You’re restoring a 1970s stereo receiver. Everything powers on, the display lights work, and you can hear music through the speakers. But something feels off about FM radio. Stations drift slightly as you tune, or you catch only parts of strong signals. You twiddle the tuning knob, and the needle jumps around the dial inconsistently. The audio quality from FM sounds thin or distorted compared to what you remember from childhood—or compared to the AM stations coming through the same receiver.
You suspect the tuner is failing. But before you condemn the whole circuit or hunt for a replacement tuner module, there’s something more fundamental to check: the FM discriminator alignment. This is one of those adjustments that separates a receiver that merely works from one that sounds and tunes like it was built yesterday. Most people never adjust it because they either don’t know it exists or assume it’s been factory-set and sealed forever. That assumption is often wrong.
The FM discriminator is the circuit that translates the frequency-modulated radio signal into audio. When it drifts out of alignment—and it will, over 40, 50, or 60 years—your tuning accuracy drops, station separation gets fuzzy, and the recovered audio loses definition. The good news: this is one of the few analog radio alignments you can diagnose, measure, and often correct yourself, using nothing more than a signal generator and a voltmeter.
What You’ll Learn and Why It Matters
This article explains how the FM discriminator actually works, why it needs periodic alignment, and what happens sonically and electrically when it drifts. You’ll learn how to diagnose discriminator misalignment using simple test equipment, and you’ll understand the difference between a misaligned discriminator and other failures that might look similar.
Understanding discriminator alignment will save you from unnecessary repair costs (replacing a working tuner is expensive) and will help you recognize when alignment is genuinely the problem versus when the issue is elsewhere in the signal chain. By the end, you’ll have a practical framework for testing and, if appropriate, adjusting your own receiver’s FM front end.
How FM Detection Works: The Physics of the Discriminator
FM radio encodes audio information by shifting the carrier frequency up and down around a center frequency. At 101.5 FM, the actual transmitted signal doesn’t sit at exactly 101.5 MHz—it wobbles around that center frequency in step with the audio waveform. A 1 kHz tone at the station might shift the carrier between 101.498 and 101.502 MHz, for example. That frequency shift is called deviation, and it’s tightly regulated (±75 kHz in the US broadcast band).
The receiver’s job is to capture that frequency-modulated signal and translate it back into voltage variations that represent the original audio. That translation happens in the discriminator, and the alignment of the discriminator determines how accurately and linearly that translation occurs.
The classic Foster-Seeley discriminator
Most vintage receivers from the 1950s through 1980s used a Foster-Seeley discriminator or a variant called a ratio detector. Let’s focus on the Foster-Seeley because it’s more common and demonstrates the alignment principle more clearly.
The Foster-Seeley discriminator works like this: the incoming FM signal is fed to a transformer (the discriminator transformer) that has a primary winding and a center-tapped secondary. The secondary creates two equal but opposite voltage signals. These are fed to two diodes, one on each half of the secondary. The output of each diode is smoothed and filtered, creating two DC voltage signals. The difference between these two voltages is what emerges as audio.
Here’s the critical part: the alignment of this circuit depends on the frequency response of the discriminator transformer. When the incoming signal is exactly at the center frequency (the tuned frequency of the transformer), both diodes receive equal voltage, and the output is zero. When the signal frequency is slightly above center, one diode receives more voltage, and the output swings positive. When the signal is below center, the output swings negative.
This relationship—input frequency versus output voltage—should be perfectly linear and symmetric around the center frequency. When it’s not, you have misalignment.
Why alignment drifts over time
The discriminator transformer has a tuned primary and secondary. These are physically wound coils with capacitors (sometimes internal, sometimes external) that set the resonant frequency. Over decades, several things happen:
- Capacitor drift: Vintage capacitors in the discriminator circuit change value. Ceramic capacitors creep upward or downward. Paper-in-oil capacitors absorb moisture. Electrolytic capacitors fail over time, losing capacitance entirely. Any shift in these capacitor values shifts the resonant frequency of the discriminator transformer.
- Coil inductance change: The physical properties of the wire and core can shift slightly due to age, temperature cycling, and mechanical stress. This is rare but not impossible.
- Tuning slug movement: Many discriminator transformers have adjustable ferrite or iron slugs inside the coil form. Vibration, thermal cycling, and even tiny mechanical shifts in the receiver chassis can cause these slugs to drift slightly. They’re usually locked with phenolic resin, but that bond weakens over decades.
