Author Topic: "Dial-a-Speed" Speed Controller for 12V Fan based Magnetic Stir Plate  (Read 2664 times)

0 Members and 1 Guest are viewing this topic.

Offline MooncabbageTopic starter

  • Newbie
  • Posts: 9
  • Country: au
Hey all!

I am working on making a magnetic stir plate for dissolving shellac in alcohol and similar. The basic idea is to glue a rare earth magnet to a computer fan, stick it in a project enclosure and power it. That part is easy enough, but I wanted to stretch myself a bit, so I've tried to incorporate a "Dial-a-Speed" speed controller from Make magazine.

I have breadboarded it out more or less as per the schematic, with only a few part substitutions.
  • I am using 1N4004 diodes instead of 1N4001.
  • I am using an IRF520 instead of IRF510, as Altronics doesn't sell the IRF510, and I had an IRF520 lying around from my arduino kit.
  • Altronics was out of 0.1uF Ceramic Disk capacitors, so they sold me 0.1uF Monolithic caps instead
  • My electrolytic cap isn't axial, but that's fine as I'm planning on building the whole thing flat on a PCB rather than all wadded up ala the original project.
The whole project is powered off a 12V/1A DC Wall Wart.

I noticed two things when I tried to run the fan with this configuration (I tested two different 80mm fans).
  • The fans require a relatively high voltage to get going (Fan 1 6.5V; Fan 2 2.7V).
  • The cutoff voltage is actually pretty low (Fan 1 0.65V; Fan 2 0.8V), but the fans both run a bit janky under 1V.

I'd like to make modifications such that the fans start and run nomatter how I turn the dial. I can think of two ways to do this, but I'm not sure how to implement either.
  • Using fan 1, somehow jump start the fan when the project is turned on, and prevent the Dial-a-Speed outputting less than 1V to the fan.
  • Using fan 2, have the Dial-a-Speed range run from 3-12V.

Any helpful suggestions would be greatly appreciated, and I'll try to reply promptly to provide any clarification required.
 

Offline kolonelkadat

  • Regular Contributor
  • *
  • Posts: 202
  • Country: us
  • Obviously, windows are central to Windows.
    • Force Project X
Re: "Dial-a-Speed" Speed Controller for 12V Fan based Magnetic Stir Plate
« Reply #1 on: March 19, 2016, 11:12:00 am »
to limit the operating range (3v-12v) you can add an additional resistor to your pot either in series or parallel depending on whether you want to  boost the low end resistance or reduce the high end.

eg if the pot is 0ohm-10kohm normally,
if 0 ohms turns the fan off,  then adding like 2,4k ohm  in series would make the pot be from 2,4k ohm to 12,4k ohm but the boost to the high end probably wont matter because full on is full on

if 10k ohms turns the fan off, then adding like 24k hm in parallel would make the pot be from 0 ohm to 7k ohm

in both cases, thats about 25% which is what 3 is from 12, so assuming a linear relationship between resistance and voltage it should work.

 

Offline MooncabbageTopic starter

  • Newbie
  • Posts: 9
  • Country: au
Re: "Dial-a-Speed" Speed Controller for 12V Fan based Magnetic Stir Plate
« Reply #2 on: March 19, 2016, 11:26:51 am »
I'm not 100% sure of the relationship between the pot resistance (0-100kohm) and the voltage over the fan, as it's PWM by the NE555. I suppose if I were to adjust the pot to the point where the fan JUST starts, that'd give me the start of the range I need. Does anyone know of any good calculator type tools for working out what resistor you need in series/parallel with the pot to set your range? I suppose I could put a trimpot or two into the circuit, so it's configurable if I ever want to re-purpose it.

What about giving it an initial kickstart?
 

Offline Synthetase

  • Regular Contributor
  • *
  • Posts: 99
  • Country: au
    • Synthetase's World of Nerd
Re: "Dial-a-Speed" Speed Controller for 12V Fan based Magnetic Stir Plate
« Reply #3 on: March 19, 2016, 01:04:16 pm »
If it's PWM using a 555 timer, the voltage over the fan will be 12V. Turning the pot will change the duty cycle of the PWM signal. It's done this way because, although you can slow the fan down simply by limiting its voltage, you also dramatically limit its torque as well, meaning it won't do anything under load. I've built stir plates as well and a simple resistive speed controller means the fan never has enough torque to actually shift the stir bar.

Offline MooncabbageTopic starter

  • Newbie
  • Posts: 9
  • Country: au
Re: "Dial-a-Speed" Speed Controller for 12V Fan based Magnetic Stir Plate
« Reply #4 on: March 19, 2016, 02:28:55 pm »
Yeah I got why it was PWM, I just wasn't sure how it would change the relationship to the potentiometer resistance. I determined experimentally that when the fan is at 3V/25% duty cycle when the Pot's wiper is 17.5/76.7kOhms. Since it's at 12V/100% at 94.2/0 kOhms, I guessed that increasing the resistance of the pair or wires that gave me the 17.5/94.2 kOhms by that 17.5, would make it a minimum of 3V/25%. I don't have a 17.4kOhm resistor, I used a pair of 33kOhm resistors in parallel, for 16.5kOhm baseline resistance. I don't really understand the mechanism, but it basically works. With the potentiometer turned to minimum speed, I now read 2.5V across the fan. If I put another 1kOhm resistor in series, it should make the minimum duty cycle closer to 25%.

It's not really the most elegant solution, but it works, I just have to figure out why. I don't really understand how the potentiometer is wired up with the two diodes, and how that effects the output. Nor do I really get the relationship between that value and the duty cycle the 555 generates. Working those things out is probably half the fun of electronics though right? That's what I'm telling myself anyway.

