Building a Cigar Box Stir Plate

Thread Starter

hrigsby

Joined Feb 1, 2018
7
So I've gotten into homebrewing and a common project is to build your own stir plate to help stimulate a yeast starter for a new batch of beer. The concept is fairly simple; you put magnets on the hub of a computer fan, add a power switch and a potentiometer to control the fan RPMs, and allow the spinning magnets to spin the stir bar in whatever your vessel of choice is.

I've ordered some parts to make one of these myself, but I've made it a little more complicated for myself and I'm struggling to understand whether or not all of these parts are going to cooperate together. My parts list includes:

- On/Off Switch
- 5mm Red LED (to be used as an on/off indicator):
- 2.1 mm Panel Mount Barrel Jack
- 680 Ohm Resistor (ran a calculator to see what I needed to power the LED, not sure I got these right):
- LED Panel Mount (to clean up the installation on the box):
- 30 W Adjustable Voltage AC/DC Power Supply:
- 20k Ohm Potentiometer (salvaged from old computer fan controller)
- 120 mm Computer Fan

I have three primary concerns:

1. How do I wire up the LED to the power switch so that it comes on when I switch the switch on? I understand the need to have a closed circuit loop, but do I put this on a separate loop from the pot and fan? Will adding in the 680 ohm resistor kill the power to my fan?

2. Is this 30 W power supply going to work with this system? I was looking at the math on the potentiometer and I saw that they have a "power rating" generally between (.25 and 2 W). Am I going to fry the pot with this power supply? Is there something else I need to add in to avoid this problem?

3. Is a 20k potentiometer way too large of a resistance spectrum? Most of my reading has said that a 1k pot is better for this application (more control because of a wider usable spectrum) but I just happened to have this 20k one that came from a fan controller in an old PC. Seems odd that they'd use a 20k if it wasn't suitable for the job, but maybe it's not the best choice?

Thanks in advance for any help y'all can offer! I'm trying to learn to read these circuit diagrams, but I'm struggling to understand how they relate to the actual hardware.
 

MrChips

Joined Oct 2, 2009
30,706
1. You put a suitable resistance in series with the LED and connect them across the output of DC power supply.

2. Your 30W power supply is likely to be inadequate for what you are attempting to do, for other reasons which are explained below. Yes, you will burn out the pot.

3. You cannot use an potentiometer alone to control the speed of the motor. You need a motor speed control circuit.

Magnetic stirrers are commonly used to agitate liquids in glass containers. The spinning motor and the magnetic fields required are much larger than what you have suggested. A computer fan is way too small for this and has the wrong rotational speed. You need a much larger motor and one that rotates at a much lower speeds, typically 60-1500 rpm.
 

Thread Starter

hrigsby

Joined Feb 1, 2018
7
Okay. I'm super naive when it comes to this, so please don't take this comment as if I'm trying to be confrontational.

There are tons of videos out there of people using the DC power supply, on/off switch, potentiometer, and computer fan with magnets (neodymium discs or HDD magnets) pathway work for this application. I believe the concept is not to control the speed, but to control the voltage supply to the fan which serves to slow it down.

Are you saying that this isn't the correct method of doing it or that it shouldn't work at all?

I definitely don't need exact RPM precision (it's just aerating yeast, after all), but I also don't want to make something that is inherently destined to fail because of poor design. I was actually worried about the 30W power supply being way too strong rather than too weak for this application, but maybe my limited understanding of how this works is flawed.
 

Alec_t

Joined Sep 17, 2013
14,280
to control the voltage supply to the fan which serves to slow it down.
That approach results in a steady reduction of torque as the voltage is reduced. It is more effective to control speed by a PWM method, which wastes less power and provides full torque in pulses.
 

MrChips

Joined Oct 2, 2009
30,706
To be honest, I have not looked at any videos or tried this myself. What is there to lose by trying?
Glue the neodymium magnet to the top of the fan. A disc magnet would not work because the orientation of the magnetic field is in the wrong plane. You need a bar market.

A typical computer CPU fan takes about 12VDC @ .2A or less. That is about 3W. A 30W power supply will not burn out the fan if you keep the voltage below the desired 12V.

