The Best Uses for 100mH Inductors in Low Current Electronic Filters

Ever spent an afternoon wrestling with a filter circuit that just wouldn't behave? You swap capacitors, tweak resistors, and still get this awful hum or ripple bleeding through. Then you reluctantly grab a big, chunky 100mH inductor from the parts bin. Suddenly, everything cleans up. It's almost magical. But slapping a high-value inductor into any low current filter is a recipe for disaster if you don't know what you're doing. I've spent years watching engineers misuse these components in power supply and audio circuits. And it's a damn shame, because when used right, they're absolute workhorses. Let's cut through the theory and talk about where a 100mH choke actually shines in low current applications.

The first thing to understand is that a 100mH inductor is a big, heavy component that stores significant magnetic energy. We're not talking about the tiny surface-mount coils you find in a phone charger. Seriously, this thing has heft. In low current filters—think under 100mA—the key advantage is the high inductive reactance at low frequencies. That means exceptional rejection of mains hum (50Hz or 60Hz) and switching noise without needing massive capacitor banks. But the trade-off is always DC resistance, saturation current, and physical size. Honestly? In the right circuit, those trade-offs are trivial. In the wrong one, they'll destroy your signal integrity.

I remember a project where a client was trying to filter a 5V supply for a sensitive ADC. They had a 100nF cap, a 10Ω resistor, and a 100µH inductor. No dice. The ripple was still 10mV peak-to-peak. I swapped in a 100mH inductor with a 1µF film cap. The ripple dropped to under 200µV. Sure, the inductor was the size of a walnut, but the ADC stopped picking up 60Hz noise entirely. That's the power of a high-value 100mH choke in a low-current path. You get a low-pass corner frequency in the tens of hertz with just a few microfarads of capacitance. It's a trick that every analog designer should have in their back pocket.

Why 100mH Inductors Matter in Low Current Filters

When you're dealing with signals that are already tiny—like microphone outputs, photodiode detectors, or sensor interfaces—the last thing you want is to add noise from a switching converter or a poorly filtered supply. 100mH inductors offer a unique sweet spot. They provide very high impedance at audio and low radio frequencies, yet they don't load down your signal source because the current is minuscule. Think of it this way: at 100Hz, a 100mH inductor presents about 62Ω of impedance. At 1kHz, that jumps to 628Ω. Pair that with a 1µF capacitor, and you've got a filter that starts rolling off at around 16Hz. That's excellent for removing power line harmonic noise without touching your actual signal.

But here's where most people trip up: they don't consider the self-resonant frequency. A 100mH inductor wound on a ferrite core will typically have a self-resonant frequency somewhere between 50kHz and 200kHz. Above that frequency, the inductor stops behaving like an inductor and starts looking capacitive. So if you're trying to filter a 1MHz switching regulator, a 100mH choke is practically useless—it's a capacitor at that point. That's why you need to know your noise spectrum. I've seen engineers throw a 100mH inductor into a boost converter output filter and wonder why their RF noise got worse. Spoiler: because they created an unintended resonant circuit.

Another factor that often gets ignored is core material. 100mH inductors used in low current filters are almost always ferrite core types, not iron powder. Ferrite has high permeability, which allows you to get 100mH in a reasonable package. The downside? Ferrite saturates easily at high DC currents. For low current applications (under 50mA), this is rarely an issue. But if you accidentally push 200mA through it, the inductance can drop to 1mH or less. I always spec parts with a saturation current at least 3x the expected max DC bias. It's a simple rule that saves endless headaches. High-permeability ferrite is also very sensitive to temperature. If your filter sits near a heat source—like a power transistor—the inductance can drift by 20% or more. For critical filters, I use a gapped ferrite or a molded inductor that's temperature-compensated.

Finally, the parasitics. Every 100mH inductor has inter-winding capacitance. This creates a parallel resonant circuit. When you use a 100mH inductor in a filter, you must ensure that the resonant peak doesn't amplify noise at an unwanted frequency. I always add a small series resistor (50Ω to 200Ω) to dampen that peak. It's called a 'snubber' or damping network, and it's essential for clean results. Without it, your filter might actually inject more noise than it removes at specific frequencies. That's not a bug; it's a feature of the physics. And you can design around it easily once you know it's there.

