AC current measure

Hi,

For a small project I need to measure the current through an AC 
lightbulb. I am using a microcontroller with AD converter and a triac to 
control it.

Any suggestions for the current sensor?

On Wed, 14 Jun 2006 14:03:07 +0200, in comp.arch.embedded Pietje Bell
<nospam538@yahoo.com> wrote:

>Hi,
>
>For a small project I need to measure the current through an AC 
>lightbulb. I am using a microcontroller with AD converter and a triac to 
>control it.
>
>Any suggestions for the current sensor?


http://rswww.com

for example:
# RS Stock no. 189-5477
# Manufacturer Honeywell
# Manufact. part no. CSNE151

But since you are using a triac you may need to do an RMS conversion
in your software or use a true RMS ic to get a meaningfull result


martin

Pietje Bell wrote:
> 
> For a small project I need to measure the current through an AC
> lightbulb. I am using a microcontroller with AD converter and a
> triac to control it.
> 
> Any suggestions for the current sensor?

First, define current.  I am serious.  Peak, RMS, etc.  Then
consider its relation to things you can measure, such as
instantaneous voltage across a resistor, bearing in mind the actual
waveform on the device.  You might want to measure the actual light
output.

-- 
 Some informative links:
   news:news.announce.newusers
   http://www.geocities.com/nnqweb/
   http://www.catb.org/~esr/faqs/smart-questions.html
   http://www.caliburn.nl/topposting.html
   http://www.netmeister.org/news/learn2quote.html

"Pietje Bell" <nospam538@yahoo.com> wrote in message 
news:448ffb04$0$15789$14726298@news.sunsite.dk...
> Hi,
>
> For a small project I need to measure the current through an AC lightbulb. 
> I am using a microcontroller with AD converter and a triac to control it.
>
> Any suggestions for the current sensor?

How about wrapping a coil or wire around the flex to the lightbulb and using 
use the current induced in the coil to produce a voltage across a resistor? 
Then you can use an A/D to measure the voltage and do some maths to relate 
it to current in the flex. Of course this would be an AC voltage so rather 
than trying to convert to DC maybe just take lots of measurements with the 
A/D and use the peak.

This is a method I'd been mulling over using so I'd be interested to see 
what the wise people on this ng have to say about whether it has a hope of 
working. 

Pietje Bell wrote:
> Hi,
>
> For a small project I need to measure the current through an AC
> lightbulb. I am using a microcontroller with AD converter and a triac to
> control it.
>
> Any suggestions for the current sensor?

Light bulb like a 60 or 100 W one at 220 V AC, well, just
use a 1 Ohm resistor to get a mV per milliamp, do some
amplification and if the MCU can sample this every few microseconds
you are done whether you need current, power etc.

Dimiter

------------------------------------------------------
Dimiter Popoff               Transgalactic Instruments

http://www.tgi-sci.com
------------------------------------------------------

"Didi" <dp@tgi-sci.com> wrote:
>Pietje Bell wrote:
>> For a small project I need to measure the current through an AC
>> lightbulb. I am using a microcontroller with AD converter and a triac to
>> control it.
>>
>> Any suggestions for the current sensor?
>
>Light bulb like a 60 or 100 W one at 220 V AC, well, just
>use a 1 Ohm resistor to get a mV per milliamp, do some
>amplification and if the MCU can sample this every few microseconds
>you are done whether you need current, power etc.

That will work, but the sampling circuitry will not be isolated from
the mains line.

Tom Lucas wrote:
> How about wrapping a coil or wire around the flex to the
>lightbulb and using use the current induced in the coil to
>produce a voltage across a resistor? 
>...
>This is a method I'd been mulling over using so I'd be
>interested to see what the wise people on this ng have 
>to say about whether it has a hope of working. 

"Wise people" say it will work, only they call it a "current
transformer"  ;-)  It also provides galvanic isolation, unlike the
previous (also working) approach. 
They are commonly implemented as transformers on a toroidal core.
The only thing to keep is mind is selecting a core/wire-loop-count
that will not saturate under the expected measurement range.

Roberto Waltman <usenet@rwaltman.net> wrote:

>>Pietje Bell wrote:
>>> ...I  need to measure the current through an AC
>>> lightbulb.
>Tom Lucas wrote:
>> How about wrapping a coil or wire ...

>...they call it a "current transformer" 

Like these:

http://www.crmagnetics.com/newprod/ProductView.asp?ProdName=CR8401
http://www.flex-core.com/2rl-7rl-photo.htm

More info:

http://en.wikipedia.org/wiki/Transformer#Current_transformers
http://leeh.ee.tut.fi/transformer/tra3.htm

"Roberto Waltman" <usenet@rwaltman.net> wrote in message 
news:bi4092dd172j8naifuurcmmk9ltljgnchn@4ax.com...
> Roberto Waltman <usenet@rwaltman.net> wrote:
>
>>>Pietje Bell wrote:
>>>> ...I  need to measure the current through an AC
>>>> lightbulb.
>>Tom Lucas wrote:
>>> How about wrapping a coil or wire ...
>
>>...they call it a "current transformer"
>
> Like these:
>
> http://www.crmagnetics.com/newprod/ProductView.asp?ProdName=CR8401
> http://www.flex-core.com/2rl-7rl-photo.htm
>
> More info:
>
> http://en.wikipedia.org/wiki/Transformer#Current_transformers
> http://leeh.ee.tut.fi/transformer/tra3.htm


Horny. Thanks for that, it's gratifying to think that I've learned 
_something_ analogue after all these years! Any ideas how much they cost of 
the top of your head?

There you go Pietje, it seems that it is possible but the series way is 
probably simpler if it is acceptable. I'd do this for the fun of it though. 

"Tom Lucas" wrote:
( about current transformers )
> ...Any ideas how much they cost of the top of your head?

No.

>There you go Pietje, it seems that it is possible but the series way is 
>probably simpler if it is acceptable. I'd do this for the fun of it though. 

To the OP: For a one of a kind project, you could improvise using any
old transformer core, specially if you are willing to spend the time
calibrating it for your particular application.
(Doesn't mater if the core is not toroidal, doesn't mater if the
transfer function is not linear, doesn't mother if the field leaks,
etc.)

Roberto Waltman <usenet@rwaltman.net> writes:
> "Didi" <dp@tgi-sci.com> wrote:
> >Pietje Bell wrote:
> >> For a small project I need to measure the current through an AC
> >> lightbulb. I am using a microcontroller with AD converter and a triac to
> >> control it.
> >>
> >> Any suggestions for the current sensor?
> >
> >Light bulb like a 60 or 100 W one at 220 V AC, well, just
> >use a 1 Ohm resistor to get a mV per milliamp, do some
> >amplification and if the MCU can sample this every few microseconds
> >you are done whether you need current, power etc.
> 
> That will work, but the sampling circuitry will not be isolated from
> the mains line.
> 
> Tom Lucas wrote:
> > How about wrapping a coil or wire around the flex to the
> >lightbulb and using use the current induced in the coil to
> >produce a voltage across a resistor? 
> >...
> >This is a method I'd been mulling over using so I'd be
> >interested to see what the wise people on this ng have 
> >to say about whether it has a hope of working. 
> 
> "Wise people" say it will work, only they call it a "current
> transformer"  ;-)  It also provides galvanic isolation, unlike the
> previous (also working) approach. 
> They are commonly implemented as transformers on a toroidal core.
> The only thing to keep is mind is selecting a core/wire-loop-count
> that will not saturate under the expected measurement range.

And frequency.