|
Home > Archive > Electrical Engineering > May 2007 > Voltage drop in a wire
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Voltage drop in a wire
|
|
|
| Recently a question came up for my group's project.
We're using a temp sensor chip that has micro-amps of max output current.
We're sending this down a long piece of wire (maybe 20awg).
I suggested a unity gain buffer because the long length of wire will create
noise but more importantly: the chip can't drive a signal through the long
wire.
Unfortunatly, I only know a buffer should be used, hoewver, I don't know
how to mathematically calculate (in theory) how long the wire can be before
the chip can't drive it.
If the wire is 2ohm for 100ft and the chip has 100uA max output current; is
the voltage across the wire 200uV (2ohm x 100uA)?
I found a chart for voltage drop per 100ft of wire, but it only listed
specific current values. Is there a better chart or an explaination on how
to calculate everything?
thanks in advance!
| |
|
|
Peter wrote:
> Recently a question came up for my group's project.
> We're using a temp sensor chip that has micro-amps of max output current.
> We're sending this down a long piece of wire (maybe 20awg).
> I suggested a unity gain buffer because the long length of wire will create
> noise but more importantly: the chip can't drive a signal through the long
> wire.
> Unfortunatly, I only know a buffer should be used, hoewver, I don't know
> how to mathematically calculate (in theory) how long the wire can be before
> the chip can't drive it.
> If the wire is 2ohm for 100ft and the chip has 100uA max output current; is
> the voltage across the wire 200uV (2ohm x 100uA)?
Information is a bit sparse here! Does the chip have a CURRENT output
or a VOLTAGE output? If it's a current source the 2 ohms of the wire
(that would be for 100 feet of 23 gauge) you get about .2 millivolts
max of drop in the wire. You are missing the fact that the wire drives
a LOAD on the far end. The resistance of that load is the critical
thing. Say it was just a DVM on the far end of the wire set for
microamps. The wire wouldn't matter so long as the chip could drive
the wire plus the meter. And that depends mostly on the METER and not
the wire. A buffer could be used but current buffers or amplifiers
aren't common and hard to deal with.
The REAL questions here are what is used to measure the output of the
chip at the far end of the wire and how fast the thing must respond
etc. (has to do with noise pickup). I think your group's project
could use some real engineering advice.
| |
| TimPerry 2007-05-23, 9:25 am |
| Peter wrote:
> Recently a question came up for my group's project.
>
>
> We're using a temp sensor chip that has micro-amps of max output
> current. We're sending this down a long piece of wire (maybe 20awg).
I suggest using shielded twisted pair such as microphone cable.
you may want to add a ferrite bead and small capacitor to help choke RF.
>
> I suggested a unity gain buffer because the long length of wire will
> create noise but more importantly: the chip can't drive a signal
> through the long wire.
the sensor must want a high Z load. if you need to drive an meter movement
the buffer is appropriate, however the gain may need to be more then 1 to
compensate for transmission line loss (your 100 ft wire).
>
>
> Unfortunatly, I only know a buffer should be used, hoewver, I don't
> know how to mathematically calculate (in theory) how long the wire
> can be before the chip can't drive it.
>
> If the wire is 2ohm for 100ft and the chip has 100uA max output
> current; is the voltage across the wire 200uV (2ohm x 100uA)?
yes, it's generally referred to as the IR drop.
>
> I found a chart for voltage drop per 100ft of wire, but it only listed
> specific current values. Is there a better chart or an explaination
> on how to calculate everything?
>
>
> thanks in advance!
type of wire makes a difference. i.e. aluminum has higher resistance then
copper.
use a chart from the manufacturer of the exact make of transmission line
(wire) that you have, if possible. if not find a way to accurately measure
it.
you will probably be much happier if you can use your buffer convert your
current source into a voltage source that operates with an output of 2 volts
or greater. just add a series resistor to the meter for calibration.
| |
|
| "TimPerry" <timperry@noaspama.net> wrote in
news:46543218$0$1428$4c368faf@roadrunner.com:
> Peter wrote:
>
> I suggest using shielded twisted pair such as microphone cable.
>
> you may want to add a ferrite bead and small capacitor to help choke
> RF.
>
>
> the sensor must want a high Z load. if you need to drive an meter
> movement the buffer is appropriate, however the gain may need to be
> more then 1 to compensate for transmission line loss (your 100 ft
> wire).
>
>
>
>
> yes, it's generally referred to as the IR drop.
>
>
> type of wire makes a difference. i.e. aluminum has higher resistance
> then copper.
> use a chart from the manufacturer of the exact make of transmission
> line (wire) that you have, if possible. if not find a way to
> accurately measure it.
>
> you will probably be much happier if you can use your buffer convert
> your current source into a voltage source that operates with an output
> of 2 volts or greater. just add a series resistor to the meter for
> calibration.
>
>
>
>
The chip is an LM70 and it's driving a microprocessor. I didn't re-
review the datasheet, however, current or voltage output isn't as
important. If it's outputting current, then I can use a transimpedance
circuit and if it's a voltage output, I can use a unity gain buffer.
How did you know my made up numbers would be for 23 gauge wire? usually
the wire lists go by 2s.
| |
| TimPerry 2007-05-24, 9:25 am |
| Peter wrote:
> "TimPerry" <timperry@noaspama.net> wrote in
> news:46543218$0$1428$4c368faf@roadrunner.com:
>
>
> The chip is an LM70 and it's driving a microprocessor. I didn't re-
> review the datasheet, however, current or voltage output isn't as
> important. If it's outputting current, then I can use a transimpedance
> circuit and if it's a voltage output, I can use a unity gain buffer.
try a little RF modulator /demod and use RG59 coax or even go wireless. that
should get you an A+
>
> How did you know my made up numbers would be for 23 gauge wire?
> usually the wire lists go by 2s.
my stage lighting rig use 13 ga. it was what was available in a
multi-conductor snake.
http://www.powerstream.com/Wire_Size.htm
|
|
|
|
|