Sense Power Line Cord, Turn-on Fan

0 views
Skip to first unread message

Jim Thompson

unread,
Aug 17, 2005, 1:29:35 PM8/17/05
to
I vaguely recall a topic where current consumption of a device was
sensed in the power line cord and a fan was turned on.

Now that I have a need for such a device I can't locate the thread (it
was probably in a.b.s.e, which isn't archived).

I need to sense the line current of a TV set and turn-on a fan.
Probably need adjustable sense current since set has your typical
standby mode.

Ideas?

If I can buy it off-shelf that's even better.

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice:(480)460-2350 | |
| E-mail Address at Website Fax:(480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.

John Popelish

unread,
Aug 17, 2005, 1:58:00 PM8/17/05
to
Jim Thompson wrote:
> I vaguely recall a topic where current consumption of a device was
> sensed in the power line cord and a fan was turned on.
>
> Now that I have a need for such a device I can't locate the thread (it
> was probably in a.b.s.e, which isn't archived).
>
> I need to sense the line current of a TV set and turn-on a fan.
> Probably need adjustable sense current since set has your typical
> standby mode.
>
> Ideas?
>
> If I can buy it off-shelf that's even better.
>

There is a product that is the electrical current equivalent of a
mechanical limit switch, usually called a current switch. You pass a
current carrying conductor through a hole in it, and either a powered
or unpowered contact closes at a preset or adjustable current.
E.g.

http://www.enercorp.com/curr/225nr.htm
http://www.electrodepot.com/itm00038.htm
http://www.crmagnetics.com/newprod/ProductView.asp?ProdName=CR9321

John Woodgate

unread,
Aug 17, 2005, 1:55:59 PM8/17/05
to
I read in sci.electronics.design that Jim Thompson
<thegr...@example.com> wrote (in
<0os6g1d8hv5l1qn39...@4ax.com>) about 'Sense Power Line
Cord, Turn-on Fan', on Wed, 17 Aug 2005:

>I need to sense the line current of a TV set and turn-on a fan.
>Probably need adjustable sense current since set has your typical
>standby mode.
>
>Ideas?

Pass *one* conductor of the mains lead through a high-permeability (low
frequency) ferrite toroid and wind say 50 turns of thin wire on the
toroid. That will give you enough signal (roughly sinusoidal pulse
waveform at mains frequency) to do with what you will, and I can't think
you need to be told how to process it to operate a solid-state relay.


>
>If I can buy it off-shelf that's even better.

I don't know of one here, but you are elsewhere.
--
Regards, John Woodgate, OOO - Own Opinions Only.
There are two sides to every question, except
'What is a Moebius strip?'
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

Frank Bemelman

unread,
Aug 17, 2005, 2:39:07 PM8/17/05
to
"John Woodgate" <j...@jmwa.demon.contraspam.yuk> schreef in bericht
news:phDAvmTv...@jmwa.demon.co.uk...

> I read in sci.electronics.design that Jim Thompson
> <thegr...@example.com> wrote (in
> <0os6g1d8hv5l1qn39...@4ax.com>) about 'Sense Power Line
> Cord, Turn-on Fan', on Wed, 17 Aug 2005:
> >I need to sense the line current of a TV set and turn-on a fan.
> >Probably need adjustable sense current since set has your typical
> >standby mode.
> >
> >Ideas?
>
> Pass *one* conductor of the mains lead through a high-permeability (low
> frequency) ferrite toroid and wind say 50 turns of thin wire on the
> toroid. That will give you enough signal (roughly sinusoidal pulse
> waveform at mains frequency) to do with what you will, and I can't think
> you need to be told how to process it to operate a solid-state relay.
> >
> >If I can buy it off-shelf that's even better.
>
> I don't know of one here, but you are elsewhere.

Googling for 'power slave switch' turned up this:
http://store.qkits.com/moreinfo.cfm/K8034

--
Thanks, Frank.
(remove 'q' and '.invalid' when replying by email)


Joerg

unread,
Aug 17, 2005, 3:43:09 PM8/17/05
to
Hello John,

> Pass *one* conductor of the mains lead through a high-permeability (low
> frequency) ferrite toroid and wind say 50 turns of thin wire on the
> toroid. That will give you enough signal (roughly sinusoidal pulse
> waveform at mains frequency) to do with what you will, and I can't think
> you need to be told how to process it to operate a solid-state relay.

That's the way to do it. #77 material comes to mind which is quite
ubiquitous (Amidon etc.).

>> If I can buy it off-shelf that's even better.
>
> I don't know of one here, but you are elsewhere.

You can buy them, look under current transformers:
http://www.alfamag.com/productindex.htm

But by the time you have finished the order you'd be done rolling your
own. If you have the toroid handy, that is. No idea if Radio Shack
carries hi-perm toroids. It used to be all #43 material there but even
that should work in a simple application like this.

Or slaughter a switcher and pick it out of there.

Regards, Joerg

http://www.analogconsultants.com

Jim Thompson

unread,
Aug 17, 2005, 4:02:09 PM8/17/05
to

Actually, I need to look in the garage... I probably have one in my
junk box.

Joerg

unread,
Aug 17, 2005, 4:37:30 PM8/17/05
to
Hello Jim,

> Actually, I need to look in the garage... I probably have one in my
> junk box.

While you are at it look at any transformers flying around. Small ones.
If you find one where there is enough space to loop a decent 12 gauge
around it once or where you could pry off the secondary to do that you'd
be in business.

Another source are those "oh s..t" toroids that get clamped onto
computer gear cords after they blew the first EMC test. If you can get
it off. Sometimes they are snap-on.

Regards, Joerg

http://www.analogconsultants.com

DaveM

unread,
Aug 17, 2005, 8:51:31 PM8/17/05
to
"Jim Thompson" <thegr...@example.com> wrote in message
news:0os6g1d8hv5l1qn39...@4ax.com...


Look at the "Smart Strip" at http://www.bitsltd.net/SmartStrip. It has
pretty much all you need to run a complete entertainment center from a
single device. Probably overkill for a single fan, but perhaps you could
use it anyway. Save some construction time.
They have a number of outlets, both controlled and always hot. Surge
protection and EMI/RFI filter. Adjustable sensitivity too.
I have a couple of these units and from personal experience, I can recommend
them. They have worked flawlessly for over 3 years. Sure beats pushing all
those darn switches.

--
Dave M
MasonDG44 at comcast dot net (Just subsitute the appropriate characters in
the address)

Never take a laxative and a sleeping pill at the same time!!


Richard Crowley

unread,
Aug 17, 2005, 10:06:23 PM8/17/05
to
"Jim Thompson" wrote ...

