aioscilloscope / Tools / Current-probe test

Current-probe test: a scope channel in amps, checked end to end

A current probe turns the current in a wire into a voltage. Set the scope channel to amps and type in the probe's ratio, and every reading on that channel should be in amps and correct. This test checks that with a real clamp-on probe on a known current: a bench supply feeding an electronic load.

Results: not yet measured

CH1Why this test

On NovaOS's EEVblog thread, Fire Doger asked:

Other than voltage probes, there are current to voltage probes with different mv/ma based on probe. It would be nice to be able to configure a custom probe multiplier for that channel, and all units for that channel change from mV, V, to mA, A.

NovaOS does this now (our answer on the thread): a channel in amps reads in amps everywhere, in the measurements, their statistics, the cursors and the histogram. You type the probe's sensitivity in mV/A or V/A, and the unit and the ratio are kept across power cycles (the request's page). This test is the proof on a real probe: we bought a current probe to check it end to end.

$479.00Micsig CP503B current probe, bought for this request (what the bench has cost)

The bench supply and the electronic load are general bench equipment, used for many tests, so they are not counted as bought for this one.

The test is a benchmark: it runs the same way on NovaOS, on Rigol's own firmware and on Kiszka's MHO900 mod. Where a firmware has no setting for the unit or the ratio, the result says so.

CH2What you need

Each figure below is quoted from the maker's own document; the page number links to our copy of it.

ItemWhat matters here
Micsig CP503B
clamp-on AC/DC current probe		 Two ranges, 5 A and 30 A Product Introduction p. 4
Output: 1 V per amp on the 5 A range, 1 V per 10 A on the 30 A range p. 4
Bandwidth 50 MHz p. 4
Accuracy ±1 % ±1 mA (5 A range), ±1 % ±10 mA (30 A range), at DC and 45 to 66 Hz p. 4
Wire up to 5 mm across; powered from DC 12 V p. 4
At least 20 minutes' warm-up User Manual p. 25; the scope input at 1 MΩ, not 50 Ω p. 14
Check yours is the newer version: range buttons for 5 A and 30 A, and a 12 V adapter in the box. An older 6 A / 30 A version, powered from USB, is still sold under the same name.
Mean Well NGE18I12-P1J
12 V adapter, in the probe's box		 12 V, 1.5 A rated, 18 W; it runs from 80 to 264 V AC, 47 to 63 Hz NGE18 Specification p. 2
Rigol DP2031
bench power supply, the current source		 We use CH1: 0 to 32 V, 0 to 3 A, over-current protection settable to 3.3 A (CH3 goes only to 6 V) DP2000 Data Sheet p. 8
Its own current reading on CH1, the reference for test 1: 0.15 % + 5 mA p. 9
Rigol DL3031
electronic load, in constant-current (CC) mode		 150 V, 60 A, 350 W DL3000 Data Sheet p. 4
CC setting accuracy on its 0 to 6 A range: 0.1 % + 0.006 A p. 4
Its own current reading, ±(0.1 % + 0.1 % FS) on its 0 to 60 A range, is recorded beside each step but is not the reference p. 5
Transient (switching between two levels) from 0.001 Hz to 15 kHz p. 4
The scopeAny DHO800, DHO900 or MHO900 with a free CH1, and CH2 for the optional power test. A 10x voltage probe for CH2.
LeadsOne red and one black lead from the supply to the load, each thin enough to fit the probe's 5 mm jaw, and long enough to keep the clamp away from both instruments.

CH3Wiring, step by step

The drawings are block diagrams, not pictures of the front panels: find each connector on your own instrument by its label.

