Skip to content

Power Consumption of Outputs

With Collin Kanofsky and Claude as my guides, I redid my testing of power consumption on my servo from week 10 and made myself a guide on power.

P = I x V

NOTE: I intentially cut off claudes answers just to report what is pertinnent.. not all the extra stuff that isn't relevant.

PROMPT: to get power do you need to measure a voltage drop or just the voltage of the circuit

CLAUDE: Just the voltage of the circuit — you're not measuring a drop, you're measuring what's available.

Put your multimeter in DC voltage mode (V⎓), red probe to positive, black to ground, and you'll see the supply voltage. That tells you power is present and what level it's at.

Voltage drop is a different measurement — that's when you measure across a specific component (like a resistor or LED) while current is flowing, to see how much voltage that component is consuming. Useful for troubleshooting but not for just confirming power.

So for your typical use case:

  • Is power present? → measure voltage at the supply rails (red to VCC, black to GND)
  • Is my 5V actually 5V? → same thing, just confirming the level
  • How much is this resistor dropping? → probe both sides of the resistor while circuit is running

When we were doing our group work, we kept measuring voltage drops over the output, one of ours had a big drop and the other was neglible. My servo had negligable power drop, so I was very confused so I asked:

PROMPT: is there a voltage drop if the right amount of power is being supplied by the circuit

4:11 PM Yes — voltage drop always happens whenever current flows through a component, regardless of whether the circuit is working correctly. It's not a sign of a problem, it's just physics (Ohm's Law).

For example in a typical LED circuit:

  • Supply: 5V
  • Resistor drops: ~1.8V
  • LED drops: ~2.0V
  • Remaining: ~0.2V (close enough, small drop across wiring)

Those all add up to roughly 5V

So a healthy working circuit has voltage drops everywhere — they should just all add up to your supply voltage. That's Kirchhoff's Voltage Law.

Voltage drop becomes a diagnostic tool when:

  • A component is dropping more than expected (possible short or wrong value)
  • A component is dropping nothing when it should - be (possible open/broken connection or burnt out component)
  • A wire or trace is dropping voltage it shouldn't be (bad connection, cracked trace, cold solder joint)

NOTE: What claude didnt talk about yet - SAG - when you component needs more than whats available.

The practical takeaway:

  1. Measure supply voltage → confirms power is available
  2. Measure across a component → tells you what that component is actually doing
  3. If your LED isn't lighting up, probe across the resistor first — if it reads 0V, current isn't flowing and the problem is upstream of it

*** To clarify on the how and where to measure, I asked a bit more.***

PROMPT: to measure power consumption , do i measure the drop or just the voltage of the circtuir

4:11 PM CLAUDE: Neither actually — to measure power consumption you need current (amps), which requires a different measurement altogether.

To measure current:

Set your multimeter to DC amps (A or mA) You have to break the circuit and put the meter in series — the current has to flow through the meter Red probe goes to the positive side of the break, black to the negative side

Then calculate power:

  • P = V × I (watts = volts × amps)

So if your circuit runs at 5V and draws 200mA, that's 5 × 0.2 = 1 watt

Important warning for your AM33D:

The amp input is a separate port on the meter — you have to move the red probe from the voltage port to the amp port.

Easy to forget and blow the fuse if you then try to measure voltage with it still in amp mode Your meter has a 10A max and a 2000µA/200mA range — for small ESP32 circuits use the mA range

Practical tip for your ESP32 projects: Rather than breaking the circuit, a lot of makers put a small resistor (like 1Ω) in series on the ground line, then measure the voltage drop across it — since V=IR, the drop directly tells you the current without rewiring anything.

Still wanting to verify my reading was correct... I asked the below which gave me what I needed to hear.. that the servo has very little resistance. The resistance is what causes the voltage drops.

PROMPT: i barely had any change in my 5 v when i ran a voltmeter across my servo, is this correct?

4:13 PM CLADUE: Yes, that's correct and expected! When you measure voltage across a component in parallel (which is what you do with a voltmeter), you'll see close to the full supply voltage — not the drop.

Here's why that confused you:

Voltmeter across a servo = sees almost full voltage

The servo is connected between 5V and GND

Your meter in parallel sees essentially the same 5V That's normal and just confirms power is reaching the servo

To actually see a voltage drop you need a component in series — like a resistor. The drop appears across the resistor because current has to pass through it before reaching the next component.

Servos don't have a significant series resistance in their supply path by design — they want the full voltage delivered cleanly.

To properly characterize your servo you'd want to:

Measure current (break the circuit, meter in series) while the servo is moving vs idle — you'll see a big difference. A servo at rest might draw 5-10mA but under load can spike to 500mA+ That current × 5V gives you actual power consumption.

The small voltage change you did see was likely just the servo loading the supply very slightly — if you saw 4.98V instead of 5.00V for example, that tiny sag is the supply working to deliver current. A bigger load would sag it more noticeably.

POWER CONSUMPTION OF the MG90S servo

So the power consumption of my servo is Ampere readings were 30mA - 170mA max.

.170 Amp * 5 V = 0.85 Watts

This is moving but no real load on it. The data sheet for the MG90S servo.

A note about SAG

But in group work, why did we have a huge voltage drop over the little fan that i had such a hard time getting to run off my esp32c3?

PROMPT: should there be a voltage drop after my servo

Claude: No — you don't want or expect a voltage drop after the servo. The servo is a load, not a component in a voltage divider. Power goes in, the motor uses it, and that's the end of the line. What you do want to check is voltage at the servo VCC pin while it's moving. If you see:

5V steady → power supply is adequate Voltage dipping when the servo moves → power supply can't handle the current draw, causing brownouts

If you're seeing weird behavior (resets, ADC going crazy, LED flickering) when the servo moves, that voltage sag is probably why —

Ah ha... SAG... I am starting to get somewhere

PROMPT: should voltage drop after an out put device in my circuit

CLAUDE: No — ideally voltage should be the same before and after any output device (LED, servo, motor). The device consumes current, not voltage. Voltage should stay constant throughout your power rail. If you do see a voltage drop across a wire or trace, that means:

  • The wire/trace resistance is too high for the current being drawn
  • Your power supply can't keep up with demand

following on this thread, I contintued to ask about voltage drop we saw.

PROMPT: whst if my power supply is too small for my output device - would it results in a drop of voltage

CLAUDE: Yes exactly — that's called the power supply being unable to source enough current, and it shows up as voltage sag.

What happens internally is the power supply has its own internal resistance, and when you demand more current than it's designed for, that internal resistance causes a voltage drop before the power even reaches your device.

You'd see:

  • Voltage looks fine with no load
  • Voltage sags when device activates (servo moves, motor starts)
  • The bigger the current demand, the bigger the sag
  • At worst the voltage drops so low the microcontroller resets or behaves erratically

This also makes sense why it would run off the 3.3 V directly, but not off the pin.

I went back to see the McWhorter Tutorial with Fan and how we wired all that up.. We had used the motor driver to control it.. not directly to a DPIO pin. I thought we only needed that for the stepper motor, but I was incorrect.

I did some more searching around the fan motor (data sheet) It requires .1mA to run without a load. An Arduino or ESP32 digital pin can safely output is more likely a maximum of 20mA continuous current. While the absolute maximum rating is 40mA, exceeding 20mA can permanently damage the microcontroller. So i am lucky I didn’t fry everything up.

The Learning

Bottom line is that you measure the voltage of the circuit (gnd to vcc). Then multiply by the current you measure in series the circuit with the device under full load. This gives you the power consumption under full load.