Andrew Morris had a problem with a small rotary tool built in China. The tool fit nicely in the hand and was useful for precision cutting, drilling, and polishing. Yet for delicate work, the tool was in bad need of a speed regulator.
Andrew had developed an analog motor speed regulator back in the mid-1990s, but this time, he wanted the benefits of a digital regulator. The digital version was just as efficient, but it was less expensive to build and easier to assemble. The digital circuit also provided more torque.
Andrew Morris' microcontroller-based DC motor speed regulator brings control for delicate work.
This gadget turns something that's almost useless into something that any gadget maker should find useful. The AC-powered rotary tools are great for heavy work, but this gadget is exactly what is needed for delicate work. The tool fits very nicely in the hand, giving the user excellent control of the drill bit, cutter or polisher. This is critical when replacing tiny surface-mount parts on printed circuit boards (i.e. cutting away old parts), where uncontrolled speed or poor manual control could tear the part off the board, causing damage. In addition to that and many other things, I've also used my older mini-drill and analog speed regulator to polish away oxidation from the tiny switch contacts in my wireless computer mouse. High speed would have torn the delicate contacts right off the tiny switches. With this gadget, the mini-drill has enough torque for most not-so-delicate work as well.
Since this article was written, the digital speed regulator has worked so well for so many projects, that the analog unit doesn't get used anymore. I keep it around since it has a lower minimum speed than the digital unit, just in case. Getting stable performance from the digital circuit at extremely low speeds might be a challenge. I haven't tried it, since the current minimum speed of the digital control has so far been perfect for my needs.
You can see in the video that when the mini-drill itself was switched off, the overload LED came on and had to be reset by switching the control unit off and then on with the tool itself on (or switched on within 3 seconds). This is caused by the fact that when the motor is disconnected, the circuit does not see the feedback from the motor and thinks that the motor has stalled. I was aware of this when I designed it, but I didn't see it as an issue until I used it for the first time on a real job, restoring an antique cigarette lighter. It became a bit of an annoyance. The switch on the tool is much more convenient to use than the switch on the control box. I solved the problem after the video was made by adding a resistor (R10) to fool the electronics into thinking that the motor is still connected and running at full speed when the motor is actually disconnected. Due to the extremely low relative resistance of the motor, the resistor is effectively bypassed when the motor is connected and has no effect on circuit performance.
I also changed the software code to boot-up at zero speed, instead of max speed as was the case in the video. You may have heard it start up at high speed and then very quickly come down to the set speed. This "up-to-speed" help was not needed due to the fact that the motor is unloaded at start-up. In the invention for which this algorithm was originally created, the motor was started up under load, requiring this extra boost. This change has no effect on the gadget's performance.
After this article was written, I created an in-line version of the digital speed regulator. Experience with the circuit so far has shown the minimum speed setting to be perfect for my needs. Sometimes it's also handy not to have a control box in the way when not working at a bench or table. The tiny in-line version has no knob, switch, or LED and runs at one set speed. This required disabling the overload timer as there is no quick way to reset it without the on-off switch. The user must take care not to keep the motor overloaded for an extended period of time. The mini-drill will probably get noticeably warm before there is risk of damage.
The circuit was built onto a tiny piece of perfboard, 4 holes by 18 holes in dimension and covered with heat-shrink tubing. Please email me if you would like construction details. My email address is in the article.
Please email me if you want to build this gadget and don't have the means to program a microcontroller. If you live in the US, I'll send you a programmed PIC for the cost of the blank PIC and estimated postage (unless that practice gets out of hand). Just let me know whether you intend to build the full-sized or the in-line version of the control, so I know whether or not to disable the overload timer. I'm sorry that I have no PCB (printed circuit board) artwork to give to you for either version. Due to the circuit's simplicity, a PCB layout should be very easy to do. I find it easier to hand-wire something than to make a PCB if I'm only making one copy of it. Hand wiring also gives me more packaging flexibility. If you create a PCB layout for it, please share it.
