Measuring power usage of device

[mribble]

I have a device that uses under 1A of power at 3.3/5V (2 different power rails) and want to measure accurately (0.1 mW would be great, but I'm not sure what to expect).  At some times the device will be in lower power modes and other times higher power modes.  So I'm thinking just using a DMM isn't good for this kind of dynamic system.  I could use a scope with a shunt resistor and calculate it, but that would be a lot of work and spreadsheets on my end so I'm looking for something to automate this.  There are lots of devices on amazon that say they measure power, but I'm not sure they meet my needs.

So basically, I'll looking for some recommendations that would measure power for this kind of device.

Thanks in advance!

[tszaboo]

An SMU/ DC power analyzer is the device that you are looking for.

[Mr.B]

Perhaps a Sparkfun Coulomb Counter.

https://learn.sparkfun.com/tutorials/ltc4150-coulomb-counter-hookup-guide

I have used this for measuring and logging power usage of a MCU that spends most of its time in sleep mode.
I used it to estimate battery life of the MCU circuit.

[mribble]

Thanks!

Looks like that sparkfun breakout board could work, but I'd want to build something with a separate microcontroller to run it.

I looked for dc power analyzers and those tools are really expensive.  Anyone out there suggest a good reasonably priced one?

Something like this would be great, but these seem to be targeted at high power usage RC applications and doesn't have find enough resolution for my needs.  And they don't specify how often they poll the current (plus a bunch of other areas that are just under specified).  https://www.amazon.com/JZCreater-Power-Analyzer-Backlight-Precision/dp/B01LVTST80/ref=s9u_simh_gw_i1?_encoding=UTF8&fpl=fresh&pf_rd_m=ATVPDKIKX0DER&pf_rd_s=&pf_rd_r=21DTH7JZV29EA7BPBBPS&pf_rd_t=36701&pf_rd_p=1cded295-23b4-40b1-8da6-7c1c9eb81d33&pf_rd_i=desktop

[Mr.B]

Looks like that sparkfun breakout board could work, but I'd want to build something with a separate microcontroller to run it.

A $4 arduino clone and the code from the Sparkfun tutorial... Job done.

[mwalker]

There is a similar one for USB:

http://www.ebay.com.au/itm/162351822315

but it also suffers from a lack of docs, and possibly resolution. It also didn't work very well on 3.3V when I tried it.

[tszaboo]

I looked for dc power analyzers and those tools are really expensive.  Anyone out there suggest a good reasonably priced one?

You said, that you want to automate stuff. I assumed it is for work. When the employer wants something reliable for production, it is not that expensive. Anyway, for home use, I would not buy one of course.
The LTC4150 is not really suited for this, at low currents it will have the same issues as any other system.

[mribble]

Thanks everyone for the suggestions.  I admit I skimmed that sparkfun article a little too quickly.  I'll go back and read that more closely.  It sounds like it might be good enough so I'll look into it some more.

You said, that you want to automate stuff. I assumed it is for work. When the employer wants something reliable for production, it is not that expensive. Anyway, for home use, I would not buy one of course.
The LTC4150 is not really suited for this, at low currents it will have the same issues as any other system.

This is for my employer, but my employer is me    I used to work at a big company that did buy tools that cost over a $1M (one example being arrays of FPGAs that simulated future chip designs pre-tapeout).  Back then I had a very similar mindset where if a tool cost $10K and it would help an engineer at his job it was an easy decision to get it.  That said when I'm going small productions I have a much higher bar for a $10K tool.  It has made me appreciate less expensive tools that still get the job done.  The smarter and cheaper tools are a big reasons kickstarter like projects can exist today.  If embedded programming still required people to get a $10K compiler a lot of cool new projects  wouldn't exist.

To me it's sounding like the exact tool I want doesn't exist, but I'll check out that sparkfun tool more.  It wouldn't be that hard to take one of those inexpensive 2 channel USB scopes and combine it with something like Dave's microcurrent.  Then write software to integrate the total power consumption by collecting something like 100K samples per second.  Sounds like a fun project if I ever have time.

[mwalker]

I just came across this on Tindie:

https://www.tindie.com/products/mux/instrument-grade-usbbattery-power-sensor/