Does anyone know if there is a bench or handheld meter that can log current consumption down to uA or even nA level and over period of days?

Thanks

Steve

 

Do you want to sample (and log) current at regular intervals or do you want to know how much energy was consumed over a period (day/s) due to current consumption?

Maxim have a Coulomb counter that I have used before. The only problem is that it can't follow sharp transients.

http://www.maxim-ic.com/appnotes.cfm...te_number/2096

Another from Linear

http://www.linear.com.cn/prod/datash...datasheet=1020

 

I built one. It depends a lot on how much current capacity you need and how much accuracy you need, so it's hard to build one without knowing the app.

 

Can't handle sharp transients? Stick a big capacitor across it it then!

 

He didn't say "can't handle", but "can't follow" - adding a big capacitor would make it far worse at doing that!.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

I want to log the total power consumed, so if I placed a load of 1mA for 1 hour then increased it to 500mA for 1 hours result shows 501mAh

There will be long period of 1mA followed by several sharp rises to 500ma and greater, I just want to stick a meter in-line and get a total reaing with a log ideally. even better if I can connect a PC.

Not looking for a home made project but a of the shelf product.

Thx

 

In our case we were asked to design something as the company could not find anything (off the shelf) on the market to do that. The problem with their application was the existence of a low level random floor current (you can say almost a noise current) of about 1 - 10uAp-p with sharp current pulses every now and then of up to 30mA - 20uS wide. They were trying to determine how long this apparatus would work on its internal long life battery 3v lithium without running it for a year to find out. It was for a remote metering application with no mains supply or the possibility to use solar power.

At the end we designed a device that was very accurate with less than 0.5% error calculated over a week. We could not use the Maxim or Linear chips.

 

Hi "TheOne"
It sound like thats what we are looking for but with a pulse of up to 2amp and only as low as 1uA, is your device a finished product I can buy?

Thanks

SCarr

 

Would like to help you out, but all rights were sold to this company in the UK (I am in the US). As far as I know they implemented this module as part of a test sets they produce for their product.

Maybe Oznog can help you out on this one.

 

Take a look at:
http://www.allegromicro.com/hall/currentsensor.asp

Kent

 

I've made one of these, it's still about somewhere. On the most sensitive range it measures 0.1 microcoulomb per step. It still has its original batteries after seven years, and there's no power switch. The only trouble is that it doesn't have a pretty custom made box. That's what stops useful things being made. No pretty boxes at affordable prices.
Incidentally, a big capacitor across its terminals will integrate spikes that the device would otherwise clip.

 

On our design the customer didn't want any range selection, which left us with the problem how to deal with the massive dynamic range, short peak transients and still maintain less than 1% error.

 

I wouldn't mind the job of designing the guts of one to a reasonable specification. I certainly took no great pains to build the one I have, and it proved quite up to the job it was made for, estimating the battery life of a weather station datalogger. However, commercial requirements, like a pretty box and the circuitry built from pure Chinese cheapening, preclude the manufacture of it.