Wiring for the current-probe test Step 1: the probe's 12 V adapter powers the probe's control module. Step 2: the control module's BNC cable goes to the scope's CH1. Step 3: the bench supply's plus terminal goes by the red lead to the load's plus input, and its minus terminal by the black lead to the load's minus input. Step 4: zero the probe at the control module. Step 5: the probe's jaw clamps around the red plus lead only, the arrow on the probe pointing from the supply to the load, the direction the current flows. Step 6, optional, drawn dashed: a 10x voltage probe from the scope's CH2, its tip on the supply's plus terminal and its ground clip on the supply's minus terminal. Oscilloscope CH1 input set to 1 MΩ CH1 CH2 12 V adapter from the probe's box Probe control module range 5 A · Zero button Bench supply DP2031, output CH1, 12 V + red − black Electronic load DL3031, CC mode + red − black + − + − current returns to the supply optional: 10x probe, tip on +, clip on − 1 2 3 4 5 6
Red is +, black is −. The dashed line is optional.

  1. Power the probe from its own 12 V adapter, plugged into the probe's control module, and let it warm up for at least 20 minutes. Never power it from a scope's USB port: Micsig forbids it User Manual p. 13.

  2. The probe's BNC into the scope's CH1, with CH1's input at 1 MΩ (the DHO800 and DHO900 have only 1 MΩ; on an MHO900 the kit sets it) p. 14.

  3. Supply to load: on the supply's CH1, the red lead from + to the load's + input, the black lead from − to the load's − input. Set CH1 to 12 V, its current to 3.0 A (the channel's maximum) and its over-current protection to 3.2 A, and leave its output off until the wiring is checked.

  4. Zero the probe: nothing in the jaws, the jaw closed until it clicks, the 5 A range chosen, then press Zero. It degausses and zeroes; if its light flashes five times, the zero failed: press it again p. 13.

  5. Clamp the + lead only, the red one. The arrow on the probe's clip points the way the current flows: from the supply's + towards the load's + p. 14. Close the jaw until it clicks, with the lead in the middle of the jaw p. 26.

  6. optional, for the power test: a 10x voltage probe into the scope's CH2, its tip on the supply's + terminal and its ground clip on the supply's − terminal.

The clamp, close up

Right: the + lead only, the arrow with the current The jaw closes around the red plus lead only. The arrow on the probe's clip points the same way as the current, from the supply towards the load. The channel reads plus 1 amp for 1 amp. ✓ Right from supply + to load + current: reads +1 A at 1 A
The clip's arrow and the current point the same way.
Wrong: both leads in the jaw Both the red plus lead and the black minus lead pass through the jaw. Their currents flow in opposite directions and cancel, so the probe reads 0 amps. ✗ Wrong: both leads + red → ← − black they cancel: reads 0 A
The current goes out on + and back on −: in one jaw the two cancel.
Wrong: the arrow backwards The jaw is on the plus lead, but the arrow on the clip points back towards the supply, against the current. The channel reads minus 1 amp for 1 amp. ✗ Wrong: arrow backwards from supply + to load + reads −1 A at 1 A
Turn the probe round: the arrow must point to the load.

CH4The four tests

  1. Amps on the channel. CH1 set to amps, with the probe's ratio: 1 V/A on the 5 A range, 0.1 V/A on the 30 A range. The load draws 0, 0.1, 0.5, 1, 2 and 2.9 A in turn; the top step stays under the supply's 3.0 A so the supply never limits the current. Each step is read twice at the same setting, the load off (the zero) and on, and the channel's mean, in amps, less that zero must agree with the supply's own current reading within 1 % + 5 mA on the 5 A range, or 1 % + 20 mA on the 30 A range. The unit A must show on the channel's label, the axis, the measurements, the cursors and the trigger level, and the unit and ratio must survive a saved setup and a power cycle.
  2. Power. With the optional CH2 probe: a math channel CH1 × CH2, in watts, against the supply's own volts times amps, within 2 %.
  3. A current step. The load in transient mode, switching between 0.5 and 1.5 A at 1 kHz, half the time at each. A rising-edge trigger on CH1 at 1 A must fire on every one of 20 single captures, the rise and fall must read in amps, and the two levels must agree with the load's settings within the same band plus the load's own setting accuracy.
  4. Zero drift. 30 minutes with nothing in the jaws after a zero, the channel's mean read every minute.
Test 1: the load's current, step by step The load's current in six steps over time: 0, 0.1, 0.5, 1, 2 and 2.9 amps, each held while the channel's mean is read. Test 1: six steps, 0 to 2.9 A 0 A 1 A 2 A 3 A 00.10.5 122.9 0 A 0.1 A 0.5 A 1 A 2 A 2.9 A time → each step held while the mean is read (time not to scale)
The reference current, in turn. Then the same on the 30 A range.
Test 3: the current step at 1 kHz The load switches between 0.5 and 1.5 amps a thousand times a second, one period every millisecond, half the time at each level. A dashed line at 1 amp marks the trigger level; the trigger fires on each rising step. Test 3: 0.5 ↔ 1.5 A at 1 kHz 0 A 0.5 A 1 A 1.5 A - - - trigger level, 1 A 01 ms2 ms3 ms4 ms time → one period every millisecond
The trigger fires on each rising step through 1 A.

