Tag Archives: soldering

Hex Light Prototype1 - Brad Ormand

12.15.2015 – Hex Light and Animation Controller

Recently, I have been making a light animation controller.  The prototype is on a breadboard and combined with a separate little 2 x 3 RGB LED matrix module I built last month.  I have the system running on 3.3v with a few test animations programmed in C on a 16-bit Pic24.

I’m excited about this project and I’m continually making it better – both in features and in the kinds of animations I’m writing.  It’s pretty wonderful to see it all come together!

I still need to mount the pieces together and stain them, but it’s taking shape.  I’m also going to carve the pieces with some more detail and perhaps put in some metal inlays – I’m still messin’ with it.  In the end, it’ll all be one piece, but I just have it sitting on the BB for the time being, to get a bearing.

Got some stills here:

Hex Light Prototype - Brad Ormand

Hex Light Prototype - Brad Ormand

I cut and sanded some wood last night and put a cut sheet of coated mylar underneath to let the colors diffuse and shine through a bit.  I thought of this design while falling asleep a few nights ago and got back up to draw it out on the wood.  So, it was nice to see it come to life.

Mainly, my workstation has been my pool table (lol), and it’s a pretty awesome little surface.  I clean it off and play pool in between projects, but yeah, it’s central and in the main room and it’s a great place to work.  Also, I have some of my other in-progress projects in the background there that I brought out to be inspired by 🙂  This captures the scene for the end of 2015, for me to look back on…

Hex Light Circuit - Brad Ormand

Hex Light Breadboard - Brad Ormand

The pins talk to multiplexer-inverter chip pairs that gives each anode and cathode of the matrix a 3-bit address that I take advantage of when I control the sequence in code.  I like this way of doing things – the complexity has been offloaded to the hardware.  I handle the current load by never having two or more LEDs on at the same time – it’s just very fast switching that makes it look “on”.  I have a rotary encoder switching the animations, properly debounced, feels great, and integrated into the user event loop to provide more features.

As I go along this month, I’ll be closer to deciding on an arrangement for a reproducible product.  I’m currently investigating putting in an OLED or Chip-on-Glass display to show the current animation and battery state, etc.  Some smaller products won’t have this, but I want other, larger ones, to have that visual piece in there.

All-in-all, I have been busy with software engineering at my day job, but when I have the time, I like to work most on this project – it’s an offshoot, or simplification, of the honeycomb light I worked on earlier this year.

I have been doing a little painting lately using a “crackle” technique, and I have been drawing and carving more human faces, too – gettin’ better with both.  One day, I’ll integrate all of them into my professional projects as well.  The practice in those other disciplines, though, also helps me come up with designs like this to do this project:

Hex Light Prototype1 - Brad Ormand

And, the next step on this Hex Light project is to design the SMD boards and get those etched and tested.  Once they work and I have the kinks worked out, I’ll send off for some properly solder-masked, dark green and gold boards (fancy), and drop my logo on it for a real nice-looking internal product.  And, yeah, I’m going to make the board a hex shape, too. 🙂   Until next time…

Comparing SMD Components

04.16.2015 – SMD In-House PCB Test & Truncated Icosahedron Paint

I painted the Truncated Icosahedron (I’ll call it by it’s shape name cuz I can’t think of anything else fancy to call it yet).  Yep – I got a supply of Metallic Blue, which glistens in the sunlight, for the outside, and painted the inside Titanium White.  What’s the result?  Well, it looks cool as hell, but now I gotta think what to do with it….

I wanna light it up soon, but I’m into 3 projects simultaneously right now and have a job, but, I’m thinking I’ll light this one up with those Philips Lumileds in series with a small dimmer control board.  We’ll see…  More thinking required.  Or, more appropriately, when I get the idea, I’ll start implementing it freely.

Truncated Icosahedron Brad Ormand

PCB SMD TEST

I’m trying to see if I can get the precision necessary to etch these micro-trace circuit boards with Ferric-Chloride, in-house.  I have a small lab set up for that, but I have no idea if the resolution from the etching phase will be sufficient.  In the past, I have had a hard time with small traces washing right away.  But, I think that’s an error in my photo-exposure process, not the chem bath.  That’s why I’m gonna try with this test.  I have my AD 654 Adapter (so I can breadboard with it), the RGB LED headers, and the MCU SMD board for my Honeycomb light all put onto a sheet that I’ll run through the process.  I might fail.

