Reference
Eight ohms is a rating, and the daisy chain is a parallel circuit
The number on the back of a cabinet is a nominal impedance: a single figure standing in for a curve that swings from a few ohms to several tens across the band. The amplifier does not see the nominal figure. It sees the curve, and it is the minimum of the curve — and how many of them are hung on the same output — that decides whether the amplifier is comfortable, clipping, or in protect.
Below: what the rating actually promises, the two rules for combining loads and the tables they produce, why unequal loads share power unequally in opposite directions depending on the wiring, what the load does to an amplifier's current, and the constant-voltage line, where impedance is derived from a tap's power rating and the wrong tap on the wrong line is a fire.
Nominal, minimum and what a meter reads
A moving-coil driver's impedance is its voice-coil resistance at DC, rises to a peak at the mechanical resonance where the motor pushes back hardest, falls to a minimum an octave or so above that, and then climbs again with the voice coil's inductance. A cabinet with a passive crossover adds the crossover's own behaviour on top. The result is nothing like a resistor, and the single figure printed on the panel is a convention for summarising it.
IEC 60268-5 defines the convention: a loudspeaker's rated impedance is chosen so that the lowest value of the impedance modulus in the rated frequency range is not less than 80 % of it. That is the entire promise. An 8 Ω cabinet may dip to 6.4 Ω; a 4 Ω cabinet to 3.2. Manufacturers who publish the minimum are telling you more than the standard requires, and those who print a nominal figure the curve does not support are the reason amplifiers have protection circuits.
A multimeter reads DC resistance, which is the voice-coil resistance alone — typically 70 to 85 % of the nominal figure, so an 8 Ω driver reads 5.5 to 7 Ω and a 4 Ω driver around 3. That is useful for telling a 4 Ω cabinet from an 8, for finding an open coil, and for checking how a multi-driver box is wired inside. It says nothing about impedance at any frequency the amplifier will drive it at, and on a cabinet with a passive crossover it may not even see the HF section, because a series capacitor is open at DC.
Series adds, parallel divides
Zseries = Z₁ + Z₂ + …
1 / Zparallel = 1/Z₁ + 1/Z₂ + …
For n identical loads that collapses to n·Z in series and Z/n in parallel. Nothing about a loudspeaker changes the rules; what changes is which one applies, and on a PA it is almost always parallel. The link socket on the back of a cabinet is wired straight across the input. Every box added to a daisy chain is another load across the same amplifier output, and the connector's name for it — loop, link, thru — does not alter the circuit.
The table gives the total for a chain of identical cabinets, and beneath it the minimum the amplifier may actually meet once the 80 % rule is applied to the total. It is the second figure that should be compared with the amplifier's stated minimum load.
| Topology, four 8 Ω drivers | Total | Wiring |
|---|---|---|
| All four in parallel | 2.0 Ω | every + to +, every − to − |
| All four in series | 32.0 Ω | a chain: − of one to + of the next |
| Two series pairs, in parallel | 8.0 Ω | the usual 4×12 guitar-cabinet wiring |
| Two parallel pairs, in series | 8.0 Ω | electrically identical to the row above |
| each cabinet ↓ cabinets in parallel → | 1 | 2 | 3 | 4 | 6 | 8 |
|---|---|---|---|---|---|---|
| 16 Ω | 16.00 Ωmay dip to 12.80 | 8.00 Ωmay dip to 6.40 | 5.33 Ωmay dip to 4.27 | 4.00 Ωmay dip to 3.20 | 2.67 Ωmay dip to 2.13 | 2.00 Ωmay dip to 1.60 |
| 8 Ω | 8.00 Ωmay dip to 6.40 | 4.00 Ωmay dip to 3.20 | 2.67 Ωmay dip to 2.13 | 2.00 Ωmay dip to 1.60 | 1.33 Ωmay dip to 1.07 | 1.00 Ωmay dip to 0.80 |
| 4 Ω | 4.00 Ωmay dip to 3.20 | 2.00 Ωmay dip to 1.60 | 1.33 Ωmay dip to 1.07 | 1.00 Ωmay dip to 0.80 | 0.67 Ωmay dip to 0.53 | 0.50 Ωmay dip to 0.40 |
Nominal total, and beneath it the IEC 60268-5 minimum. Amber where the minimum is below 3.2 Ω — the floor for an amplifier rated to 4 Ω — and red where it is below 2 Ω, which is below the rating of nearly everything. Three 8 Ω cabinets on one channel is the classic case: 2.67 Ω nominal, 2.13 possible, and fine on an amplifier rated for 2 Ω and nothing else.
Who gets the power
Identical loads share equally either way. Unequal ones do not, and the direction depends on the wiring. In parallel every load sees the same voltage, so power goes as V² / Z and the lower impedance takes more. In series every load carries the same current, so power goes as I² · Z and the higher impedance takes more. Two rules, opposite senses, and the second is the one that catches people.
