Skip to content

What to measure before you buy anything

Most power purchases are made on an estimate that turns out to be wrong by a factor of two. These five instruments replace the estimate with a reading, and between them they cost less than the mistake.

The estimate that is always wrong

How much does the fridge use. How much is the van drawing when everything is off. How much did the solar actually make today.

Every one of those gets answered with a guess, and the guess sizes a battery, a panel or an inverter. A reading costs less than the second purchase.

1. The plug-in meter

Start here. No installation, no risk, and it answers the appliance question directly.

The figure that separates a useful one from a useless one is the lower end of its range. Standby draw sits between a third of a watt and a couple of watts, and a meter with a one-watt floor reports most of it as zero.

Resolution and accuracy are different things, and a good listing publishes both.

2. The DC clamp meter

For everything the plug meter cannot reach: a battery feed, a solar string, a parasitic drain.

Around a conductor rather than in series, so nothing gets disconnected. The three figures that matter are the DC range, the jaw opening — a battery cable with a lug is thicker than it looks — and whether there is a zero function.

Not one clamp meter in this catalog publishes a safety category, which is the rating that says which circuits an instrument may go on.

3. The DIN rail meter

For a circuit rather than an appliance. A heat pump, a workshop, a charger, a solar feed.

If the reading has to divide a bill, MID certification is what gives it standing. If it only has to inform you, it does not matter at all, and the cheaper meter measures just as well.

A pulse output with a published rate is what gets the reading into software.

4. The USB power meter

Small, cheap, and it settles arguments about chargers, cables and power banks.

Four values at once — volts, amps, elapsed time, accumulated capacity — is what makes a charge legible while it happens. Running a power bank flat through one gives a real figure to set against the number printed on the case.

5. The battery monitor

A shunt in the negative lead, counting amp-hours in and out.

That is the only honest state-of-charge measurement. On lithium in particular, voltage is nearly flat from 20% to 90% and a voltmeter tells you almost nothing.

Install it once, and every other number in the system becomes checkable.

The order these get bought in

One and four on a whim. Two when something is wrong. Three and five when the system is big enough to have questions that last longer than an afternoon.

Which station covers your load?

You do not know what you consume
No product page can answer in your place.
Consumption swings with the seasons
A week of readings beats an annual average.
One appliance dominates
Measure that one first; the rest can wait.
The need is occasional
A peak gets measured; it cannot be deduced from a label.

Picking by the load

Sizing a storage system

Without readings you buy a capacity at random, and find it too small or paid for in vain.

Understanding a bill

A per-appliance reading says where the energy goes, which no overall bill shows.

Checking a label

Stated consumption holds for a precise regime; the real regime is usually somewhere else.

The numbers that decide run time

What the instrument records
An instantaneous display is no use for sizing; it is the running total over time that speaks.
The power range covered
An instrument meant for heavy loads reads standby badly, and the other way round.
How the data is read back
A reading that cannot be exported ends up copied by hand, and therefore never reread.

Which figure for which load

What you want to knowWhat you measureOver what period
Baseline drawPower at restSeveral days
Start-up peakMaximum powerAt the moment of starting
Daily needAccumulated energyA full week
SeasonalityAccumulated energyTwo opposite seasons

Where the sizing goes wrong

Trusting the rating printed on the appliance
That is a maximum, not a consumption. A 2,000 W appliance can use very little over a day.
Measuring one day and concluding
A single day catches an exception and misses the usual regime. A week is the useful minimum.
Forgetting standby
It never shows in use but adds up continuously, and it weighs on a storage system.

By what you need power for

Starting from nothing
Measures the three biggest appliances first, and decides afterwards.
Heading for solar
Needs the hourly profile, not just the total: producing at midday only serves what draws at midday.
Hunting a leak
Compares standby appliance by appliance rather than watching the main meter.

Cycle life and how long the pack lasts

A measuring instrument is only worth the regularity of its use. A stray reading in a drawer serves nobody: what counts is the series, and therefore the habit of writing it in the same place.

  • Keep readings in the same format from one time to the next.
  • Note what was running during the measurement, or the figure is unreadable six months later.
  • Take a fresh reading after any change of appliance.
  • Check the instrument was not unplugged in the middle of a running total.

Safety

Measuring means handling sockets and sometimes clamps on live conductors.

  • Measurements on an opened circuit are not made without competence and purpose-built equipment.
  • This site publishes stated specifications and describes no live-measurement procedure.
  • Do not exceed the ranges published by the instrument’s maker.
  • A damaged instrument gets replaced, not taped up.

Questions this guide gets asked

Which one should I buy first?

The plug-in meter, almost always. It is the cheapest, it needs no installation, and it answers the question most people actually have: what is this appliance costing me. Everything else on this list answers a question you only get to once the first one is settled.

Why does a DC clamp need a zero function?

Because a Hall-effect sensor reads a small offset with no current flowing, and that offset drifts with temperature. On a battery feed carrying 200 A it is noise; on a parasitic drain of half an amp it is the entire reading. One clamp meter in this catalog names a zero function.

Is a battery monitor a measurement instrument?

More than a voltmeter is. A shunt counts what actually went in and out over time, which is what tells you state of charge. Battery voltage is a poor proxy for that on lead-acid and close to useless on lithium, where it stays almost flat across most of the discharge.

How this order was worked out

The stations listed come from the catalog of this site. Specifications are those the manufacturer and the Amazon listing publish; prices are what Keepa recorded on the date below, and both move. The order follows declared specifications, stated strengths weighed against stated limits, and price. Nothing here has been charged or discharged by us, and no page claims it was.

Last checked on