However it turns out, I’ll find out what capabilities I *do* have after it’s done.  And, perhaps, I’ll see where the bottleneck is and be able to overcome, we’ll see.  But, it could be awesome and hold electrical integrity just fine!  I hope that’ll be the case.  Then, I’ll be able to pump out usable products from this run.

Comparing Circuit PCB Components

At the top of this pic, I put in my old RGB LED header for comparison which is twice as big as the ones I have slated to make.  That’s good.  These will be nice and tidy.  Also, you can see in the pic, next to the quarter, the tiny TQFP chip that I have to be able to make precision traces for and be able to solder onto it.  And, then there are the SOIC-8 parts at the top left to go onto the other boards.  Most of the passives are 1206, but a few are 0805’s.  I bunched the rest of the packaging next to it because it sets the scene 🙂 .

I’m glad I did a test print on paper because the first one was crappy resolution @ 72 DPI (fixed it), and also, in this one, I left one of the soldermask layers on the design (fixed that, too).   BTW, the moisture test (Cobalt Dichloride Free) turned pink in a matter of minutes taking it out of the package, but was blue just before.  I’m not going to reflow these, but it’s interesting to see the effects, especially here in Texas.  Shipped from Mouser, also here in Texas.  It’s humid today – 89%, reported.

Comparing SMD Components

Anyeeewayees… We’ll see how everything comes together.  I see it like this: If I’m able to fabricate these SMD boards here, more power to me for testing prototypes with small surface-mount components – mainly the chip leads.  But, if I cannot get consistent results, then I’ll either go back to prototyping with through-hole parts, improve my fab process, or just get them made when I need ’em.  But, I *do* wanna find out if I have the resolution necessary, in-house.  Exciting!  I’ll report on what I find.

Honeycomb Light First Circuit BradOrmand

04.08.2015 – Synth, Job & Lights

New, in the sphere of my workings is a job that I will be starting at soon as a Sr. Software Engineer.  The recruiting process was long and pretty taxing due to just the management of all of the leads and people to stay in close contact with over the last few weeks.  My goal was to find a “match”, and that’s what I found.  The people are great, the job calls for what I have to offer, and it appeals to me as a developer, as well.  I’m ultra-excited about it!

SYNTH

In other news, I ordered some components to start building a real, accurate, and high-potential analog synth module!  I have the basics spec’d out and ready for a test setup.   So far, it’s going to feature the Analog Devices 654 Voltage-to-Frequency converter as the main oscillators and LFO, and Texas Instruments Active Filters throughout.  I just received them in today.

Synth Keys Idea - Brad Ormand

I posted my thoughts on an idea for a frequency mangler / modulator using digital logic (as an analog signal effect) a while back, and after I get them running, I’ll try to implement that idea with these as I go along.  It’ll produce a kind of ring-modulated effect.  So, there’s plenty to do before I get it up and running as I have to design circuits around them, but I imagine I’ll have something going pretty soon.

I hung out with friends and family over the Easter holiday, and I probed some mechanically-minded acquaintances’ minds about putting together a wooden keyboard (with my electronics inside) to hook into the synth module.  I got plenty of ideas.  I’m visualizing a few scenarios: One, as an all-in-one solution for peeps to just get and start playing as a standalone fun-toy, and two, as a separate controller for hooking into various modules.  I’ll keep that on the back burner, but for now, I just wanna build something that works and makes awesome sound.  The first synth will be on the breadboard, and the first keyboard will probably just be a crazy in-shop, spring-loaded array of chopped wood hitting tac switches.

LIGHTS

Honeycomb Light First Circuit BradOrmand

And, for the lights…  The thing I was waiting for was to write the software to be as fast as possible with the on-chip oscillator, and to see if it’d be fast enough for my scanning.  Well, I did that, and it’s not.  It falls short by at least a few MHz.  So, I have now, in-stock, an external oscillator that will push the MCU to 20MHz, and I’m waiting to see what kind of stability I can get with that.  As the experiments go forward, I’ll know if the solution I have already designed will work.  One more home-made PCB test, and I’ll have the answer.

