Re: Low Current Measurement (without ammeter)


Chris wrote:
ewilliams2@xxxxxxxxx wrote:
thank you so much John, I will try that and post my success message
soon....

Eric Williams
(left the name off last time...thanks again)

John O'Flaherty wrote:
ewilliams2@xxxxxxxxx wrote:
Hello Everyone,

I have an application that uses 3 displacement sensors with analog
current output from 0 to 20mA. I have a real time DAQ system that only
measures voltage from 0 to 10V. I have read the previous response to
Pradheep's question and it seems like the current values are too high
to get the kind of precision that I need. I was wondering if I could
use the resistor in series method and if there was some trick that
could help me develop this setup? Any help at all would be appreciated,
because I've never had to measure currents like this before so I'm a
little beside myself trying to figure this one out...Also if the
resistor "shunt" is placed in series what kind of accuracy can be
achieved?

If you put the current into an opamp inverting input, with a 500 ohm
feedback resistor, and the non-inverting input grounded, it will
convert 0-20mA to 0 to -10V, and the output will be stiff even if it's
loaded a little. You'd have to invert the output to +10 V. If the opamp
needs to be protected from voltages, you can use a series resistor with
diodes to plus and minus supply on the opamp input to limit voltage
excursion, but you'd have to make sure the current source has enough
compliance to handle the extra resistance. The source will only need
enough voltage compliance to handle the drop across that protective
resistance.You could also just use a 500 ohm resistor, if the DAQ
system has high input impedance. Using just a 500 ohm resistor, the
current source would have to have the compliance to produce the 10V.
--
john

Look before you leap, John -- there are a few potential gotchas here.

First, not all op amps are capable of sourcing/sinking 20mA (and that
has to be a min., not typ. on the data ***).

That's a good point.You could put a transistor buffer on the opamp
output, or, in a one-direction application like this, you could hang a
resistor to the minus supply, pulling 10mA out. Then the opamp could
switch between sinking and sourcing 10 mA.

Second, you really have to look carefully at your system before you
just plug it in. Many sensors that output a 4-20mA (or 0-20mA) current
operate on another supply voltage, and have a simplified internal
circuit like thios (view in fixed font or M$ Notepad):

|
| +24V
| .------------o-----.
| | | |+
| | '-----|------o
| | |4-20mA
| | |
| | .-----|------o
| | |\| | |-
| | -|+\ |/ |
| | | >--| |
| |.---|-/ |> |
| || |/| | |
| |.-----------o |
| | | |
| | .-. |
| | | | |
| | | | |
| | '-' |
| | | |
| | | |
| | | |
| '------------o-----'
| COM
|
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)

If your sensor common is connected to DAC GND, you may have an issue
connecting the external power supply (which may exceed the DAC power
supply or the input common mode voltage) to the DAC input. And if you
connect the negative sides of more than one sensor together, it won't
work.
Another difficulty is power-up sequencing. If your sensor supply comes
up before the DAC's, you may get latchup when the DAC turns on. Not
pretty.

Do you think using a series resistor with diodes to prevent excursion
more than V+ plus .7 or V- or ground minus 0.7 would prevent that
problem? With a hookup like that, wouldn't an early start of the sensor
supply at worst just try to power up your circuit?

For current output sensors, you typically place a 250 ohm (0-5V) or 500
ohm (0-10V) load resistor at the receiving end of your signal, and then
use a diff amp to get the input voltage to the DAC. That's how it's
done.

That seems like the best solution if you have 24V available from the
sensors.
--
john

.