Why Lead and Probe Condition Is Part of the Measurement

Why this matters

The leads are the only part of an instrument that gets slammed in a truck door, dragged across a roof, pinched under a panel cover and wound around a hand a hundred times a week. They are also the part no calibration certificate covers and no tech thinks of as part of the instrument. Put those two facts together and you get the most common way a well-maintained meter produces a confidently wrong number: the box was in tolerance and the wire in front of it was not. A degraded lead set does not fail obviously. It shifts a low-resistance reading up, shifts a high-resistance reading down, and holds perfectly still while it does either one.

Before you check or use them

29 CFR 1910.334(c)(2) requires a visual inspection of test instruments and equipment and all associated test leads, cables, power cords, probes and connectors for external defects and damage before use, and requires anything defective or damaged to be removed from service until repaired. That is a duty, not a suggestion, and it is the only OSHA rule in this whole subject that is about the wire rather than the box.

Any reading taken on a live circuit sits behind the energized-work gate at 29 CFR 1910.333(a)(1), with the electrical protective equipment 29 CFR 1910.335(a) requires for the exposure and the arc-flash boundary and PPE your employer's electrical safety program sets under NFPA 70E-2021 in the edition that program adopts. The instrument, the leads, the tips and any adapter must all be rated for the circuit under 29 CFR 1910.334(c)(3); a sibling article owns how those ratings work (see References).

Resistance and insulation tests are de-energized tests and the de-energizing is the hazard control, not a convenience. De-energize, apply electrical lockout under 29 CFR 1910.333(b)(2), or 29 CFR 1926.417 on construction, and prove dead with the live-dead-live sequence in NFPA 70E-2021, 120.5 before a lead touches a winding. An insulation-resistance tester then deliberately applies hundreds or thousands of volts to that winding, which charges it capacitively: use the instrument's discharge function, confirm the winding is at zero volts with a separate meter, and ground it before it is handled. Disconnect drives, control boards and any solid-state device from the circuit first, because the test voltage that proves a winding destroys electronics.

What a certificate covers, and the list it leaves out

A calibration certificate establishes agreement with a reference on the functions and at the points tested, on the day tested, under laboratory conditions. That is a genuinely useful document and it is narrower than most techs read it as. Unless you explicitly sent the accessories and the certificate names them, the following are not covered, and the exclusions are not oversights:

• The test leads and the probe tips. A lab cannot certify a consumable whose condition changes within a week of leaving the bench. Anything they said about it would be false by the time you used it.
• Clips, spring hooks, magnetic hangers, extension leads and any adapter. Same reason, plus most of them were bought separately and never went.
• The clamp jaw faces on a clamp instrument. A nicked or contaminated mating face changes the magnetic circuit, and it is a mechanical condition that happens in the field.
• The battery, the case seal, and the fuse in the current input. Two of those affect readings and the third affects whether the instrument fails safe.
• Everything that happened after it shipped back to you. The certificate is a snapshot, and the drop that matters is always the one after the snapshot.

Which means the accessory chain is a self-inspected item by design. Nobody is coming to check it for you, and the check is short.

What leads add to the number: series resistance

Every lead has resistance, and it sits in series with whatever you are measuring, so it adds. On voltage and on anything above a few tens of ohms, that addition disappears into the rounding. On low-resistance work, motor windings, ground bonds, contactor contacts, shunts, connections, it is often the largest error in the measurement by a wide margin.

Measure your own rather than looking it up: short the two tips together on the resistance range and read the pair. Record the number on the case. Then either subtract it from every low-ohm reading or use the instrument's relative or zero function, which does the subtraction for you and has to be re-taken each time because the value changes with how the leads are lying.

A lead pair that reads meaningfully higher than it did the last time you checked has lost strands somewhere. A pair that reads differently when you flex it has a break that is currently making contact.

What leads add to the number: leakage, capacitance and the wrong alloy

Leakage runs the other way. On high-resistance measurements, an insulation-resistance test being the obvious one, a film of moisture, salt or flux across the probe body and its insulator makes a conductive path in parallel with the thing you are measuring. Parallel resistances combine as the reciprocal of the sum of reciprocals, so the pair always reads lower than the lower of the two. A clean, dry, warm probe body is a measurement requirement on any high-resistance test, not housekeeping.

Capacitance matters where the source cannot drive it. Long leads run together have capacitance between them, which on a high-impedance node or a fast signal changes what you see. That is a loading question and a sibling article owns it.

Temperature leads have a chemistry problem instead of a resistance problem. A thermocouple works because a junction of two dissimilar metals produces a voltage, which means any other junction of dissimilar metals in that circuit is also a thermocouple, generating its own unwanted voltage. Extension wire has to be the correct alloy for the thermocouple type and connected with the correct polarity, and its connectors are made of the alloy for that reason. A plain copper splice in a hot junction box, or a reversed pair, produces a reading that looks plausible at room temperature and moves the wrong direction as the process heats.