Now that I have cleared the schedule a bit, I can begin to develop more projects.

 

Brad Ormand HoneycombLight Main MCU Board

03.25.2015 – Main MCU Board & ComCath RGB LED MaTRIX

After hours and hours of taxing my brain for what approach to use to streamline the “guts” of the HoneycombLight (for real – it deadened me for a bit from getting stuck several times and trying to overcome these obstacles.  I lost sleep.), I finally came to a conclusion for the architecture of the matrix routing (and then I was ok).

MATRIX BOARD

So, I figured, I’ll just place daughter boards, matrix mixer boards, every 4 pixels and let the pixel driver boards run straight to it, 4 at a time, raw, (R, G, B, and cathode) instead of “mixing upstream” by soldering wires together right on the drivers (like I did last time).  Mix as I go – 4 times downstream to the final MCU – it’ll all be prepared once it arrives there.   …Because the entire thing is really messy if done with twisting or soldering wires together onto connectors at the pixels, believe me.  It’s a “grid”, a Matrix, a mess.  I really needed the double-sided planes, here.  I did it in two “phases” (those 9 via holes in the center), alternating top and bottom copper.  Reminds me of the butterflies in Fast Fourier Transforms, but physical.

This allows the entire path to the MCU boards to be nice and clean with dedicated plugs for easy service and replacement, and I admit, also for the cleanliness of the way it looks – fresh, not cluttered.  I like that. 🙂  That was one of the tradeoffs I made.  The way that “under the hood” looks is important to me.

Brad Ormand RGB LED Matrix1x4 MixerBoard

MCU BOARD

And, as for the MCU board, most of its architecture is the same as the first through-hole rendition, but it’s just double-sided, solder-masked, and silk-screened, now.  Loads of improvement!  But, the concept is the same…  However, I can’t have it looking like it’s still a prototype.  (I have already built that and it worked (with wires all over the place), so now it’s time to do it for real).  So, every component is now surface-mount, minus the row and column connectors (for rigidity).  It’s not the *final*, final version – I have already thought of things that I’ll have to do to revise it, but it’s a good start.

I am yet to send it off to be manufactured, so I haven’t tested it yet, but I’m really excited to get a small batch back and run it through the ringer!  I mean, I’m almost as excited about how it’s going to *look* as I am about how it’s going to work!  Haha – to me, it’s like a work of art!  Especially with that ground plane in there surrounding the traces – looks dope.

Brad Ormand HoneycombLight Main MCU Board

I don’t know if putting “BradOrmand.com” on there is “forced” or not, actually.  I want a “maker’s mark” on there, but am still deciding if that’s the direction I wanna go or not.  I might put in a bitmap with my logo or even transfer it over to the company name I’m thinking about using for the marketing of the lights, etc.  I don’t know, yet.  Decisions, decisions.

All-in-all, though, this board was weeks in the making and took a lot out of me.  I didn’t think I could do it there for a while…  But, I was also kind of stressed out with all of the Portfolio and job stuff, recently.  But, I have it quite in-hand, now.  Let’s hope the electrical connections all check out and that those tiny-ass traces can handle the current I’ll give it.  It’ll be multiplexed, but still only milliseconds between blasts.  We’ll see.  This is all part of the learning.  I’m determined to do this.

FINAL THOUGHTS

Last thing…  The main thing I am concerned about right now with this is soldering the central ground pad underneath the 0.8mm pin pitch TQPF without a reflow oven or a good heat station.  I don’t know how that’ll work.  I’ll be thinking about it.  In fact, I think it has solder-mask under it right now.  I’m going to have to investigate, but the datasheet *does* recommend tying it to ground.  Onward we go…

 

Hex Light - Front Assembly - Brad Ormand

01.18.2015 – Hex Light Wiring

I made some progress with the wiring for the Hex Light.  I have ’em set up in Common-Cathode-Column configuration, with RGB rows, and started the wiring on the art itself (like I said before, instead of routing all of this on the circuit board 🙂 ).  There are 16 mini-PCBs mounted to house the light, the passives, and the connector headers.