To reduce the voltage (and hence the current) you would need a power pot of about 100Ω rated for about 5W. It would be better to control the power to the fan by other electronic means such as a series pass transistor or by PWM using a power MOSFET.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
If I were involved in designing one of these, I think I would start with a small DC gear motor, fit a disc to the shaft that has a couple of bar magnets or a strip of the Very powerful button magnets that are available, this way you can use one of the $4.00 PWM controllers (or build a 555 ver) and this way you will not be unbalancing a fan etc. that was not designed for the task.
The gear box on the motor will give you the efficiency you want with this low RPM.
Max.
 

Thread Starter

hrigsby

Joined Feb 1, 2018
7
To be honest, I have not looked at any videos or tried this myself. What is there to lose by trying?
Glue the neodymium magnet to the top of the fan. A disc magnet would not work because the orientation of the magnetic field is in the wrong plane. You need a bar market.

A typical computer CPU fan takes about 12VDC @ .2A or less. That is about 3W. A 30W power supply will not burn out the fan if you keep the voltage below the desired 12V.

To reduce the voltage (and hence the current) you would need a power pot of about 100Ω rated for about 5W. It would be better to control the power to the fan by other electronic means such as a series pass transistor or by PWM using a power MOSFET.
Okay, this is starting to make more sense. It's not necessarily about the most efficient or purpose driven method of making this work, but more about using what I have on hand until I get better parts to make a dedicated version.

Most of what I don't understand correlates to the amps, volts, watts part of the equation. When I see that I have a constant 30W on a variable voltage power supply, I get worried that I'm going to burn out these components (because I'm delivering, at a minimum, 2.5 amps, correct?) I don't have to use this power supply if I need one that puts out fewer amps, nor do I have to use this 20k ohm pot that I have; I'm just trying to understand the relationship between them so that I know what to return, what would work better/more efficiently, etc. If you have time, can you explain how you came to the 100 ohm pot rated for 5W? I think understanding the genesis of those numbers will help me grasp how this works.

Oh, and I should have mentioned that I'll be using two disc magnets at a minimum. One on each end of the stir bar to keep it aligned.

If I were involved in designing one of these, I think I would start with a small DC gear motor, fit a disc to the shaft that has a couple of bar magnets or a strip of the Very powerful button magnets that are available, this way you can use one of the $4.00 PWM controllers (or build a 555 ver) and this way you will not be unbalancing a fan etc. that was not designed for the task.
The gear box on the motor will give you the efficiency you want with this low RPM.
Max.
This makes sense too. I know the most common issues with these homemade devices is that the fan gets out of balance (I think it's more of an issue when people use the curved magnet out of a hard drive). It's obviously not built for the purpose of spinning magnets on its hub. It also runs too fast for the proper scientific application of a stir plate, but for yeast starters it really doesn't matter, as long as you can control the speed enough to not throw the stir bar in the solution.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
This makes sense too. I know the most common issues with these homemade devices is that the fan gets out of balance It also runs too fast for the proper scientific application of a stir plate, but for yeast starters it really doesn't matter, as long as you can control the speed enough to not throw the stir bar in the solution.
Also if the No of fan blades are an odd No.
Magnets are not at 180°!
You can get small gear motors for the same price ($3.00) as a fan.
Max.
 

Thread Starter

hrigsby

Joined Feb 1, 2018
7
Also if the No of fan blades are an odd No.
Magnets are not at 180°!
You can get small gear motors for the same price ($3.00) as a fan.
Max.
Well in this case we aren't mounting the magnets to the blades, we're mounting them to the flat hub in the middle where the motor is. The stir bar is only 1" long, so it fits on there okay at 180°. I think this solution came about because most people have old computer fans, but not everyone has a small gear motor laying around!
 

MrChips

Joined Oct 2, 2009
30,706
The only thing you need to know about electricity and electronics in this case is Ohm's Law.

In algebra, Ohm's Law tells us:

I = V/R

Two derivatives of Ohm's Law by simple mathematical manipulation are

V = I x R
R = V/I

Power is calculated from one of the three formulae:

W = I x V
W = I x I x R
W = V x V / R

Let us say, for example, the fan takes 0.2A @ 12V, i.e. 2.4W
Let us say, for discussion sake, that the fan take 0.1A @ 6V. (We don't know this is true and this could be totally incorrect for various reasons.)
(Here we make an assumption that the fan looks like a 60Ω load at all voltages, and this may be an incorrect assumption.)

