Solid State Stir Plate

[kwikfivo]

Check out this link to my YouTube video:



It is still in the prototype phase, but it works!

I am using a custom PCB that I designed and purchased through expressPCB. The board is using an Arduino Pro Mini 16MHz microcontroller. The H-Bridges are SN754410NE units from TI. I have suppressor diodes installed on each output channel to suppress the flyback voltage spikes from the electromagnets.

The electromagnets are actually coils that I salvaged from old pneumatic valves, they are rated for 12Vdc/8.5 Watts.

Right now the system works best with the small 1" stir bar. Approximate RPM is 1500-1800 (which I'm working on increasing through a different control method in software).

The only major concern I have right now is heat. The coils do warm up, and I think I can sink a lot of this heat with an aluminum plate that is mounted on the bottom of the coils (opposite the side that the flask sits on to keep heat away from the yeast).

I will admit, this is totally overboard for a stir plate, but brewing is all about learning and having fun anyways, right?

I'll post more once I get it finished.

 

[RocketBrewer]

You electronic geeks make the coolest toys!

 

[kcmobrewer]

That's really cool lol. I wish I knew about this kind of stuff.


Sent from my iPhone using Home Brew

 

[kwikfivo]

You electronic geeks make the coolest toys!

I'll take this as a compiment!

That's really cool lol. I wish I knew about this kind of stuff.


Sent from my iPhone using Home Brew

If you're ever interested in electronics a good place to start would be Radio Shack. They have some good books available that start with the basics, and for probably $50/$60 you can buy an Arduino (cheap and easily programmable controller), some LED's, a breadboard, and wires. Once you learn the basics the sky is the limit.

The basic concept of the system I built is to apply +Vdc and 0Vdc to the two wires on each electromagnet in a specific pattern. When you swap the wires to an electromagnet (wire that used to have +Vdc now has 0Vdc) it changes its polarization (north/south poles) and therefore will attract or repel the stir bar magnet (one side of which is a north pole, the other south), and if you do this fast enough and in the right pattern the stir bar will spin. No rocket science, just fun.

 

[ChefRex]

Very cool!

 

[Facinerous]

Couple questions. Are you driving your coils a pair at a time to get opposite poles, or just one at a time?

Second, what is the power consumption of the unit? Is it comparable to the ol' computer fan w/ a magnet that I see so often?

 

[kwikfivo]

Great questions!

I am actually powering all coils at the same time. I discovered through some testing that if I repel the stir-bar from one location while attracting it at the next it improves speed and response. Since I have (4) magnets each coil will be energized at any time. Although, I may change this for low speed operation and start-up to improve stir-bar control.

Power consumption is approximately 2.5A @ 14Vdc. Probably a lot higher than the computer fan, but as I said in my original post, my intent was to over-build and have fun.

 

[BlackenedBrew]

Really cool! I look forward to seeing updates on this as you make changes!

Where did you get that clear project box you're using? I want to rebuild my stir plate with a clear box, but I haven't found one that I think would survive being drilled.


Sent via HBT on iOS!

 

[Facinerous]

Slap me if I'm wrong, but I was thinking about it. I don't know if your limitations are in the software for increasing your rpm.

On my (not limited but mediocre) knowledge of inductors your limitation would be hardware. Once magnetically charged you are looking at a time constant for discharge. EG the necessity for your voltage kickback diodes.

I was thinking of ways around this. Couldn't you put a high impedance drain to ground in series with the inductor? Just a 1000 Ohm resistor would do the trick. I think. Even if the voltage spiked to 140V like the times 10 rule implies for inductors you would only be looking at 0.14 amps for a few milliseconds.

Am I wrong in thinking this?

Being able to reduce your CEMF quicker should allow you to generate a higher rpm.

 

[kcmobrewer]

Hey thanks. Do you have an exact book in mind from radio shack you would suggest? I would definitely like to learn about wiring electronics.

Would learning this stuff translate to wiring my own brewery control panel?

Sent from my iPhone using Home Brew

 

[bu_gee]

Don't start at radio shack. They're absolutely useless sacks.

Start by watching youtube videos. An Arduino Uno also shouldn't cost you more than $25.

 

[bu_gee]

Did you set up a resonant circuit or are you driving the coils fresh every time?

 

[kwikfivo]

Really cool! I look forward to seeing updates on this as you make changes!

Where did you get that clear project box you're using? I want to rebuild my stir plate with a clear box, but I haven't found one that I think would survive being drilled.


Sent via HBT on iOS!

It is manufactured by a company called Hoffman. You should be able to find them on Digi-Key's website. I found mine at a local electronics shop.

Slap me if I'm wrong, but I was thinking about it. I don't know if your limitations are in the software for increasing your rpm.

On my (not limited but mediocre) knowledge of inductors your limitation would be hardware. Once magnetically charged you are looking at a time constant for discharge. EG the necessity for your voltage kickback diodes.

I was thinking of ways around this. Couldn't you put a high impedance drain to ground in series with the inductor? Just a 1000 Ohm resistor would do the trick. I think. Even if the voltage spiked to 140V like the times 10 rule implies for inductors you would only be looking at 0.14 amps for a few milliseconds.

Am I wrong in thinking this?

Being able to reduce your CEMF quicker should allow you to generate a higher rpm.

Good idea! I wanted to make sure my software was optimized before I dug into the hardware. I can breadboard this pretty easily, it makes enough sense. We'll see what happens!

Hey thanks. Do you have an exact book in mind from radio shack you would suggest? I would definitely like to learn about wiring electronics.

Would learning this stuff translate to wiring my own brewery control panel?

Sent from my iPhone using Home Brew

They have a book called "Getting Started with Arduino". It goes over a bunch of simple projects and gets you familiar with the platform.

I saw a negative comment about Radio Shack, and I can't say I disagree with it. I should've said "Radio Shack or equivalent", there are a lot of places you can get this stuff.

Did you set up a resonant circuit or are you driving the coils fresh every time?

I am driving them fresh every time. I will admit, I am not an electronic circuit guru, and I'd never even read about an LC circuit until now. I think I need to read about this further and do some more math. It looks like I can gain some efficiency if I go down this path, but the one thing I can't get my head around right now is how I would size the capacitor if I want to retain my variable RPM. Doesn't the capacitor need to be sized so that it resonates at a specific RPM? Again, I need to read more about this, so maybe that was a dumb question.

Thanks for the input everybody!

 

[kcmobrewer]

Sorry to restate the question, I think it got lost in the mix. But would this arduino project be a good primer to learn about wiring? Like would it be applicable to eventually learning about wiring PIDs and switches and such? I would really love to gain the knowledge to wire my own electric brewery control panel.

Very cool project. I would love to make one of these.


Sent from my iPhone using Home Brew

 

[kwikfivo]

If I were you I would start out with the basics, just so you understand the relationships between the arduino hardware and the software you will write.

Get yourself an Arduino, a breadboard, some LED's, and a few switches (sometimes you can find all of these items in a prepackaged kit) and work through the beginner projects (project 1: turn on an LED with a switch, project 2: flash an LED when the switch is activated, project 3: analog inputs...). Once you are comfortable with how the software works you can move on to more difficult projects. Do a Google search for beginner Arduino projects, you should find plenty.

There are several PID examples on the internet. I will say though, PID is a little beyond what I would consider the "beginner" end of the spectrum, you will need to know the basics first, then move into PID control.

Good luck! You won't regret getting into electronics, it is a lot of fun to see your ideas move from concept to reality.

 

[kcmobrewer]

Cool I'll definitely do that. Thanks!


Sent from my iPhone using Home Brew

 

[podz]

Hey thanks. Do you have an exact book in mind from radio shack you would suggest? I would definitely like to learn about wiring electronics.

Would learning this stuff translate to wiring my own brewery control panel?

Sent from my iPhone using Home Brew

No need for Radio Shack, this site has all you need. Nothing being sold there as far as I can tell, just useful information.

http://www.allaboutcircuits.com

 

[bu_gee]

No need for Radio Shack, this site has all you need. Nothing being sold there as far as I can tell, just useful information.

http://www.allaboutcircuits.com

That may not be the level of user-friendly he needs right now. I use that site when I need to learn why something works or if I need a refresher on some complicated electrical concept.

 

[Facinerous]

Answering your capacitor question would be a little complex. Creating a resonant LC circuit is not to bad when you are dealing with a stable frequency, or in your case rpm. With a changing rpm it can get a little more complicated. A good approach would be to find the rpm that you want to keep your stir plate continuously running at, lets say 2400 rpm. Divide that by 60 and you will get a frequency. 2400/60 = 40 hertz. You are essentially making an AC circuit by flip flopping the voltage across your inductors.

[edit]
I since you have 4 poles you would want to multiply that by four. So 160 hertz.

Here is a link to essentially what you are creating. Explains it perfectly.
http://www.allaboutcircuits.com/vol_2/chpt_6/2.html

The reason I am making another post is to contradict what I had wrote earlier. Basically I am slapping myself. It took a three hour drive today, boredom, and so I thought about what I posted earlier.

I came to conclude that you do not want to drain your charged inductor. Basically doing so would slow down your ability to switch your polarities effeciently.

I was thinking about it and remembered that many small motors have roughly 6 poles. In your stir plate configuration you basically have 4, and you have in a small way made the essence of a motor. You have a magnet, and you have coils driving the magnet in a direction.

So rather than trying to drain the CEMF from the inductors, you would actually want to try to hold onto it for when you change polarity. Its very easy for an iron core (part of the coil) to switch polarity. It does, on the other hand, take effort to charge the inductor to operating capacity. Rather create the magnetic pole that will force the stir bar in your desired direction.

So where the hell am I going with this. I think thats as far as I should go unless there is more interest. Two points of note though. Adding a capacitor in parallel with each of your coils would increase your efficiency and even help your coils hold their charge until it get to its next cycle, and second to slap myself in the face for my post earlier. I was thinking about it wrong.

[edit]
Oh and BTW. Keeping your inductors charged will help keep those coils cooler.

 

[bu_gee]

I am driving them fresh every time. I will admit, I am not an electronic circuit guru, and I'd never even read about an LC circuit until now. I think I need to read about this further and do some more math. It looks like I can gain some efficiency if I go down this path, but the one thing I can't get my head around right now is how I would size the capacitor if I want to retain my variable RPM. Doesn't the capacitor need to be sized so that it resonates at a specific RPM? Again, I need to read more about this, so maybe that was a dumb question.

In a lot of ways, right now you just have an incredibly small capacitor which drives your resonant frequency high. Just because you have a resonant frequency, it doesn't mean you have to run at that frequency; it is just easier if you do. So if you pick a capacitor sized for your steady-state frequency, you should be in good shape.

Also, I have seen digitally-controlled variable capacitor ICs. You might be able to vary your resonant frequency using one of those if they can handle the power levels you're throwing around.

 

[kcmobrewer]

Man I wish I understood half of this lol. I'm gonna learn this arduino stuff just to build one of these. I love the idea and how quiet it must be.

Once you get everything wired up inside the box will it need a fan for cooling or will it use heat sinks? Also, is a potentiometer applicable here or is that not necessary due to just programming the thing to a certain rpm?

Also, sorry to bug you with all this. Let me know if you would rather I pm you.


Sent from my iPhone using Home Brew

 

[podz]

Also, is a potentiometer applicable here or is that not necessary due to just programming the thing to a certain rpm?

Potentiometer (POT) slows a motor down by reducing voltage. This also causes torque drop.

Pulse-width modulator (PWM) slows a motor down by rapidly switching voltage on and off. This causes no torque drop.

You choose your weapon based on your motor and your application. Does a POT outputting low voltage leave enough remaining torque in the average PC fan motor to sufficiently turn a vortex in a 2000ml flask? Probably. What about a 5000ml flask? Probably not. That's why you see 30 buck price differences in the stir-plates at the online homebrew shops - the ones capable of using up to 5000ml flasks are about 30 bucks more because PWM is more expensive due to increased amount of and different components plus more labor.

 

[Facinerous]

To answer kcmobrewer's question I don't believe a fan would be necessary to cool everything off. The heat generated by this contraption shouldn't affect the actual wort in any major way.

A pot could actually be useful in this application. Setting the pot at 0 ohms would all allow maximum current flow to get your stir plate up to speed (to reiterate podz you would have maximum torque). After that setting the pot to some value larger would limit the current flow to the coils, but you have already developed the intertia to required to keep the liquid moving and therefore impose less opposition to the stir bar. The pot could act as a cruse control so to speak for keep the fluid moving using less power.

 

[kcmobrewer]

Ah ok. Please excuse my ignorance on the subject lol. I just always see heat sinks on SSRs in people's control panels. Wasn't sure if that would be the case here?


