-\\ Completion - and it works! (04.01.2010)

---

I took on this project to investigate how a DC motor really works - beyond the vague, textbook descriptions which say something about coils of copper wire and magnets. In the process, I learned how to cast aluminum at home. That was an adventure all it's own starting with a home-made, charcoal-fired furnace, home-mixed casting sand and a mold-making process that was entirely new to me (sand casting). I purchased and learned how to use a metal lathe, read about and tried many new shop techniques and read a fair bit of interesting history - not to mention quite a bit about motors, how they work, how they are built and a lot of other related information about electromagnets, inductors and electricy in general, and even a bit of geometry and handy shop math.

Below: The last step was the brush assembly, which is a pretty tricky little piece. It has to be able to insulate the two brushes from eachother, which is why it is made of wood. I also opted to made it adjustable - the whole unit can be rotated around the axis of the armature to find the optimal timing/advance on the commutator - so you can't just bolt it in. It has to be made as a unit unto itself, which slip-fits into the rest of the motor and can be loosened, adjusted, and then tightened as the motor is running. Add to that the tendancy of spring-loaded parts to fly across the room and you could say that designing/fabricating/installing this part was more challenging than it looks.

A fixed brush assembly would probably work fairly well if you happened to nail the position just right. But I found that I can dramatically improve the performance (speed) of the motor by making very small adjustments to the exact position (rotation around the commutator) of the brush assembly. Also, you'll note in the photo that the brushes are both attached to the same side of the wooden ring and arc towards each other almost in an "A" shape. I think this was a bad idea. They don't contact the commutator at points exactly 180 degrees apart from each other. Next time I'm going to attach one brush at the opposite side (180 degrees apart) of the wooden ring so that even if I miscalculate the arc of the brush, the points of contact with the commutator will be perfectly 180 degrees apart. As it is, my brushes contact the commutator at positions of 10 o'clock and 2 o'clock. I imagine that if I made this adjustment and got them closer to 9 o'clock and 3 o'clock there would likely be a dramatic improvement in the performance/speed/efficiency of the motor. I suppose I'll have to give it a try.

-\\ LESSONS FOR NEXT TIME (04.02.2010)

---

Although it runs pretty well - even a fair bit of torque - I'd do a few things differently next time, mostly:

• More powerful stator magnets, probably electromagnets, with a more comlpete magnetic circuit.
• Ball bushings to easily prevent the armature shaft from binding. As it is when you bolt down the bearing blocks (end plates) the shaft binds unless you fiddle with it for way too long.
• I'll probably use ball bearings on the next one, but still need to address the binding issue.
• The brush assembly needs improvement - better alignment of tangent points (they are not really 180 degrees apart on this one). And maybe leaf springs instead of coil for less hassle.
• I might try a 6-pole armature next time.
• More magnet wire in the coils, they could be twice as big, and more neatly wound - maybe using the lathe for tidy winding.

-\\ IGNITE UPDATE (03.05.2010)

---

Here is the list of links & leads, as promised during my talk at Seattle's Ignite 9.

- BackyardMetalCasting.com (All the info you need to setup your home aluminum foundry.)

- MetalCastingZone.com (This is a fairly active forum with lots of opportunity to Q&A)

- How to Cast Small Metal and Rubber Parts (Good intro into sand casting)

- Secrets of Green-Sand Casting (A very old book, but full of useful info.)

- The Complete Handbook of Sand Casting (This book should have "Bible" in the title.)

- The Complete Metalsmith (Casting, soldering & more for jewelers - silver, gold, etc.)

- Ignite Seattle (In case you don't know what Ignite is in the first place.)

- Phinney Bischoff Design House (Where I work in graphic design)

- Metrix Create Space (Broadway, Seattle. They have Makerbots & laser cutting!)

- Suggested Reading (At the bottom of this page is more related reading)

Drawings & plans (01.14.2010)

---

All of these parts and process photos are pretty meaningless out of context, so I fancied up a couple of my sketches to show the big picture a little bit better. For the most part, I am sticking to my plans as they were originally drawn on paper. However a few details change once you get your hands on real metal/wood/etc.