The practical consequence is that a 4 Ω cabinet and an 8 Ω cabinet on the same parallel output is not "a bit unbalanced" — the 4 Ω box gets two thirds of everything, and if the two are similar in sensitivity the 8 Ω one is 3 dB quieter. Mixing impedances on one output is a level decision, and usually one nobody meant to make.
| Loads | In parallel | In series |
|---|---|---|
| 8 Ω and 8 Ω | 50 % / 50 % | 50 % / 50 % |
| 8 Ω and 4 Ω | 33 % / 67 % | 67 % / 33 % |
| 8 Ω and 16 Ω | 67 % / 33 % | 33 % / 67 % |
| 4 Ω and 16 Ω | 80 % / 20 % | 20 % / 80 % |
What the load does to the amplifier
| Load | Power | Current | At the IEC minimum |
|---|---|---|---|
| 8 Ω | 200 W | 5.0 A | 6.3 A into 6.40 Ω |
| 4 Ω | 400 W | 10.0 A | 12.5 A into 3.20 Ω |
| 2.67 Ω | 600 W | 15.0 A | 18.7 A into 2.13 Ω |
| 2 Ω | 800 W | 20.0 A | 25.0 A into 1.60 Ω |
An amplifier holding 40 V RMS at its output (200 W into 8 Ω), if it could hold it into anything. Real amplifiers cannot: the power supply sags and the current limit engages, which is why a spec sheet's 4 Ω figure is less than double its 8 Ω one and why a 2 Ω figure, where given, is less again.
An amplifier is a voltage source with limits. Halve the load and, if the voltage holds, the current doubles and so does the power — that is the appeal of running low impedances. The cost is that every part of the output stage carrying that current dissipates more, the supply rails droop under it, and the protection circuitry is watching. An amplifier's minimum-load rating is the impedance at which it can deliver its specification and survive doing so, and it is judged against the minimum of the load's curve, not the nominal figure.
The cable is part of the load, and it matters more as the load falls. Twenty metres of 2.5 mm² is 0.28 Ω there and back, which throws away 3.3 % of the amplifier's output into an 8 Ω load, 6.5 % into 4 Ω and 12.1 % into 2 Ω — the last of those is 1.1 dB. Running 2 Ω to save an amplifier channel and then losing a decibel in the loom is the false economy the voltage-drop page puts numbers to.
Constant-voltage lines: impedance from a power rating
Distributed systems — ceiling speakers, paging, anything with dozens of small loudspeakers on one run — do not use low-impedance loads, because the cable losses above would be intolerable at 4 Ω over a building. They step the amplifier's output up to a nominal 100 V (Europe and most of the world) or 70.7 V (North America) at full power, and every loudspeaker has a transformer with taps marked in watts. The tap's impedance is not printed because it is implied:
Ztap = Vline² / Ptap
Loads on the line are in parallel as before, so the line's total impedance is the line voltage squared over the sum of the tap wattages — and the design rule collapses to one sentence: the sum of the taps must not exceed the amplifier's rating. No reciprocals, no minimum-load arithmetic. That is the whole reason the scheme exists.
The trap is the two voltages. A transformer's 10 W tap is 10 W on the line it was wound for. Put a 70.7 V-rated tap on a 100 V line and the same 500 Ω sees a voltage 1.41 times higher, so it draws 2.0 times its rated power — 20 W through a 10 W transformer and driver. The inverse mistake, a 100 V tap on a 70.7 V line, is merely 3 dB quiet.
| Tap | On 100 V | On 70.7 V | 70.7 V tap fed 100 V |
|---|---|---|---|
| 2.5 W | 4000 Ω | 1999 Ω | 5 W |
| 5 W | 2000 Ω | 1000 Ω | 10 W |
| 10 W | 1000 Ω | 500 Ω | 20 W |
| 20 W | 500 Ω | 250 Ω | 40 W |
| 30 W | 333 Ω | 167 Ω | 60 W |
| 60 W | 167 Ω | 83 Ω | 120 W |
Your cabinets, your amplifier
Low impedance first: how many cabinets of what rating, wired how, on an amplifier with what minimum load and what output voltage. The tool gives the nominal total, the IEC minimum, the current at that minimum, the verdict, and what each cabinet receives.
Then the constant-voltage line: the line voltage, the amplifier's rating, and the taps in use. Enter the taps as a comma-separated list of wattages.
These calculators need JavaScript. The tables above cover the common cases.
The low-impedance tool applies IEC 60268-5's 80 % rule to the nominal total. A manufacturer's published minimum is better information than that rule and should be used instead where it exists. Amplifier output voltage is the RMS figure at rated power into 8 Ω — √(P × 8) — and the current shown assumes the amplifier holds it, which at low impedances it will not.
Sources
- IEC 60268-5, Sound system equipment — Part 5: Loudspeakers, the rated-impedance definition: the lowest value of the modulus of impedance in the rated frequency range shall be not less than 80 % of the rated impedance. Paywalled; the clause is quoted verbatim in secondary sources including Audioholics' measurement standard and Dynaudio's note on impedance, and the 80 % figure is consistent across every one checked.
- Small, R. H., "Direct-Radiator Loudspeaker System Analysis", J. Audio Eng. Soc. vol. 20 no. 5, 1972 — the shape of a driver's impedance curve and why the minimum sits above resonance. Background; no figure on this page is taken from it.
- Series and parallel combination, power sharing, constant-voltage tap impedance — Ohm's law and Kirchhoff's laws. Every total on the page is computed from those and checked for the identities that must hold.
- The cable figure — 20 m of 2.5 mm² at 1.724 × 10⁻⁸ Ω·m, the same arithmetic as the voltage-drop page and the multipin looms page.
Assembled 8 September 2026 with AI assistance. The arithmetic is exact and checked at build time; the 80 % rule is the standard's floor and a real cabinet may be better or, from a careless manufacturer, worse. Nothing here has been measured on a cabinet, and the DC-resistance ranges are typical rather than guaranteed.
Companion pages: voltage drop by cable length and conductor size for the loom's share of the load, and multipin looms for the damping-factor view of the same resistance.