3
\$\begingroup\$

I'm trying to do a makeshift small grinder/sander out of a car radiator fan, but I want to power it up with a laptop charger for convenience.

Since it's a salvaged fan, I wasn't able to look up its schematics, I only know it takes 12VDC input (as most car radiator fans do). These fans typically go from 10-15 Amps while running, with up to 25 Amps inrush.

The laptop power brick is rated at 18.5 Vout, 6.5 Amps, 120 Watts.

I connected it as-is for a short period, and everything works fine, I get satisfactory RPM and torque on the fan. However, I wouldn't want to leave that setup running for a long time, for multiple obvious reasons.

I'd like to be able to have some sort of potentiometer as well, in order to be able to regulate its speed.

So far, I've only come up with this:

  • Limit 16 Volts on the laptop output via triac voltage regulator or similar (at this point around 16v)
  • Static resistor rated at 100 Watt / 3 Ohm maximum (at this point making sure it'd be around 16V 5A)
  • Variable resistor (pot) rated at maybe 100 Watts / 100 Ohms (resistance is less important than wattage at this point)

The problem with this setup, is that those kinds of resistors and pots are hard to come by, and usually expensive.

Can anyone come up with a better / more intelligent solution and suggestions?

Most answers I found are related to more common uses, at usually lower amperage.

I have moderate experience with electrical appliances, what I don't know I'm willing to research, so don't hesitate to go wild.

\$\endgroup\$
  • 1
    \$\begingroup\$ Have you measured the voltage and current on the fan while running? I can't think of a solution that will be safer and cheaper than just buy/build a proper speed controller. \$\endgroup\$ – pjc50 Oct 16 '15 at 12:32
  • \$\begingroup\$ A 100 watt pot isn't just difficult to come by - I doubt it exists at all! \$\endgroup\$ – Nick Johnson Oct 16 '15 at 12:35
  • \$\begingroup\$ I'm not recommending this @NickJohnson, but: resistors.co.in/rheostat.html \$\endgroup\$ – Icy Oct 16 '15 at 12:41
  • \$\begingroup\$ An easy solution would be to use an Desktop PC power supply. They provide a lot of power on 12V with low costs. \$\endgroup\$ – Botnic Oct 16 '15 at 12:47
  • \$\begingroup\$ You won't be able to limit voltage output using a triac! - triacs are only useful on AC. And if you try to put it on the AC input the power laptop supply (being a clever switch mode device) would just take more current when the triac is on. \$\endgroup\$ – Icy Oct 16 '15 at 12:50
1
\$\begingroup\$

Is there a reason you're not controlling it via PWM? The main argument against is the complexity, but would give you the effective reduction of voltage you're after. Though there are potential issues with using PWM on an input to (what I assume is) a DC brush-less motor, you may want to end up smoothing it out with capacitance, but then it becomes a DC-DC converter...

Alternatively, as your current plan is to change the voltage across the motor, it isn't insane to considering using a DC-DC converter with a potentiometer on the voltage setting pin.

Third option would be to put a known fixed resistor in series with the fan, and a lower rated pot across the pins of the motor.

enter image description here

So R1 is a high wattage (100W), low resistance device, R2 is a high resistance POT (you should really put a high value resistor to go in series with the POT to make sure you can't set it too low). But his is a pretty wasteful way of controlling the fan, burning energy in the resistors rather that switching off as you would with a switched DC-DC or PWM option.