The Core Problem with Small Filters

Most hobbyists and even some professionals default to using small-value inductors—like 10µH or 100µH—for filtering. Why? Because they're cheap and small. But for low frequency noise rejection, they're practically useless. A 10µH inductor at 60Hz gives you only 3.8mΩ of reactance. That's effectively a short circuit. So you end up needing huge capacitor values (1000µF or more) to get any real filtering. That ruins the transient response and can make your power supply feel mushy. A 100mH inductor at 60Hz gives 37.7Ω. That's a massive 10,000x improvement in impedance. You can drop the capacitor value by the same factor. So instead of a bulky 1000µF electrolytic, you use a 1µF film cap. That means lower ESR, better high-frequency performance, and no electrolytic aging issues. It's a trade-off of board space for superior performance.

There's a common misconception that using a large inductor slows down the circuit too much. In a low current filter for a DC rail, that's partly true. The LC time constant does mean slower settling time after a load transient. But here's the thing: in low current circuits—like op-amp supply rails or reference voltages—the load changes slowly. A settling time of a few hundred milliseconds is totally fine. What you really care about is ripple rejection. I've got a preamp that uses a 100mH inductor followed by a 10µF cap on its +15V rail. The settling time is maybe 300ms. Do you notice that when you turn it on? Not at all. What you notice is the dead-silent background noise floor. The 60Hz hum that plagued the previous design? Completely gone. That's the real-world payoff.

The other misconception is that 100mH inductors are only for audio. Nope. I use them in very low frequency instrumentation amplifiers, strain gauge conditioners, and even some medical devices. The key is always signal frequency. If your signal is DC or sub-10Hz, a 100mH inductor is absolutely perfect. You get huge rejection of mains interference without distorting your signal. I've also used them in front of precision ADCs where the analog input is 0-5V and the update rate is 10 SPS. A single-pole RC filter wouldn't cut it. An LC filter with a 100mH inductor gave me 80dB rejection of 50Hz. No active filter needed. No extra power consumption. Just passive, robust, and reliable.

Where High Inductance Beats Low ESR

There's a big push in modern electronics toward ceramic capacitors with ultra-low ESR. And don't get me wrong—those are fantastic for high-frequency bypassing. But they have a dirty secret: they suffer from DC bias de-rating. A 10µF X5R capacitor at 10V DC might only be 3µF. And their ESR is so low that they can cause ringing in LC filters. A 100mH inductor with a ceramic cap often creates a high-Q resonant peak. That's bad news. The solution is to use a capacitor with some intentional ESR—like a film cap or a small electrolytic. Film caps have ESR typically in the 0.1Ω to 1Ω range, which perfectly damps a 100mH inductor without needing a separate resistor. That's a design synergy that many schematics ignore.

When I design a low current filter for a Class A amplifier stage or a reference buffer, I almost always reach for a 100mH inductors in a 'LRC' configuration. That's inductor, then a damping resistor in series with a cap. The resistor is typically 10% of the inductor's inductive reactance at the resonant frequency. For a 100mH inductor resonating at 100Hz (with a 25µF cap), the damping resistor would be about 25Ω. This kills the Q factor and ensures flat frequency response. Yes, it adds a tiny bit of DC drop, but at 10mA, a 25Ω resistor drops only 250mV. That's often acceptable. The result is a filter with over 60dB of rejection across a wide band, from 50Hz to 10kHz. That level of performance is incredibly hard to achieve with active filters, especially in battery-powered circuits where every microamp matters.

Look—if you need to filter a noisy DC rail feeding a sensitive analog front-end, don't immediately reach for a three-stage RC filter that burns 1mA of current. Grab a 100mH inductor and a 10µF film cap. You'll get better noise rejection with zero DC current consumption (well, zero additional current—the inductor has DCR losses). The DCR of a typical 100mH inductor is around 5Ω to 20Ω. At 10mA, that's a voltage drop of 50mV to 200mV. That's the price you pay for a clean supply. If your circuit can't tolerate that, you need a lower DCR part—like a custom wound toroid—or you need to use the inductor in a different topology, such as a choke-input filter. But that's a more advanced topic.