- Component tolerance drift near the discriminator: The diodes themselves can shift in characteristic slightly with age. More commonly, resistors in the audio output filtering stage change value, which affects the DC operating points of the diodes.
The result: the peak of the discriminator curve (where the transformer is resonant) shifts away from the nominal FM channel frequency. Suddenly, the relationship between input frequency and output voltage is no longer symmetric around the center frequency you’re trying to receive.
What Misalignment Sounds and Measures Like
A misaligned discriminator produces several recognizable symptoms, and they’re often confused with tuner problems, IF strip issues, or speaker problems.
Tuning instability and drift
When the discriminator is misaligned, the center frequency of its response curve shifts. This means the frequency at which the receiver responds best is no longer the frequency of the station you’re trying to hear. The AFC (automatic frequency control) circuit—a feedback loop that tries to keep the tuning locked—becomes confused. It over-corrects and under-corrects, causing the needle to hunt around the dial instead of settling cleanly.
Physically, this manifests as drift: you tune to a station, and 30 seconds later, the signal seems to slide slightly off-frequency. The audio might briefly cut out or sound distorted as the AFC circuit chases the moving frequency.
Reduced sensitivity and fuzzy station separation
A misaligned discriminator’s response curve is not only shifted but also often flattened. The peak becomes less pronounced, and the sides of the curve are less steep. This means the discriminator is less responsive to frequency changes, so small deviations in the incoming signal (which carry the audio information) produce smaller changes in the output voltage.
The practical consequence: weak stations vanish from the dial. Strong stations come in, but they’re mushy—lacking clarity. The spacing between adjacent channels feels compressed, and you might hear adjacent-channel bleed-through where two strong stations are close to each other.
Audio distortion and loss of high-frequency content
FM audio bandwidth is determined by how much the carrier can deviate (±75 kHz in the broadcast band). The discriminator is supposed to respond linearly to this full ±75 kHz swing. If misaligned, one side of the discriminator curve might be steep and the other side shallow, or the whole curve might be shifted so that only part of the ±75 kHz range is captured linearly.
This causes asymmetric distortion: peaks clip while troughs don’t, or vice versa. You hear this as a subtle but persistent grittiness in the audio, especially on strong stations. High-frequency content—cymbals, sibilants—often suffers first because those frequencies are encoded as higher deviations in the FM signal. If the discriminator is only capturing part of the deviation range linearly, the highs get compressed and distorted.
Mono/stereo switching instability
Stereo FM receivers use a 19 kHz pilot tone to switch the multiplex decoder between mono and stereo mode. The discriminator’s linearity affects the output signal-to-noise ratio, which indirectly affects whether that pilot tone is cleanly decoded. A misaligned discriminator can make the stereo indicator flicker or prevent stereo locking entirely, even on strong stations.
How to Diagnose Discriminator Misalignment
Before you go chasing this problem, you need to rule out other failures. A weak or distorted FM signal could be caused by a failing IF strip, a bad limiter stage, a defective tuner module, or even a misaligned front-end. Here’s how to isolate discriminator alignment as the culprit.
Step 1: Verify the basic tuner function
First, confirm that the tuner is actually locking and receiving stations. Tune to a strong local FM station and listen. You should hear audio, and the tuning meter (if present) should indicate you’re on frequency. If there’s no audio at all, or if no stations appear across the dial, the problem is likely upstream of the discriminator—bad tuner module, dead oscillator, or broken IF strip.
To test this more objectively, you need an RF signal generator. This is a piece of test equipment that outputs a clean FM signal at any frequency. If you don’t own one, many community maker spaces, electronics repair shops, and college electronics labs have them available for use or will rent time on them for a reasonable fee.
Step 2: Generate a known-frequency FM test signal
Set your RF generator to output an unmodulated (or low-level modulated) FM signal at a popular FM frequency: 98.0 or 101.5 MHz. Set the generator’s output level to about –20 dBm (a standard test level that approximates a weak but tunable station). Inject this signal into the receiver’s antenna input using a simple capacitive coupling (a 10 pF capacitor in series works fine).
Tune the receiver to the exact frequency you’re generating. You should see a clear signal lock on the meter. Now, slowly vary the generator frequency up and down around that center frequency—say, from 97.98 to 98.02 MHz—while watching the output on a voltmeter connected across the discriminator output (ask yourself: where exactly is this point in your receiver’s schematic? See the service manual). You’re not looking for audio yet; you’re looking at the raw DC discriminator voltage.
