1
\$\begingroup\$

This question might seem a bit alarming and it did come up because of an inadvertency on my part.

I tried to reuse old PC fans for cooling and grabbed an old phone power supply that wasn't being used anymore, this specific one:

Old AC power supply

I soldered what I had thought was positive (center) to a fan terminal and the other cable to the negative side. Since the fan just started spinning when I plugged it in, I assumed everything was correct. Later, I noticed that the output is given as 9.5V AC, not DC. This was only after I had tested multiple fans with it.

How is it possible that all fans worked perfectly fine (apparently), even over longer periods of time in this configuration? I was under the impression that DC motors could not work on AC power at all and found no helpful info about this online.

Or is it maybe just a declaration issue and the power supply does, in fact, output DC power? I am also unaware that phones use AC power.

If not, is this actually fine or are there issues to be expected at some point, if not already?

\$\endgroup\$
5
  • 2
    \$\begingroup\$ It could be that the fans have a diode in series with the input in order to protect against reverse polarity, effectively converting the AC to DC. \$\endgroup\$
    – John D
    Commented Jul 14, 2022 at 17:51
  • 1
    \$\begingroup\$ PC fans aren't standard DC motors, they're bushless DC motors (which really means they're kinda-AC motors with driver electronics built-in). You're lucky that your AC supply didn't fry the internal electronics, but it's not unlikely that there's some protection built in too. \$\endgroup\$
    – brhans
    Commented Jul 14, 2022 at 17:55
  • 1
    \$\begingroup\$ We don't know how your fans are designed. Some might work, some might not, some might get damaged. PC fans usually have driver chips in them. It is also possible the internal circuiry shorts out the negative half wave and the 300mA is too weak to damage it. \$\endgroup\$
    – Justme
    Commented Jul 14, 2022 at 17:55
  • 2
    \$\begingroup\$ Just because it works doesn't mean it's a good operating condition to put the fan in. \$\endgroup\$
    – Voltage Spike
    Commented Jul 14, 2022 at 17:56
  • 1
    \$\begingroup\$ Don’t chance it. No one can certainly say it will work. Get a DC power supply. \$\endgroup\$
    – winny
    Commented Jul 14, 2022 at 18:07

2 Answers 2

1
\$\begingroup\$

Fascinating. My theory would be, the control electronics act as a reverse shunt diode, momentarily shorting out the adapter's output (it has a high impedance, due to winding resistance and leakage inductance). On the forward swing, it operates as normal, spinning up somewhat, but not very fast, or smoothly (50Hz pulsating torque).

A lot of adapters are "impedance protected", meaning they have enough such impedance that -- while they get stonking hot when shorted out -- it's not quite not enough to be a hazard. Very inefficient, brute force sort of limitation. But eh, it's low power, it works.

I notice a "θ102°C" mark on your part, which seems to imply thermal fusing rather than impedance protection. (Afraid I'm not familiar with this mark, or what standards body is responsible for it.) This could imply it's capable of much more current (and thus heat) when shorted, hence the need for an extra cutout.

So it's not clear if the short circuit current would be near 300mA, or many times higher; and I don't know offhand what kind of reverse current (or voltage, if equipped with a series diode as @JohnD suggested) a typical fan can withstand before blowing up.

In any case, it does appear you got lucky, that it didn't simply blow up right away; it might if given more time (overheating), or if connected to a lower impedance transformer.

\$\endgroup\$
0
\$\begingroup\$

"Old PC fans" have a reverse blocking diode that half-wave rectifies the AC to DC.

The internal Hall winding current sensors detect the magnet positions and commutate the FETs which drive the dual phase fan BLDC motor to accelerate to its maximum RPM with the average DC voltage.

The 9.5 Vac is rms and has a peak 1.4x or 13.3V then may rise 10% more with no load then drop some with diode and sag depending on the internal RC time constant of the fan.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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