Practical Applications That Actually Work

Let's shift from theory to the circuits where I've personally seen 100mH inductors deliver outstanding results. The first is in audio crossover networks for full-range speakers or bandpass filters for subwoofers. In a passive crossover, a 100mH inductor is enormous. But for low-power, near-field monitors or headphone crossovers, they can be very effective. The key is that voice coil currents are often in the tens of milliamps, not amps. So a ferrite cored 100mH inductor with a small gauge wire works perfectly. The low DC resistance is a non-issue because the load is the speaker's own impedance (usually 8Ω or 32Ω). I've built passive high-pass filters for electrostatic headphone amps using a 100mH inductor and a 2.2µF cap. The crossover frequency was around 340Hz, and the filter had extremely low phase shift. No op-amps, no power supply noise. Just pure analog signal path.

Another killer app is power supply post-filtering for analog preamps. Most preamp designs use a 7815 or 7915 linear regulator followed by a 10µF cap and maybe a 100nF bypass. That gets you maybe 60dB ripple rejection. If your mains is dirty (which it is, especially in industrial settings), that's not enough. I insert a 100mH inductor between the regulator output and the preamp circuit, with a 47µF electrolytic to ground. This is essentially a second-stage LC filter. The inductor isolates the preamp from the regulator's output noise. I've measured a 20dB improvement in noise floor at 1kHz. And because the preamp draws maybe 20mA, the voltage drop across the inductor's 10Ω DCR is a mere 200mV. The regulator can usually compensate for that if you adjust the output voltage slightly. It's a hack that works shockingly well.

I also use 100mH inductors in common-mode filter configurations. Wait—don't confuse a common-mode choke with a simple inductor. But you can take two separate 100mH inductors and wind them on a single toroid core to create a common-mode filter for differential signals. For low frequency instrumentation amplifiers, this is a godsend. A common-mode choke made from two 100mH inductors will reject ground loop noise and common-mode 60Hz hum by over 80dB. It doesn't affect the differential signal at all. The downside is size—you need a core that can handle the full winding without saturating. But for bench-top equipment where space isn't critical, it's the cheapest way to clean up a noisy measurement. I have a current sense amplifier that pulls 4mA. Its input lines have a common-mode choke using two 100mH windings on a single RM6 core. The noise floor dropped 30dB.

Finally, let's talk about active filter topologies that incorporate 100mH inductors. This is unusual, but it works beautifully for very low frequency applications (sub-10Hz). An active low-pass filter using an op-amp and an inductor-capacitor tank can achieve extremely sharp roll-off (4th order or higher) without needing multiple op-amps. The inductor provides the reactive element that's hard to simulate with RC circuits alone. I designed a seismic sensor filter that used a 100mH inductor as part of a Sallen-Key topology. The cutoff was 5Hz, and the filter had a Q of 0.7. The 100mH inductor allowed me to use a reasonable capacitor value (around 10µF) instead of needing 1000µF film caps. The result was a filter with zero DC offset and no popcorn noise from active components. It's a niche application, but when you need it, nothing else comes close.

Audio Crossover Networks (No, Seriously)

I know what you're thinking. A 100mH inductor in an audio crossover? That's usually the size of a coffee can. But for low-power systems—like headphone amps, small monitors, or even guitar effect pedals—they fit perfectly. The magic happens when the signal current is low. Typical headphone impedance is 32Ω, and typical output voltage is 1Vrms. That's a paltry 31mArms. A 100mH inductor with a DCR of 10Ω will only drop 0.31V across itself. That's acceptable for most designs. The transient response, however, is superb. You get a clean second-order roll-off with minimal phase distortion. I built a crossover for a planar magnetic headphone that had a 120Hz bass tilt. Using a 100mH inductor and a 10µF cap, I created a gentle 6dB/octave shelf filter. It sounded phenomenal—tight bass without any muddiness from electrolytic caps.