On a properly aligned discriminator, that DC voltage should sweep from negative (when the signal is below the center frequency) to positive (when the signal is above), and the transition should be symmetric. The steepness of the slope and the center point where the voltage crosses zero tell you about alignment.
Step 3: Plot the S-curve (or observe it visually)
This is the real diagnostic test. The graph of discriminator output voltage versus input frequency is called the S-curve, and its shape tells you everything.
To generate this curve:
- Set the RF generator to your test frequency (e.g., 98.0 MHz).
- Set a sensitive DC voltmeter to the 0–1 V scale and connect its probes across the discriminator output load resistor (or the point before the audio coupling capacitor—see the schematic).
- Starting from about 2 MHz below the center frequency, slowly increment the generator frequency in small steps (e.g., 0.05 MHz), recording the voltmeter reading at each step.
- Continue until you’re about 2 MHz above the center frequency.
- Plot these readings on graph paper or a spreadsheet.
An ideal S-curve for a 98 MHz discriminator should:
- Cross zero voltage at exactly 98.0 MHz
- be symmetric: the voltage swing above 98 MHz should mirror the swing below 98 MHz
- Be steep in the middle: the slope through the zero-crossing should be sharp (typically 1–2 mV per kHz, depending on design)
- Have gentle tails on either side: the curve should flatten out away from center, not droop or rise suddenly
If your S-curve shows any of these deviations—zero crossing shifted away from the nominal frequency, asymmetry, shallow slope—you have discriminator misalignment.
Step 4: Interpret the S-curve to identify the specific misalignment
If the zero crossing is shifted but the curve is symmetric: The discriminator transformer’s primary and/or secondary resonant frequency has drifted. This is almost always a capacitor value shift. The transformer itself is probably fine.
If the curve is asymmetric but centered correctly: One side of the secondary coil is resonating differently from the other, or one diode is failing. This could indicate a bad capacitor in only one half of the circuit, or a diode age-related shift.
If the slope is too shallow everywhere: The overall Q (quality factor) of the transformer has dropped. This could mean the transformer is damaged, shorted turns, or the load impedance has changed (bad resistor or diode).
If the curve crosses zero properly but shows distortion away from center: The transformer might be partially shorted, or the adjustment slug has moved mechanically. This is less common but happens when vibration or thermal cycling loosens the slug.
Adjusting the Discriminator (If You’re Qualified to Do So)
Discriminator adjustment is straightforward in principle but requires care and proper equipment. Before you attempt this, understand what you’re doing and what can go wrong.
Safety first
Your receiver likely has high voltage (400–600V) in the power supply and on the screen grids of vacuum tubes. The discriminator itself operates at low voltage, but you’re working inside an energized chassis. Wear an ESD wrist strap grounded to the chassis. Discharge yourself frequently. If you’re uncomfortable working inside a powered receiver, have a professional technician perform the adjustment. The cost is typically $75–200, which is reasonable peace of mind.
If you proceed, do so with the receiver powered on (you need signal for the adjustment) but be aware of shock hazard and never touch the high-voltage rails or capacitors.
The adjustment procedure
Most vintage receivers have a small trimmer capacitor or variable inductor (a slug) that adjusts the discriminator transformer resonance. This is usually labeled “FM Discriminator” or “FM Disc” on the schematic, and it’s located near the discriminator transformer itself.
- Generate a test signal at the nominal frequency (e.g., 98.0 MHz) at about –20 dBm injected into the antenna.
- Connect a DC voltmeter to the discriminator output (the point just before the audio coupling capacitor).
- Locate the trimmer adjustment (usually a small screw or slug on the discriminator transformer).
- Slowly adjust the trimmer while watching the voltmeter. Turn the adjustment screw in small increments—a quarter turn at a time. You’re trying to move the zero-crossing of the S-curve to align with your test frequency.
- Check your work by varying the generator frequency slightly (e.g., ±0.2 MHz) and confirming that the DC output voltage swings up and down symmetrically around zero.
- If available, repeat the full S-curve plot to verify that the curve is now centered and symmetric.
If your receiver doesn’t have a readily accessible trimmer, the discriminator transformer might have an internal adjustment (a slug inside the coil form) that requires opening the shielded can. Proceed with extreme caution; these cans are sometimes sealed with epoxy, and forcing them open can break the transformer.