>I vaguely recall a topic where current consumption of a device was
> sensed in the power line cord and a fan was turned on.
>
> Now that I have a need for such a device I can't locate the thread (it
> was probably in a.b.s.e, which isn't archived).
>
> I need to sense the line current of a TV set and turn-on a fan.
> Probably need adjustable sense current since set has your typical
> standby mode.

The "Mini Power Minder" is shown on www.froogle.com as
available from several different vendors for around US$ 15-20

ehsjr

unread,
Aug 17, 2005, 10:22:13 PM8/17/05
to
Jim Thompson wrote:
> I vaguely recall a topic where current consumption of a device was
> sensed in the power line cord and a fan was turned on.
>
> Now that I have a need for such a device I can't locate the thread (it
> was probably in a.b.s.e, which isn't archived).
>
> I need to sense the line current of a TV set and turn-on a fan.
> Probably need adjustable sense current since set has your typical
> standby mode.
>
> Ideas?
>
> If I can buy it off-shelf that's even better.
>
> ...Jim Thompson

You can roll your own cheaply with either cat# VT-27 or VT-5
from http://www.allelectronics.com/

Ed

Steve

unread,
Aug 17, 2005, 10:32:13 PM8/17/05
to
Jim Thompson wrote:

> I vaguely recall a topic where current consumption of a device was
> sensed in the power line cord and a fan was turned on.

Sears sells an auto-switch designed for power tools. I'm not sure how well
it will work with your TV.

<http://www.sears.com/sr/javasr/product.do?BV_UseBVCookie=Yes&vertical=TOOL&pid=00924031000&tab=des#tabs>

--
--Steve

Watson A.Name - "Watt Sun, the Dark Remover"

unread,
Aug 17, 2005, 11:17:02 PM8/17/05
to

"Jim Thompson" <thegr...@example.com> wrote in message
news:0os6g1d8hv5l1qn39...@4ax.com...
> I vaguely recall a topic where current consumption of a device was
> sensed in the power line cord and a fan was turned on.
>
> Now that I have a need for such a device I can't locate the thread (it
> was probably in a.b.s.e, which isn't archived).
>
> I need to sense the line current of a TV set and turn-on a fan.
> Probably need adjustable sense current since set has your typical
> standby mode.
>
> Ideas?
>
> If I can buy it off-shelf that's even better.
>
> ...Jim Thompson
> --

Radio Shaft used to have a power strip that turned the monitor and
printer, for example, on when the PC was switched on. It sensed the
current on the master outlet and switched the others on. It wasn't
cheap, tho. There are still other companies that sell a similar outlet
strip, probably much cheaper than RS. Check around, on the 'net and at
Home Depot, etc.

Watson A.Name - "Watt Sun, the Dark Remover"

unread,
Aug 17, 2005, 11:23:43 PM8/17/05
to

"Joerg" <notthis...@removethispacbell.net> wrote in message
news:hrMMe.285$GV7...@newssvr25.news.prodigy.net...

[snip]

> Or slaughter a switcher and pick it out of there.

But aren't switch supplies using ferrite mat'l that's going to do okay
at 80kHz or more, but would saturate easily at 60Hz? I guess it's no
big deal if it saturates, since it just passes the current thru easier.
As long as there is enough out of the sense winding to turn on the relay
driver.


> Regards, Joerg
>
> http://www.analogconsultants.com


Ptaylor

unread,
Aug 17, 2005, 11:22:33 PM8/17/05
to
John Woodgate wrote:
> I read in sci.electronics.design that Jim Thompson
> <thegr...@example.com> wrote (in
> <0os6g1d8hv5l1qn39...@4ax.com>) about 'Sense Power Line
> Cord, Turn-on Fan', on Wed, 17 Aug 2005:
>
>> I need to sense the line current of a TV set and turn-on a fan.
>> Probably need adjustable sense current since set has your typical
>> standby mode.
>>
>> Ideas?
>
>
> Pass *one* conductor of the mains lead through a high-permeability (low
> frequency) ferrite toroid and wind say 50 turns of thin wire on the
> toroid. That will give you enough signal (roughly sinusoidal pulse
> waveform at mains frequency) to do with what you will, and I can't think
> you need to be told how to process it to operate a solid-state relay.
>
>>
>> If I can buy it off-shelf that's even better.
>
>
> I don't know of one here, but you are elsewhere.


Radio Shack (atleast *used to* carry them),and probably other local
places (Circuit City,etc)sells power/surge strips for home entertainment
systems with a "sense" outlet or two,and the rest were switched on/off
depending on the load sensed on the "sense" outlet

Turn your TV on,and it senses it,and turns on the stereo/vcr/dvd player
for you.

Watson A.Name - "Watt Sun, the Dark Remover"

unread,
Aug 17, 2005, 11:31:17 PM8/17/05
to

"ehsjr" <eh...@bellatlantic.net> wrote in message
news:phSMe.1020$zb.20@trndny03...

One thing that I don't believe anyone has yet suggested is a heat
sensor. The fan comes on when it gets hot. and then turns off when the
sensor gets cool. Trick is to get the sensor near the point where it
can reliably sense heat. Another thought is to use a simple light
sensor on a power LED of the nmaster equipment.


Michael A. Terrell

unread,
Aug 17, 2005, 11:39:26 PM8/17/05
to
Jim Thompson wrote:
>
> I vaguely recall a topic where current consumption of a device was
> sensed in the power line cord and a fan was turned on.
>
> Now that I have a need for such a device I can't locate the thread (it
> was probably in a.b.s.e, which isn't archived).
>
> I need to sense the line current of a TV set and turn-on a fan.
> Probably need adjustable sense current since set has your typical
> standby mode.
>
> Ideas?
>
> If I can buy it off-shelf that's even better.
>
> ...Jim Thompson


Some of the kits were a low voltage relay in series with the load,
and had back to back clamping zener diodes to limit the current flow.
SOme used a full wave bridge in line with the load to reduce the voltage
drop, and used an opto-isolator to turn on a relay. Heathkit sold one,
and several hobby magazines had construction projects. You might turn
up the Heathkit schematic online if anyone has the model number.

--
Link to my "Computers for disabled Veterans" project website deleted
after threats were telephoned to my church.

Michael A. Terrell
Central Florida

Walter Harley

unread,
Aug 18, 2005, 1:30:13 AM8/18/05
to
"Jim Thompson" <thegr...@example.com> wrote in message
news:0os6g1d8hv5l1qn39...@4ax.com...
> [...] I need to sense the line current of a TV set and turn-on a fan.