Above: This is the big idea. It's an electric (DC) motor. The desgin is entirely my own, based on basic DC motor principals. In this drawing you can't see the (permanent) stator magnets or the brush assembly because they would obscure the view of the armature/shaft, and this drawing was mostly to workout the details of the shaft.

Below: I designed the armature magnets around a standard hex nut. Magnet cores have to be steel/iron, so I used readily-available steel bolts from the hardware store, and brazed them onto a steel nut. As yet, I have not wound the magnet wire around the cores, but hope to do so soon. In this drawing I left the wire off one of the bolts so you can see what it looks like underneath. All three will be wound with wire.

An Apparatus for Self-Acting Mechanical Rotation through Electromotive Force

---

Currently, I'm building a smallish, permanent-magnet, DC motor 100% from scratch. The joy of this project is the combination of disciplines involved. I've been at it for a few weeks and have enjoyed some woodworking, metal casting, machine shop tool use, geometry, electromagnets and more. I'm doing it because it's fun. A lot of friends wonder if I want to have a motor for some purpose like to power a fan or pump or something. That is not the case. If I simply wanted to have a motor, I would buy one and save myself countless hours of reading, testing, trying out, failing, sourcing materials, buying tools, etc. For this project I do not want to have a motor. I want to make one. It's a hands-on study of a device that fascinates me. It's a kinetic sculpture. It's a chance to play with tools and explore some modern and bygone manufacturing and shop techniques that interest me.

I'm more or less fascinated with the development of technology through history - and life as we know it - especially those pivotal inventions that changed the course of civilization. It wasn't very long ago that the world was transformed by electricity, steam power, chemicals, plumbing, sanitation and other scientific advances. For 99.9% of human history we lived without these things. I'd like to explore some of them in their primitive forms, hands-on. In short, I'm a huge nerd and this is one of the ways I have fun.

Casting the Bearing Blocks (01.02.2010)

---

A few days ago I completed my first successful sand casting using aluminum. I've cast silver, gold and copper before using the "lost wax" technique and plaster/investment molds, but never using a sand mold. I feel this is a breakthrough for the advantages of making sand molds: They are cheaper, more quickly made and your original pattern is re-usable not destroyed (that's the "lost" in "lost wax"). Also patterns can be made of virtually anything including wood, metal, high-desity stryofoam, etc - which really opens up the possibilities. And I can make much bigger parts now, than my little jewelry casting setup would allow.

Below: I began my pattern on the lathe. Wood machines well on the metal lathe, which is nice for making smooth/flat/precise patterns for casting.

Below: Here is the almost-finished pattern. This part will be one of two exact duplicates set back to back, as the main housing for the motor armature. The part is designed with tapers in a few certain dimensions, so that it can be easily removed from the mold.

Below: Here the wood pattern (now painted) is sitting in the mold frame, awaiting sand packing around it. After that step, the mold is split open (along the seam being held together by the latches), and the pattern is removed, leaving and cavity into which the molten aluminum will be poured. I'm melting the metal in a furnace I made at home. You can find plenty of related info if you do a web search for "flower pot furnace". Odds are, you will find this great site: Backyard Metal Casting.

Below: This is after the top of the mold has been filled with sand and compacted firmly with the end of a stick. Then the mold is opened and the pattern is carefully removed using a wood screw which is inserted in a pre-existing hole in the bottom of the pattern. This leaves a nice mold cavity, ready for hot metal. In this one you can see words imprinted in the mold, which were added to the pattern after the other photos were taken using puffy, stick-on letters from a craft store. The white stuff is talcum powder used as "parting dust" to prevent the damp clay/sand mixture from sticking to the pattern.

Frequently Asked Question: Doesn't the sand fall apart after you remove the pattern? Is it solidified? Crumbly? Answer: It's not solid or crumbly, it's a tightly packed, damp sand/clay mixture, like a sandcastle is made from - only a touch more clay in the mix. Too little clay and the sand crumbles, won't stick together. Too much clay and the sand is not permeable by steam, so when you pour in the hot metal, the steam will blow your mold to smithereens or retain air/steam bubbles. They say that 5-10% clay is the right amount. I couldn't get the mixture to stick together until I added about 12% clay.... judging by the cups-full that I was mixing. Then you really ram it it all together with the end of a stick, so the sand/clay mixture is packed as tightly as possible.

