Tag Archives: Electrical

Impedance River

Good luck with trying to fill the Mississippi using a fish-tank pump.

Please Remember:

The opinions expressed are mine only. These opinions do not necessarily reflect anybody else’s opinions. I do not own, operate, manage, or represent any band, venue, or company that I talk about, unless explicitly noted.

Want to use this image for something else? Great! Click it for the link to a high-res or resolution-independent version.

Yup – there are a million-billion analogies about impedance. I’m going to do one more anyway. Maybe this will be the one somebody reads – you never know.

The electrical underpinnings of this business range from simple concepts to alien geometries. Looking at the “EE” side of audio is a great way to realize just how much you don’t know. It’s particularly tough on non-technical folk and newbies, for whom the “alien geometries” bar is very low. They’re trying to get things all plugged in and playing nicely, but they’re also faced with these mysterious pronouncements on rear panels: “Minimum Load: 4 Ohms.”

Seriously…what do you mean, “minimum?” A load is something you can carry, and you want to avoid having too much, right? WHAT INSANITY IS THIS?

Anyway.

Impedance is an important topic for those of us who use electricity to create various noises. I personally think that the basics ARE fairly intuitive, but are rarely presented in an intuitive way. So, I’m going to do my best.

Up A Creek

Water and electricity are very analogous to one another. Voltage is like pressure, and current is like, well, current. Flow, you might say. The amount that’s “going by” per unit of time. Impedance is opposition to flow in an AC (Alternating Current) circuit. This opposition to current varies with frequency. Some circuits have very high impedance at low frequencies, but pass high frequency material readily. Other circuits do the reverse. Other circuits easily pass a specific range of frequencies, and offer higher impedance on both the high and low sides of that range. (This is how you make analog EQ, by the way.)

Rivers and streams are imperfect analogies, because they are really examples of DC (Direct Current). Opposition to DC flow is resistance, and it’s much simpler than impedance overall. Nevertheless, simple is a good way to start.

Consider two waterways. One is a creek. The creek is seven feet wide and five feet deep. The other is a major river that’s 500 feet across and 50 feet from the surface to the bottom. Here’s an SVG to help you visualize the difference in scale:

creek_vs_river

What if you could get 25,000 cubic feet of water to flow down both waterways each second. Which one would be likely to knock you off your feet and slam you into a rock?

The creek, of course.

The creek has a very small cross-section when compared to the river. In order to get 25,000 cubic feet of water down the creek every second, the fluid would have to flow at a speed of 714 feet per second. That’s about 487 MPH. (!) The big river, on the other hand, is going less than 3/4 of a mile per hour. The narrow, little creek offers a proportionally high opposition to flow, so getting the same amount of current as the river requires a lot of pressure – or voltage, if electricity is our thing.

What you can begin to see here is the relationship described by Ohm’s law. If the impedance of a circuit rises, maintaining the same flow requires greater “motive force/” pressure/ voltage. If the impedance drops, maintaining the same voltage creates more flow. (If you could run a main sewer pipe at the same pressure as a power-washer, you would have a LOT of water going down that pipe.)

What This Means To You

Let’s say you have a pump. It’s built for a home aquarium. It has no trouble at all pressurizing a 1/8″ tube and keeping water flowing along.

Now…connect that pump to a large sewer pipe, and try to pressurize THAT.

Good luck.

Even if you did something rather dangerous (do not try this at home, or AT ALL) and found a way to make the pump run harder, all you’d do is burn out the poor thing. The unit simply wasn’t designed to put that kind of flow down a pipe.

This basic principle is why amplifiers have “minimum” load ratings. Loudspeakers connected in parallel are effectively being attached as a larger and larger “pipe.” The overall circuit impedance goes DOWN, because there are more possible paths for the electricity to take. It becomes easier and easier for electricity to flow somewhere. The problem is that your amplifier is a pump that attempts to create a constant pressure. That is, if you have one load attached, and you send an input signal that should result in, say, 50 VRMS (Volts RMS) at the amplifier outputs, the amp will attempt to swing 50 VRMS at the outputs if you change the load. To fill the larger pipe, the amplifier has to supply proportionately more energy to maintain the voltage.

