Actions
Forum Thread Details
  • State Verified Answer
  • Replies 7 replies
  • Answers 3 answers
  • Subscribers 341 subscribers
  • Views 4381 views
  • Users 0 members are here
Related

How to measure PCB current consumption

mike24

Hello guys,

 

I am trying to measure how much current does my pcb consumes, it is a USB powered device which logs data from a temperature sensor, and I wanna make it a portable device by adding a battery to it and take it anywhere I want while logging data from the temperature changes.

The thing is I need to know how much current does the device consumes in order to decide what capacity to choose for the battery, I do not want to just buy a 3000 mah battery and hope for the device to last long, I want to know how much would it take to discharge a certain capacity battery in order to make a final decision. I also would like to add an OLED display and some buttons/switches/LED's

 

My pcb does not have test points, and I lost the pcb/schematic files nor did I made any simulation or measurements for power/voltage information.

 

The only thing that I have is the schematic but only as a screenshot image.

 

Hope that I could get help or ideas from you guys.

  • 0

    Without knowing all the details of your setup, it's hard to give a perfect answer. I assume the device works without a USB host connected (the device hasn't enumerated, and only with 5V power over USB)? Do you have another power input where a battery can connector, or will you need to redesign it to include that?

     

    What voltage of battery do you need? In other words, what voltages are required on the PCB? Presumably these are derived from the 5V USB supply.

     

    You can start with a good estimate just by measuring the current on the 5V supply (make up a USB host connector with flying leads to connect to your power supply). A good power supply (such as this one E36313A Triple Output DC P.S. ) will give you accurate current measurements. If you don't have that, you'll need a multimeter inline with the 5V power line. If your device has different power modes (e.g., a run and a sleep mode), you usually can't measure this on a multimeter because you might need to switch between low and high current modes which can break the connection. The power supply I linked in does this automatically, so you can easily measure run currents, and very small sleep currents and estimate a good power profile of your PCB during all its operating states.

     

    When you switch to batteries, remember they don't have a flat discharge curve (the voltage drops as they discharge). If your PCB has switching regulators (good for efficiency) you'll find the battery seems to discharge faster as it gets to the end because it needs more current as the voltage drops to maintain the same power to your PCB. So it might be best to think of battery capacity in mW.hr rather than mA.hr.

  • 0 in reply to dr-jhc

    As of now it only works with a USB host connected (5V, 500mA according to specifications for USB type a). I have two boards attached together, the main board is at the bottom and the adapter board (this one I have to redesign to add OLED display, battery connector, buttons, etc) is at the top, it kinda looks like this

    image

    When the USB is connected I need to downgrade the voltage to 3.3V and 1.8V for certain devices (most devices use 3.3V), therefore I have thought of a 3.7V rechargeable battery (Fully charged it outputs 4.2V which would be downgraded to 3.3V and 1.8V, and when it is almost discharged the battery outputs 2.6V and the battery charging/manager IC works fine with this voltage range)

     

    For recharging the battery I am gonna use BQ24075, it is a power manager and recharging device which outputs regulated 5V to supply my system

    The LDO's output 800 mA to the devices but I do not know how much of it is being used, the device that would consume more current would be the microcontroller.

  • 0 in reply to mike24

    A 3.7V Li-ion/Li-Po battery is a good option. You probably don't want to let it go as low as 2.6V as this will likely start to damage it. The protection chips prevent the biggest failures are such deep discharge, but for best longevity (recharge cycles) try to keep it above 3.0V (check the battery datasheet for recommendations, they're not all the same).

     

    Will your 3.3V devices run at 3.0V? If so you might be able to extend operating time a bit as it the battery level drops (using linear regulars, rather than a boost/buck converter).

     

    To measure the power required in all parts of your circuit, it's a good idea to have small shunt resistors in line with the power. You can them just probe across these with a multimeter to measure the voltage, and so calculate the current.You can always replace these with shorts (0 ohm links) later. Adding 2-pin 0.1in headers inline with the power can also be useful. You can them put a jumper across it to short it, or use it to connect a multimeter in-circuit to measure the current. You can also use it to supply power from an external supply to test various parts: for example supply 3.3V from a bench supply to measure only that current, and also vary the voltage and see what your circuit is doing as the voltage drops. It makes experimenting a bit faster so you can work out what's going on and how your circuit is responding to changes in voltage.

  • 0 in reply to dr-jhc

    Hello again,

     

    I have successfully measured how much current my device consumes thorough your method and it is about 40mA (if I add a 1500mAh battery it would last up to 37.5 hours) then I have to add the other devices which would be an OLED display, buttons, 2 LED's, and 2 sensors which I know don't consume that much current, the only device that I would say consumes a lot of current would be the LED's.

     

    What is a good program to simulate and measure my design?

  • 0 in reply to mike24

    Good work!

     

    An OLED can consume a reasonable amount of current, depending on the size of the display. The smaller OLEDs often have built-in charge pumps to generate the higher voltages it may need (e.g., 5-7V from a 3.3V supply), but larger ones typically need you to provide that. So you might been a higher voltage supply if you use a large OLED... good to check what's required.

     

    Another thing to watch with OLEDs that have multiple power supply pins (i.e., digital power and display power) is you need to make sure you sequence these correctly: don't allow the display power to be active before the digital power or it may fry the display.

     

    How much simulation do you want to do? If you want to just analyse the power/battery trade-off, it can often be done very well using a spreadsheet, building the simulation from measurements you have done. If you want to simulate everything including the code, that's a bigger deal. For circuit simulation, I use Proteus, and this also does MCU simulation too, although I've never use it for that.

  • 0 in reply to dr-jhc

    The OLED that I am gonna use is a 0.96in I2C OLED display and the datasheets show that the maximum current it can consume while the whole screen is being used is about 15 to 20 mA and it works with 3.3V, it only has 4 pins (Vcc, GND, SCL, SDA)

     

    For simulation I only want to analyse the overall current/power.

  • 0 in reply to mike24

    I have used those OLED displays! They are very easy to use and have low current: 15-20mA sounds about right from memory.

     

    I would just create a spreadsheet that lists all your current drains and add equations if you need any. For example if you're using any switching regulators you can put in numbers for efficiency and work out your current consumption as the battery level drops. If everything is linear, the current will stay the same you'll just burn energy in the regulators.