Hex Light - Mini-PCBs on the art - Brad Ormand

 

After a few days of work, and with some in-situation design, I got the whole board wired and connected up to the main MCU board.  I continuity tested all of the connections, and lit each up with voltage at their respective addresses. I found a short and some mechanical failure, fixed those, and then it tested out perfectly.  I think it’s go for a test with the MCU!   But, that’ll be for another day – Whew!  I’m beat…

Once I test it out with the whole system, I’ll get the wires all harnessed correctly, mount the main PCB, put some standoffs on there, clean it up a bit, and start experimenting with writing nice animation programs.  I programmed a test pattern, flashing each on and off individually every 200ms, but I’m yet to hook it up to the real system –  I did that from a breadboard with a sub-sample of 12 lights.  On the system, there are 48.  We’ll, see if it scales up…

By doing all of this, I clearly see that I could optimize several things – from wire routes, to PCB size and routing, to channels for each harness, etc.  Just had to set this prototype up first to see where I’m gonna have to take it in the future.  Gonna make a lot more. 😉

Hex Light - Initial Wiring - Brad Ormand

But, yeah – despite the ideas for optimization and all of the underlying engineering stuff – I don’t get too hung up on it… I just want to create awesome art and programs and fun products!  In the end, the kinks will work themselves out if I just treat it all with love, care, and patience.

… And, the front view…  ( looks a lot better than the back – Haha )

Hex Light - Front Assembly - Brad Ormand

 

Hex Light Chassis and PCB Transparency - Brad Ormand

01.13.2015 – Hex Light

The Hex Light is in the spotlight today.  I mentioned that I ran the RGB LED CC boards through and that I had fun doing it.  And now, I have assembled 12 of 16.  It takes a bit of time.  Plus I have job stuff all day to do…

Hex Light Chassis and PCB Transparency - Brad Ormand

Hex Light RGB LED CC PCBs - Brad Ormand

I cut ’em and drilled them.  My smallest drill was actually too big to accommodate what I needed.  It was about a millimeter and a half, but I’m gonna need about a millimeter.  I did have a circular end bit that was about a half mm, but that was too small.  So, I’ll need to reconfigure my tooling.  I need the precise size because soldering to holes with no copper pads is horrible (because the too-big drill bit cut right through the pads), and not being able to get the component leads through is also horrible because.. well, they can’t get mounted at all! 😉

So, another trip to the Depot is warranted.  I have a list.  Plus, I love my trips there. Just wanted to say…  It’s probably my favorite place to go (And, then I hit up the grocery store next door for that grubbbz, yum – tacos, lettuce, noodles, sparkling water, pears, beef jerky – all an essential part of the dev process for damn sure 🙂 ).

Soldering the RGB LED CC Boards - Brad Ormand

And then, I soldered these small boards – 12 of them of 16, and stuck them on the Hex Light.  Next will be getting the remaining 4 (when I get the time) and then testing the shit out of all of them.  And then, then next step would be, of course, applying the main board that I fabricated a few days ago to the entire install.

I have almost got the firmware written already, but I’ll need a few more tests and tweaks.  It builds on everything I have ever done before of this nature.  I’m going to make it sing and, yeah, of course, shine. 😉 (light)

I’ll get the remaining boards and apply them to the chassis in the coming days.  I’ll be nearly 80% there.  Once I get all of this rolling, I’ll have the beta to move on to determining a good way to smooth out the entire process for future builds.  But, for now, it’s going intensely well, and I’m lovin’ it, mon.  Irie!  Life is good in the shop.

 

Brad Ormand PCB Fab REV A for the Hex Light

01.11.2015 – Hex Light & PCB Fab

Wow – I have done quite a lot of work on the Hex Light since the 6th.  I’ve managed to cut and assemble the top pieces, paint the backing board, I designed circuit boards for the whole thing, started manufacturing them, and then testing and assembling them.

I spent a few days designing the main board as I wanted to get as close to production values as possible.  I still have a few things to work out (like drill bit size and some trace widths ), but I had to go through it to see what errors would pop up.  Now I know, and now I can move forward.