Out of idle interest (since you've gotten lots of real answers already):
what for? Is the fan to cool the TV (why does it need cooling?) or for
something else?


default

unread,
Aug 17, 2005, 11:43:44 PM8/17/05
to
On Wed, 17 Aug 2005 10:29:35 -0700, Jim Thompson
<thegr...@example.com> wrote:

>I vaguely recall a topic where current consumption of a device was
>sensed in the power line cord and a fan was turned on.
>
>Now that I have a need for such a device I can't locate the thread (it
>was probably in a.b.s.e, which isn't archived).
>
>I need to sense the line current of a TV set and turn-on a fan.
>Probably need adjustable sense current since set has your typical
>standby mode.
>
>Ideas?
>
>If I can buy it off-shelf that's even better.
>
> ...Jim Thompson

As already pointed out, ferrites designed for the KHZ range make lousy
50-60 cycle current sensors.

Cheap and easy to add a few turns to a transformer and use one of the
existing windings to sense current. The "flat pack" style cores on
small 60 HZ power transformers are easier to use in that respect
(square iron structure with separate bobbins for primary and
secondary). A hacksaw can easily remove one set of windings without
damaging the other.

I've used them to do things like turn on a horn when a warning light
fails, or turn on the fan when some chemist in the lab powers up the
an ultrasonic cleaner in an un vented enclosure - at work. So simple,
I never drew the schematics. Basically a current sensing transformer
made from a square core power transformer. Usually the 12 VAC winding
works best to drive a solid state relay through a rectifier bridge and
small filter cap.

These things where all the rage in the 50-60's. and were used to turn
off the vacuum tube amplifier when the last record in a stack had
played - a circuit or two appeared in Poptronics at the time. (and a
lot more complicated with no solid state relays)

I built a little gizmo for my water heater. Sounds a buzzer for a few
seconds when the heating element turns on or off, and blinks an LED
while the element is on.

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Jim Thompson

unread,
Aug 18, 2005, 10:34:13 AM8/18/05
to

The TV is tucked under a kitchen cabinet, the contents of which are
getting hot... coffee beans, vitamins, daily medications, etc. ;-)

Richard Crowley

unread,
Aug 18, 2005, 10:41:39 AM8/18/05
to
"Jim Thompson" wrote ...

> The TV is tucked under a kitchen cabinet, the contents of which are
> getting hot... coffee beans, vitamins, daily medications, etc. ;-)

Might be a good excuse to replace it with a modern LCD
TV receiver! :-)

Jim Thompson

unread,
Aug 18, 2005, 11:15:19 AM8/18/05
to

But it already is... 21" Sony WEGA... produces a phenomenal amount of
heat. But, admittedly, it's in a confined space, kitchen counter
corner, cabinet bottoms with just 1.5" clearance from top of set.

Jim Thompson

unread,
Aug 18, 2005, 11:24:42 AM8/18/05
to
On Wed, 17 Aug 2005 20:51:31 -0400, "DaveM"
<mason...@comcast99.net> wrote:

>"Jim Thompson" <thegr...@example.com> wrote in message
>news:0os6g1d8hv5l1qn39...@4ax.com...
>>I vaguely recall a topic where current consumption of a device was
>> sensed in the power line cord and a fan was turned on.
>>
>> Now that I have a need for such a device I can't locate the thread (it
>> was probably in a.b.s.e, which isn't archived).
>>
>> I need to sense the line current of a TV set and turn-on a fan.
>> Probably need adjustable sense current since set has your typical
>> standby mode.
>>
>> Ideas?
>>
>> If I can buy it off-shelf that's even better.
>>
>> ...Jim Thompson
>

>Look at the "Smart Strip" at http://www.bitsltd.net/SmartStrip. It has
>pretty much all you need to run a complete entertainment center from a
>single device. Probably overkill for a single fan, but perhaps you could
>use it anyway. Save some construction time.
>They have a number of outlets, both controlled and always hot. Surge
>protection and EMI/RFI filter. Adjustable sensitivity too.
>I have a couple of these units and from personal experience, I can recommend
>them. They have worked flawlessly for over 3 years. Sure beats pushing all
>those darn switches.

I can't find the specification for adjustability. Is the LCG3 what
you have, or a different model? Thanks!

Walter Harley

unread,
Aug 18, 2005, 12:19:21 PM8/18/05
to
"Jim Thompson" <thegr...@example.com> wrote in message
news:tv69g1hq51ajrsk4f...@4ax.com...
> [...]

> The TV is tucked under a kitchen cabinet, the contents of which are
> getting hot...


Ah, I see.

Maybe you should approach the problem the other way. You could use a PID
controller to turn on or off the TV, and thus precisely regulate the
temperature of the cabinet ;-)


Jim Thompson

unread,
Aug 18, 2005, 12:21:34 PM8/18/05
to

Now *there's* the trick! Sno-o-o-o-o-ort ;-)

Joerg

unread,
Aug 18, 2005, 1:45:50 PM8/18/05
to
Hello Watson,

> But aren't switch supplies using ferrite mat'l that's going to do okay
> at 80kHz or more, but would saturate easily at 60Hz? I guess it's no
> big deal if it saturates, since it just passes the current thru easier.
> As long as there is enough out of the sense winding to turn on the relay
> driver.

You just have to calculate it so that the core is staying clear of
saturation at the maximum current. I usually maintain a 30% distance
from that point. So the core has to be big enough that one turn at
whatever current Jim's stuff is drawing at the peaks of the 60Hz cycle
is in the kosher range. Plus some margin.

For this app you don't need a large core. But you can, depending on
what's in the garage boxes.

Regards, Joerg

http://www.analogconsultants.com

Ian Stirling

unread,
Aug 18, 2005, 2:58:08 PM8/18/05
to

Compare the power consumption of a modern LCD and CRT monitor of the
same sizes.

They are often very comparable, due to the small fluorescant tubes having
lousy efficiency even before the 50% loss in the polariser, and 66% loss
in the colour filters.

Glenn Gundlach

unread,
Aug 18, 2005, 4:06:35 PM8/18/05
to

OK, here's a trick from one of us 'low tech' guys. Use a filament
transformer in reverse. 0.5 ohm resistor in parallel with secondary, in
series with TV. 'Primary' runs into a bridge rectifier and RC to drive
a solid state relay. Whole thing fits into a standard electrical box. I
used mine for 10 years to turn on the audio amp with the TV.
GG

Joerg

unread,
Aug 18, 2005, 4:24:11 PM8/18/05
to
Hello Jim,

In case all you need to do is turn on the fan when it gets toasty you
might be able to do that with an attic fan thermostat. Every large
hardware store has these. At HD usually way in the back behind the
lumber, where the roofing accessories are. Last time I looked they were
around $10 and are adjustable. I was thinking about using one to turn on
the pool pump when frost sets in but they can't be adjusted this low.

Regards, Joerg

http://www.analogconsultants.com

Joerg

unread,
Aug 18, 2005, 4:54:43 PM8/18/05
to
> As already pointed out, ferrites designed for the KHZ range make lousy
> 50-60 cycle current sensors.

I had no problems with ferrites here. They can even be used to check
whether or not a DC current is there by running them into saturation.
The DC trip level can be set by the number of turns.

