Maker Pro
Maker Pro

method to detect current flow

P

PDRUNEN

Jan 1, 1970
0
Hi Group:

Looking for suggestions:

1: I have a light bulb load can be as low as 60 watts or as high as 600 watts.
The power source is the AC main at 120 Vrms 60Hz. I don't require knowing the
exact current, only if some current is being carried to the load.

I need to detect if circuit has current flow and these are some of the ways I
have approched.

1: low value current sense resistor into a comparator.

2: current-xformer: Very expensive and large.

3: Heating of the PCB trace carrying the current (maybe hard to determine for
all cases if the switch is outside in the cold etc.

Is there a hall-effect method or other method? IR sensor?
Must be something cheap and simple. So far only method one appears cheap and
simple.

Thanks Group!!!

Paul
 
S

Sir Charles W. Shults III

Jan 1, 1970
0
Well, the resistor as a current shunt is truly a simple way to measure it.
However, I might suggest something. Use a current sensing relay.
In essence, this is usually a reed relay that has a low resistance coil. It
is placed in series with your load and only when a certain current threshold is
exceeded will the relay become active.
Check around for current sensing relays.

Cheers!

Chip Shults
My robotics, space and CGI web page - http://home.cfl.rr.com/aichip
 
R

Robert Baer

Jan 1, 1970
0
PDRUNEN said:
Hi Group:

Looking for suggestions:

1: I have a light bulb load can be as low as 60 watts or as high as 600 watts.
The power source is the AC main at 120 Vrms 60Hz. I don't require knowing the
exact current, only if some current is being carried to the load.

I need to detect if circuit has current flow and these are some of the ways I
have approched.

1: low value current sense resistor into a comparator.

2: current-xformer: Very expensive and large.

3: Heating of the PCB trace carrying the current (maybe hard to determine for
all cases if the switch is outside in the cold etc.

Is there a hall-effect method or other method? IR sensor?
Must be something cheap and simple. So far only method one appears cheap and
simple.

Thanks Group!!!

Paul

A current transformer is *not* expensive; a filament transformer can
be used and is not too large (depending on what "large" means for this
use).
Heck, a sub-minature "open frame" audio transformer could be pressed
into service after re-winding (to allow just a few turns for the current
primary).
 
J

John Woodgate

Jan 1, 1970
0
2: current-xformer: Very expensive and large.

You can use any small step-down transformer 'backwards' as a less-than-
precise current transformer. But it is true that even a 1 V secondary
that carries 6 A implies a not so small lump. You MUST put a resistor
across the 120 V winding if you do this; if you can accept 100 mV across
the low-voltage (1 V) winding at 6 A, the voltage across the 120 V
winding is 12 V and the current is 50 mA, so you need 240 ohms.

Overall, the low-value series resistor is better. To get an isolated
output, connect a much smaller transformer 'backwards' across it. This
transformer doesn't need to carry the whole load current through its
input winding.
 
R

Robert Lacoste

Jan 1, 1970
0
2: current-xformer: Very expensive and large.

You're wrong, if you don't care on the precision a current transformer is
the solution. Just use a small wire, turn it ten or twenty times around one
of the conductors going to the lamp, add filtering and a high gain amplifier
and you do have a $1 current transformer...

Robert
 
J

John Woodgate

Jan 1, 1970
0
I read in sci.electronics.design that Robert Lacoste <see.www.alciom.com
[email protected]> wrote (in <[email protected]
lub-internet.fr>) about 'method to detect current flow', on Fri, 31 Oct
2003:
You're wrong, if you don't care on the precision a current transformer
is the solution. Just use a small wire, turn it ten or twenty times
around one of the conductors going to the lamp, add filtering and a high
gain amplifier and you do have a $1 current transformer...

No, you CAN'T make a transformer by winding one wire round another one.
Oh, you'll probably get an output, but that is because you made a
capacitor and it's sensing the *voltage* on the load wire.
 