Below: Here's my new home-made furnace lighting up the neighborhood with a crucible full of molten metal inside. More pics of this farther below.

Below: Pouring the hot, liquid metal into the mold. Immediately after this, I am beside myself, wondering what's inside. I let it cool as long as I could stand (probably about 5 minutes) before opening the mold to look inside.

Below: This is the casting, fresh out of the mold - it worked! This is my first attempt at sand casting, and very exciting. There is plenty of roughness to work on, but not bad for the very first test run. Again, this one doesn't have the words because the words were on attempt #2, and this pic shows attempt #1. I'm fine tuning the process now...

Below: A little grinding and most of the ugly stuff is gone. I have a few tactics in mind to make the next casting a little cleaner in the first place - although that rough sand-cast texture is not objectionable, that's part of it, and I think it's where "Hammerite" spray paint takes inspiration for their paint texture. I like the texture, but could do without the big, chunky lumps at the inside of the circle where the sand stuck to the pattern.

An Inspirational Book (01.05.2010)

---

Part of the joy of this project has been reading one book in particular, which was originally published in 1903, which is called Dynamos and Electric Motors: How to Make and Run Them. This book is a fascinating look into the technology and mindset of the time, which at times seems surprisingly advanced and at times seems pretty backward. Here are a few pages from the book.

Below: This is the same basic commutator design that I am using, except that this one is for a two-pole motor. Maybe three-pole motors weren't common at that time. I'm not sure. I really enjoy the materials and processes they detail in the text: silk ribbon, melted paraffin, lamp black, real state-of-the-art stuff in 1903!

Below: They really go into detail about a lot of stuff we take for granted these days, like these "small clamps, sometimes called connectors"

Below: This book has some really practical information for the novice motor designer. You can't even hardly find much of this information on the internet today. I guess nobody is building motors at home these days, you just buy 'em. Maybe in 1903 this was a much more useful book for the average guy-who-needs-a-motor.

Armature & Commutator (01.05.2010)

---

Last night I began the main armature shaft and the commutator. This isn't yet complete, but you can see the beginning of the process here. The fun part is finding existing materials that can be re-worked into something new, as you will see below...

Below: The three poles on a tri-pole motor must be spaced perfectly evenly from each other for balance and proper function. I decided to start with a standard hex nut as a base since that comes with 6 evenly spaced sides. Using every other flat surface as a base for my electromagnetic poles will give me a great head start on even spacing. To those three flats will be welded some modified steel bolts as magnet cores. We'll get to that later. To start, I drilled out the threads, leaving a perfect 0.25" hole centered in the nut.

Below: That nut was a perfect (very tight) press fit onto a piece of 0.25" steel rod. I had to heat the nut and hammer it into place, it's not going anywhere.

Brazing the magnet cores (01.06.2010)

---

Last night I finally assembled the steel electromagnet cores and installed them onto the armature shaft. This is the same shaft as shown above.

Below: I found some cool bolts with contoured heads, so I chose those as my magnet cores. An electromagnet core has to be steel (iron), otherwise I would have cast something myself. Since I can't cast steel (yet), I'm fabricating this part. I cut off the bolts and ground their faces at 90 degrees. In the background is the shaft and nut, as assembled yesterday.

Below: Photos with fire are the most fun, so I had to include this. I setup the bolts, one-at-a-time in a sort of vise/clamp jig to hold them while welding. I'm finding that for most work, it's the preparation that matters most. You might spend more time preparing than actually "doing" - but that's how you do a good job. I fiddled with each bolt forever getting them aligned perfectly in the clamp. The brazing took only seconds. The same concept has applied to every part of this project so far.

Below: Here is the finished steel rotor (since this photo, I re-fitted the nut-and-core to a longer shaft). Onto the bolts will be wound matching lengths of solid copper wire to form the electromagnets. Onto the long end of the shaft will slide some spacers and the finished commutator, which is currently awaiting some tiny (#1) brass screws ordered on Amazon. I am happy (and amazed) to report that the spacing between the bold heads, edge-to-edge, is perfectly consistent. As is the "tilt" measured from the end of the shaft to the top of the bolt. I'm really using those little dividers, which I ordered randomly because they looked interesting and were cheap. I never knew how handy they would be.