CH1Running it

current_probe_kit is one Python file (Python 3.8 or newer, standard library only, MIT licence) that runs the four tests over SCPI and writes each result to a JSON file. It sets CH1 to amps with the probe's sensitivity, 1 V/A on the 5 A range and 0.1 V/A on the 30 A range, and the 20 MHz bandwidth limit Micsig suggests for small currents User Manual p. 26. Where a firmware takes the ratio the other way round, the kit measures that and records it.

Our own runs read the reference current from the supply over its LAN port. With --source manual, the kit asks you to type the reference current at each step instead, from your own meter.

python3 current_probe_kit.py amps  --host SCOPE --range 5A  --source manual
python3 current_probe_kit.py amps  --host SCOPE --range 30A --source manual
python3 current_probe_kit.py power --host SCOPE --range 5A --vch 2 --source manual
python3 current_probe_kit.py step  --host SCOPE --range 5A --source manual
python3 current_probe_kit.py drift --host SCOPE --range 5A --minutes 30 --interval 60
python3 current_probe_kit.py persist-setup --host SCOPE --range 30A
python3 current_probe_kit.py persist-arm   --host SCOPE --range 30A --state s.json
    (now power-cycle the scope)
python3 current_probe_kit.py persist-check --host SCOPE --range 30A --state s.json
python3 current_probe_kit.py report *.json

Add --dry-run to any of them to run against a simulated scope: it prints every command it would send and touches nothing. The last three check that the unit and the ratio survive a power cycle; report puts the runs from several scopes in one table. The README has every option, and METHOD the arithmetic and the limits.

Files

Version 1.0.0, MIT licence (LICENSE).

FileBytesSHA-256
current_probe_kit.py86,926ceccf3cf3bcaef811b0f7f9516fe68e7a86f0be2fa0b8f12e4ba96ababa37f83
test_current_probe_kit.py12,627f70439c926e58caa97a660aec7795fe97a6b0d46f876b26279bb9f7247227dd0
README.md13,0541c3bc72fb65da4c862a628fc25edf87b830ca5506d4ddf21c2c7f0f28a533f02
METHOD.md10,458f2a1164e194a20a1ebb2a2b798c612e7435eb45fa71692e6b90a97ca98fdc060
LICENSE1,0690f248fd800e1876da972b234d751e04f8ac6730364c6f9bc04bd0e1f89b62b06

The same sums are in SHA256SUMS; check them with sha256sum -c SHA256SUMS. The kit's own tests run against its simulated scope: python3 -I test_current_probe_kit.py (23 tests).

CH2Results

Not yet measured. The results will be on NovaOS vs stock, each with its date, its method and the raw data, once a run on our scopes is done.

CH3Safety

The probe is isolated from the wire it clamps, but the supply and the load are live: 12 V at up to 3 A. Before you turn the supply's output on, check that no lead, terminal or clip touches another.

Power the probe only from its own 12 V adapter, never from a scope's USB port.

Keep the current within the probe's range, and press Zero again after an overload: an overload magnetises the probe User Manual p. 12.