mdszy's electrical engineering blog

I have a bunch of old hard drives lying around, all IDE, one of which is only 80GB, the other is 250GB. Since they’re entirely useless to me as small IDE drives, I’ll be turning one of them into a magnetic stirrer. It’d be super helpful to have one around for making PCBs, so I can agitate the developer/etchant solution. Purpose-built magnetic stirrers cost $100-$200 and I don’t feel like spending that amount of money on one.

First step is getting the build sort of planned out, and seeing if I can get the hard drive motor spinning. Since a hard drive motor is just a brushless DC motor, I’m using a cheapo 10A RC ESC to drive it. The ESC is controlled by an Arduino and a potentiometer to vary the PWM frequency sent to the ESC. ESCs “want” a servo-like PWM signal, so I’m using the Arduino Servo library to drive it. You can’t just immediately give the ESC the speed setting you want, though. You have to give it some specific signals as it boots:

1. Boot the ESC
2. Give it a signal that corresponds to some slow speed.
3. Wait a few seconds.
4. Give it whatever speed you want.

This is a safety feature of the ESC that prevents it from spinning up a motor if you start your airplane or whatever with the throttle stick all the way forward. The ESC will wait for the throttle to be brought back before it starts spinning up the motor.

The Arduino code to accomplish this is shown here:

So now we can drive our motor and use a potentiometer on analog pin A0 to vary the speed, exactly what we need!

The motor draws about 60mA at its lowest speed, and around 1.6A at its highest, so a 10A ESC is complete overkill for this, but that’s just fine, ESCs are pretty cheap.

There is a bit of weirdness with the motor that applies to all BLDC motors: at low speeds, the motor doesn’t spin up cleanly and will sit around and “twitch” until it’s driven at a suitably high speed. I’ll fix this by changing the minimum throttle value that can be sent to the ESC and it’ll work just fine.

Next up will be getting an ATTiny and using that to drive the motor, and putting everything on a protoboard inside the hard drive enclosure, and somehow getting a knob and power connections outside of the device. Or maybe I could go the “Internet of Things” route and use an ESP8266 or Bluetooth board to control it from my phone.

On the other hand, I despise IoT garbage.

We’ll see…

For now, enjoy this video that basically goes over everything I just mentioned!

Just a quick video addressing a project I’ll be working on, but have been delayed making a video about.

The delay comes down to a couple of things: a dislike of making videos with a breadboard for demonstration, and delays in actually getting my own circuit boards made.

I hate how messy and crappy breadboards look for demonstrations, so I made up my own circuit board for this nixie tube power supply. However, getting it made in my lab has been a huge pain. I purchased all the things I need to make my own boards, but they haven’t been turning out at all. The problem mainly lies in the UV sensitive film I’ve been using. It’s absolutely impossible to apply this stuff nicely without it bubbling up or failing to stick properly to some part of the board, resulting in broken traces, crappy spots that don’t develop properly, etc. So I ordered some UV sensitive paint that will be arriving soon (from China, unfortunately, takes a while to get across the ocean!). I also decided, just for the fun of it, to spend $13 and get 5 boards made by an uber-cheap Chinese board fab. So we’ll see how those turn out.

Once I get the board made, I’ll be making a video about this. It’s simply a power supply that takes ~12VDC in and gives anywhere from ~70-300VDC out for power nixie tubes, which require approximately 180V to light up. The circuit is a very simple switched mode boost converter, and I’ll explain in detail how it works once I make the video.

At my university there’s a resale store that sells/auctions off old equipment. I decided to try to snag a pizza oven that was being auctioned off, $5 starting bid. The listing said it was previously used by the electrical engineering department for reflow soldering, perfect! The only issue was the knob for the timer was missing. The oven had two knobs: a bake/broil/off knob and a timer knob. Bid on it, set a $10 max or something and a week later got an email telling me to pay for the thing I had just won. I paid $7 or so for the thing, since someone else had bid on it and raised the price, but didn’t bid past $7.

After I got it, I took it home and dismantled it. Here’s my first video on checking it out and planning out my mod.

From here I discovered a few things: the bottom switch simply chooses which of the heating elements to turn on. One, both or none of them. The timer switch is just that, a switch. When the timer is “expired,” the switch is open but it closes as long as it is ticking down. Once it’s ticked down, the bell rings and the switch opens again to turn the oven off. At this point, I already started planning out that I’ll use a cheap PID temperature controller from ebay to control the oven, with some cheap thermocouples that were also bought from ebay.

Now, part two where I do a bit more decision making as to the design of the control system.

At this point, I’ve realized that the relay inside of the temperature controller is not going to cut it at all. The temperature controller’s relay is rated to 3A, while the oven can draw around 12A. So I’ll be using the 3A relay inside of the temperature controller to switch a larger, 13A-15A relay which will actually switch the oven’s heating elements. What I’ll end up doing is replacing the timer switch with the relay, since the timer switch is just that… a switch! Finally, on to part 3:

So now I have it all put back together and all functioning! I run it through a quick test and it does, in fact, work. It holds a temperature pretty well and its temperature readings are pretty comparable to those from my multimeter’s thermocouple.

I ended up using a pair of tin snips to cut a larger hole in the front where the timer switch used to be, and shoving the temperature controller into there. This causes one issue: the cavity where the temperature controller sits gets extremely hot when the oven is in operation, which caused some pretty weird behavior when the oven was cooling down. The thermocouple attached to the temperature controller would be fully cooled down, but the controller would still read ~50+C for a very long time, even though the thermocouple was cool to the touch and very clearly not that hot. However, the oven itself was still pretty hot, which is probably what caused the issue. At some point I might shove a PC fan into the casing of the oven to help remedy this, but for now it’s not the biggest issue.

Another design flaw with the project is the fact that it requires three separate connections to the mains! One for the oven, one for the temperature controller, and a third for a 12V plug pack that powers the big 13A relay. Ideally, I could tap off the oven’s mains connection internally and use that to power the temperature controller and the plug pack, but it’s extremely difficult to get that deep into the oven’s internals without destructively disassembling it.

I’ve only reflowed a little breakout board with it, and it went pretty terribly. I used way too much paste and had a lot of trouble securing the thermocouples to the board and the board to the oven. However, I’ll be getting some boards soon from a PCB fab and will hopefully be able to use those in the oven to give a better demonstration.

Overall, this project was a success and I’m really happy with how it turned out, with room for improvement in the future!

Welcome to my humble blog.

Basically, I’m a EE student and I have a little basement lab I tinker in. This is a place so I can document my projects and another place to post my videos. Check out my YouTube Channel for those, and if you like what I’m doing, feel free to support me on Patreon so I can fund keep providing helpful content to anyone who finds it beneficial! Peace!