|improve this answer|||||
\$\endgroup\$
  • \$\begingroup\$ There is no specific reason I'm not controlling it via PWM, other that I haven't come across it before as I'm not very experienced with motors/fans. I'm not sure if it'd protect my supply however, the ones I looked at online look like they'd either try to draw more amps from the supply, or pop a fuse if the fan tries to get the amperage. Is this wrong? Also, could you explain why the third option would work, and how it's different from my suggestion? (the mentioned 100ohm fixed resistor would limit it to 5Amps, the pot would be for the rest but it still needs to be rated at high wattage) \$\endgroup\$ – Mr.M Oct 16 '15 at 13:02
  • 1
    \$\begingroup\$ The power supply is probably current limited anyway. What normally happens is that when the rated power is exceeded, the voltage output droops, reducing current and therefore power out. A PWM controller would give you good control over the motors speed. \$\endgroup\$ – Icy Oct 16 '15 at 13:26
  • \$\begingroup\$ The third option can be looked at with the fixed (high wattage) resistor and the POT giving you a potential divider, but the high (relatively speaking) resistance of the POT would mean that it won't have much current going through it, and hence wouldn't need to be of high power. The POT would then set the voltage across the motor, which would then give you the speed control. \$\endgroup\$ – Puffafish Oct 16 '15 at 14:41
  • \$\begingroup\$ Could you give me an example for the POTs rated resistance / wattage? Also if I understood correctly, this would be the setup: Vin -> Motor lead -> POT -> Motor lead -> Vout. I still fail to understand why the current will be lower across the POT, but this is due to my lack of knowledge on the matter. \$\endgroup\$ – Mr.M Oct 16 '15 at 14:47
  • \$\begingroup\$ @AndyHall Thanks for the schematic! You're the man! \$\endgroup\$ – Mr.M Oct 16 '15 at 15:05
0
\$\begingroup\$

Cheap, Simple, but no over current protection.

Several resistors (resistor bank) in series. Put a switch across each resistor bank. Opening or closing the switches individually will change the overall resistance.

I will leave all the ohm's law calculations for resistance and power rating of resistors up to you.

EDIT : Probably better to use the resistor banks in parallel (with switch then in series with each resistor bank. This allows use of higher value resistors.

|improve this answer|||||
\$\endgroup\$
  • \$\begingroup\$ If left to a single resistor bank, it would still have to be rated at a high wattage. Does wattage split between resistors in a bank? E.G. 16V @ 5A will be 80 Watts. To get it to 1A, I'd need a total resistance of 16Ohms. If it does split, I'd still need several higher rated wattage resistors, which could prove to be more expensive... \$\endgroup\$ – Mr.M Oct 16 '15 at 13:48
  • \$\begingroup\$ The watts consumed are as you say. 10 ohm, 1 watt (or 2 watt) are relatively inexpensive. You could build many variations in resistance and wattage using 10 ohm resistors. I just thought it would be cheap and Very simple to build. \$\endgroup\$ – Marla Oct 16 '15 at 13:55
  • \$\begingroup\$ The power sharing and voltage drops can be calculated using Ohms Law. With endless variations in the resistor network, I didn't attempt to do any calculations. \$\endgroup\$ – Marla Oct 16 '15 at 13:57
  • \$\begingroup\$ Yes, but to get 16V @ 5A, I'd only need 3.2 Ohms, and get 80 watts. I've no idea where I'd find several resistors to fit 3.2, rated at 25-30 watts each... I feel like I could be misunderstanding something here? \$\endgroup\$ – Mr.M Oct 16 '15 at 13:57
  • \$\begingroup\$ I was demonstrating a solution using passive components so you wouldn't have to design using active components. Also was demonstrating how to construct larger power rated resistor (bank) using smaller inexpensive resistors. Using the passive component approach does still leave you with large amounts of dissipated power. \$\endgroup\$ – Marla Oct 16 '15 at 14:11
0
\$\begingroup\$

Just thinking a power transistor with a potentiometer as voltage divider/ base current control should work. Just never fully open the transistor or something will die. ;-)

|improve this answer|||||
\$\endgroup\$
  • \$\begingroup\$ Perhaps, but that will turn quite a bit of power into heat. Effectively it is the same as putting a power rheostat in series with the motor. \$\endgroup\$ – Chris Stratton Dec 12 '15 at 2:42
  • \$\begingroup\$ There is the difference that a power transistor will probably cost next to nothing and a massive rheostat doesn't usually come cheap (or easy). \$\endgroup\$ – XTL Aug 23 '16 at 19:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.