Since I'm still at the breadboard stage, any other suggestions are more than welcome. In the mean time I'm going to try to work out how to add some indicator lights without exploding them. Thanks for the prompt feedback guys!
 

Offline alsetalokin4017

  • Super Contributor
  • ***
  • Posts: 2055
  • Country: us
Re: "Dial-a-Speed" Speed Controller for 12V Fan based Magnetic Stir Plate
« Reply #5 on: March 19, 2016, 10:21:35 pm »
The source article linked in the Makezine page describes how the 555-based PWM actually works.

http://www.dprg.org/tutorials/2005-11a/

This should answer your questions.

The easiest person to fool is yourself. -- Richard Feynman
 

Offline MooncabbageTopic starter

  • Newbie
  • Posts: 9
  • Country: au
Re: "Dial-a-Speed" Speed Controller for 12V Fan based Magnetic Stir Plate
« Reply #6 on: March 20, 2016, 10:39:31 am »
Argh, went to Altronics today to pick up some 500ohm resistors for the LEDs, and something to replace my parallel 33k resistors. For some reason I fluffed the math when I did my shopping list, and ended up picking out 1.8kohm resistors instead :palm:.

Finally soldered most of the major external components. Got me to thinking though, what's best practice for soldering to casemount devices with the tags, like big pots and switches? I'm only using 1.5A 7/0.16 hookup wire for most of the connections, and I'm not really sure what the best method is. I looked around online and didn't find anything relevant to using light duty hookup wire with these things.
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: "Dial-a-Speed" Speed Controller for 12V Fan based Magnetic Stir Plate
« Reply #7 on: March 20, 2016, 12:26:56 pm »
A couple of fairly random thoughts...

1. Are you going to have more success and less frustration by abandoning the fan and switching to a small brushed motor.

2. I wonder if one of the more modern PC fans with built-in PWM speed control signal might have better speed / torque characteristics than trying to do variable voltage / PWM externally. They also have a tachometer output signal that might be usable for a feedback loop.
Best Regards, Chris
 

Offline Synthetase

  • Regular Contributor
  • *
  • Posts: 99
  • Country: au
    • Synthetase's World of Nerd
Re: "Dial-a-Speed" Speed Controller for 12V Fan based Magnetic Stir Plate
« Reply #8 on: March 20, 2016, 02:39:55 pm »
Finally soldered most of the major external components. Got me to thinking though, what's best practice for soldering to casemount devices with the tags, like big pots and switches? I'm only using 1.5A 7/0.16 hookup wire for most of the connections, and I'm not really sure what the best method is. I looked around online and didn't find anything relevant to using light duty hookup wire with these things.

I usually tin both the tags and wire first, then put them together and touch with the iron to reflow the solder on each and make the connection. You'll find that large tags suck the heat from the iron, so I rest the tip on them for a couple of seconds to heat them, then add the solder.

Occasionally, I'll use heat-shrink tubing around the connection if it's likely to get close enough to something else to short out. Otherwise, I don't bother.

Light duty hookup wire is fine for what you're doing. An ordinary 12V fan won't draw more than a couple of hundred mA.

Offline MooncabbageTopic starter

  • Newbie
  • Posts: 9
  • Country: au
Re: "Dial-a-Speed" Speed Controller for 12V Fan based Magnetic Stir Plate
« Reply #9 on: March 20, 2016, 03:23:41 pm »
A couple of fairly random thoughts...

1. Are you going to have more success and less frustration by abandoning the fan and switching to a small brushed motor.

2. I wonder if one of the more modern PC fans with built-in PWM speed control signal might have better speed / torque characteristics than trying to do variable voltage / PWM externally. They also have a tachometer output signal that might be usable for a feedback loop.

There are two main reasons for using the PC fans:
  • I already have them, so they are basically free.
  • They mount easily to my project box, and the magnet is easy enough to glue onto the fan housing.

If the torque is insufficient, then I'll look at replacing the fans with an alternative. That's why I've as far as possible tried to design the rest of the project to work independent of the motor.

Finally soldered most of the major external components. Got me to thinking though, what's best practice for soldering to casemount devices with the tags, like big pots and switches? I'm only using 1.5A 7/0.16 hookup wire for most of the connections, and I'm not really sure what the best method is. I looked around online and didn't find anything relevant to using light duty hookup wire with these things.

I usually tin both the tags and wire first, then put them together and touch with the iron to reflow the solder on each and make the connection. You'll find that large tags suck the heat from the iron, so I rest the tip on them for a couple of seconds to heat them, then add the solder.

Occasionally, I'll use heat-shrink tubing around the connection if it's likely to get close enough to something else to short out. Otherwise, I don't bother.

Light duty hookup wire is fine for what you're doing. An ordinary 12V fan won't draw more than a couple of hundred mA.

I think the whole circuit comes in at about 200mA, I measured it  8)

I didn't try tinning the tags first, I just tinned about 15mm of lead, folded it through the hole in the tag, and then flooded the whole thing. That said, should there not be some kind of mechanical bond beyond just the solder? Just kind of sticking it on there doesn't seem right to me, but then, what do I know.
 

Offline MooncabbageTopic starter

  • Newbie
  • Posts: 9
  • Country: au
Re: "Dial-a-Speed" Speed Controller for 12V Fan based Magnetic Stir Plate
« Reply #10 on: March 21, 2016, 09:49:35 am »
I've built stir plates as well and a simple resistive speed controller means the fan never has enough torque to actually shift the stir bar.

So my circuit is like, 99% complete, and I was testing the fan/magnet combo out in the enclosure. I can't seem to get anything to couple well with the magnet on the fan. I don't actually have a stirrer bar as yet, so it could be the things I am using to test. I wonder if you had any similar problems in your own stir plates? Any idea what I should be looking at as the likely cause? Do I need more magnets?
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf