch for turning on the sound and lights system. With both switches off, no power is consumed and the TARDIS can sit for months without needing to be charged thanks to the Powerex Imedion batteries which are charged with a simple LM317 constant current source that trickle charges at 90mA.



The door is lockable via a turnable screw head through the door and another screw bolt soldered to the other end that rotates parallel to the back of the door, locking into a piece of metal that prevents the doors form opening when “locked”. It works nicely, and Kim likes how a screwdriver is needed to open the door.


Kim loves her new TARDIS, and I had a lot of fun building it for her! I love you, Kimberly! :D


Having already wrestled to get the LED matrix to stop ghosting at 12V into the LM317 over the course of the past week, I managed to tweak my matrix scanning code on the Arduino to work with only a 9V supply into the regulator. Despite only having to supply power to the ATMEGA328p and a lone CD4017, the already heavily heatsinked TO-220 package was heating enough to be painful to touch and it even caused apparent heat-smell. I was about to give up and just install a small fan inside the assembly, but I decided to bear down on my code and try some new things. I am glad that I did.

Thus is the life of an engineer. Today, I hunted a most desirable game: efficiency.

I believe this the point in time at which I must don a pith helmet and my work boots. The JB Weld beckons me.

DNA Model

I made a DNA model with steel wireframe, LED base pairs, and control circuitry to flash individual pairs randomly at a variable rate – X-Mas gift for my Mother. Just another week of Winter break!

This is, unfortunately, the only good picture I have that is representative of the final form. While I started with the intention of documenting the build in its entirety, I quickly realized it was taking me longer than expected and I had many other things to take care of simultaneously, so I pretty much stopped taking pictures after this one.

Synthesizer Power Supply

Been a while since my last post; been busy stressing myself over school and trying to actually relax a little with my Winter break. Fall 2010 went well in the end; I was quite worried about two of my classes going into the last few weeks, but I pulled everything around and somehow made a decent finish with three B’s and an A. X-Mas was good, even managed to get a Wii! Of course, I spent the last few days working on getting it soft-modded (fairly easy process at this point in the maturity of the product; relies on some crappy programming in a Lego Indiana Jones game that someone managed to exploit with a modified save file). I slapped together a 2.5″ SATA HDD and an external enclosure that runs it off of USB power, and now I have all the games I want! Thank you, Internets! Now Kim and I just need to find people to come over and play with us! >.<


As I mentioned a few months ago, I built a simple power supply into my synthesizer. I also finished putting in all the components on the front panel, which looks nice and spiffy; the rectangles of leftover acrylic that I glued to the edges to hold the panel to the case are still rather hackish (and always will be, because I didn’t give that mechanism any thought when I designed this thing!), but overall, I still think it came out just as well as I had intended it to over 4 years ago. Here is the overview of the internals of the power supply, which was pretty much ripped straight from the MFOS Adjustable 1.5A Bipolar Power Supply, but still required a good bit of my engineering construction know-how:

The placement of the components had to be very carefully scoped out in order to prevent any possible problems with interaction with the electronics above (both physical and electromagnetic) – while this overall project is pretty much mid-fi, I have been trying to spend at least a bit of my time on paying attention to hi-fi aspects like noise reduction and high-quality components. Of course, most of the wiring shown here is done with scraps that I needed an excuse to get rid of anyway, hence all the splicing (although even a lot of that is equally due to needing to move the transformers around after having put what I thought to be “just enough” wire length to keep things neat):

The larger transformer (which is pretty much the only product I actually WANT to buy from Radio Shack) is what is driving the regulator circuitry. I realized that a fan would be nice to have to cool those heatsinks, but discovered that the fan was making a mess of my power rails (even after putting in a huge bypass capacitor), so I took apart a wall-wart and ran it in parallel with the main power transformer to power the fan by itself (I know it is overkill, but it works and people pretty much give these things away!). I made the heatsinks with a good bit of hacksaw work on an old winged heatsink I had in my collection of dumpster-diving goodies from back in High School:

My oscilloscope shows this to be a very clean supply with excellent regulation; I did quite a bit of experimentation to make sure of this because, well, I could. It was fun! :D
On the outside of the case, I had a little geometry fun and hand-drilled about a hundred holes in a pattern for the fan intake:

It is a little noisy (acoustically-speaking, while using the machine), so I think I’m actually going to build a small regulator for the fan and place some thermistors to sense the temperature of the heatsinks and only turn on the fan when they actually need it on. However, it might be better to simply put in a lower-voltage wall-wart supply – the airflow is also good for the rest of the electronics inside.

New Car Stereo

I’ve been clunking along with the factory installed stereo and 6-CD changer in my 2001 Isuzu Rodeo for years now. It finally got bad enough (not loading cds, probably a bad stepper motor based on the loud clicking and whirring sounds coming from it) for me to consider and go about purchasing a new CD player. I got a relatively cheap one from Auto Zone, and then had the honor/fun of using my multimeter to determine which wires on the wiring harness went to what. After an hour in the afternoon heat trying to figure it out, I did a little more exhaustive online searching and finally found a diagram that seemed to match what I had.
Ten minutes later, I finally had the thing wired up and installed properly. Works nicely and actually plays CDs. I’m happy. :D

I also did quite a bit of work on the synth this weekend; built and installed a power supply and a fan for the heatsinks on it. I had been previously using a pre-built linear supply that I purchased many a year ago, and decided I’d rather go the more simplistic route and build my own. It works excellent! I had to rip apart a 12-V wall-wart and wire it in parallel with the main power transformer in order to power the fan separately, though, because it was leaving a strong bit of noise on my power lines (even with a 10mF bypass capacitor; talk about overkill!).

More on that later though, because school is taking up most of my time now; this semester is quite busy, but I’m learning a lot about electromagnetic theory and various other very interesting things right now.