Regards, Joerg

http://www.analogconsultants.com

default

unread,
Aug 18, 2005, 9:53:27 AM8/18/05
to

Have you been getting enough power to run an LED? (or a solid state
relay)?

It is easy enough to get signal to drive a high impedance, but when
one tries to drive lower impedances, the energy isn't there.

I didn't spend much time with ferrites since I was interesting in
directly driving LED's or SSR's with the output. With my iron cored
transformers I plotted out the power for given loadings.

I rigged a selector switch with 47 ohm resistors (easier than a pot)
and measured the secondary voltage and current with 5 turns to 25
turns on the primary.

With 5 turns (and my 40 watt soldering iron on the input) I got 1.3
milliamps, .6 volts into 470 ohms. Secondary was the 120 volt
winding.

With 25 turns (and 40 watts in) 1128 ohms loading 5.26 volts @ 4.56
milliamps. Enough to run some LEDs and do useful work directly, no
amplifiers, no other power source. Elegant.

The peak power points for 5 and 25 turns was 470 and 1128 ohms
respectively (with the 40 watt load)

Peak current was 3 milliamps 5T into zero ohms and 11 milliamps with
25 turns. Peak voltage was 1.04 volts 5T open circuit, and 11.15 for
25 turns.

I plotted this out for 5-10-15-20-25 turns with the 40 watt load and
that gave me a quick reference to use for current sensing relays (at
least with that particular transformer)

With the ferrites I tried the power wasn't there to drive an led (at
least in a reasonably small size current transformer). The output
waveform wasn't sinusoidal either (probably due to saturation).

Joerg

unread,
Aug 18, 2005, 8:13:08 PM8/18/05
to
Hello,

Could you give us a first name? It's just nicer to communicate then.

>>I had no problems with ferrites here. They can even be used to check
>>whether or not a DC current is there by running them into saturation.
>>The DC trip level can be set by the number of turns.
>

> Have you been getting enough power to run an LED? (or a solid state
> relay)?

That wasn't the task at hand here. Jim wanted an easy (and I suppose
safe) method to measure the current in an AC circuit. You do not need
any power to do that and ferrites can perform nicely. Plus they are very
cheap. All you need in this case is a voltage across a load resistor on
the secondary that is proportional to the current flowing though that
one turn of primary. I guess in Jim's case this voltage would be
rectified and fed into a Schmitt trigger that controls a fan.

> With the ferrites I tried the power wasn't there to drive an led (at
> least in a reasonably small size current transformer). The output
> waveform wasn't sinusoidal either (probably due to saturation).

If you want to transfer power at 60Hz the classic lamination core of the
usual power transformer is the ticket. Ferrites excel when you are above
several kHz. But for current sensing it doesn't really matter.

Regards, Joerg

http://www.analogconsultants.com

Graham W

unread,
Aug 18, 2005, 2:54:11 PM8/18/05
to
Jim Thompson wrote:
> On Thu, 18 Aug 2005 07:41:39 -0700, "Richard Crowley"
> <rcro...@xpr7t.net> wrote:
>
>> "Jim Thompson" wrote ...
>>> The TV is tucked under a kitchen cabinet, the contents of which are
>>> getting hot... coffee beans, vitamins, daily medications, etc. ;-)
>>
>> Might be a good excuse to replace it with a modern LCD
>> TV receiver! :-)
>
> But it already is... 21" Sony WEGA... produces a phenomenal amount of
> heat. But, admittedly, it's in a confined space, kitchen counter
> corner, cabinet bottoms with just 1.5" clearance from top of set.

IMHO, you'd be better off putting some foil-lined polystyrene 'tiles'
on the under side of the affected cabinets to try to isolate them from
the TV heat. I think putting a fan in there will only increase the chance
of blowing warm air into the cracks around the door.

Swap around the cabinet contents to place stuff which could benefit
from a warmer/dryer environment in the warmer places.


--
Graham W http://www.gcw.org.uk/ PGM-FI page updated, Graphics Tutorial
WIMBORNE http://www.wessex-astro-society.freeserve.co.uk/ Wessex
Dorset UK Astro Society's Web pages, Info, Meeting Dates, Sites & Maps
Change 'news' to 'sewn' in my Reply address to avoid my spam filter.

DaveM

unread,
Aug 18, 2005, 9:11:42 PM8/18/05
to
"Jim Thompson" <thegr...@example.com> wrote in message
news:sv99g15d7n09gr45p...@4ax.com...
> | James E.Thompson, P.E.\


Yes, I have the LCG3 models. According to the "documentation" that came
with the strips, at max sensitivity, they will switch on with a 40W
incandescent lamp load. They state that the strips will work reliably with
home entertainment eqpt, TVs, computers, medium & large shop eqpt, fans,
heaters, pumps, and "some kitchen appliances", etc.
No specs for the current sensitivity are given, but I can say that they live
up to the claims. I use one on my computer system to turn on the monitor
and sound system connected to the computer. The other one is connected to
the home entertainment system. When I turn on the TV, the sound system
turns on in unison. No need to collect all the remotes to switch everything
on.
You might find a better deal on Ebay... maybe not.

I remember the unit that Radio Shack sold a number of years ago.. very
similar in operation, but none of the sockets were always hot. Sensitivity
was not adjustable either.
Heathkit sold one too... it was the model GD-1295. You can get a copy of
the schematic at
http://web.archive.org/web/20010618185923/www.tom80.freeserve.co.uk/diagrams/gd-1295.gif.
I was thinking about building a copy of that unit, but just never seemed to
get the time. From what I can remember from the schematic, the current
sensing device is R1, and is likely to be something like a Thermometrics
current limiting thermistor. I seem to remember doing the calculations and
came up with a Thermometrics model CL-80 for the sensor. I think I was
using something like 0.5A as the threshold for the device to switch on. The
relay is a 115 VDC coil, and the 417-834 transistor is equivalent to an
MPSU10 (that kinda dates this design). The 56-44 is a 1N4653 4.7V zener.
D1 and D2 are 1N4004 units. The varistor R4 is an everyday MOV for a 117V
line. That's about as much as I can remember about it. I have all the
parts but not the time to build it. Maybe that will be still on my agenda
when I retire <sigh>.

Oh, BTW, if you're interested in building one of these strips using a
current transformer, I have several units that I had left over from a
project. Be glad to send you a couple if you want them. They're 200 turns,
rated at about 10A AC. About 1.25" dia, and 0.625" high with a 3/16" hole,
just about big enough for a single turn of 14 ga wire. Would work out great
for one of these strips.

Cheers!!!

default

unread,
Aug 18, 2005, 1:28:05 PM8/18/05
to
On Fri, 19 Aug 2005 00:13:08 GMT, Joerg
<notthis...@removethispacbell.net> wrote:

>Hello,
>
>Could you give us a first name? It's just nicer to communicate then.

