8-May-2021 (Sat)
Wherein we break a mighty wind.

We finally got our fan controllers replaced. And there was much rejoicing.

We have two gigantic fans on the roof, and about a decade ago the controllers that let us operate them at variable speed broke. Those controllers were many thousands of dollars each, so we just tore them out and replaced them with a pair of on-off switches, meaning the fans were either on at 100%, or off. This kind of sucked (see what I did there).

Well it turns out, the price of the controllers has dropped a lot since then, so we finally got new ones, and now we can run our fans more sensibly. We now have choices lying on the continuum between "sweatbox" and "arctic hurricane".

Why so expensive? Why aren't these controllers just a potentiometer knob? Because that trick only works for DC motors. DC motor speed is controlled by the voltage, so a resistor slows them down. AC motor speed is based on the input frequency, which is typically 60 Hz, so to run them at a different speed you need to vary the frequency, which is complicated. Also, big AC motors like the ones on our roof usually run off three phase power, meaning a separate feed to each of the motor's three coils, with both the input signal and the physical coils being 120° out of phase, because science.

So that's why it takes more electronics than a pot. The controllers also do other stuff like having an attack/decay ramp to avoid stressing out the motors, and for some reason there's a 5kHz "carrier frequency" which I don't understand. Some kind of in-band signaling for fancier motors than ours?

One thing that occurred to us, now that we have these new controllers, is that one technique for clearing away the years of confetti that are gumming up the outtake vent above the stage would be to just run that fan backwards and blast it all out into the room again. Then we can pack it all up in a box and mail it to Adriana.

14 Responses:

  1. Joe Loughry says:

    Those look exactly like something you'd find installed on the Death Star.

  2. Christian Vogel says:

    Yeah! VFDs are awesome, but... driving a fan isn't actually the hardest work for a VFD, are yours insufficiently cooled so that they failed early, and have they been setup by a competent guy?

    I bought quite a few of them used, and never got a broken one (over in Europe, 1.5kW and 3kW), light fan(!) and servo duty over a few years.

    "Carrier Frequency": That's the rate at which the output transistors switch. Higher frequency makes for higher-pitched, but quieter whining. Some VFDs also have a mode where they spread the frequency spectrum, it becomes less of a whistling noise, but more like scratching noise - less obnoxious, and less EMI emission (but ask someone who knows your system what settings are appropriate).

  3. Roo says:

    that sounds a lot like audio geek speek.
    mail it back, haha.

  4. Adriana says:

    I'm sure I could find a way to re-use and upcycle all that confetti!

  5. tfb says:

    The very existence of devices like this, where solid-state devices are used to switch mains voltage directly, is one of the hidden wonders of the world. These things aren't switched-mode power supplies but they're related to them, and switched-mode PSUs are just everywhere. The only people who still use linear PSUs are HiFi cultists and guitarists. I can just remember computers which had linear PSUs and they were big, very heavy and hot. Inevitably, some of the earliest switch-mode PSUs went to The Moon in 1969.

    (Entertainingly some guitar amps not only use linear PSUs, they use valve (tube) rectifiers, which are just objectively awful: they have limited current output and sag horribly under load. That's good for a guitar amp as it makes it very responsive to how you play and guitarists want their amps to clip. But HiFi cultists also use valve rectifiers, because, you know, they cause the HT to droop in loud parts of the music thus making the amp clip and sound even lower-fidelity than a valve amp already does. And they'll spend days swapping rectifiers to find, I guess, the crappiest one (I am sure they sound better if their heaters have their atoms properly aligned, too). Some of these things cost tens of thousands of pounds/dollars.)

    • Nick Lamb says:

      HiFi will never be the same after Peter Belt died (in 2017).

      Normal HiFi enthusiasts of the sort you're referring to are only interested in doing things that could in principle make some sort of difference to the sound, sure, it's probably either unnoticeable or making the sound worse but at least they're doing something right? You can sometimes ABX a $5000 HiFi setup and a $10000 HiFi setup and conclude that yup, I can tell these two things apart, they're both awful but they're different.

      But Belt's PWB Electronics specialised in Morphic Resonance, so often their "HiFi components" couldn't possibly make any difference at all to anything, unless you accepted this supernatural theory about the nature of our reality. Forget buying cables with gold connectors to run between a CD player and separate DAC, PWB would recommend interventions like putting a foil sticker that you paid $10 for on the lamp you're using to light your music room, or using "special" $5 pens to write the word "GOOD" on the album covers of the other albums by the same band as the one you're listening to.

      • tfb says:

        So at various points I've parodied HiFi cranks as believing in the supernatural. If only I'd known...

        • ssl-3 says:

          You weren't around audiophile circles in the 90s, then:

          Purists were gluing expensive "Mpingo Dots" onto their walls in carefully-scrutinized positions in order to optimize the resonant qualities of the listening room. These were literally chunks of wood (allegedly from Mpingo trees), cut into round shapes and neatly finished.

          Source devices (like a CD player) were stacked above the pre-amplifier, which was in-turn stacked above the amplifier, and the whole conglomeration was placed above the speakers (so the electrons never needed to go up-hill).

          Cables were separated from the floor to isolate them from vibrations from the floor on specialty devices called "cable elevators."

          In one published case that I read a in a paid-for periodical about a quarter of a century ago, the afflicted addict carefully draped the wiring over a particular spot on a big Krell amplifier so he could find the best impartation of magnetic bias from that amplifier's transformer into that particular wire.

          It wasn't even pseudo-science and ignorance. It was just sheer, unashamed lies and grift. (There was also some good published science around that time -- plenty of it, in fact -- but there was a ton of utter bullshit, too.)

    • K says:

      Use in spacecraft was a big driver for early SMPS development. There's an interesting timeline summed up here.

      • tfb says:

        Thanks! As someone who worked in electronics companies (in school holidays, I'm old but not that old) in the late 1970s the 'Apple II invented the switch-mode PSU' claim made by Jobs & debunked at the start of the article is ... farcical.

  6. K says:

    To conceptually understand a VFD I find it helpful to think of it as an audio amplifier that puts out a LOT of power but only up to about 100Hz. You could use a Class AB or even Class A topology but it would be horribly inefficient. Instead you use what is effectively a Class D topology. Your switching (carrier) frequency needs to be a multiple at least a handful of times higher than your desired highest output frequency. In this case 5Khz is high enough that even though the power to the motors is made up of a pattern of ON and OFF it's filtered by the windings in the motor well enough to effectively look the same as a 60Hz or so signal. That way you can use software to make the signal in the controller, and everything stays efficient because modern power switching transistors are very good turning power on and off.

  7. ceedub says:

    I would like to push the button labeled "DSP FUN."

Leave a Reply

Your email address will not be published. But if you provide a fake email address, I will likely assume that you are a troll, and not publish your comment.

You may use these HTML tags and attributes: <a href="" title=""> <b> <blockquote cite=""> <code> <em> <i> <s> <strike> <strong> <img src="" width="" height="" style=""> <iframe src="" class=""> <video src="" class="" controls="" loop="" muted="" autoplay="" playsinline=""> <div class=""> <blink> <tt> <u>, or *italics*.