If you connect a 12V 3A power supply to the fan, the fan will take 0.2A @ 12V. The load (the fan) takes what it needs, not the full 3A.
Now suppose we want to give the fan 6V instead of the 12V from the power supply in order to run the fan at a lower speed. We put a resistor in series with the fan. The fan gets 6V, the resistor gets 6V and the current through both drops to 0.1A.

The value of the resistor is R = V/I = 6V/0.1A = 60Ω.
The resistor has to dissipate this power as heat, equal to V x I = 6V x 0.1 = 0.6W

If we make the resistor 100Ω, the current is I = V/R = 12V/160Ω = 0.075A
Power dissipation of the resistor = I x I x R = 0.6W

Ok, 5W may be overkill. But you should go with 1W wire-wound pot to avoid burning out the pot.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
I think this solution came about because most people have old computer fans, but not everyone has a small gear motor laying around!
Maybe not, but I was a surmizing the project was worth spending at least $2 or $3 dollars on it!
Plus $2.50 for a PWM controller.
Max.
 

Thread Starter

hrigsby

Joined Feb 1, 2018
7
I think I'm starting to understand. I'll have to try to run the numbers again after work.

Changing direction back to the LED power indicator, where does this go in the circuit again? I was thinking that I'd wire everything as mentioned above (power supply to on/off switch to pot to fan and then I'd made a separate loop off of the on/off switch that was just resistor to LED.

My fear was that if I were to put the resistor and LED between the switch and the pot, then I'd be cutting the power to my fan way down. And I can't just put an LED without a resistor between the switch and the pot, because then I'd be putting too much power into the LED without a way to cut it down, right? This is the kind of stuff that confuses me, because I can't seem to grasp how these loops work (other than that they need to be closed).

I'm also thinking now I might want two LEDs: one to indicate power, and one that dims/brightens with the fan speed. I assume this second type would need to go between the pot and the fan, but I still don't get the resistor part!
 

Thread Starter

hrigsby

Joined Feb 1, 2018
7
Maybe not, but I was a surmizing the project was worth spending at least $2 or $3 dollars on it!
Plus $2.50 for a PWM controller.
Max.
Right, but I think the problem is not that people don't think the project is worth the money, but that they don't know exactly what parts to buy or where to source them from. I certainly don't know where to start when it comes to a small DC gear motor...when I googled it I found a bunch of stuff about "high torque"!
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
Ebay.
Gear motor 311707230569
PWM controller 182304262165
Total $5.00.
No real torque required to figure out when using a G.B. to rotate a disc with a couple of magnets
Max.
 

MrChips

Joined Oct 2, 2009
30,706
There is not enough torque from a computer fan. Computer fans are designed to move air, not magnets and liquids.
You need a toy DC motor with gears.
I will post a circuit for you when I have the spare time.
 

Thread Starter

hrigsby

Joined Feb 1, 2018
7
Well, this thing kinda works. I tried to diagram out how I wired it up, but it could be inaccurate. There also might be a better way to wire it up (actually, I feel pretty certain there is).

Anyhow, I've noticed two main issues, which actually might be one issue manifesting itself in two different applications:

1. The potentiometer I have isn't suited for this purpose, both in ohms and as an actual hardware piece. I took it out of an old PC fan controller and it looks like they broke the prongs off at the PCB, so it's almost impossible to test without soldering the wires to what remains of the leads. I'm currently getting really a really spotty connection, which I think has to do with the quality of my solder joint to the fan and to the second LED circuit. If I move this connection around, both the fan and the right-most LED (on the actual box) work. You can see the box lid holding the wire in place in the image.

2. The LED does dim/brighten by twisting the pot's dial, but it doesn't have a very wide control area. I presume this is due to the 20k ohm pot and the fact that I need a much smaller resistance spectrum. This would also help with dialing in the fan to a slower speed (as it's currently super touchy).

My solution to these issues has been to order a wide range of pots (500, 1k, 2k, 5k, and 10k ohms) to test. I also ordered a variety pack of resistors so that, once I figure out the viable voltage range of the pot for the fan speed, I can use that to choose my resistor for the LED rather than basing my calculation off of the full 12V, as I did for the power switch LED.

Soooooo....how bad is this for a first attempt? I'm fairly happy that it works, but I'm sure this is a cringeworthy circuit schematic.


Stir Plate Circuit Schematic.jpg 20180205_231455.jpg
 

Attachments

Top