From: jswayze on
I'm working on an amp-hour meter (actually milliamp-hour) and am
planning on using an Allegro current sensor to measure current from a
discharging battery under load. I plan on displaying the accumulated
amp-hours on an LED display, and a PIC will be running the show.

Is the Allegro the best method for doing this, or is there another way?
I haven't gotten into it yet, but it looks like the Allegro takes some
calibration to make it accurate.

Any suggestions? Thanks.

From: antedeluvian on
>I'm working on an amp-hour meter (actually milliamp-hour) and am
>planning on using an Allegro current sensor to measure current from a
>discharging battery under load. I plan on displaying the accumulated
>amp-hours on an LED display, and a PIC will be running the show.
>
>Is the Allegro the best method for doing this, or is there another way?
>I haven't gotten into it yet, but it looks like the Allegro takes some
>calibration to make it accurate.
>
>Any suggestions? Thanks.
>
>

Since you are measuring mA (you don't mention how many) you should check
out the FAQ on the allegro part
http://www.allegromicro.com/faq/acs704faq/index.htm#Q6

Given the low resolution and accuracy issues, and assuming that you are
sub 1Amp, I would reccomend that you measure the voltage across a current
shunt. Maxim has a part that will allow an external current shunt,
National has one with an internal one. I think TI (Burr Brown) & Linear
Tech also have products. Here's a start at Maxim.
http://para.maxim-ic.com/compare.asp?Fam=CS_Amp&Tree=PowerSupplies&HP=PowerSupplies.cfm&ln=


But you don't need a dedicated part- you simply need a difference
amplifier that can handle input voltages beyond the supply rail. You can
make your own with resistive dividers at the input, or use a product like
the AD626, an Instrumentation amplifier from Analog Devices. Of course
this approach is not isolated.

I am currently using the Allegro ACS704xx-005 current sensor and I have
developed I problem where the device "locks up". I am awaiting tech
support to find out where the problem lies.

-Aubrey

This message was sent using the comp.arch.embedded web interface on
www.EmbeddedRelated.com
From: TheDoc on

<jswayze(a)gmail.com> wrote in message
news:1125081413.077909.242660(a)g47g2000cwa.googlegroups.com...
> I'm working on an amp-hour meter (actually milliamp-hour) and am
> planning on using an Allegro current sensor to measure current from a
> discharging battery under load. I plan on displaying the accumulated
> amp-hours on an LED display, and a PIC will be running the show.
>
> Is the Allegro the best method for doing this, or is there another way?
> I haven't gotten into it yet, but it looks like the Allegro takes some
> calibration to make it accurate.
>
> Any suggestions? Thanks.
>

Maxim has a few battery "fuel gauge" devices.. real low power..


From: Hershel Roberson on

On 26-Aug-2005, jswayze(a)gmail.com wrote:

> I'm working on an amp-hour meter (actually milliamp-hour) and am
> planning on using an Allegro current sensor to measure current from a
> discharging battery under load. I plan on displaying the accumulated
> amp-hours on an LED display, and a PIC will be running the show.

About a million years ago, I built a mA hour (actually a uA hour) meter
using a voltage to frequency converter. There was an app note on a Analog
Devices part that we used in a configuration for measuring current (I don't
remember the part number). The neat thing is that it will integrate any
peaks or spikes between output pulses, and has a very wide dynamic range
(especially if you trim any ofset errors). We wanted to know the long time
power consumption of a device that drew uAs in sleep-mode, and mAs when it
had something to do. If it's for lab use, just connect a frequency counter
and a few passive parts and you're done.

-Hershel