Sent from my iPhone using Home Brew

 

[ndsgr]

Question about the TI H-Bridges. I have used these in a lot of projects, but have never drawn much current from them. You said you are drawing over 2.5A. Assuming the microcontroller draws ~2mA, that still puts you above the peak current rating of that IC (~2A peak, 1.1A continuous). Am I missing something or are these a lot more durable than I thought? I know some folks have stacked them (parallel), is that what you are essentially doing? I know the current is being divided among two, but I think that is still over the rated current. Do they get pretty hot?

This is an awesome project, thanks for sharing! I'll be ripping you off as soon as I get a chance. I have a ton of magnet wire laying around and needed an excuse to try out a better acid etch PCB technique a friend uses.

 

[kwikfivo]

Question about the TI H-Bridges. I have used these in a lot of projects, but have never drawn much current from them. You said you are drawing over 2.5A. Assuming the microcontroller draws ~2mA, that still puts you above the peak current rating of that IC (~2A peak, 1.1A continuous). Am I missing something or are these a lot more durable than I thought? I know some folks have stacked them (parallel), is that what you are essentially doing? I know the current is being divided among two, but I think that is still over the rated current. Do they get pretty hot?

This is an awesome project, thanks for sharing! I'll be ripping you off as soon as I get a chance. I have a ton of magnet wire laying around and needed an excuse to try out a better acid etch PCB technique a friend uses.

I am using two H-Bridges, each H-Bridge drives 2 Pairs of coils. The data sheet I have says each driver can handle 1A, which I am under (total current for each IC is ~1.2A). I might migrate this project to a 24Vdc supply, instead of 12Vdc to get the current down (I have access to the same coil in 24Vdc).

The H-Bridges do get warm, nothing too bad though. I will take some temp readings tonight and post them. Once I get my heat sink on the the coils I am going to turn it on and let it run, I have a circuit on my board that will use the DS18B20 temp sensor, so I can log a little data, and shut it off if things go awry.

 

[kwikfivo]

Man I wish I understood half of this lol. I'm gonna learn this arduino stuff just to build one of these. I love the idea and how quiet it must be.

Once you get everything wired up inside the box will it need a fan for cooling or will it use heat sinks? Also, is a potentiometer applicable here or is that not necessary due to just programming the thing to a certain rpm?

Also, sorry to bug you with all this. Let me know if you would rather I pm you.


Sent from my iPhone using Home Brew

I'm going to put heat sinks on the bottom of the coils so they conduct heat away from the platform (where the flask sits), not sure how good this will work. The heat from the coils isn't too bad. Since I have a temp sensor available I've thought about using that to create a control loop for running the coils (drive them less to drop heat or keep it consistent).

I am using a pot to change the speed at which my coils are driven on and off. The coils are being powered by the same voltage every time, the potentiometer just controls how frequently they are turned on and off. The video I posted shows software ramping speed control, but I have since added a pot so I can vary it easily.

 

[kcmobrewer]

Any chance of a new video with the pot?

This is such a cool project! I'm totally hooked. Gonna get an adruino and get started learning this stuff. It will be my mission to make one of these lol.

Would putting a fan in the box and drilling some vent holes work for cooling the coils? Does a tiny, something like 1x1 inch, fan even exist?

 

[podz]

Does a tiny, something like 1x1 inch, fan even exist?

Sure, you can find them inside of decommissioned network switches, for example Cisco.

 

[Sudz_Makenzie]

Any chance of a new video with the pot?

This is such a cool project! I'm totally hooked. Gonna get an adruino and get started learning this stuff. It will be my mission to make one of these lol.

Would putting a fan in the box and drilling some vent holes work for cooling the coils? Does a tiny, something like 1x1 inch, fan even exist?

Crap .... I just threw out an old motherboard that had a tiny fan on the cpu.

 

[podz]

Crap .... I just threw out an old motherboard that had a tiny fan on the cpu.

Never, ever throw away working electrical motors! From anything!

 

[Sudz_Makenzie]

Yeah .... the SWHMBO has me tagged as a hoarder and was nagging the bejeebus outta me, so I reluctantly tossed it

 

[kcmobrewer]

I found some 30mm fans on newegg if you're looking for one. Don't know much about them though, dunno if they would work. They're dc.

