Measuring battery consumption over time

[Prime]

Hi,

Hoping someone can point me in the right direction:

We're building a battery based wireless device and need a way to measure current consumption over time as well as instantaneously so that we can estimate battery life. The batteries are primary cells and fitting a fuel gauge to the boards is going to make them too expensive and requires me to go make a second iteration on the prototype - something I don't have time for right now.

Since it's in uAmps-hours and milliAmps-hours that I need, can someone recommend an instrument to log current and voltage over time? (I realise I can measure the current through a shunt but since that involves extra fiddling, I'd prefer a reliable instrument that produces repeatable results).

I looked at a uCurrent Gold but they're largely out of stock

Any advice would be appreciated.

Thanks in advance

[mariush]

Get two multimeters with logging option

Get a very low value resistor, like 0.01 ohm .... put it in series with the battery ...
Measure the voltage across the resistor AND the voltage after the resistor and you get the current and voltage... log them.   V = IxR ... you have the R, you have the measured V, you have the instant current value.

Alternatively, you can get one of those programmable power supplies which can also report voltage and current to PC or some logging device (through serial, usb , network).
You will have a pretty much constant voltage, unlike a battery that will lower voltage as it discharges, but you can calculate the power consumption over a long period of time and determine the average power used over that time.
You can also fake a battery by lowering the DC output every 30 minutes or every hour by some amount, which would account for  the efficiency of your dc-dc converters (if your circuit has any)

[bdunham7]

Hi,

Hoping someone can point me in the right direction:

We're building a battery based wireless device and need a way to measure current consumption over time as well as instantaneously so that we can estimate battery life. The batteries are primary cells and fitting a fuel gauge to the boards is going to make them too expensive and requires me to go make a second iteration on the prototype - something I don't have time for right now.

Since it's in uAmps-hours and milliAmps-hours that I need, can someone recommend an instrument to log current and voltage over time? (I realise I can measure the current through a shunt but since that involves extra fiddling, I'd prefer a reliable instrument that produces repeatable results).

You can easily find GPIB/RS232/USB based DMMs that will log for you, but first you need to know what the characteristics of your draw are so that you can get a proper setup.  The primary issues I see are a potentially large range of current draw and possibly very short current pulses.  So what are the approximate maximum and minimum currents and how short in duration can current pulses be?  And how precise do you need the results to be--are you just trying to size the battery or determine approximate battery life?

[gslick]

Maybe something like an HP / Agilent 66311B/D, 66309B/D "Mobile Communications DC Source" would be suitable for this task?

In the Low Current range -20 mA to +20 mA they have a Measurement Accuracy of 0.1% + 2.5 μA and a Measurement Resolution of 0.6 μA.

Those are > $3000 new, but can sometimes be found $100 - $200 on eBay, although the shipping cost to South Africa might be prohibitive.

Keysight Model 66111A Fast Transient DC Source
Keysight Model 66311B/D, 66309B/D Mobile Communications DC Source
User’s Guide
https://literature.cdn.keysight.com/litweb/pdf/5964-8125.pdf

Keysight Model 66319B/D, 66321B/D Mobile Communications DC Source
User’s Guide
https://literature.cdn.keysight.com/litweb/pdf/5964-8184.pdf

[trobbins]

A primary battery capacity is going to have quite a variation, sample to sample.

You could run your device off a somewhat similar secondary battery, and then discharge that test battery to a known end voltage after a period of operating time, and repeat say 10 times, and in between each test do a standard CC or CP discharge as a capacity benchmark, keeping battery temp constant all the time, and using a CC or CP discharge rate that is similar to your circuitry application.

[jeremy]

Unfortunately I think the microcurrent may not do what you need as it does not do range switching. The devices I test can go from a few hundred nA to 0.5A in less than a second and then back down again.

I do this a fair bit, and I use a N6781A. It’s really good, but also really expensive. I’ve also used a second hand keithley battery simulator (model is 2306), you can sometimes find them on eBay for pretty cheap. The trick is to write some software which logs the data from the system and calculates the energy usage in joules. Then unless you have an excellent battery model, just take the capacity of the cell in joules, multiply it by some safety factor depending on how badly you need it to be within your estimates, and use that in your calcs.