Thermostat/Fan

[Enzo]

A lot of fans have circuits inside, they are not simple motor windings.  Pop a dead one open sometime and explore.

[Roly]

I had assumed this to be a very simple fan, but you speak of something more sophisticated.  Did you deduce this based on the photo or the resistance measurement?

I deduced this from working on computers from before MSDOS and ripping more than a few fans apart to lubricate their bearings to prolong the life a bit (until I could get a new fan).  {as I keep saying, I'm one of those people who open up sub-assemblies that were never intended to be opened.*} 

Back in the days of CP/M fans were six inches square, made out of die-cast ali, and were simple "impedance protected" induction motors powered by the mains.

These were displaced by low voltage fans, typically 12 volt, and right from the outset these were electronically commutated (and often proudly marked as such).  Never once have I seen a muffin fan with a brushed DC motor (the only kind that behaves anything like a resistor) and there are a couple of good reasons for this; first is operating lifetime - the brushes would wear out long before the bearings, and that is quite soon enough (just having replaced the squealing fan on my video card).

{in fact since I mostly work on older computers these days replacing distressed fans is something like half of all repairs, certainly a major proportion.}

The second is that some sparking and therefore electronic hash is generated by the brush/commutator interaction and this is the last thing you need inside your computer.  The electronic noise generated by electronic commutation is predictable and tractable, the fan manufacturer can easily kill it at source (at least more easily than brush hash) so it's no problem for the PSU manufacturer or computer assembler.

Early versions used a fairly simple circuit with a Hall Effect magnetic sensor returning the position of the rotor (the outer ring is actually magnetised into a number of poles).

Later versions seem to have lost the Hall sensor in favor of the cunning trick of using the inactive drive coil(s) as the rotation sensor.  These are tiny PCB's we don't give much thought to, but they are quite a bag of tricks these days, and all that extra silicon complexity must still be worth it over brushes and a mechanical commutator from the fan manufacturers PoV.


Surprise!   :crazy2:

Pop a dead one open sometime and explore.

And the interweb being what it is, somebody has already done it for us.

I see a bridge rectifier (4 legs), 5 diodes (orange glass), a couple of transistors (3 legs) and a couple of IC's de mystery, 8 legs each which could be anything from a simple sub-systems to a full-on microcontroller...or two;





{* SWMBO's car (Subaru Liberty) developed a nasty little pause.  Several mechanics tore their hair out over it for weeks, computer? ($800), sundry sensors? ($80 a pop).  But they wisely held off because nobody could be sure.  Then my son (the supervising mechanic) took it for a blast and pointed to the air intake sensor.  "The salmon mousse" he said.  Cleaning the plug and socket seemed to cure it for a while, but it would creep back over time and SWMBO got so frustrated she was talking about selling the best handling car I've ever driven (all-wheel drive tracks like F-1, astonishing on highway or dirt).

Finally I ripped the sensor assembly off and opened it up for a look-see, and presto, right behind the socket was a row of solder joints, and one was obviously broken and just touching.  A dab with a hot iron and the case was closed.  Totally cured.  Finito.}

[g1]

  Never once have I seen a muffin fan with a brushed DC motor

Gotcha, thanks!  I probably quit taking fans apart around the time the DC versions became popular.  Up til then they were often expensive enough to warrant a repair attempt.
  So I had assumed they were brushed type but now I know better. 

[galaxiex]

Hummm, are these fan motors not similar to the brushless motors that are commonly used in radio control model aircraft (and other R/C models)?

R/C brushless motor >>> http://www.hobbyking.com/hobbyking/store/__25082__NTM_Prop_Drive_28_26_1000KV_235W_short_shaft_version_.html

...and a brushless speed controller >>> http://www.hobbyking.com/hobbyking/store/__2164__TURNIGY_Plush_30amp_Speed_Controller.html

[Roly]

Well I've not had one of these little wonders apart but my understanding is that yes, they are also two or three-phase permanent magnet induction motors (but using high currents through only a handful of turns, and "scary" neodymium magnets in the outrunner rotor.  A 235 watt motor in about a cubic inch is quite astonishing  8| (but I do wish they would use "K/V" rather than "KV" which is already spoken for  :grr ).

{I did a bit of aeromodelling when I was a teen, cutting balsa and fingers, getting high on tissue dope  xP, control lines and glow-plug motors.  The modern stuff that's available like autopilots/flight controls, cameras, etc, is quite boggling, and I'd love to get back into it, but I already have quite enough stranded projects as it is.   :lmao:}

[J M Fahey]

Probably.

DC motors do not have a rotating electromagnetic field available (easy to get from sinusoidal AC mains) so next best was to switch windings sequentially as they turned, (or a multipole magnet turned) , by using brushes and a commutator ... as crude a solution as can be.

That said, it has worked well for way over 100 years .

Big problem is sparking, so if there was another way to switch, welcome.

These motors use transistors to sequentially switch windings, which are driven by some clever circuitry, from simple extra coils to more complex Electronics, up to dedicated microprocessors.

I *have* repaired some, not because of cost which is very low, but because of Murphy's Law which DICTATES that such problems happen on Saturday afternoon , all of Sunday or some National Holiday , with all shops closed for good.

It always was some dead TO92 NPN transistor which I replaced by BD639??