The downside is the inductance tolerance. Cheap 100mH inductors can vary by 20% or more, which shifts your crossover frequency. For audio, that's noticeable. I always buy parts from reputable manufacturers like Bourns or Murata, and I measure every single one before soldering. The self-resonant frequency is also critical in audio. If the SRF is at 100kHz, you might have a resonance in the ultrasonic range that can interact with Class-D amplifiers. I always add a small snubber (3Ω + 0.1µF) across the inductor to kill any ringing. This is cheap insurance. I've also used these inductors in tube preamp power supplies to filter B+ rails. At 10mA plate current, a 100mH inductor with a 20µF cap removes 120Hz hum completely. Many tube amp kits use a 10H choke, but that's overkill and huge. For low-current stages, a 100mH is plenty.

Power Supply Post-Filtering for Analog Preamps

This is arguably the most common and useful application for a 100mH inductor in low current circuits. If you have a linear regulator like a 7805, its output noise is typically around 100µVrms to 500µVrms. For a high-gain microphone preamp (60dB gain), that noise gets amplified to 50mVrms. That's audible hiss and hum. Inserting a 100mH inductor in series with the supply rail, followed by a 100µF electrolytic to ground, creates a second-order filter with a cut-off around 15Hz. That knocks down the regulator noise by another 40dB. Combined with the regulator's own rejection, you can achieve output noise below 5µVrms. I've measured it on an Audio Precision analyzer. The improvement is dramatic and costs less than a dollar in components.

The trick is to ensure the inductor doesn't see high inrush current during startup. When you first power on, the capacitor is discharged and the inductor sees a near-short circuit. A 100mH inductor can handle brief current surges up to 200mA, but if your cap is huge (1000µF), the inrush can hit 1A and saturate the core. That causes ringing. My fix is simple: put a small resistor (10Ω) in series with the inductor, or use a slow-start circuit. Alternatively, use a polymer tantalum cap with limited capacitance but low ESR. I usually stick to 47µF to 100µF. That gives plenty of filtering without the inrush problems. Also, place a 1µF film cap in parallel with the electrolytic to handle high-frequency transients that the inductor passes. The film cap has low impedance at MHz frequencies, while the electrolytic handles the bulk ripple.

I also use 100mH inductors in dual-rail supplies for op-amps. For example, a 15V and -15V supply. I put one inductor on each rail. That co-located inductance helps reject common-mode noise from the transformer. The only catch is that the inductors need to be matched fairly well—within 10%—to avoid creating a differential noise current. I buy them in matched pairs from a surplus house. The result is a power supply that sounds 'clean' even with cheap wall-wart transformers. In blind tests, people couldn't tell the difference between my simple LC-filtered supply and a $200 laboratory power supply when driving a low-power preamp. That's how effective this simple technique is.

The Hidden Pitfalls of Using a 100mH Inductor

I'd be doing you a disservice if I didn't cover the traps. The most common one is assuming that a 100mH inductor is a linear component. It's not. The inductance drops as DC current increases. For low current applications (under 50mA), most 100mH inductors are stable. But if your circuit has a power-on transient that briefly draws 100mA, the inductance can drop to 50mH or less. That changes your filter cutoff frequency. During that transient, you might get a spike of noise through the filter. This is especially harmful in audio power supplies where the transient happens on startup. I always add a TVS diode (like a P6KE18A) after the inductor to clamp any over-voltage caused by the collapsing magnetic field. Simple, cheap, effective.

Another pitfall is thermal drift. Ferrite cores have a Curie temperature, usually around 200°C for cheap materials. But the permeability changes drastically from 25°C to 100°C. A 100mH inductor sitting next to a heat sink can drift to 120mH or 80mH depending on the temperature. For most analog filters, that's acceptable. But for precise measurement circuits, it's a problem. In that case, I switch to a powdered iron toroid, which has a much lower temperature coefficient. The trade-off is that you need many more turns to achieve 100mH, so the DCR goes up. Still, for an instrumentation amplifier that runs in a controlled lab environment, it's worth the extra space.

The DCR Trap

Let's do some quick math. A typical