Sure Bob


>
>>>I had no problems with ferrites here. They can even be used to check
>>>whether or not a DC current is there by running them into saturation.
>>>The DC trip level can be set by the number of turns.
>>
>> Have you been getting enough power to run an LED? (or a solid state
>> relay)?
>
>That wasn't the task at hand here. Jim wanted an easy (and I suppose
>safe) method to measure the current in an AC circuit. You do not need
>any power to do that and ferrites can perform nicely. Plus they are very
>cheap. All you need in this case is a voltage across a load resistor on
>the secondary that is proportional to the current flowing though that
>one turn of primary. I guess in Jim's case this voltage would be
>rectified and fed into a Schmitt trigger that controls a fan.
>

Schmidt trigger would require a separate power supply. I was using a
$5 part for my current transformer but factor in the cost of a supply
for the signal amplification of a ferrite and there's no advantage to
using it. Can't get much simpler than a small transformer driving a
SSR via a diode and cap.

Safety . . . I didn't fool around with any open circuit transients on
the high turns ratio iron lamination transformer. Presumably a high
voltage spike could be generated if it were unloaded and power
applied. One MOV would fix that. I did use a pair of zeners to clamp
my "X-ray warning beacon back up horn" since I couldn't control what
wattage lamp someone might install in it.

I didn't take any safety precautions on the output side of my
transformer test rig and didn't get zapped, but that might be luck.
The load switches only had one open circuit position, and I only took
five readings OC (or into the 10 Mohm meter).

>> With the ferrites I tried the power wasn't there to drive an led (at
>> least in a reasonably small size current transformer). The output
>> waveform wasn't sinusoidal either (probably due to saturation).
>
>If you want to transfer power at 60Hz the classic lamination core of the
>usual power transformer is the ticket. Ferrites excel when you are above
>several kHz. But for current sensing it doesn't really matter.

Well. if I had the circuitry and supply voltages there for some other
purpose a ferrite is smaller and might be a better choice.

Is it possible to get an accurate RMS voltage from a ferrite current
trans at 60 HZ? What load impedance?

Regards Bob

James D. Veale

unread,
Aug 19, 2005, 11:00:03 AM8/19/05
to
I had a Radio Shack strip, and I did adjust it.

Joerg

unread,
Aug 19, 2005, 12:55:17 PM8/19/05
to
Hello Bob,

Thank you for giving us your first name.

> Schmidt trigger would require a separate power supply. I was using a
> $5 part for my current transformer but factor in the cost of a supply
> for the signal amplification of a ferrite and there's no advantage to
> using it. Can't get much simpler than a small transformer driving a
> SSR via a diode and cap.

A Schmitt needs power but very little in the case of HC or CD4000 logic.
If you want to directly drive the power device or a relay the classic
transformer is certainly better here.

> Safety . . . I didn't fool around with any open circuit transients on
> the high turns ratio iron lamination transformer. Presumably a high
> voltage spike could be generated if it were unloaded and power
> applied. One MOV would fix that. I did use a pair of zeners to clamp
> my "X-ray warning beacon back up horn" since I couldn't control what
> wattage lamp someone might install in it.

Just keep in mind that MOVs are like bank accounts. Every zap takes away
some of the assets (Joules). Then one day .... kablouie.

> Is it possible to get an accurate RMS voltage from a ferrite current
> trans at 60 HZ? What load impedance?

Magnetics, Inc. offers SW to calculate all that but I have not used it
yet. I am old-fashioned in this respect, using the graphs and the old
HP11C calculator. When I feel nostalgic I crack out the slide rule
calculator. Anyway, here is the link:
http://www.mag-inc.com/software/transformer.asp

Getting an accurate RMS voltage is possible. However, while ferrites are
cheap and easily available (or even already present in the junk box)
they aren't ideal. Strip-wound cores (tape cores) are better but more
expensive. Here is a link with more detail:
http://www.metering.com/vac/

Impedance with ferrite cores at low frequencies is low, you can't get
too many turns on them. But in a current transformer you don't have to.

Regards, Joerg

http://www.analogconsultants.com

Charlie Edmondson

unread,
Aug 19, 2005, 12:59:19 PM8/19/05
to
Jim Thompson wrote:
> On Wed, 17 Aug 2005 22:30:13 -0700, "Walter Harley"
> <wal...@cafewalterNOSPAM.com> wrote:
>
>
>>"Jim Thompson" <thegr...@example.com> wrote in message
>>news:0os6g1d8hv5l1qn39...@4ax.com...
>>
>>>[...] I need to sense the line current of a TV set and turn-on a fan.
>>
>>
>>Out of idle interest (since you've gotten lots of real answers already):
>>what for? Is the fan to cool the TV (why does it need cooling?) or for
>>something else?
>>
>
>
> The TV is tucked under a kitchen cabinet, the contents of which are
> getting hot... coffee beans, vitamins, daily medications, etc. ;-)
>
> ...Jim Thompson
Hey Jim,
Time to rearrange the cabinets! Put your plates and coffee cups in this
one. Always nice to have a pre-warmed plate and cup... 8-)

Charlie

Jim Thompson

unread,
Aug 19, 2005, 1:05:06 PM8/19/05
to

But it's the closest to the RO water, necessary to take the pills ;-)

(I've ordered the LCG3 Smart Power Strip that "DaveM" recommended.
That will resolve the issue.)

JeffM

unread,
Aug 19, 2005, 5:52:40 PM8/19/05
to
>>>>I vaguely recall a topic where current consumption of a device
>>>>was sensed in the power line cord and a fan was turned on.
>>>>Now that I have a need for such a device I can't locate the thread
>>>> Jim Thompson
>>
>>You can buy them, look under current transformers:
>> Joerg
>
>Actually, I need to look in the garage...
>I probably have one in my junk box.
> Jim Thompson

I think I found that thread.
http://groups-beta.google.com/group/sci.electronics.design/browse_frm/thread/b0067293cc192a1b/bfbd91b3b6c37e9e?q=current-transformer+When-I-turn-the-saw-on-I-want-the-dust-colector-to-automaticaly-come-on
Look at who the author of the solution is. 8-)

default

unread,
Aug 19, 2005, 8:50:27 AM8/19/05
to
On Fri, 19 Aug 2005 16:55:17 GMT, Joerg
<notthis...@removethispacbell.net> wrote:

snip


>
>Just keep in mind that MOVs are like bank accounts. Every zap takes away
>some of the assets (Joules). Then one day .... kablouie.
>

Are you sure about this? MOV's do fail, and sometimes quite
spectacularly, but if the energy they are asked to dissipate is within
their ratings (and repetitive hits, heat build up or a steady state
over voltage condition isn't encountered) they survive to do it again.