Here

I'm gonna be obsessed with this until I get one made lol. I found the Hammonds clear project boxes, but can't find one with smooth sides like the OP had. Just doing some reading I think I would want to make one with a LCD display for a stopwatch. So you know how many hours/days it's been on the plate. With just a push button to start the timer. Also a potentiometer if I can mange to make it work. Although if I can just program it to spin at a certain rpm I would be ok with that as well. Kinda overboard I know but sounds really fun! Looking forward to buying my kit and getting started.

I assume the custom PCB an essential part of this? I kinda figured just an adruino couldn't accomplish this. That seems like it'll be way over my head, and something I just can't read up on and figure it out. Maybe the OP would be willing to share the design?

Otherwise maybe I can just use one of those motors that come with the kits to spin a set of rare earth magnets.

 

[kwikfivo]

I'll try to get a video with the potentiometer up this evening.

I think any fan would work, but my goal is to keep the moving parts inside the flask. I think an aluminum heat sink sized properly should pull enough heat off of them.

I like your ideas for using an LCD, I need to find one of those so I can track some key variables.

The custom PCB is not necessary, I just did it to clean up the circuit. My original setup consisted of a bread board with a lot of wires. I would recommend using a solderless breadboard for all of your projects until you get the circuit figured out (it is a lot easier to relocate a wire or component on a breadboard than to buy another custom PCB).

The fan with magnets is always a good fall back option, it might be a good place to start if you want a beginner level electronics project.

Cheers!

 

[MrNatural]

Way cool. Almost like a maglev for brewing.....

 

[bu_gee]

I think any fan would work, but my goal is to keep the moving parts inside the flask. I think an aluminum heat sink sized properly should pull enough heat off of them.

I like your ideas for using an LCD, I need to find one of those so I can track some key variables.

Try out that tanking circuit. You might be able to get away with just passive cooling if you can reduce the power draw. Do you have a schematic of your circuit posted anywhere?

If you're looking at getting a screen, here is some maker stuff I've been working on for my company:

[ame]https://www.youtube.com/watch?v=5Dc3u0rnO48[/ame]

It only costs about $5 if you make it yourself. It very similar to an OLED board you can get at sparkfun for about $20.

I've been needing incentive to post more about it. If you're interested, I can get you more info.

 

[bu_gee]

Wow. I didn't expect it to import the video.

 

[kwikfivo]

So I was going to take a new video for someone on here who was curious how the potentiometer worked in the system, but I had a minor electrical meltdown.... I had my scope hooked up and was watching the waveform and the probe arced from one of the outputs to ground, and it literally blew the center of the h-bridge chip out, time to figure out how I protect those circuits. Once I get the board repaired I'll post a new video.

bu_gee, I'm interested in the display you have. Where can I find one? Do you have a data sheet for it, how I can make it work? Since I'm running the arduino pro-mini I don't have a lot of I/O left, not sure if this will be able to control that or not.

 

[passedpawn]

Very cool. I know what I'm doing with the broken sprinker valves I have in the garage.

Whoever said you should add a fan to keep them cool made me laugh. Add a fan and you don't need the electromagnets at all.

I messed with emags some at one time. They do suck a bunch of current and get warm.

I wonder if the current through them is dependent on load... I'm guessing yes, just like any other winding. If so, you could add some feedback (haptic control) to measure the current and then manage the speed in order to not lose synch with the stir bar.

 

[bu_gee]

So I was going to take a new video for someone on here who was curious how the potentiometer worked in the system, but I had a minor electrical meltdown.... I had my scope hooked up and was watching the waveform and the probe arced from one of the outputs to ground, and it literally blew the center of the h-bridge chip out, time to figure out how I protect those circuits. Once I get the board repaired I'll post a new video.

bu_gee, I'm interested in the display you have. Where can I find one? Do you have a data sheet for it, how I can make it work? Since I'm running the arduino pro-mini I don't have a lot of I/O left, not sure if this will be able to control that or not.

I've blown out H-bridge ICs with spectacular results before. If you can put a small resistor on your supply through the H-bridge, you'll limit the current and make it less likely to blow. That isn't always doable so the other option is a fast blow fuse or a polyfuse like they have on the Arduino.

As for the display, it has an I2C interface (2-wire) plus one extra pin for external control. The biggest problem you're likely to run into is that these displays aren't 5V tolerant and a 3.3V level is right on the cusp of working for the Arduino so it could prove a bit of an issue although, others claim that they've used it with no trouble.

I'll work on getting the docs an sort posted on my wiki and I'll let you know.