Something's wrong with my fan controller!

[xxxxxxxxxxxxxx]

Hey guys - Ok, so I'm using a brushless 12vDC fan driven directly with 12v from a 12v, 1.5A regulator. Now, a while back, this circuit was working fine - no dramas at all. But then, one day I accidentally broke the lead of the original 12v regulator which the fan uses. No biggie - just replace the thing, right? Apparently not.

You see, I did and here's what happenes. The fan starts up at full power and works fine for about 1-2 minutes. Then, it switches off, tries to start itself (but fails) a few times, comes back to life then dies again. Each on-off cycle bit lasts about 30-40sec.

Hmmm... So I replaced the regulator. And the same thing happened again. Perplexed, i put my scope on the fan positive line and visually saw the voltage rise and fall like I described earlier. Next I tested stuff. The regulator regualtes fine (both under a recent test and under past use), the fan works on a different 12v line (from the same power supply) and wiring should be Ok. The device normally gets warm (is heatsunk) but during this problem it was a little hotter. The shutdown temp of the reg is 150 C - and the max temperature (incl headroom) found was 70C. So this doesnt appear to be the problem either.

I attached a video here to show the voltage fluctuation - the fan starts up when the line on the scope transitions from high-to-low.

What have i done this time???

 

[Boncuk]

Hi Michael,

are you sure you found the exact replacement for your voltage regulator?

The 78XX are manufactured also as "M-types" which means medium. The package is TO-220. Their max. current is 500mA, whereas the normal 78XX can provide currents up to 1.5A depending on the brand. (nominal 1A)

The malfunction as described points very much in the supposed direction. If the heatsink has reached a temperature of 70 degrees the chip is considerably warmer than the heatsink. Depending on the size of it most of the heat will be transferred to the ambient air. So if the heatsink is large it gets less hot.

Assuming the fan is a 12V-type you might try an experiment. Use two paralleled 5W filament lamps, usually used for car backlights and connect them to the regulator output. The measured current should be close to 833mA, not too much for the chip. If the lamps turn into flashlights you can be sure, that you either purchased the wrong regulator or a piece of junk.

If the experiment turns out positive there is only one possiblity left: The fan must have a short circuit within the winding which means:

Replace the fan.

Hans

 

[xxxxxxxxxxxxxx]

hmmm Thanks Hans - too much current is what I suspected, but I'm fairly certain they are all the same regulator (I have a tube of them from Ti) However I will check the part No. of them and get back to you with the results.

Thnaks!

 

[xxxxxxxxxxxxxx]

OK - Sorted! The problem was indeed an inferior regulator. The parts I actually had was a 78M12 - can't believe I didn't check! Anyway, I replaced it with a 1.5A regulator and the fan works perfecto.

Thanks a lot for your suggestions hans!

 

[Boncuk]

Always welcome