MOV's can be degraded by repetitive strikes, no argument there. But
the joule rating, isn't the number of strikes. e.g. 150 joule rating
doesn't mean it fails if asked to take 300 hits of one joule each time
(assuming heat build up isn't a factor).

The typical failure mode, as I understand it, is when the device
reaches a thermal runaway condition and shorts, shunting the power
supply. Well within its energy rating and thermal limits, it is
expected to last for a very long time.

Some parts will always fail the first time . . . infant mortality.

>> Is it possible to get an accurate RMS voltage from a ferrite current
>> trans at 60 HZ? What load impedance?
>
>Magnetics, Inc. offers SW to calculate all that but I have not used it
>yet. I am old-fashioned in this respect, using the graphs and the old
>HP11C calculator. When I feel nostalgic I crack out the slide rule
>calculator. Anyway, here is the link:
>http://www.mag-inc.com/software/transformer.asp

I always trust the empirical data . . . "Never argue with empirical
data."


>
>Getting an accurate RMS voltage is possible. However, while ferrites are
>cheap and easily available (or even already present in the junk box)
>they aren't ideal. Strip-wound cores (tape cores) are better but more
>expensive. Here is a link with more detail:
>http://www.metering.com/vac/

Particularly interesting because we are supposed to get electronic
meters in the next year or so. The mechanical ones suck -
particularly with things like proportioning heaters - the meter seems
to spin up in less than a 1/8 revolution and coasts back in a 1/4 to
1/2 revolution, and their accuracy is limited in the best of
conditions. At least with an electronic one they can use a look up
table to correct for environmental conditions and loading.

I do wonder about the loss of privacy - the ones we are getting can
report when power is consumed, how much, and the power factor of the
load - I think you could tell a lot by that (like when someone's home
and what they might be doing).


>
>Impedance with ferrite cores at low frequencies is low, you can't get
>too many turns on them. But in a current transformer you don't have to.
>

I guess I'd take issue with the statement: "in a current transformer
you don't have to" (have low impedance - or high permeability) That
all depends on what you plan to do - what load impedance you want to
drive.

Regards
bob

Joerg

unread,
Aug 19, 2005, 8:41:37 PM8/19/05
to
Hello Bob,

>>Just keep in mind that MOVs are like bank accounts. Every zap takes away
>>some of the assets (Joules). Then one day .... kablouie.
>
> Are you sure about this? MOV's do fail, and sometimes quite
> spectacularly, but if the energy they are asked to dissipate is within
> their ratings (and repetitive hits, heat build up or a steady state
> over voltage condition isn't encountered) they survive to do it again.

They do age with every strike above a certain threshold. Here is a brief
explanation:
http://www.mikeholt.com/mojonewsarchive/LSP-HTML/HTML/TVSS-Protection-Questions-and-Answers~20040708.php

"Other surge suppressors rely on "sacrificial components," Metal Oxide
Varistors (MOVs) or gas discharge devices that shunt or divert power
surges from the hot line to the neutral or ground line. Sensitive
digital equipment systems depend on a clean and uncontaminated ground.

SurgeXŽ power filters offer an all-new, revolutionary Series Mode
technology, which captures surge energy and slowly releases it onto the
neutral line. Surge energy will not contaminate ground lines or wreak
havoc with your system peripherals. With every surge above a modest
level that hits your system, the MOVs in ordinary surge protectors
degrade until they eventually fail. Unfortunately, it is impossible to
tell when a MOV has become ineffective, leaving you exposed to potential
equipment destruction."

Anyway, they are indeed sacrificial components. The detailed mechanism
should be explained in mfgs app notes. I have seen it but don't recall
where, since I rarely use MOVs.

> I guess I'd take issue with the statement: "in a current transformer
> you don't have to" (have low impedance - or high permeability) That
> all depends on what you plan to do - what load impedance you want to
> drive.

Well, usually a current transformer is used for current sensing and not
to convert energy. There are certainly exceptions but then you'd rather
need a "real" transformer.

Regards, Joerg

http://www.analogconsultants.com

Fred Bloggs

unread,
Aug 19, 2005, 9:01:56 PM8/19/05
to

DaveM wrote:

> Heathkit sold one too... it was the model GD-1295. You can get a copy of
> the schematic at
> http://web.archive.org/web/20010618185923/www.tom80.freeserve.co.uk/diagrams/gd-1295.gif.

[snip]

> From what I can remember from the schematic, the current
> sensing device is R1, and is likely to be something like a Thermometrics
> current limiting thermistor. I seem to remember doing the calculations and
> came up with a Thermometrics model CL-80 for the sensor.

Looks like it should have been a PTC type to get the circuit past
turn-on surges on the control load, with sensitivity settling to that of R2.

> I think I was
> using something like 0.5A as the threshold for the device to switch on. The
> relay is a 115 VDC coil,

What kind of 115VDC coil is that that stays on throughout a half cycle
without current? My guess was that it was an AC coil, large inductive
reactance and a shading turn on the coil to time smooth the flux intensity.

ehsjr

unread,
Aug 20, 2005, 12:43:05 AM8/20/05
to
default wrote:
> On Fri, 19 Aug 2005 16:55:17 GMT, Joerg
> <notthis...@removethispacbell.net> wrote:
>
> snip
>
>>Just keep in mind that MOVs are like bank accounts. Every zap takes away
>>some of the assets (Joules). Then one day .... kablouie.
>>
>
> Are you sure about this? MOV's do fail, and sometimes quite
> spectacularly, but if the energy they are asked to dissipate is within
> their ratings (and repetitive hits, heat build up or a steady state
> over voltage condition isn't encountered) they survive to do it again.
>
> MOV's can be degraded by repetitive strikes, no argument there. But
> the joule rating, isn't the number of strikes. e.g. 150 joule rating
> doesn't mean it fails if asked to take 300 hits of one joule each time
> (assuming heat build up isn't a factor).
>
> The typical failure mode, as I understand it, is when the device
> reaches a thermal runaway condition and shorts, shunting the power
> supply. Well within its energy rating and thermal limits, it is
> expected to last for a very long time.
>

Here's a link to a paper that is related to this, but does not
directly address the question, which boils down to:
if an MOV operates to supress a surge, is it or is it not
always degraded?

The link:
http://eprint.uq.edu.au/archive/00001980/01/The_effects_of_electrical_.pdf

Here's another link, that discusses MOV operation/structure at
the atomic level, and is more closely related to the question:
http://vpd.ms.northwestern.edu/Publications_files/Prad/JACS_82_3_485_1999.pdf

Watch out for line wrap in the links.

Now addressing the question directly:
It is my understanding that while it is theoretically possible
for an MOV to operate (ie switch from very high resistance
to very low resistance) without damage, for all practical
purposes, an MOV is degraded any time its threshhold is exceeded.