When to stop and call a technician
If you adjust the trimmer and the S-curve doesn’t improve, or if it gets worse, stop. Multiple possible causes are at play:
- The discriminator transformer itself is damaged and cannot be adjusted
- A capacitor in the discriminator circuit is failing and needs replacement
- A diode is leaking or failing
- The adjustment range of the trimmer has been exceeded (meaning the original design frequency has drifted so far that the trimmer can’t correct it)
Any of these require component-level repair. At that point, you’re better served by a professional, or by consulting a detailed service manual and being prepared for recapping a vintage receiver or replacing specific components.
Discriminator Alignment vs. Other Tuner Problems
Misalignment is easy to confuse with other failures. Here’s how to distinguish them.
Misaligned discriminator vs. failing tuner module
A dying tuner oscillator or varactor tuning diode will cause the receiver to be unable to lock on any frequency or to lock very weakly. You’ll hear no audio or barely audible audio across the entire dial, even at strong local stations. The tuning meter will barely deflect.
A misaligned discriminator allows the tuner to lock and produce audio, but the audio quality or tuning stability is degraded. Strong stations come in relatively clearly; weak stations are lost. The tuning meter responds normally.
Test: Use your RF generator at two very different frequencies (e.g., 88.5 MHz and 107.5 MHz). If the receiver locks and plays audio at both, the tuner oscillator is working. If one end of the band is dead or very weak, you have a tuner problem, not a discriminator problem.
Misaligned discriminator vs. failing IF strip
The intermediate frequency (IF) strip amplifies and filters the signal after the mixer. If an IF stage is failing (bad transistor, blown capacitor), you’ll see a dramatic loss of sensitivity across the entire band. Weak stations disappear; even moderately strong stations sound weak.
Discriminator misalignment affects tuning lock and audio quality more than raw sensitivity. You can still receive strong stations clearly; you just have to tune them precisely.
Test: Compare the tuning behavior of your receiver with a reference receiver (if you have one, or a friend’s) on the same strong station. If your receiver is noticeably less sensitive, the IF strip is probably the culprit. If sensitivity is similar but tuning is fussy or audio distorts, the discriminator is suspect.
Misaligned discriminator vs. limiter stage failure
The limiter stage in an FM receiver flattens the amplitude variations in the IF signal, leaving only frequency information for the discriminator. A failing limiter (usually a transistor or tube circuit with an automatic gain control loop) allows amplitude variations to reach the discriminator, where they can cause distortion.
The result sounds similar to discriminator misalignment—distorted audio, especially on strong stations. The difference is that a bad limiter will show rapid gain fluctuations in the audio output as signal strength changes; you’ll hear the audio level pump up and down. A misaligned discriminator produces steady distortion regardless of signal strength.
Test: Use your RF generator to create a strong, steady test signal. Slowly vary the generator’s output level. A working limiter will keep the audio level nearly constant. A failing limiter will show the audio level rising and falling with the input signal.
Advanced Diagnostic: Measuring THD and Frequency Response
If you have access to a distortion analyzer and audio spectrum analyzer, you can quantify discriminator performance more precisely.
Total harmonic distortion (THD) test: Feed your FM generator a modulation signal (e.g., 1 kHz sine wave at 25 kHz deviation, a standard FM test signal). Measure the THD of the recovered audio output. A properly aligned discriminator should show THD below 2% on this test. Misalignment typically raises THD to 3–5% or higher.
Frequency response test: Use an FM generator with variable modulation frequency (sweep from 30 Hz to 15 kHz) and measure the output amplitude at each frequency. The response should be relatively flat across the audio band. Discriminator misalignment often causes a presence peak (a boost around 2–4 kHz) or a rise in high-frequency content, reflecting the asymmetry of the S-curve.
These tests are more involved than the basic S-curve plot, but they provide clear, numeric evidence of the problem and proof that your adjustment worked.
Why Alignment Matters More Than You Think
You might be tempted to dismiss discriminator alignment as a minor tweak—something that affects only weak-signal performance or fringe cases. It’s worth understanding why that’s not quite right.
FM stereo reception depends on discriminator linearity more than any other factor. Stereo separation (how well the left and right channels are isolated) is determined by how well the receiver rejects crosstalk from the multiplex subcarrier. That rejection is entirely dependent on the linearity of the discriminator S-curve. A misaligned discriminator with an asymmetric S-curve will cause stereo crosstalk—you’ll hear the opposite channel bleeding through.