J

John Smith

Jan 1, 1970
0
Baphomet said:
determine cheap

I'm not sure I am understanding your question correctly but if you only need
to know whether or not current is flowing, almost by definition, if the
light bulb lights, current is flowing.


So... use a phototransistor? That way you'll know when the filament burns
out.
 
F

Fred Bloggs

Jan 1, 1970
0
Looking for suggestions:

1: I have a light bulb load can be as low as 60 watts or as high as 600 watts.
The power source is the AC main at 120 Vrms 60Hz. I don't require knowing the
exact current, only if some current is being carried to the load.

I need to detect if circuit has current flow and these are some of the ways I
have approched.

1: low value current sense resistor into a comparator.

2: current-xformer: Very expensive and large.

3: Heating of the PCB trace carrying the current (maybe hard to determine for
all cases if the switch is outside in the cold etc.

Is there a hall-effect method or other method? IR sensor?
Must be something cheap and simple. So far only method one appears cheap and
simple.

Your question is too broad . You have some simple problem like lighting
a LED or sounding a buzzer when a security light comes on, and you have
managed to transform it into an open ended metaphysical study of the
essence of current flow and the various detectable and measurable
properties associated with it. Your question has no answer apart from
the specific application, and is a total waste of time.
 
J

Jim Meyer

Jan 1, 1970
0
Robert Lacoste said:
You're wrong, if you don't care on the precision a current transformer is
the solution. Just use a small wire, turn it ten or twenty times around one
of the conductors going to the lamp, add filtering and a high gain amplifier
and you do have a $1 current transformer...

Robert

This suggestion will NOT create a current transformer. Wires in
any working transformer must be parallel to each other. Your
suggestion will place the wires at right angles to each other and that
will ensure that NO transformer action takes place.

I would use some diodes in series with the circuit to create a
small voltage due to the current flow. Then I would use that voltage
as the input to an opto-isolator. The total cost would probably be
less than a dollar and the size would be tiny.

Three diodes in series would make a little more than 2 volts. That
would be enough to light the LED in an opto-isolator. Three more
diodes in parallel with the first three and in the opposite direction
will allow current to flow in both directions. A current limit
resistor between the diodes and the opto would be a good idea. Using
diodes would make sure that the voltage developed due to current flow
will be reasonably constant whether the current is a half an amp or
ten amps.

Jim
 
B

Baphomet

Jan 1, 1970
0
PDRUNEN said:
Hi Group:

Looking for suggestions:

1: I have a light bulb load can be as low as 60 watts or as high as 600 watts.
The power source is the AC main at 120 Vrms 60Hz. I don't require knowing the
exact current, only if some current is being carried to the load.

I need to detect if circuit has current flow and these are some of the ways I
have approched.

1: low value current sense resistor into a comparator.

2: current-xformer: Very expensive and large.

3: Heating of the PCB trace carrying the current (maybe hard to determine for
all cases if the switch is outside in the cold etc.

Is there a hall-effect method or other method? IR sensor?
Must be something cheap and simple. So far only method one appears cheap and
simple.

Thanks Group!!!

Paul

I'm not sure I am understanding your question correctly but if you only need
to know whether or not current is flowing, almost by definition, if the
light bulb lights, current is flowing.
 
L

Luhan Monat

Jan 1, 1970
0
PDRUNEN said:
Hi Group:

Looking for suggestions:

1: I have a light bulb load can be as low as 60 watts or as high as 600 watts.
The power source is the AC main at 120 Vrms 60Hz. I don't require knowing the
exact current, only if some current is being carried to the load.

I need to detect if circuit has current flow and these are some of the ways I
have approched.

1: low value current sense resistor into a comparator.

2: current-xformer: Very expensive and large.

3: Heating of the PCB trace carrying the current (maybe hard to determine for
all cases if the switch is outside in the cold etc.

Is there a hall-effect method or other method? IR sensor?
Must be something cheap and simple. So far only method one appears cheap and
simple.