Brad Ormand Designing a Circuit

Next, I went ahead and set up the manufacturing process in the workshop.  I have always had all of the parts and chemistry for it, but haven’t set it all up in years.  But, now it’s rolling again for board fab.  I printed the traces and pads out to a transparency,  exposed the photo-sensitive board to fluorescent light through the traces, removed the resist with the photo-developer, and etched the copper right off the fiberglass with ease.  Went very smoothly.  All of the techniques I used to do are coming back to me – like how long to agitate the board in the Ferric Chloride bath and how crucial it is for the transparent ink to lay *exactly* flat when being exposed.

Brad Ormand Circuit Board PCB Fabrication

Then, I drilled and populated the board with the components I specified.  There were a few complications along the way, but I was prepared and I adapted.  I tested that board up and down, with and without power.  And, in the coming days, I’ll take the rest of the daughter boards through the exact same process and test and populate them as well.  Then, it’ll be system hook-up time.  The moments of truth.  🙂

I also managed to put a hex design on the PCB and experimented with a few trace widths and logos and labeling.

Brad Ormand PCB Fab REV A for the Hex Light

 

The breadboard had all of that clutter on it, hooked up in a small space (+the 1×4 Matrix), and so half of that clutter was reduced to the clean board, half the size – Haha.  Cool to see a comparison and nice clean header connectors instead of a nest of wires.

Brad Ormand - Circuit Comparison

 

Sure, there are a few things I am going to do differently for the final version, but this setup is functional and will allow me to test the system in-situ.  And, when I go back to design the next one, I’ll know just what to do.  I know I can get a little closer with the traces, I can go bigger on some of the labeling, and I have to get a smaller drill bit for some of those holes- whew! I’m already drilling pretty small.  And, of course, down the road, I can start on some of the SMD versions, single-sided, for now.  If I get a really rocking design, I’ll double side it and send them off for fab with silkscreen and everything.  But, that’ll be when I come up with a rock solid REV A or REV B.

So, stepping back a few hundred meters….  I’m making this piece of art (that happens to have electronics in it).  And, I want it to look and feel and act right, first and foremost.  So, that’s a “project requirement”, but also, the engineering side of me wants the parts to be replaceable and serviceable and modular, too.  I’m trying to make the craftsmanship very tidy and tight, but while also making it cost-effective and modular.  Haha – I’m feeling the pull between the two.  But…  That’s the game.  That’s what I like about it.  It’s engaging and requires some thought, over time.  And, it’s something I want to become as expert as possible in.  But, all-in-all, I’m delighted to switch back and forth between roles and processes – getting both the art stuff and the tech stuff in.  I require them both.  It has really been a great exercise in balancing those points.

Until next time…

Brad Ormand - ARM 0.5mm pin pitch soldered and tested

11.22.2014 – Potluck & ARM Cortex

The day before yesterday was a social day.  I made Pico de Gallo from scratch for a Potluck I went to.  And, the conversations were flowing.  We talked about all kinds of stuff and ate, and, went back for more.  It was with a lot of people from work, and it was great to be able to just chill out and talk about our personal stuff for a while.  So, lots going on that day.  Didn’t work on projects much that day, but had a lot of needed social fun.

The Pico de Gallo was amazing and I want to make it again for Thanksgiving, if possible.  And, it was fun to make.  I have a slight tweak in the recipe that I want to try.  I guess I treat cooking creation just like I would do any other project.  Iteratively refining.  🙂

Pico de Gallo - Brad Ormand

 

And in other news, I did end up soldering that ARM chip in juuust fine.  Oh my god, though – it was a big challenge!  Yeah, please excuse my amazement if you are a veteran eyelash-pin-pitch solderer, but I had a bit of trouble starting up.  I had never gone *that* small before!

Brad Ormand - ARM 0.5mm pin pitch soldered and tested

Ok, enough of the drama…  I did it.  It’s done.  I checked the continuity of every inner lead and outer lead (board connectors) individually, and it’s fine 🙂  I’m getting the development  toolchain ready to go, and I’ll be on my way to (finally) doing some more coding again, and to hop back on that FFT testing with my wonderful array of displays.  Love it!

