Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

PWM linear-esque fan controller

Status
Not open for further replies.

technogeek

New Member
I need to power a DC fan off of rectified 120VAC (169V DC). I found one that takes 12V/0.4A.

Here is what I was thinking of doing... If I take a PWM signal and throw it into a smoothing capacitor, I'll have a nice linear DC signal. So, say I take a CMOS 555, something that draws low current. Have a linear regulator to take the 169V down to 12 volts to power the 555. Have the 555 switching a mosfet with the drain on the Vcc, source going to a reverse biased diode (prevent flyback) and the load and a smoothing capacitor.

In theory ( ;) ) if I adjust the duty cycle to say 10%, I should be able to get a 16V DC output (with a fair amount of ripple)

It's kind of like a switching regulator, but no coils. Will it work?
 
I wonder your linear regulator can handle 169 V input voltage? If it is able, how about the power dissipation of it? You might need a huge heatsink.
I suggest to step down the AC to 12 VAC, across the rectifier and filter, then to a 12 V linear regulator (since your DC motor is rated at 12 V)
Then you can control the speed by PWM.
 
yes, but only at a current of a few mA (i.e. to power the 555). Otherwise power dissipation is too much
A transformer isn't an option because I have run out of board space. I can fit a 555 and other misc components, but not a 2"x2" transformer.
 
There would be a very large voltage across the MOSFET. Also it is potentially quite unsafe. In a Switching PSU design book I have, it mentions that voltage inputs of more than 40VDC should be isolated using a transformer. 12V at 0.4A is less than 5Watts, which would require a 5VA transformer. (Maybe slightly more just to be safe) A 5VA transformer and the required smoothing / rectifying gear would probably be much smaller than your linear regulator / 555 / MOSFET + heatsink combination.
 
Don't do it, this it's a really bad idea, use a small transformer/mains adaptor or buy a fan that's rated to 120VAC.
 
Hero999 said:
Don't do it, this it's a really bad idea, use a small transformer/mains adaptor or buy a fan that's rated to 120VAC.

I've tried 120VAC fans, but they are poor performers (20cfm vs 40+cfm) and loud compared to the DC brushless fans. (60hz hum and all)

yngndrw said:
A 5VA transformer and the required smoothing / rectifying gear would probably be much smaller than your linear regulator / 555 / MOSFET + heatsink combination.

All of the transformers are 2"x2"x1".... pretty large. If they were a 1" cube, I could do it. I have maybe 2 square inches of real estate to fit something in. I can't imagine something drawing <0.5A through a switcher would require much if any heat sinking.

Something similar to this:
https://www.electro-tech-online.com/custompdfs/2007/08/HV7100.pdf

Is probably what I'll need. This has a Vin of 90V, I would need something closer to 200V to be safe.

I'm not sure I understand the objections... This is no different from a switching power supply. They run directly off rectified mains.

edit: I guess I might need the inductor afterall. This is pretty much what I'm after:
http://www.ecircuitcenter.com/Circuits/smps_buck_vm1/smps_buck_vm1.htm
 
Last edited:
hi,

>> I'm not sure I understand the objections... This is no different from a switching power supply. They run directly off rectified mains.

An isolated mains switcher has a isolation transformer, its a small transformer, as its designed to transform/switch at a high frequency, the method you are proposing is NOT mains isolated.

Eric
 
I still think you should do isolation from the mains as all switching PSUs do. You probably could get away with powering the 555 with a capacitor coupled AC supply, but I doubt you can get the really narrow pulses for the fan, and in any case, you still need a rather large inductor for the LC filter to get rid of the transient nature of 10% duty cycle and filter out the high frequency(high voltage pulses) component. If you omit the inductor you'll probably fry the fan as it'll happily try to absorb all of the power each pulse and produce magic smoke for that short period of time it's on. Plus you'll lose a lot of the efficiency you would have gotten through pulsing.

This is very dangerous on a lot of aspects, try to find some other way.

Is there any other sources of power that you can tap off of? Why is it so limited in area?
 
If size is a problem then use a mains adaptor.

You could try building an isolated switching power supply, here's the simplest one I could find but you sill need to be very careful with the transformer construction - use plenty of insulation tape betwen the primary and secondary.
**broken link removed**
 
technogeek said:
All of the transformers are 2"x2"x1".... pretty large. If they were a 1" cube, I could do it. I have maybe 2 square inches of real estate to fit something in. I can't imagine something drawing <0.5A through a switcher would require much if any heat sinking.
Keep looking, I'm sure you can find a smaller one or just put the transformer off the board.