Thanks Group!!!

Paul

Paul,

I like an optocoupler in this application - one that responds to AC
input. Select a resistor accross the input side and a load resistor on
the output side so that your smallest light bulb makes the output switch
state. Cheap, simple, small, reliable.
 
J

james

Jan 1, 1970
0
Baphomet said:
I'm not sure I am understanding your question correctly but if you only need
to know whether or not current is flowing, almost by definition, if the
light bulb lights, current is flowing.

So all you need is a photodiode, a few resistors and a comparator

James
 
D

default

Jan 1, 1970
0
Hi Group:

Looking for suggestions:

1: I have a light bulb load can be as low as 60 watts or as high as 600 watts.
The power source is the AC main at 120 Vrms 60Hz. I don't require knowing the
exact current, only if some current is being carried to the load.

I need to detect if circuit has current flow and these are some of the ways I
have approched.

1: low value current sense resistor into a comparator.

2: current-xformer: Very expensive and large.

3: Heating of the PCB trace carrying the current (maybe hard to determine for
all cases if the switch is outside in the cold etc.

Is there a hall-effect method or other method? IR sensor?
Must be something cheap and simple. So far only method one appears cheap and
simple.

Thanks Group!!!

Paul
I concocted a little gizmo that tells me if my water heater is sucking
power. I took one of those (cheap surplus) side-by-side power
transformers. (core is square and primary on one side secondary on
the other) I cut out the secondary (24 V winding) with a hack saw and
that left plenty of room to run the heater wire through it. I
verified the operation with my 40 watt soldering iron.

Run the power lead through the core - need more voltage/current? wrap
a few turns of wire around the core.

I used the 120 volt winding for my circuit on the heater but found the
24 V winding is better for higher current like lighting leds. My
little warning circuit only required some microamps to run.

It blinks a LED when it is using power and briefly sounds a horn when
the power turns either on or off.

A toroidal power transformer would also work well -

Adapted the same idea to a situation at work where there was a
rotating beacon that told people the X ray was hot, if the beacon bulb
burned out it allowed a solid state relay, connected to a mechanical
relay, sound a horn.
 
A

Allen Windhorn

Jan 1, 1970
0
Hi Group:

Looking for suggestions:

1: I have a light bulb load can be as low as 60 watts or as high as
600 watts. The power source is the AC main at 120 Vrms 60Hz. I
don't require knowing the exact current, only if some current is
being carried to the load.

I need to detect if circuit has current flow and these are some of
the ways I have approched.

1: low value current sense resistor into a comparator.

2: current-xformer: Very expensive and large.
...

Tou may be able to take apart a defunct GFI outlet and get a current
transformer from it.

Allen
 
B

Baphomet

Jan 1, 1970
0
John Smith said:
So... use a phototransistor? That way you'll know when the filament burns
out.

Why? If the filament burns out, then no current is flowing. If the filament
is lit, current is flowing. The lightbulb is the load, the OP's question has
been answered, and no additional parts are necessary. Now if the OP really
wanted to know if A.C. mains voltage was present, that's a horse of a
different color.
 
A

Allen Windhorn

Jan 1, 1970
0
Baphomet said:
...
I'm not sure I am understanding your question correctly but if you only need
to know whether or not current is flowing, almost by definition, if the
light bulb lights, current is flowing.
I think he wants to know if a light bulb is lit SOMEWHERE ELSE. I
have a gadget like that; it turns on a light upstairs when a light
downstairs is turned on, without having to run 120V wire. I used a
current transformer, which I wound myself (I have a bunch of toroid
tape cores from a previous job).

Allen
 
B

Baphomet

Jan 1, 1970
0
Allen Windhorn said:
I think he wants to know if a light bulb is lit SOMEWHERE ELSE. I
have a gadget like that; it turns on a light upstairs when a light
downstairs is turned on, without having to run 120V wire. I used a
current transformer, which I wound myself (I have a bunch of toroid
tape cores from a previous job).