At some point of decreasing load impedance, the amp just can’t keep up. It can’t deliver that much energy on a continuous basis. It’s running hotter and hotter, with greater stress on everything from the power supply to the output devices. Eventually, depending on the amp’s sophistication, it might “thermal” and shut itself off, current-limit by throwing a resettable breaker, or drop its output to keep giving you something whilst recovering.

This also connects to the whole issue of impedance bridging, which is what enables maximum voltage transfer. Maximum voltage transfer is what we want in pro-audio, and it happens when low output impedances drive high INPUT impedances. To keep the water analogy going, it’s like connecting a city water line to a house. The city water is a big pipe (low impedance), which feeds the rather smaller house inlet (high impedance). As long as everything is working correctly, the city line has no trouble keeping the house inlet fully supplied. There’s plenty of pressure available to the house, because of good pressure (voltage) transfer.

The opposite of this is when you connect something like a piezo pickup to a “vanilla,” passive DI box. The piezo transducer actually makes a good bit of pressure, but its output impedance is very high. It’s like one of those tiny little capillary tubes that the folks at the doctor’s office use when drawing blood from a finger stick. The input impedance of the passive DI is actually pretty high, but it’s proportionally low compared to the piezo output impedance. The piezo drives the passive DI poorly, resulting in very low level, and the circuit configuration causes a noticeable loss of low-frequency material.

Solve the impedance bridging problem by connecting an active DI with very high input impedance, and your problems go away.

The overall point is that you can’t fill an infinitely large conduit with a finite supply. That’s why audio devices have appropriate loads for their outputs, and why you have to be mindful of those loads.


Why Buy An Active DI

An active DI box can cost a bit more, but they have big advantages.

Please Remember:

The opinions expressed are mine only. These opinions do not necessarily reflect anybody else’s opinions. I do not own, operate, manage, or represent any band, venue, or company that I talk about, unless explicitly noted.

The beat-up device up there with a missing knob is one of my well-loved passive DI boxes. It’s sounded plenty decent on a number of sources, solved at least its own share of ground-loop issues, and has never had any problems (beyond losing its attenuator knob).

Passive DI boxes are very handy creatures. They solve connectivity problems with almost no fuss at all, and the well-designed models are highly resistant to both stupidity and malice. My guess is that, about 80% of the time, they’re a perfectly decent choice.

The thing is, though, that active DI boxes let you cover the full 100% at all times. They’re also cheap enough now that there’s really no reason not to go active (if you’re starting from scratch).

The “cheap enough” bit is pretty self-explanatory. Head on over to your favorite music-gear retailer – Sweetwater, PSSL, Zzounds, whoever – and find their direct box category. Sort by ascending price, and you’re almost sure to find active units before you leave the $30 price point. (Some of the really cheap units are junky, but to be fair, I own two Behringer DI800 units that have never let me down…and at $120 a pop, their per-channel cost is $15.)

What isn’t so self-explanatory is why passive units don’t quite cover 100% of the direct-input situations you’ll encounter. There’s a bit of science involved.

A Few 10s of kOhms Is Usually Enough

Modern audio is all about voltage transfer. Voltage transfer is all about connecting an output device to an input device with an impedance (opposition to current flow) that is high when compared to the output circuit.

Okay, that sounds like gobbledygook. An analogy would be helpful.

Think of a bunch of cars on the freeway. Traffic is flowing nicely. Everybody’s just flying along without a care in the world. This is low impedance. There’s very little opposition to traffic flow.

Now, we construct an exit to the freeway. The exit leads to a one-lane road. The one-lane road, in comparison to the freeway, is a high-impedance device. Fewer cars can flow down that one lane road, and as a result, the freeway has no trouble keeping the little road supplied with cars.

