Knowledge hub / PostProgMetal / gear
Speaker Impedance and Power Split Calculator
Music knowledge / Guides and practical tools
About this reference
From the music-learning collection, adapted for Philojain Music Muse. Referenced sources remain credited in the article.
Wiring cabinets together changes the load your amplifier sees and decides how the power divides between them. Enter the impedances and this works out the total load in series and in parallel, and how much of the amplifier’s output each cabinet actually receives.
The calculator
You enter the impedances and the amplifier rating. Nothing here states what load your amplifier wants — that is on the back of the amp, it differs between valve and solid-state designs, and getting it wrong on a valve amp is expensive.
Series and parallel, and why the split reverses
In parallel every cabinet sees the same voltage, so the one with the lowest impedance draws the most current and takes the most power. In series the same current flows through all of them, so the one with the highest impedance drops the most voltage and takes the most power. The two wirings load an unequal pair in opposite directions, which is the part that surprises people.
Equal cabinets are the easy case
Two identical cabinets in parallel halve the impedance and split the power evenly; in series they double the impedance and still split it evenly. It is only when the impedances differ that the share becomes lopsided, and the arithmetic above is worth running before rather than after.
What the measurements actually say
The arithmetic above uses nominal impedance, because that is what cabinets are sold by. Measured impedance is a different quantity, and the gap between them is not small. A measured 8 Ω Eminence driver reads exactly 8 Ω at precisely three frequencies — 50, 200 and 500 Hz. It peaks near 70 Ω just above 100 Hz at mechanical resonance and climbs back to around 50 Ω by 20 kHz. Even at DC it is not nominal: a Celestion Vintage 30 rated 8 Ω measures 7.3 Ω. Nominal impedance is a mean, not a value, and the output transformer is working into a load that changes with every note. Verified — component-level measurement.
The speaker is also doing more filtering than the amplifier. Celestion publish the Vintage 30 as 70–5,000 Hz and the G12M Greenback as 75–5,000 Hz — devices sold as stopping at 5 kHz. Clipping has no upper harmonic limit, which is why digital amp simulators run their clipping stages at 8× oversampling, around 353 kHz: a nonlinearity cannot be instructed to generate only harmonics below half the sample rate. The raw electrical signal genuinely contains ultrasonic content and a listener hears essentially none of it — independent measurement puts the 7–20 kHz band 20–25 dB down on a real Vintage 30.
There is no such thing as “the V30 curve”. A 2018 Mesa-branded Vintage 30 measured 13 dB more energy at 6.5 kHz than a 2022 Chinese-made one. And a Greenback measured in an actual Marshall 4×12 showed its −3 dB point at 106 Hz — above the open low E’s own 82.41 Hz. A low E through a 4×12 is already partly a psychoacoustic illusion, reconstructed by the ear from harmonics the cabinet does reproduce.
How these figures are graded. Each is marked verified where it comes from a manufacturer datasheet or a published component-level measurement, practitioner consensus where credentialed builders agree but no measurement was found, and unverified where a claim is repeated widely and traceable nowhere. Where a figure could not be sourced it is said so rather than quietly omitted.
What it does not do
Speaker impedance is nominal, not fixed — the real impedance of a driver rises steeply at its resonance and again at the top of its range. This page computes with the nominal figures, which is what matching is conventionally done on, but it means the load your amplifier sees varies with frequency. It also has no view on what load your amplifier can safely drive: that is on the amp, and on a valve amp it matters a great deal.
Keep exploring