Picking a multimeter by CAT rating — not price. Here are the models we actually use and recommend for electronics, automotive, and solar/120V work.
CAT III 600V, True RMS, auto-ranging, NCV, includes temperature probe. Best multimeter under $50.
CAT III 600V, True RMS, electrician's favorite, built-in voltage detector. Industry standard for professional electricians.
CAT IV 600V / CAT III 1000V, True RMS, waterproof IP67, high accuracy, 4-year warranty. Real industrial tool.
400A DC current clamp. Essential for solar and automotive work where you need current without breaking the circuit.
Buy the right CAT rating for the work environment. Not the one with the prettiest display.
Safety first
A CAT II meter on a CAT III panel is a real arc-flash hazard. When in doubt, buy the next rating up. Your test leads need to match the CAT rating too — cheap leads ruin a good meter.
| CAT Rating | Nominal Voltage | Typical Use Case | Recommended Meter |
|---|---|---|---|
| CAT I | Up to 600V | Low-voltage electronics: breadboards, battery packs, USB circuits. No mains-connected equipment. | Any handheld or bench meter |
| CAT II | 600V / 1000V | Single-phase household outlets, small appliances, portable tools plugged into wall receptacles. | KAIWEETS HT118A, Fluke 101 |
| CAT III | 600V / 1000V | Three-phase distribution, commercial/industrial wiring, building mains panels, solar arrays up to 1000V, HVAC equipment. | Fluke 117, KAIWEETS HT118A |
| CAT IV | 600V / 1000V | Utility service entrance, outdoor conductors, pole-top work, lightning-strike-prone locations. The highest transient-energy environment. | Fluke 87V MAX, Fluke 88V |
The spec sheet checklist we actually use. Ignore marketing copy — focus on these.
Required for accurate measurements on non-sinusoidal loads — inverters, LED drivers, motor drives, dimmers. "Averaging" meters will read 20-30% wrong on these. If your work involves anything with a power supply or motor, you need True RMS.
Auto-ranging saves time and reduces mistakes. Manual-range meters force you to dial in the correct scale every time — useful for precision work, slower for everyday diagnosis. Most modern meters are auto-ranging with a manual override option.
Your meter must be rated for the highest-energy environment you work in. If you work on service panels, you need CAT IV. Solar arrays up to 1000V need CAT III. Don't use a CAT II hobbyist meter on CAT III work — it's a real safety gap.
At least 10 MΩ (10 million ohms) on voltage ranges. Lower than that and the meter itself starts loading the circuit you're measuring, giving incorrect readings — especially on high-impedance sensors and microcontroller circuits.
4,000 count is fine for most work — shows 0.00 to 3.99 on a 4V range. 20,000 count (like the Fluke 87V) gives you an extra decade of precision. Bench meters go to 6.5 or 7.5 digits — useful for calibration, overkill for field work.
Useful for electronics (heat sink temps, battery packs) and automotive (catalytic converter, coolant). Look for a meter that includes a K-type thermocouple — or at minimum has a thermocouple input jack.
A loud continuity beeper saves you from staring at the screen while probing. Diode test mode should show voltage drop at a reasonable test current (1 mA is standard). Combined with resistance mode, these three cover half of all troubleshooting.
Capacitance range lets you test capacitors and start caps without an LCR meter. Frequency (Hz) mode is essential for working on generators, VFDs, and signal circuits. Duty cycle measurement is a nice bonus for motor and SMPS debugging.
Backlight is non-negotiable for attic, crawlspace, and panel work. Data hold (max/min is even better) lets you capture a reading and move it to a safer location before writing it down. Bar graph feature gives you visual trend info you lose on a numeric display.
Ceramic fuses with sand filling for arc suppression. Cheap meters ship with glass fuses — which are worse than useless because they can shatter and continue to arc. Fluke and other professional-grade meters use HRC ceramic fuses on current ranges.
| Model | Approx. Price | CAT Rating | Counts | DC Accuracy | AC Accuracy | True RMS | NCV | Capacitance | Temperature | Our Rating |
|---|---|---|---|---|---|---|---|---|---|---|
| Fluke 87V MAX | $480-520 | IV 600V / III 1000V | 20,000 | ±0.05% | ±0.7% | Yes | Yes | 9,999 µF | Yes | 9.8 / 10 |
| Fluke 117 | $170-195 | III 600V | 6,000 | ±0.5% | ±1.0% | Yes | Yes | 9,999 µF | Optional | 9.2 / 10 |
| Fluke 101 | $85-100 | III 600V | 6,000 | ±0.5% | ±1.5% | Yes | No | 1,000 µF | No | 8.2 / 10 |
| KAIWEETS HT118A | $35-45 | III 600V | 6,000 | ±0.8% | ±1.2% | Yes | Yes | 9,999 µF | Yes (included) | 9.0 / 10 |
| KAIWEETS HT206D (Clamp) | $50-75 | III 600V | 4,000 | ±1.5% | ±2.0% | Yes | Yes | 400 mF | Yes | 8.8 / 10 |
| Uni-T UT61E+ | $55-70 | III 1000V | 22,000 | ±0.06% | ±0.8% | Yes | No | 220 mF | Yes | 9.1 / 10 |
| Klein Tools MM700 | $140-170 | III 600V | 6,000 | ±0.5% | ±1.2% | Yes | Yes | 100 mF | Yes | 8.9 / 10 |
| Tektronix DMM4020 (Bench) | $1,800-2,300 | I (lab only) | 200,000 | ±0.024% | ±0.12% | Yes | N/A | 100 mF | Yes | 9.9 / 10 |
Residential rooftop arrays operate at DC voltages up to 600V (sometimes higher on commercial systems).
