Magnetic Stir Plate is a Hack

If you’ve ever spent any time around a lab, you’ve doubtless seen one of those awesome combination magnetic stirrer and heater plates that scientists use to get liquids mixed and up to temperature. If you’ve ever etched your own PCBs using ammonium persulfate, you’ve experienced the need for both heating and agitation firsthand. Using a stirrer plate for PCB etching is putting two and two together and coming up with four. Which is to say, it’s a good idea that’s not amazingly novel. [acidbourbon] built his own, though, and there’s almost no part of this DIY heater/stirrer that isn’t a hack of some kind or another.

Start off with the temperature controller. Instead of buying a thermocouple or using an LM75 or similar temperature-measurement IC, [acidbourbon] uses a bog-standard 1n4148 diode. The current passed through a diode, at a given voltage, is temperature dependent, which means that adding a resistor and a microcontroller’s ADC yields a quick hacked temperature sensor. [acidbourbon] glued his straight onto the casserole that he uses as an etching tray.

Does the type of person who saves $0.25 by using a diode instead of a temperature sensor go out and buy a stirrer motor? No way. Motor and gears come from a CD-ROM drive. The “fish” — the magnetic bar that spins in the etchant — is made of neodymium magnets lengthened by shrink-wrapping heat-shrinking them together with some capacitors. Who knew that shrinkwrap heat-shrink, fused with pliers, was waterproof? Is that a wall-wart in that box, with the prongs wired to mains electricity?

Anyway, this just goes to show that etching equipment need not be expensive or fancy. And the project also provides a showcase for a bevy of tiny little hacks. And speaking of [acidbourbon]’s projects, this semi-automatic drill press mod has been on our to-do list for two years now. Shame on us!

18 thoughts on “Magnetic Stir Plate is a Hack

  1. For PCB etching, I just use my 3D printer’s bed. Gets nice and warm, and with a little Python script, agitates the etching solution for as long as I want it to. Cool build, though!

    1. OMG. I used to do that too. But I wrote the gcode by hand. Literally this:

      G28 ; home
      G90 ; use absolute coordinates
      G21 ; set units to millimeters
      G1 Z60 ; raise up the z-axis so that it doesn't collide with the etching tray
      ; If you've got a heated bed, you can speed up etching by increasing temp a bit     
      ; Heat will also increase fuming.  I use a closed tupperware.
      M140 S25 
      G1 Y10 F4500 ; move to y=10, setting movement speed to 4500.  You can tweak this.
      G1 Y45       ; move to y=45.  You can tweak this too if you need longer travel. 
      G1 Y10 
      G1 Y45 
      G1 Y10 
      G1 Y45

      But not everyone has a 3D printer sitting around underutilized.

    2. I really like the stirring magnet thing. First time I’ve seen it was when I was a kid visiting my dads lab at work. Lots of pots on heaters and stirring magnets. Gets the imagination of a youngster going :)

      My rods are rusting already as it is, greased them up to stop it from rusting any more. I bet acid fumes won’t help either. But I guess if you use some kind of ventilation it might be ok.

      I’d rather use an NTC, and when choosing the divider resistor carefully gets you much more voltage swing to work with than a diode might give, I’d imagine. NTCs are very cheap too. I used one in a diy heater table going up to 180 degrees C with PID control. Stable within 0.1 degrees C and calibrated with a thermocouple. Heats up very quickly using a standard 3D printing hotbed at 24V and 15A PSU.

      1. Exactly. I’ve seen things corrode in the presence of etchant. They were parts that hung above the etchant. There was no visible splashing, no visible vapor and no smell and yet there was enough acid in the air to cause corrosion. I wouldn’t do this any where near a 3d printer!

    1. If it’s temperature controlled, it doesn’t matter. I did the same with a 12V heated bed for 3D printing, running nicely on 24V. Just make sure the temperature sensor is glued against the copper tracks. Twice actually, both are still working fine after a few years.

    2. Twice the voltage and a quarter of the PWM duty cycle. It still the same power dissipation so you just need to watch the current limit for the heater.

      I have used diodes before. A 1N419 will work just as well. Any diode will work but the small glass encapsulated ones have a much lower thermal mass and better response time. The pullup resistor needs to be fairly high to prevent too much self heating.

      The voltage variation from temperature is small but usable. Common AVRs like ATmega’s have an internal 1.2 Volt reference that helps the resolution.

      Another simple solution is a DS18B20 ‘one wire’ temperature sensor IC (3 pin, TO-92)

      1. This is true… the main worry about over-volting is current, and if it’s a closed loop system (which this is), then the likelihood of burning out the element diminishes significantly, with the benefit being it’ll heat up faster.

        Unless of course you take it to extremes and try to run a 12V element on 240V mains… the problem being insulation not being up to scratch.

        The ATTiny[248]5 mentioned has a temperature sensor built in that could be used instead of too, unless your target temperature is above the operating temperature of the MCU, you could also calibrate and use that.

  2. As far as i know, the etching works better if you use an air bubbler as agitator. I think it has something to do with adding oxygen to the mix, but i’m not shure about the chemical details. And i think it would not hurt to make a cover for the etching tank, to keep the fumes and evaporation in check.

  3. Apropos “The current passed through a diode, at a given voltage, is temperature dependent,”, what is actually temperature dependent is the diode’s forward voltage drop. It changes by close to 2 mV per degree Celcius. To achieve the 10 mV per degree, provided by commercial sensors, just put 5 in series, if needed.

    It is often written that a transistor with base & collector shorted, so that we use the BE diode, gives better results. (Perhaps because it’s closer to the 2 mV? If so, it’s less of an issue when we have a uP for correction.)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s