Fabricating the Commutator (01.11.2010)

---

Over the weekend I more or less completed the commutator, which is an under-appreciated little piece of genius, if you ask me - if only for it's simplicity. It's the commutator, with the brushes, that gets the "juice" to the magnet windings - perfectly switched every revolution, tens of thousands of revolutions per minute in most cases. I have some rough brush designs in my head right now, and am very much looking forward to starting on those. For now, the commutator:

Below: To begin the commutator, I press-fit a copper pipe cap over a 0.75" oak dowel - nice tight fit. So far so good.

Below: Dowel and pipe are trimmed together on the lathe, and the ends are faced flush. It's important to note that the "cap" end of the copper pipe was trimmed off, so this is just a ring of copper now, like a section of regular pipe. Both ends look like this:

Below: Here's the commutator, center-drilled and test fit to a mock armature shaft. Looking good, so far.

Below: To continue work on this, the commutator is removed from the mock armature shaft and mounted to a threaded spindle temporarily. Here I have drilled holes for 6 screws, which will hold the segments of copper pipe to the dowel after they have been slit/separated from each other (see two pics down). Two screws per segment. After this step, the holes were countersunk, threaded, and fitted with tiny (#2) countersunk machine screws. Once the screws were properly seated, I ground off their heads flush and smoothed the whole thing on the lathe again.

Below: Here the commutator-to-be is in a cross-slide vise (an awesome tool). The slitting saw is in a drill press. Using the cross slide allowed me to slit this in a perfect, straight line, and advance the cut slowly to avoid binding. That old book (above) suggests using a hacksaw for this operation. That was in 1903. This is much cleaner.

Below: Here is the finished commutator, it spins true, and has been tested with a multi-meter. The three plates are electrically isolated from each other. After the magnet coils are wound, the ends of the coil wires will be soldered to these plates (one coil bridging each gap between the three copper commutator plates). Also, here you can just make out the heads of three of those six brass screws discussed above.

Thought for the Day (01.11.2010)

---

It amazes me how much time, energy, effort and tooling this project requires - especially when one considers when/how this amazing little device (the electric motor) was first invented. Any inventor must have first been a master of shop practices and had quite a workshop in which to create. I've used just about every tool at my disposal so far in this little project, and had to build quite a few along the way - especially in regard to casting equipment - from the furnace itself to mixing the sand, etc. You can hardly buy half of this stuff. And many of the tools I have used are powered by electric motors, so tell me this - what kind of tools were they using to build the first electric motors?? I've come across treadle-powered tools in my reading and research, but it must have been 100 times the effort. Amazing. It's hard enough to pull this together with all of the handy tools and materials I can buy at Home Depot and order on the internet. Imagine pulling this off - for the first time ever - in 1850 or so. Amazing. It must have taken a year (and a lot of resolve) to just build the first one, even after the concept was established.

The model (01.12.2010)

---

How you make the "master"" model (AKA pattern) for your casting is very important. I made this one out of wood, as detailed above. It's best to avoid parallel geometry, angles and tapers are better, oriented in such a way that you can pull the pattern from the sand smoothly, without breaking the mold. This is called giving the part a "draft". You have to design the part knowing how it will sit in the sand/mold. After sorting the angles out and assembling the part, I filleted the seems with putty and glue to help the sand release from those tight spaces and then painted the whole thing to give it a smooth overall texture. Finally, I found some puffy letter stickers at a craft store, to see if I could add a little detail. After those were applied, I shellaced the whole thing to make it as slick as possible to further ensure that it won't stick to the casting sand. It just so happens that the puffy letters I found did not contain numerals, so I had to improvise and display the year in Roman Numerals.

Casting update (01.13.2010)

---

Last night's casting was almost perfect! Very little grinding/finishing required and a nice sand texture on all sides. There is still one small hole, but I think I know how to eliminate it next time. Lot's of learning this time through.