Modern cars run on 12V but have sensors that require millivolt-level accuracy.
Breadboards, 3.3V microcontrollers, signal circuits. Resolution is everything.
The Uni-T UT61E+ is 22,000 count — which means it displays millivolt-level precision on the 20V range. DC accuracy is ±0.06%, comparable to Fluke's professional series at a fraction of the price. Includes capacitance, frequency, duty cycle, and a serial/USB data logging option.
Downsides: no NCV, case is less rugged, fuses are ceramic but harder to replace. For bench electronics work rather than field work.
Tektronix DMM4020 or equivalent Keysight models (DMM4020, DMM4040, DMM4050). 6.5-digit resolution, ±0.024% DC accuracy. These are the tools you use when you need to verify another meter's reading — or when you're characterizing precision reference circuits.
Not a field tool — these are 2-3kg bench instruments. Used for characterizing power supplies, sensor output verification, and checking the calibration of your handheld meters.
The $10 no-name Amazon meters. Here's why.
Cheap meters ship with glass fuses instead of ceramic — glass fuses can shatter under transient overvoltage and continue to arc inside the meter. The thin test leads don't have the insulation thickness required for CAT II work.
A CAT-II-labeled no-name meter may arc-flash when exposed to even moderate transients on 120V. Not worth the risk to save $30. We have personally watched a $15 meter smoke when connected to a 240V circuit under load.
The specs listed on cheap meters are often fictitious. We've tested $12 meters against lab reference standards and seen errors of 5-10% on DC voltage ranges. The internal shunt resistors drift with temperature and age — good meters use trimmed references that stay stable for years.
Display segments die. Rotary switches break. Input protection is nonexistent (reverse probe a cheap meter and it's dead). A $40 KAIWEETS HT118A is the minimum bar for actual work — anything below that is a toy.
Yes if you're measuring anything driven by an inverter, LED driver, variable-frequency drive, or dimmer — basically anything with a switch-mode power supply or solid-state control. "Averaging" meters assume a pure sine wave and will read 10-30% low on distorted waveforms. If your work is purely resistive (incandescent bulbs, heating elements on direct mains), an averaging meter is technically fine — but modern True RMS meters cost barely more, so there's no reason to buy one.
It's the resolution indicator. A 4,000-count meter can display 0 to 3,999 on its main range, giving you 3 full digits plus a partial first digit. A 20,000-count meter gives you 4 full digits — enough precision to see millivolt changes on a 12V rail. Higher counts give finer resolution but don't directly improve accuracy (which is determined by the internal reference and ADC). For field work, 4,000-6,000 count is adequate. 20,000 count is for precision and lab work. 200,000 count (6.5-digit bench meters) is for calibration and component characterization.
Only if your meter is rated for it — CAT III 1000V or CAT IV 600V minimum. The CAT rating is what matters here, not the display range. Some meters display "1000V" on the dial but are only CAT III 600V rated — meaning the display can show 1000V but the insulation wasn't certified for sustained 1000V in a CAT III environment. Read the actual safety rating, not just the dial position.
Annually for professional use if you need ISO 17025 traceable calibration — this is required for commercial solar installers, ISO-certified labs, and most industrial maintenance programs. Every 2-3 years for hobby and personal use, or whenever you suspect a reading is wrong. Most handheld multimeters hold their calibration very well — the big drift happens in the first few months of use, after which a good reference chip is remarkably stable.
(1) CAT-rated test leads with silicon insulation — match or exceed your meter's rating. (2) Insulated alligator clips for hands-free probing. (3) A K-type thermocouple if your meter supports it — the included one is usually fine, but upgrading to a beaded probe gives faster response. (4) Spare fuses of the correct type — most people lose a meter forever because they shorted a current range and didn't have spare fuses on hand.
Full technical specifications for EcoFlow, Jackery, Anker SOLIX, Bluetti, and Goal Zero stations. Battery chemistry, inverter output, real-world runtime.
Toyota Prius, Camry, Highlander HV battery packs. Module specifications, voltage ranges, and fault code diagnostics.
200+ fault codes for hybrids, power stations, drones, and smart home devices. With actual diagnostic steps, not just "contact support."