So, it’ll take me a few days to get all of this going.  Also, there are a lot of port names I have to learn, and the pinout of the chip has to be mapped to JTAG and so on.  I have to do the ground work.  Yeah, I could buy a ready-made development board, but nah – that’s not what I’m doing this game for.  I want to learn how to build around a chip in my projects.  There’s a chip – it’s pinouts are like this, and its inputs require filtering and it gets x-and-x regulated power and outputs to this other stage that I need to amplify, etc, etc.  That’s the fun part. 🙂  And, plus, I have specific space-sensitive applications for this development and I want to make applications that actually embed tightly in cases…  So, the getting used to devving with the dev boards is convenient, and you don’t have to (or maybe don’t *get* to) reinforce those skills and practicalities of like soldering the chip, pinning out the JTAG, etc, but that’s where the product space is at.

Anyway, so that’s the direction I want to keep going in.

Brad Ormand Strat

11.19.2014 – Songs, ARM & Strat

Ugh.  Major Fail.  I tried soldering my ARM chip to a board that I wanted to fit on my breadboard and I nearly stuck all of the pins together.  It’s my first venture into 0.5 millimeter pin pitch soldering, and I sucked very badly at it.  But, I haven’t quit, yet 🙂 .

Brad Ormand - Botched ARM 0.5mm chip

Yeah, I hesitate to even show this pic at all since this is public, but…  Oh well, I started this whole thing to detail out the progress and happenstance events of my various journeys into art and engineering, and well…  The fails are part of it, I guess.  But….  It’s like soldering eyelashes together!!  Like connecting the ridges of a dime!!  (rant)  Whew, okay… Hey, I’m a software guy (cop out).  I mean, I’m just a musician (cop out).  Damn.  Ok, I’m not very good at this yet.  For real. Dammit.

Ok, moving on (quickly).  I finally have my guitar tuned inside and out!  I, of course, put those new strings on there, and adjusted the bridge and neck so that the strings would be as close to the neck as possible, and retuned like 12 times, and finally came out with something stable.  It plays better than ever!  I mean, one of the things I didn’t like about this Strat (’91 vintage, USA) was that the strings stood off of the neck quite a bit more than an Ibanez metal guitar or Les Paul (which are other guitars that I like 🙂 ).  But, this Strat is my primary axe and my favorite.

Brad Ormand Strat

So, now, I’m working on transcribing out the solo I want to accomplish on the guitar out into musical notation from MIDI screenshots so that I have a good view of it, and can read off of it while I’m learning it.  I’m starting with the first 2 bars and then I’ll learn it 2 bars at a time, from the beginning of it to the place I’m at.  Keep in mind that I wrote this solo, and it has already been recorded.  But, I used an instrument other than the guitar to capture and post-process the final solo.  I used the Roland JP-8000 to assist with that 🙂

And, speaking of the songs, I did get a chance to mix down “Fine” and “Kitchen” with a new kick drum arrangement, for ’em both.  They are massively punchy, and I am incredibly glad!  There’s not much to do left.  The mixdown session well.  The testing session went well.  It’s just failing less and less.  Which is a good thing.

And….  back to the ARM chip, if you’re still with me.  Okay, yeah, uhm, I will get out a heat gun to desolder that chip, and really, I shoulda used that in the first place to solder it.  I just had too big of a solder tip and not enough flux when I did it.  I’m going to have to get a smaller tip and a flux pen or something to start soldering that kind of shit, manual.  I’ll work it out. The pins are sooooo small!!  I keep thinking: “eyelashes”!

And, I’m doing these projects fresh off of work, most days, where I’m still somewhat thinking about our project iterations there, and next to my social life where I have several motivating things to think about right now.  So, it’s hard to switch over, I find. …Okay, I might be making more excuses…  However, I do have some vacation time in about a week, and I do want to lay out some time between the holiday outings to get me and 0.5mm ARM on good terms. And, I’ll launch this blog as well during that time.  Music player and painting gallery are almost finished..

Eventually, I’d like to be able to get this manual small-pitch soldering down.  It’s not like slipping a chip into a DIP socket or through-hole shit – it’s a whole different ball game.  People have become awesome at this shit, and I know some of these tiny-part, microscope-soldering guys, personally – and they brag about it, too.  And, now I know why.  Props to them.  I know now.  LOL.  But, my experiment went haywire, and I learned a few good lessons.  Cheers to the new revolution of tiny eyelash-spaced parts!!! YaY. (mumbled facetiousness).