4
\$\begingroup\$

This schematic is part of a fan controller.

schematic

simulate this circuit – Schematic created using CircuitLab

FAN_POWER: 3.3 V (or 5 V for higher loads)

PW_FAN: goes to the fan

What kind of filter is this on the right hand side of the PNP-BJT? I'm not very familiar with LC circuits in this way.

\$\endgroup\$
  • 1
    \$\begingroup\$ For high frequencies caps are like shorts and inductors are like opens, so guess what kind of frequencies get through and which not \$\endgroup\$ – PlasmaHH May 23 '18 at 11:09
4
\$\begingroup\$
What kind of filter is this

It's a "Oh crap, this thing failed FCC testing again. My boss is already pissed, and if it doesn't pass next time, I'll be fired. If only I had actually paid attention in school instead of getting homework answers from EE.SE and copying test answers from the guy next to me. I remember something about inductorators and capacitators getting rid of high frequencies. [23 seconds on Google] Ah, here's something. It must be right. Someone named Anne Onimus says so. She must be an expert or something. I'm not sure what those 4.7 µH and 470 µF mean exactly (and why can't they write "u" right?), so I'd better copy it exactly. Hmm, wasn't there something about types of capacitators for different frequencies? Something about ceramics for stuff that radiates? Nah, ceramics are for pottery. I'll just do what Anne says here. Oh oh, what if this isn't enough? I know, I'll use two of them! Wait, I can't afford to fail again. I'll use 3. No, wait, if 3 are good, 4 must be better!" filter.

\$\endgroup\$
  • \$\begingroup\$ And a googled soft start in the left. \$\endgroup\$ – Dorian May 23 '18 at 13:45
  • 2
    \$\begingroup\$ I can't believe I fell for Anne Onimus, I even googled it. \$\endgroup\$ – Harry Svensson May 23 '18 at 13:51
0
\$\begingroup\$

It's just a cascade of LC low pass filters. This looks like a "added filter stages until it worked" thing, not something designed for a specific purpose.

I think Q1 - R1 - R2 form some kind of linear voltage regulator; I'm not quite sure why you'd want to do any of this. Is FAN_POWER maybe more something like a PWM?

\$\endgroup\$
  • \$\begingroup\$ R1, R2, C1 and C2 are forming a turn-on delay for the fan. There is now PWM in here. The 3.3 V or 5 V from FAN_POWER are switched by a relay. \$\endgroup\$ – Michael Koeppen May 23 '18 at 11:17
  • \$\begingroup\$ Ah! interesting, yes, makes sense – but wouldn't the steady-state emitter-base diode still drop the input voltage significantly? \$\endgroup\$ – Marcus Müller May 23 '18 at 11:36
0
\$\begingroup\$

It's definitely unique. Apparently the designer was extremely concerned with maintaining a constant current to the fan. The R1/R2/C1/C2 looks like an attempt to minimize the effect that supply ripple or noise will have on the current through Q1 (when FAN_POWER shows noise, the voltage across R1 [and therefore the base current] is kept stable), and the Chain O' Chokes will also impede any changes in the current. It may be an attempt to protect the supply from motor noise or flyback in the fan itself.

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

The configuration in signal processing is known as a "comb filter". The passive circuit version consists of iterative "pi filters", which are a low-pass.

Here is a single pi filter:

Pi Filter

Successive pi filters give you what is seen in that fan circuit.

Pi filters are fairly common methods of removing noise, especially the electromagnetic radiation kind that interferes with wireless signals. The advantages of the larger comb filter are debatable, and depend of course on the tuning. As with many filters, while they can dampen noise at some frequencies, there may be increased noise at others.

\$\endgroup\$

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.