I'm checking current of kitchen timers to estimate operation time.

The obstacle here is that in a "passive" mode (no blinking/beeping/key pushes, only working counters + LCD) devices consume 8-12 uA but my multimeter breaks line if current out of 20 uA range and this happens during power on & menu navigation by pushing buttons.

So I started with 200mA range, altered device state into desired mode and started to lower range via mechanical switch on the multimeter (VC830L): 200mA => 20mA => 2mA => 200uA => 20uA.

I saw some "blinks" on device LCD, fortunately I am able to lower range for all tested devices without ruining their operation.

I'm an amateur and the question how such kind of measurements are performed "professionally":

  • is there special auto auto-adjustble measuring devices (that jumps through the ranges)?
  • is there a special design for current measurement that doesn't break a line when ranges are switched?

I saw:

and they are about schematic design. My question is about "industry practice" to measure in lower current ranges with market available multimenters.


3 Answers 3


Use a shunt resistor, but use a large enough shunt resistor that you can see a meaningful voltage at the currents of interest.

As in your original experiment, this shunt resistor may present too much of a burden voltage to get through some more power-hungry tasks you need to pass through to get to the phase of operation you are interested in measuring, so temporarily bridge the resistor with a clip lead until you get to the situation of interest.

In some cases it's also useful to add a capacitor to the circuit, possibly by putting it across the resistor.

These techniques can be extended to some types of time-varying behavior by using a scope instead of a simple voltmeter. Sometimes it's useful to use a fairly high resistance shunt to determine the actual sleep current, then a lower one and a scope (or MCU datalogger) to determine the percentage of active vs. sleep time - even if the sleep current is misread in such case, if you combine the duty cycle from one study with the carefully measured sleep current from another, you can form a good picture with only cheap tools.

  • \$\begingroup\$ Did you suggested a scope because it allows to see dynamics of power consumption? I am fascinated by the idea to have "adjustable" shunt resistor )) \$\endgroup\$
    – gavenkoa
    Dec 6, 2020 at 14:35

This behavior is characteristic of cheaper multimeters with a range switch. You'll usually not find those in serious electronics design labs since they're rather outdated and "cheap" (read: inaccurate and/or unsafe). There are multimeters available which can "find" the range for you (called "autoranging") and just use their big switch for changing the mode. Having this feature is almost an expectation of professional multimeters.

Multimeters which have an "autoranging" capability (most Fluke, Extech, Keysight, Tek/Keithley, etc multimeters) will most likely implement some kind of "make before break" circuit when they change range so that they don't cut power to the device being tested (SSRs/MOSFETs are used to swap out the current measurement shunts, and the "next" shunt is turned on before the "previous" shunt is turned off). These kind of multimeters at a minimum are at least $50-$200, if not much much more.

These kind of measurements are very commonly performed in industrial/professional situations (R&D, production test, etc) to verify current-based performance of the device being designed or manufactured. Multimeters used for that kind of characterization are at least $1000 usually, but can reliably measure in uA and nA.

Here are some example products which you might find in a professional situation:

Another device you'll encounter is called a "source-measure unit" (SMU). These are essentially a very precise, very fast, power supply hooked up to a very precise and accurate multimeter. These take the place of both the battery/power source and the DMM. You set a voltage on your SMU, then ask it how much current the device is drawing. These are actually what are used to characterize startup current and things like that. Here are some more examples:

These tend to be several thousand dollars at least, but are essential for many types of electronics design and verification situations.

  • \$\begingroup\$ Thanks for writing! My budget and application is definitely our of listed professional equipment. Another person suggested to use "shunt resistor" with the scope. I think it gives some pure man solution to see power consumption in dynamic. Still you provided with good information that I couldn't know as I am not working in EE fields. \$\endgroup\$
    – gavenkoa
    Dec 6, 2020 at 14:43

Put a shunt resistor in place of your ammeter, then read voltage across the resistor instead of current. This way, whatever your meter does will not affect the circuit under test. A common value is 1 ohm because you can read from the meter directly (1V measured equals 1A current)

Note this is what your meter is already doing (adding a series shunt resistor and measuring voltage across it) You just can't see it.

The trick is using a value resistor that won't adversely affect your circuit. For your 1uA to 10mA range, the 1ohm suggestion is probably pretty good

10ohms would give you 10uV=1uA and 100mV=10mA


I have dozens of these shunts in our lab...we use them to measure 10-100A circuits just as I describe above.


(Yours doesnt need to be that fancy.... A common resistor costing pennies is probably good enough for you)

If this is confusing to you, post again. Happy to explain further.

  • \$\begingroup\$ I didn't believe that "professionals" bother with resistor for the task (as I don't have degree or professional experience). "shunt resistor" is new thing to me but I understand the reason and underlying math. Thanks for explaining concepts! \$\endgroup\$
    – gavenkoa
    Dec 6, 2020 at 14:39
  • 1
    \$\begingroup\$ Well I'm a "professional" and that's how we do it ;) Solutions don't need to be complex to be effective \$\endgroup\$
    – Kyle B
    Dec 6, 2020 at 17:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.