What leads add to the safety envelope

The category rating of an assembly is the lowest-rated piece in it, so a lower-rated lead set downgrades the whole instrument regardless of what the meter is marked. Hand-held probe assemblies have their own product standard, IEC 61010-031, and the rating is marked on them.

Beyond the rating, three physical features on a probe are doing safety work and are worth insisting on: finger guards that stop a hand sliding forward onto the tip, shrouded or short exposed tips that cannot bridge two adjacent terminals in a crowded panel, and intact insulation right up to the strain relief, which is where flex fatigue cracks first and where a cracked lead puts line voltage under your palm. A long bare probe shank is the feature that turns a routine reading into a phase-to-phase fault.

The 60-second lead check

Do this at the start of the day and again any time a reading surprises you.

1. Look at the whole length, both leads, both ends, under 1910.334(c)(2). Cracks, cuts, stiffness, exposed strands at the strain relief, a loose or wobbly tip, a jack that no longer grips.
2. Short the tips on the resistance range and read the pair. Compare against the value written on the case.
3. Flex the full length while watching the display. Any jump, flicker or open is a failed lead, and it is failed even though it reads correctly when lying still.
4. Check the markings. Category and voltage on the leads and on the tips, against where you are about to work.
5. Seat the plugs fully and confirm the tips are tight in the probe bodies.

A lead that fails any of these is removed from service, which under 1910.334(c)(2) is the required response, not the cautious one.

The case: one degraded lead set, two opposite wrong answers

A tech is checking out a motor that tripped its overload. Two tests, one after the other, with the same leads. The unit is de-energized, locked out and proven dead first, and the drive is disconnected from the motor leads before either test.

Test one: winding resistance. Nominal for that winding is 1.20 ohms. The instrument's resistance accuracy is on the order of half a percent of reading, which at 1.20 ohms is about 0.006 ohms. The lead set shorts at 0.35 ohms; when it was new on this instrument it shorted at 0.12 ohms.

The tech reads 1.55 ohms and does not zero. That is 0.35 ohms of lead sitting on top of a 1.20 ohm winding, or 29 percent high, and it points squarely at a partially open winding or a bad internal connection. Note the scale of it: the lead contribution is roughly 58 times the instrument's own accuracy specification at that reading. Zero the leads against the value measured today and the winding reads 1.20 ohms, exactly nominal. Zero against the remembered 0.12 ohms instead and it reads 1.43 ohms, still 19 percent high and still pointing at a fault that does not exist.

Test two: insulation resistance, winding to frame. The probe body has been handled with damp gloves in a mechanical room and has a surface film across the insulator. Say the winding's genuine insulation resistance is 200 megohms and the surface path across the contaminated probe is 20 megohms. In parallel, the reciprocal of one over 200 plus one over 20 is one over 0.055, which is 18.2 megohms. The instrument reports about 18 megohms and the motor looks like it is on the way to a ground fault.

The two errors point in opposite directions and both come from the same accessory. Series resistance adds, so a low-resistance reading comes out high and looks like a bad connection. Parallel leakage subtracts, so a high-resistance reading comes out low and looks like failing insulation. A tech who knows only the first one will be baffled by the second, because the mental rule "bad leads read high" is exactly half true.

How it was caught. The winding readings across the three phases were 1.55, 1.56 and 1.54 ohms, all high by the same amount. A genuine winding fault is not symmetrical; a common-mode offset across every phase is an artifact of something shared, and the only shared thing was the leads. That check costs nothing and should be automatic on any set of readings that ought to match each other.

The failure mode. The motor gets condemned twice over on the same visit, from two tests that appear to corroborate each other, when in fact they share a single cause. Corroboration between two measurements taken with the same accessories is not corroboration.

How to verify you got this right

• Write your shorted-lead resistance on the instrument case and re-measure it monthly. A value you remember is a value that is out of date.
• Re-zero for every low-ohm reading, not once at the start. The value moves with lead position and temperature.
• Wipe and dry the probe bodies and insulators before any high-resistance test, and if a reading is unexpectedly low, clean and re-read before you believe it.
• Look for common-mode offsets. When several readings that should match are all off in the same direction by about the same amount, suspect the shared accessory before the equipment.
• Check that a suspect lead fails while flexed, not only while still. An intermittent that passes a static check is the one that will ruin a reading in the field.

References

• 29 CFR 1910.334(c)(2) and (c)(3) - pre-use visual inspection and removal from service of defective instruments, leads, probes and connectors, and the requirement that they be rated for the circuit
• 29 CFR 1910.333(a)(1) and 1910.335(a) - the energized-work gate and required electrical protective equipment
• 29 CFR 1910.333(b)(2), 29 CFR 1926.417 and NFPA 70E-2021, 120.5 - electrical lockout in general industry and construction, and the live-dead-live proving sequence, binding through the employer's electrical safety program
• IEC 61010-031 - safety requirements for hand-held probe assemblies, binding through the product listing
• See related: What a Meter Category Rating Is Protecting You From; What Loading Does When You Connect an Instrument