I'll describe a way to think of an MOV to illustrate puncture
within it. This is not an accurate description - only an easy
way to visualize what goes on. For the real description, read the
second link. In what follows, references are made to the
paper found at that site.

Picture an MOV as a huge number of parallel paths, consisting
of 2 conductors separated by an insulator. For a variety of
reasons, these paths are not 100% identical, despite
manufacturing efforts to make them so. This leads to a
phenomenom the paper in the 2nd link refers to as "local Joule
heating" in the presence of a surge. That means that a pulse
of energy will heat some paths more than others, causing
a puncture in the ZincOxide material. The paper states:
"...puncture failure is a progressive form of irreversible
degradation..." Thus, one (or more) of the paths can be
thought of as having a hole torn in the insulator (with
resultant heat) by the surge energy, allowing the two
conductors to short out, thus producing more heat due to
I^2R. The result is that the path is destroyed - the
irreversible degradation referenced above - by thermal
runaway. But there are an enormous number of paths, so
if only one is gone, the MOV is damaged, not totally
destroyed. Of course, with a long enough surge, or enough
surges of shorter duration, all of the paths will go.

There is another material failure mode within the
MOV - fracture of the crystaline microstructure -
discussed in the paper. It occurs with larger amplitude
shorter duration surges, as exhibited by the graph in
fig 11 in the paper. The MOV material simply cannot expand
fast enough as it heats due to conduction.

Ed

default

unread,
Aug 19, 2005, 10:05:52 PM8/19/05
to
On Sat, 20 Aug 2005 00:41:37 GMT, Joerg
<notthis...@removethispacbell.net> wrote:
snipped

>
>They do age with every strike above a certain threshold. Here is a brief
>explanation:
>http://www.mikeholt.com/mojonewsarchive/LSP-HTML/HTML/TVSS-Protection-Questions-and-Answers~20040708.php
>
>"Other surge suppressors rely on "sacrificial components," Metal Oxide
>Varistors (MOVs) or gas discharge devices that shunt or divert power
>surges from the hot line to the neutral or ground line. Sensitive
>digital equipment systems depend on a clean and uncontaminated ground.
>
>SurgeXŽ power filters offer an all-new, revolutionary Series Mode
>technology, which captures surge energy and slowly releases it onto the
>neutral line. Surge energy will not contaminate ground lines or wreak
>havoc with your system peripherals. With every surge above a modest
>level that hits your system, the MOVs in ordinary surge protectors
>degrade until they eventually fail. Unfortunately, it is impossible to
>tell when a MOV has become ineffective, leaving you exposed to potential
>equipment destruction."
>
>Anyway, they are indeed sacrificial components. The detailed mechanism
>should be explained in mfgs app notes. I have seen it but don't recall
>where, since I rarely use MOVs.
>
I buy that. (minus the hyperbole about "all-new revolutionary
yada-yada). MOV will eventually fail with repetitive strikes and
there's no way to tell when.

They are useful components, but they are no substitute for LC filters
if one is serious about cleaning up hash (with a spark gap and/or MOV
there for safety). Most of my applications for current sensing were
to operate some other gizmo when there was current flowing in the
monitored line, and usually for some safety related application.

For example, tell when a warning system fails, when a compressor dies,
a battery charger fails, etc.. Most of the time I'm asked to go in
after the equipment is installed, operational, and a problem crops up
or some underwriting agency or quality control person is unhappy about
the fail/fail safe modes that may be encountered.

There's an advantage to being able to install a device into existing
enclosures and not tap into power sources unnecessarily (clean sterile
environments, for the most part, where its much better if something
can work without maintenance or failure for decades or at least longer
than the life of the equipment, or CEO whichever comes first)

Sometimes merely running a wire takes damn near an act of God - dozens
of protocols to observe, and people who are paid to find reasons to
deny the feasibility of doing something/anything.

>> I guess I'd take issue with the statement: "in a current transformer
>> you don't have to" (have low impedance - or high permeability) That
>> all depends on what you plan to do - what load impedance you want to
>> drive.
>
>Well, usually a current transformer is used for current sensing and not
>to convert energy. There are certainly exceptions but then you'd rather
>need a "real" transformer.
>

Those tape wound iron-nickel alloy cores are for "real transformers"
in my opinion.

take care

default

unread,
Aug 20, 2005, 1:04:23 AM8/20/05
to
On Sat, 20 Aug 2005 04:43:05 GMT, ehsjr <eh...@bellatlantic.net>
wrote:

No arguments there. We've sort of gotten off on a tangent.

My original reason for citing an MOV was to clamp any HV spike that
might be present in the secondary of a current transformer while it
was not connected. That possibility might exist during a switching
transient in the monitored line, or a sudden short in the monitored
load. Current transformers, by their nature, usually have high turns
ratios, so it could happen . . .

Given that the transformer cores I'm using have around 0.15 square
inches of iron in them and any HV transient that might occur would be
of very short duration and probably far less than a joule an MOV would
probably last far longer than any equipment or personnel it protected.

I don't have to be sold on the limited protection MOV's offer. Seen
too many of them fail for that. But the applications where they do
tend to fail are when connected across the power line or line to
neutral, with lightning strikes or big inductive loads.

One place I worked had some "Reliance" DC drives. 460 volt three
phase input with 10-100 HP motors on them. If the mains fuses blew,
the grossly underrated MOV's would vaporize. There'd be a piece of
lead soldered to a ring terminal where the MOV's used to be. Never
found any shrapnel either, just some carbon and copper flashed on to
the chassis.

Fred Bloggs

unread,
Aug 20, 2005, 1:42:44 PM8/20/05
to
I don't think that PTC is necessary because the load is always on in
this case. The main thing is the ratio of standby of run current. I like
the AC relay idea, may require a little experimentation, but it should
go with a trigger not far from 90o point, maybe the 60o point. So Rs is
selected to trip the SCR at that point with safe margin at running load
current. The circuit drops out in standby. It's a Radio Shack project:
View in a fixed-width font such as Courier.


.
. +---------------------o/ o-------+-----+
. | | |
. | 1A | | |
. | 400V _____ | |
. | SCR 1N4007 ----- | |
. +-----|>|---|<|---+---KKKKK--+ | |
. | | | | | |
. | | +---|>|----+ | |
. | | 1N4007 | | -|-- N
. | | | | | - | E
. |1N4007| >10xRs | | | o| O
. +-|>|--+----/\/\--+ | | |--- | N
. | | | | -|--
. ___ | | | | |
. | \ | Rs | | | |
. L ---| |--+-----/\/\--------+ | | |
. | | | | | |
. N ---| |--+-----------------|----------|---|-----+
. ---/ | | | |
. | | | |
. | -------- | | | --------
. | | | | | | | |
. | | __ | | | | | __ |
. | | |------+ | +-----| | |
. | | -- | | | -- |
. | | ____ | | | ____ |
. +------| |----------------+ +----| | |
. | | ---- | | | ---- |
. | | | | | |
. | | | | | |
. | -------- | --------
. | TRIP | AUX
. | LOAD | LOAD
. | SKT | SKT
. +--------------------------------|
.