This condition, when applied to electrical connections, is called “bridging impedance.” An output device with low impedance is like a freeway, and an input device with a comparatively high (10x or more) impedance is like a one-lane road. For audio types, we’re not concerned with preserving the amount of electrical flow, so much as we’re concerned with preserving electrical force (voltage). Bridging impedance lets us do that.

Most passive DI boxes have an input impedance that’s in the range of several tens of thousands of Ohms. Some can even be in the 100,000 Ohm range. Connect a device with an output impedance of a few thousand Ohms or less, and – no problem! A lot of devices are perfectly suited to interacting with a passive DI, because a lot of the gear and instruments that get connected are active units. Keyboard outputs are low-impedance creatures. Guitar-processors have low-impedance outputs.

Heck, a lot of acoustic-electric guitar outputs are low impedance. The actual pickup might be anything under the sun, but quite often you’ll find some sort of preamp sitting between the pickup and the output jack.

In a lot of cases, you can even get away with connecting a bass or electric guitar with passive pickups to a passive DI. It’s not theoretically ideal, but it usually sounds fine.

This covers the “80% of the time” thing. The 20% comes in when you encounter an instrument with a very high impedance pickup, and no preamp. Plug one of those into a passive DI, and…yuck.

Easy As Pie-zo. (Yeah, That Was A Cheesy Pun…)

The ur-example of the high-impedance pickup is the piezo. Piezo pickups are neat because they’re small, put in direct contact with the instrument (which makes them resistant to external noises, insofar as the instrument resists those noises), affordable, and simple.

The problem with piezos is that they are passive devices with a very high output impedance – so high that getting into impedance bridging territory requires millions of Ohms or more.

So, you plug one of the little darlings into a passive DI, and what happens?

First, you probably get a weak signal out of the pickup. Poor impedance bridging means poor voltage transfer, and voltage transfer is how we ensure good signals in the world of pro-audio.

Second, the instrument probably sounds terrible.

Why?

A piezo pickup (when connected to another audio device and viewed as a set of electrical building blocks) is a capacitor, inductor, and load resistor in series, with a capacitor connected in parallel before the load resistor.

What all of that means is that passive EQ is happening – the capacitor, inductor, and load form a classic resonant circuit. The capacitor and inductor in series allow a range of frequencies through, and the parallel capacitance acts as an additional low-pass filter. (Whether or not this low-pass is significant after the capacitor-inductor bandpass is a whole other issue.)

The issue with passive filter circuits is that everything has an effect on everything else. If the load impedance is adequately high, then we get a nicely damped, wideband filter that sounds natural. If the load impedance is too low, however, the filter gets narrow and odd sounding. This effect can become so pronounced that string instruments start to sound like horns(!)

The obvious fix, then, is to connect the piezo pickup to a very-high impedance device. An easy way to do this is to use an active DI box.

The Buffer Zone

Active DI boxes solve the piezo impedance problem because they can employ buffer amplifiers. The great thing about a buffer amplifier is that its input impedance is very, very high (millions or even billions of Ohms). It also does this in a very small package. You could probably construct a passive DI box with an input impedance in the millions of Ohms, but the size and weight of the thing (not to mention the cost) would be really off-putting.

The downside of using a buffer amplifier is that it requires a power supply. This means batteries, or engaging phantom power from the console. In practical reality, though, this downside is almost negligible. Almost any modern console that’s capable of mixing a full band will have phantom available, and a battery in a DI box will probably last for tens (if not a hundred or so) hours.

So – all of this is just a very long way of saying, “Buy active DI boxes.” They’re pretty much guaranteed to work with any kind of instrument output you encounter, and they can be powered by any half-decent console or mic pre. They remove any need for guesswork, and they can even have nifty extras like signal boosters and guitar cab emulations.

Passive direct boxes are the right choice most of the time, but a reliable, full-featured, active DI is the right choice all the time.

No contest.