Signal-to-noise ratio in FM is also tied to discriminator linearity. The quieting curve—the relationship between input signal strength and recovered audio noise—is optimized only when the discriminator is properly aligned. A misaligned discriminator reaches its quieting point at a higher signal level, meaning weak stations sound noisier.
This is why vintage receivers that have been properly maintained and aligned sound noticeably better than ones that haven’t. It’s not magic; it’s engineering working as intended.
When to Align vs. When to Replace the Whole Tuner
Not every tuner problem is worth fixing with alignment alone. If you’ve confirmed that misalignment is the issue but the discriminator transformer itself is damaged (shorted turns, mechanical damage), or if the discriminator adjustment has maxed out its range with no improvement, you’re looking at component replacement or tuner module swap.
Modern tuner modules (DSP-based software-defined receivers) are cheap—$30–80 for retrofit kits. But they may not sound or feel like the original and often require wiring modifications. For a proper vintage HiFi setup, keeping the original tuner is preferable if it can be salvaged.
If the discriminator trimmer has reached its physical limit (it’s turned all the way in or out) and still doesn’t center the S-curve, you likely have a capacitor value shift that’s too large to compensate with the existing trimmer. At this point, you have three options:
- Replace the discriminator capacitors: If you can identify which capacitor has drifted (usually by comparing the measured capacitance to the schematic value), replacing it is a straightforward repair. Cost: $5–20 for the part and an hour of labor.
- Replace the entire discriminator transformer: If the transformer itself is damaged or if the capacitor is internal to the can, this is the better option. Cost: $20–50 for the part, plus labor.
- Replace the tuner module or the entire receiver chassis: If the cost of repair approaches the value of the receiver, or if you prefer not to work at the component level, replacement might be pragmatic.
The decision depends on the value of the receiver, your skill level, and your commitment to original restoration.
Practical Workflow: Diagnosis to Decision
Here’s a straightforward decision tree for tackling FM tuner problems.
Does the receiver receive any FM stations clearly?
No: The problem is in the tuner, oscillator, or IF strip. Discriminator alignment is not the issue. Consult a service manual for the specific tuner circuit and test the oscillator frequency and IF response.
Yes, but with one or more of these symptoms:
- Tuning is fussy (needle drifts after locking)
- Strong stations sound distorted or thin
- Weak stations are inaudible even though AM works fine
- Stereo indicator flickers or won’t lock
Proceed to the next step.
Is the problem frequency-dependent (worse at one end of the band)?
Yes: You probably have a tuner or IF problem, not discriminator misalignment. Misalignment affects the entire band uniformly.
No (affects all frequencies equally): Continue to the next step.
Do you have access to an RF signal generator?
No: Without a signal generator, you cannot conclusively diagnose discriminator misalignment. You can infer it based on the symptoms above, but you won’t have numeric proof. Consider borrowing or renting a signal generator for a few hours, or taking the receiver to a technician who has one.
Yes: Plot the S-curve as described above. If the curve is shifted, asymmetric, or shallow, discriminator misalignment is confirmed.
Is the discriminator adjustment accessible and within its adjustment range?
No, or it’s maxed out: Component-level repair is required. Identify and replace the misaligned capacitor, or replace the discriminator transformer, or decide that the repair cost is not justified.
Yes: Proceed with adjustment as described above. Re-plot the S-curve to confirm the fix.
Final Thoughts: Alignment as Preventive Maintenance
If you own a vintage receiver that you plan to use regularly, periodic discriminator alignment is worth considering as preventive maintenance. Unlike restoring sticky potentiometers, which degrades over time with use, discriminator alignment typically needs attention only once every 20–30 years, and only if the receiver hasn’t been recapped or the discriminator transformer replaced.
If you’ve recently recapped your receiver or replaced the discriminator transformer, you have a perfect opportunity to align the discriminator while you have the RF generator and voltmeter out. It takes 30 minutes and ensures that your receiver will deliver optimal FM performance for years to come.
The alignment of an FM discriminator is not a mysterious black art reserved for factory technicians. It’s a straightforward application of feedback control and resonant circuit tuning—the same principles that govern countless other aspects of vintage electronics. With the right equipment and a methodical approach, you can diagnose and correct it yourself, and in doing so, you’ll restore a critical part of your receiver’s original performance.