0.5A @ 200V is 100Watts ! I know that a switching PSU would give out a lot less heat than that but it's still a lot to get rid of. Added to the fact that what you're asking is plain dangerous, do NOT do it.
 
It draws 0.5A @ 12V, barely 6W. I would not use a linear regulator because of that high dissipation. A switching regulator of this size would probably use 1-2w.

Isolation is NOT an issue. You guys were just recommending using a 120V AC fan. If the wire got nicked or cut there would be the same electrocution hazard.

**broken link removed**

Mains switcher ^^^ Not isolated.

**broken link removed**

Mains switcher ^^^ Not isolated.

Heck, 90% of their products are mains/line voltage powered and all of their demo boards are not isolated. Why? It's not an issue. If it's in an enclosure, there is no danger.
 
technogeek said:
It draws 0.5A @ 12V, barely 6W. I would not use a linear regulator because of that high dissipation. A switching regulator of this size would probably use 1-2w.
Yes, switching regulators are much more efficient about 75% to 95%.

Isolation is NOT an issue. You guys were just recommending using a 120V AC fan. If the wire got nicked or cut there would be the same electrocution hazard.
But the insulation on a 120VAC fan's wires would be thicker and have a higher degree of mechanical protection.

Heck, 90% of their products are mains/line voltage powered and all of their demo boards are not isolated. Why? It's not an issue. If it's in an enclosure, there is no danger.
So there's an enclosure now, you didn't state that before. I see your point, but in order to ensure adequate safety, make sure it's either double insulated or earthed properly.

Have you found a regulator that works with 160V input?

If you're planning to use a straight forward buck you're better off using a fan with as higher input voltage as possible, go for 24V or even 48V if possible. Although switching regulators don't burn up extra voltage in the same way that linear regulators do, simple buck regulators are more efficient when the output voltage is nearer to the input voltage, the only way round it is to add a transformer but it won't be a simple buck anymore.

You could adapt the circuit on this site for 160V input, you'll need to play around a bit, I'd increase the resistor values and use a darlington for the driver and obviously make sure the transistors, capacitors and resistors are all suitably rated.

Also make sure your circuit is fused to smoething like 100mA.
 
Hero999 said:
But the insulation on a 120VAC fan's wires would be thicker and have a higher degree of mechanical protection.

Very true. That could probably be addressed by some heat shrink tubing around any exposed wiring.


Hero999 said:
So there's an enclosure now, you didn't state that before. I see your point, but in order to ensure adequate safety, make sure it's either double insulated or earthed properly.

Well goodness, I'd think that would be a requirement for any mains circuitry.


Hero999 said:
Have you found a regulator that works with 160V input?
Buck or linear? I've found both.

Hero999 said:
If you're planning to use a straight forward buck you're better off using a fan with as higher input voltage as possible, go for 24V or even 48V if possible. Although switching regulators don't burn up extra voltage in the same way that linear regulators do, simple buck regulators are more efficient when the output voltage is nearer to the input voltage, the only way round it is to add a transformer but it won't be a simple buck anymore.

You could adapt the circuit on this site for 160V input, you'll need to play around a bit, I'd increase the resistor values and use a darlington for the driver and obviously make sure the transistors, capacitors and resistors are all suitably rated.

Also make sure your circuit is fused to smoething like 100mA.
Thanks, I'll take a look. The differences between the 120VAC and 12V fans are just amazing though. The AC fan makes all this racket and barely makes a tissue wave in it's breeze. The DC fan just makes this "whooshing" noise and starts scooting itself across the desk from the thrust! :D And the AC fan draws 12W, DC fan 6W. Amazing!!
 
hi,

>> Isolation is NOT an issue. You guys were just recommending using a 120V AC fan. If the wire got nicked or cut there would be the same electrocution hazard.


What are you going to reference your PWM 0V/Gnd to?

Unless you use opto isolators, if this device gets plugged into a mains source where Line and Neutral are reversed you will get a electrical hazard.

The attached gif is an extract from one of your links.
 
Last edited:
Still, if you stick the whole thing in a plastic case and only have the fan connected to the switching regulator, it can float at whatever voltage it wants, providing it's isolated from you and the other components it doesn't matter.
 
On the issue of isolation - there are many ways to provide it but if you don't and some one gets hurt the lawyers eat you alive.
 
I hope he's joking.
 
Status
Not open for further replies.

Latest threads

Back
Top