I also slimmed up the pattern this time around. The previous castings were very heavy/thick. In the process of removing material off the back of my pattern, I inadvertently removed the nice little hole that I had used to twist in a wood screw when removing the pattern from the fresh sand mold. Now, without that screw to hang on to, it's difficult to get the pattern out of the mold. That little hole is more important that I thought. In the books, they call that the "draw screw" - to draw the pattern out of the mold, I guess.

Below: You can see where the metal is poured in, via the sprue (which is cut off during finishing). This is not only how you get the metal into the mold, but it is an important part of the process for a successful casting for other reasons. Look at the top of the sprue (left side of photo), where the metal looks like it as been sucked in on itself. It has. As the hot, liquid metal in the middle of the mold cools, it contracts and it needs a reservoir of molten metal to draw from. If it doesn't have this metal to suck into the contracting casting, it will shrink and look like it imploded on itself when you take it from the mold, ruining the casting. After you pour the hot metal, you can watch as that little puddle of metal sucks slowly into the mold.

Below: The previous flaws in the back of the casting (lots of huge bubble holes, shrinkage, etc) are 100% solved in this one. I'm almost certain that's because in the previous castings, I was using the cement floor of the garage as the back half of the mold - convenient, but not permeable. This time I used a solid sand backing - permeable so the hot steam/gasses can escape.

Below: Contrast that with some of the previous attempts. These are the worst, full of bubbles, shrinkage and all kinds of problems.

Casting update (01.19.2010)

---

Thus far, I've been melting scrap "pot metal" (a cheap alloy of Zinc, Tin, Aluminum and who-knows-what) on a hotplate - which can reach the required 700 or so degrees (F) to melt that kind of metal. Now I'm moving forward, and have successfully melted and cast more-or-less pure Aluminum for the first time, which happens at about 1200 degrees F. I'm doing this in my home-made, air-injection, charcoal-fueled furnace. Tonight I hope to reach the 1600 degrees required for brass, and will attempt to cast that.

Below: Here is the furnace that I built from junk. It's constructed of some very cheaply-obtained steel ventilation pipes, and insulated with a cement-perlite mixture. All told, cost about $40. The pipe on the side is where air is injected, which is required to heat the (standard BBQ) charcoals way beyond typical BBQ temperatures. Air is supplied via a once-discarded hair dryer blower.

Below: After about 20 minutes (just enough time to pack a sand mold), the embers are glowing and the aluminum is melting. I poured this quite successfully and made my best bearing block casting yet.

Casting update (01.20.2010)

---

Below: Last night's melt was amazing. Using the more typical "modern" hair dryer, I more than doubled the air flowing into the furnace, and reached what I believe to be about 1700 degrees Farenheit. The brass melted, and so did a lot of other miserable impurities and fillers. That's what I get for using brass from a thrift store candle holder. Next time I'll use good metal. Meanwhile, here are some pics of the hottest melt yet, and my new pipe-nipple crucible which performed nobly amongst all the heat.

Below: Grabbing the crucible with fireplace tongs. I believe the yellow light/gasses/smoke coming out of the crucible are the result of lots of unknown impurities and filler in the cheap, commercial casting metal which was probably made to look like brass, but likely contained all sorts of other nonsense.

Thought for the Day (01.25.2010)

---

I think I figured out the common thread here which unites the appeal of casting metal, making electricity, magnets and motors, etc: I'm fascinated by the roots of technology from the very beginning. Specifically that point where a guy walking in the woods with no tools could start the industrial revolution all over again, using only the materials provided by mother nature. That means smelting metal from stones by forcing air into a fire, making electricity by spinning some coils of wire around, taking away pain with naturally-occurring chemical compounds, harnessing the power of steam, making alcohol (for drinking and sanitation!) by fermenting sugar, etc. It was barely 100 years ago that the Ford Model T was born - some people alive at that time are still alive today. The world has changed drastically in one person's lifetime, yet most of us don't know how these things were invented, discovered or work. We have already collectively "forgotten" what was only so recently discovered, because most of us don't really need to know. These things have so drastically transformed the world, that I can't imagine not having a thorough, first-hand understanding of how they work and how they were really developed by the people who only had what nature offered them at their disposal.

Armature wound! (01.27.2010)

---

Last night I wound 18 feet of solid-core copper wire onto each of my three armature magnet cores. I was hoping to source some old-fashioned, cloth-covered wire, but that stuff is considered "vintage" now, I guess and it's expensive. So I opted for FREE and used wire from the dead fan motor that I recently removed from our water heater ventilator. Although this wire looks bare, it actually has a very tough, thin and heat-resistent insulating coating on it - like it was dipped in some kind of flexible shellac. That's fine, anything is better than thick, colored, plastic insulation which not only bulks up the windings with useless dead space, it just looks ugly. Currently my fingers are cracked, cut and bleeding from unwrapping and re-wrapping wire for over an hour.

Below: Here's the armature, mostly complete except for a couple bushings and spacers. The coils are soldered to the commutator segments in series with each other, and all wound the same direction. This is not my finest soldering job - sloppy and too much solder. I had a really tough time getting the solder to stick to that stupid wire, even after scraping it clean and using flux. Oh well, I removed the extra solder with the lathe, and the commutator is clean and smooth once again. I was also hoping for a tidier winding job, with every turn laying perfectly next to the previous, but I soon found that to be near impossible. Maybe next time. They are very tight, so that will have to do. I gave them 12 Volts and they become quite magnetic, so it should work.

Below: I've drilled the bearing blocks and set the armature in there as a quick test fit. It's starting to look like something now! Those holes will soon be drilled larger to fit bronze bushings and the whole works will be fastened to a nice hardwood base plate. Then only the stator magnets and brushes will remain to be designed and fabricated.

Below: Here's a sort of schematic of how the coils are connected to the commutator. Of course the direction of rotation depends on the polarity, but this just shows the general axis of rotation. With this setup, the brushes would contact the commutator on opposite sides from each other. It doesn't matter exactly where, so long as the stator magnets are 90 degrees offset from the points contacted by the brushes (in theory). I'll experiment with this a bit. I think I'll mount my stator magnets first, then experiment with the best brush position before installing those permanently.

Wooden Base (01.28.2010)

---

Below: I had a big chunk of American Walnut left over from a couple Christmas presents I made this year so I cut a piece of that down to size, planed it smooth and cut a fancy edge with the router. I didn't finish it yet because I still have some more drilling to do, so as yet it is still pretty dull. Here are all the parts so far, sitting together loose. Now this is looking like something from the mid nineteenth century and so I will have to come up with a proper name for this device, perhaps: An Apparatus for Self-Acting Mechanical Rotation through Electromotive Force (I just re-titled my page by this name). Naming and general styling inspired by some language I've heard in reference to early steam engines and some of the early motors, shown at The Spark Museum.

Assembly (04.01.2010)

---

Below: I've been busy with other projects for a while, but now am back to the motor... I've assembled the permanent magnet housings, and completed some brass screw terminals. The main assembly is done and it now only awaits brushes, which I hope to design and build tonight.

-\\ SUGGESTED READING

---

- Electric Motors and Drives Fundamentals and Types. (Amazingly technical-yet-practical info about motor applications. Probably for engineers)

- Dynamos and Electric Motors: How to Make and Run Them (Published in 1903, great insights into motor design including armature and magnet winding)

- How to Cast Small Metal and Rubber Parts (Good intro into sand casting and basic metallurgy/alloying)

- The Complete Metalsmith: An Illustrated Handbook (Wonderful book about casting, soldering, metals melting and metalworking for jewelry specifically)

- Power from Steam: A History of the Stationary Steam Engine (An awesome history of the steam engine including plenty of physics and contextual history)

- Tesla: Man Out of Time (Very interesting biography of Tesla)

- BackyardMetalCasting.com (A great website with lots of photos and info)

- Secrets of Green-Sand Casting (Another very old book, but full of useful info.)

- The Code Book (Not exactly related, but if you like this stuff, you'd like this one. About codes and ciphers in history.)

- The Boy Who Harnessed the Wind (An inspirational true story about an African farm boy who builds an electric windmill from junk and transforms the lives of his family and village)

- Gaviotas: A Village to Reinvent the World (A true story about an experimental community in Colombia, totally off the grid, with a closed-loop energy and waste system. Very cool.)

- (Gutenberg Book)

- (The Alarming History of Medicine)

- (A History of Chemistry)

- (Batteries, Leyden jars and dynamos)