Allen

Ah...the dawn breaks. Thanks for the clarification.
 
W

Wieslaw Bicz

Jan 1, 1970
0
PDRUNEN said:
Hi Group:

Looking for suggestions:

1: I have a light bulb load can be as low as 60 watts or as high as 600 watts.
The power source is the AC main at 120 Vrms 60Hz. I don't require knowing the
exact current, only if some current is being carried to the load.

I need to detect if circuit has current flow and these are some of the ways I
have approched.

1: low value current sense resistor into a comparator.

2: current-xformer: Very expensive and large.

3: Heating of the PCB trace carrying the current (maybe hard to determine for
all cases if the switch is outside in the cold etc.

Is there a hall-effect method or other method? IR sensor?
Must be something cheap and simple. So far only method one appears cheap and
simple.

Thanks Group!!!

Paul

your post has caused, that I have apparently invented a new method for current
sensing. I have checked, if somebody already proposed this idea, and it seems, that
nobody has made it.

This has caused, that now I am considerating what I should do: The method is based
on a relatively simple priciple, and it seems, that nobody has proposed it until
now (I have checked it). But it requires some development effort. The result can be
interesting. It seems, that such an element can be cheap, very small, useful for
both AC and DC currents and can be also used for current measurement.

If you have an idea, what I can do with my idea, I would be gratefull for each
suggestion. Maybe somebody knows a company, that would be interested in such idea.

Kind regards

Wieslaw Bicz

---------------========== OPTEL sp. z o.o. ===========---------------
------===== R&D: Ultrasonic Technology/Fingerprint Recognition ====------
ul. Otwarta 10a PL 50-212 Wroclaw Tel.:+48 71 3296854 Fax.:+48 71 3296852
--------==== mailto:[email protected] -=- http://www.optel.pl ====-------
 
J

Jim Thompson

Jan 1, 1970
0
[snip]
your post has caused, that I have apparently invented a new method for current
sensing. I have checked, if somebody already proposed this idea, and it seems, that
nobody has made it.

This has caused, that now I am considerating what I should do: The method is based
on a relatively simple priciple, and it seems, that nobody has proposed it until
now (I have checked it). But it requires some development effort. The result can be
interesting. It seems, that such an element can be cheap, very small, useful for
both AC and DC currents and can be also used for current measurement.

If you have an idea, what I can do with my idea, I would be gratefull for each
suggestion. Maybe somebody knows a company, that would be interested in such idea.

Kind regards

Wieslaw Bicz

---------------========== OPTEL sp. z o.o. ===========---------------
------===== R&D: Ultrasonic Technology/Fingerprint Recognition ====------
ul. Otwarta 10a PL 50-212 Wroclaw Tel.:+48 71 3296854 Fax.:+48 71 3296852
--------==== mailto:[email protected] -=- http://www.optel.pl ====-------

Do a patent search to make sure you haven't just reinvented the wheel.

If nothing shows from a search, then patent it; then approach a
company.

...Jim Thompson
 
K

Klave

Jan 1, 1970
0
John Woodgate said:
I read in sci.electronics.design that Robert Lacoste <see.www.alciom.com
[email protected]> wrote (in <[email protected]
....

No, you CAN'T make a transformer by winding one wire round another one.
Oh, you'll probably get an output, but that is because you made a
capacitor and it's sensing the *voltage* on the load wire.

Actually you CAN! A Rogowski Coil. Not quite the way the OP described
it but technically possible. Of course it's used for AC and it gives a
90 degree phase shift for sin waves (it differentiates) but you can
measure amplitude directly for sin waves. Dynamic range is excellent,
bandwidth is amazing (top end anyway) and you can buy one off the
shelf - or you can make one by... winding a coil around a wire.

Let us know your results - I can't see this one getting away without
some experimentation.

GK
 
Top