My lifelong interest in motor speed regulators (also called governors) began in high school in the 60's, when I had a cassette recorder, a portable record player and later, an 8-track tape player, all with faulty mechanical governors. I eventually figured out how the cleverly designed Philips (Norelco) governor circuit worked. Most manufacturers wouldn't pay Philips for the rights to use their patented circuit. I repaired cassette recorders and 8-track tape players for other people by bypassing the mechanical governor and adding my version of the Philips governor circuit. The simple two-transistor circuit was easy to build and tuck away inside the set. Nowadays, cassette recorder motors have built-in governor chips.
Back in 1987, I designed a "bang-bang" motor speed regulator using a 555 timer and was disappointed to discover that it was no more efficient than a linear regulator. For a fixed speed and load, the current draw remained constant with changes in the power supply voltage. The motor got warm instead of the output transistor. The 555 timer turned the motor on for a fixed period of time whenever the motor's BEMF dropped below a certain level. The hardware configuration was very similar to that used in this gadget. The patent application referenced in this article explains "bang-bang" motor control in much more detail. BTW, the aforementioned mechanical governors are examples of "bang-bang" motor control.
This gadget is not just about motor speed regulators, it's also about patents. I worked for 5 years in a R&D (research and development) center, where I developed the motor control algorithm and where we worked with patents every day. I had been a member of an inventor's club for about 15 years before that. There may be a lesson here for other gadget makers who hope to patent their inventions. I hope to spark some conversation on the subject of patents, as well as the gadget itself.
I believe that when my former employer closed the facility where I worked before the final step in the patent process could be completed, they effectively served my invention to the sweeper manufacturer on a silver platter. The timing is right and the patent office is the first place someone would look, who was looking for a cheap motor speed regulator for a consumer product. Even if the sweeper manufacturer intended to develop their own motor control, they almost certainly would have done a patent search first in order to avoid creating a costly patent infringement. But then, something like this probably happened: "Hey, check this out! Here's just what we need, and for some reason it didn't get patented. It's now in the public domain and a prior-art search has already been done. Wow! Let's grab it!"
I'm glad the patent application failed. I wouldn't have gotten any money from it anyway, but at least now, people are able to use my idea. My former employer would have used it in a very limited application, if at all. It's common practice for big companies to patent ideas that they're not going to use and then sit on them to prevent competitors from using them. That's probably why they tried to patent my invention, even though they didn't really want or understand it. If they really wanted it, they would have assisted me somehow in getting witness signatures on the disclosure form (explained later) when I couldn't find someone, rather than let it become public domain a year later. The purpose of the invention was to replace a piece of mechanical hardware with software, saving some recurring cost in a mass-produced product. This proved to be much more complex than just the speed regulator.
The make and model number of the floor sweeper that I believe uses my speed regulator algorithm was not given out in order to avoid potential problems from the manufacturer for publicly disclosing details of their design. Anyone interested can contact me individually for that information. If you like, I'll send you a JPG copy of the hand-drawn schematic and a PDF copy of the user manual (or the link to it). It's a good example of another use of the speed regulator algorithm, and how to interface with it. For example, the floor sweeper uses a pushbutton to sequentially select one of three fixed speeds and has three speed-indicating LEDs, using just an 8-pin PIC. I don't have access to the sweeper's code-protected software, but I used a digital oscilloscope to observe its operation.
I did not give out the name of my former employer, or what the invention was that employed my speed regulator algorithm, in order to avoid potential problems from them for publicly disclosing "company confidential information". Instead, I made reference to a public document that contains that information. The patent application also supports my claim that I invented the speed regulator algorithm. The patent lawyers did a "prior-art" search and would have discovered if someone had patented or tried to patent it before me. For those not familiar with patent searches, here is the link to the aforementioned document:
Ever wanted to see light beyond what's detectable by the human eye? You can with DOLPi - a homemade Raspberry Pi-based polarization camera. You can even use it to detect unseen objects like landmines, IEDs, pollutants, and maybe even UFOs.
A Design News contributor takes on the challenge of building an old-fashioned metric clock that uses French Revolutionary time, which divides the day into decimal units, and shows you how to build your own.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.