Put it in a plastic project box and if you want adjustablility , include
a little header with a jumper to select different Rs- all available from
RS. If your running load is say 50W then Iac is maybe 500mA so that the
trip can be set at 500mA (45o) - then you want 20mA*10*Rs + 0.7V=
500mA*Rs or Rs=2.2 ohms @ 1W. The circuit is completely dead when
Ipk*2.2<0.4V or Ipk=180mA, which is something like 15W, on the outside.
The relay provides all the hysteresis- and SCR chatter is immaterial.

Fred Bloggs

unread,
Aug 20, 2005, 1:53:02 PM8/20/05
to

Would help to get the SCR in right:

Fred Bloggs

unread,
Aug 20, 2005, 2:06:05 PM8/20/05
to

Fred Bloggs wrote:
>
>
> Fred Bloggs wrote:
>
>> I don't think that PTC is necessary because the load is always on in
>> this case. The main thing is the ratio of standby of run current. I
>> like the AC relay idea, may require a little experimentation, but it
>> should go with a trigger not far from 90o point, maybe the 60o point.
>> So Rs is selected to trip the SCR at that point with safe margin at
>> running load current. The circuit drops out in standby. It's a Radio
>> Shack project:
>>

Shaking the Heathkit idea, a TRIAC should be much better on the
conducted EMI with coil inductance doing the basic filtering.


View in a fixed-width font such as Courier.

.
.
.
.
. +---------------------o/ o-------+-----+
. | | | |

. | 0.001 100 | |
. +---||--/\/\------+ | | |
. | | | |
. | 400V | | | |
. | TRIAC | _____ | |
. | MT1 |>| | ----- | |
. +-----|<|---------+---KKKKK--+ | |
. | | | | |
. | | | | |
. | | >10xRs | | -|-- N
. +------+----/\/\--+ | | | - | E
. | | | | | o| O
. | | | | |--- | N
. +-----LVTVS-------+ | | -|--

Fred Bloggs

unread,
Aug 20, 2005, 2:09:23 PM8/20/05
to
View in a fixed-width font such as Courier.

.
.
.
.
. +---------------------o/ o-------+-----+
. | | | |

. | 0.001 100 | |
. +---||--/\/\------+ | | |
. | | | |
. | 400V | | | |
. | TRIAC | _____ | |
. | MT1 |>| | ----- | |
. +-----|<|---------+---KKKKK--+ | |
. | | | | |
. | | | | |
. | | >10xRs | | -|-- N

. | +----/\/\--+ | | | - | E


. | | | | | o| O
. | | | | |--- | N

. +-----LVTVS-------+ | | -|--

Joerg

unread,
Aug 20, 2005, 4:18:13 PM8/20/05
to
Hello Bob,

> There's an advantage to being able to install a device into existing
> enclosures and not tap into power sources unnecessarily (clean sterile
> environments, for the most part, where its much better if something
> can work without maintenance or failure for decades or at least longer
> than the life of the equipment, or CEO whichever comes first)
>
> Sometimes merely running a wire takes damn near an act of God - dozens
> of protocols to observe, and people who are paid to find reasons to
> deny the feasibility of doing something/anything.

In my domain (med electronics) the situation is similar. Everything must
be isolated and often up to defibrillator safey level, meaning it must
withstand a 5kV jolt without flinching. Lots of agency involvement and a
new agency to deal with in every market, US, EU, Asia plus some
"non-connected" countries who run their own regulatory scenarios. This
is where transformers and mostly of the ferrite type play a major role.
The actual defibrillator test usually impresses the regulatory folks.
You can see their faces become a bit pale and the chairs slowly rolling
backwards when I get ready to press that big button.

>>>I guess I'd take issue with the statement: "in a current transformer
>>>you don't have to" (have low impedance - or high permeability) That
>>>all depends on what you plan to do - what load impedance you want to
>>>drive.
>>
>>Well, usually a current transformer is used for current sensing and not
>>to convert energy. There are certainly exceptions but then you'd rather
>>need a "real" transformer.
>
> Those tape wound iron-nickel alloy cores are for "real transformers"
> in my opinion.

IIRC they are also used in GFCI outlets that you find in bathrooms and
kitchens. I haven't seen small ones used for power conversion yet, just
the larger versions, >1" or so.

Regards, Joerg

http://www.analogconsultants.com

Joerg

unread,
Aug 20, 2005, 4:21:23 PM8/20/05
to
Hello Bob,

> My original reason for citing an MOV was to clamp any HV spike that
> might be present in the secondary of a current transformer while it
> was not connected. That possibility might exist during a switching
> transient in the monitored line, or a sudden short in the monitored
> load. Current transformers, by their nature, usually have high turns
> ratios, so it could happen . . .

Old rule: Never let a current transformer secondary run open circuit.
Not even for a brief moment of switching.

Regards, Joerg

http://www.analogconsultants.com

ehsjr

unread,
Aug 21, 2005, 12:29:50 AM8/21/05
to

Absolutely. If there's no burden in the design, go
to jail, go directly to jail, do not pass go, do
not collect $200.00. An MOV doesn't replace the
burden R.

Ed

Jim Thompson

unread,
Aug 22, 2005, 11:08:20 AM8/22/05
to
On Fri, 19 Aug 2005 10:05:06 -0700, Jim Thompson
<thegr...@example.com> wrote:

[snip]


>
>(I've ordered the LCG3 Smart Power Strip that "DaveM" recommended.
>That will resolve the issue.)
>
> ...Jim Thompson

Got this back from the vendor...

The adjustment range is limited on the LCG1/2 models & recommended for
standard computers only. (use LCG3/4/5 or SCG4/5 for Apple, IBM & Dell
computers).

The LCG3/4/5 & SCG4/5 can be adjusted to turn on with any load 7 watts
to 1000 watts. Turn off is always set to 10% lower then turn on.

DaveM

unread,
Aug 22, 2005, 7:26:11 PM8/22/05
to
"Jim Thompson" <thegr...@example.com> wrote in message
news:2fqjg1ho6ab9sgio7...@4ax.com...

> On Fri, 19 Aug 2005 10:05:06 -0700, Jim Thompson
> <thegr...@example.com> wrote:
>
> [snip]
>>
>>(I've ordered the LCG3 Smart Power Strip that "DaveM" recommended.
>>That will resolve the issue.)
>>
>> ...Jim Thompson
>
> Got this back from the vendor...
>
> The adjustment range is limited on the LCG1/2 models & recommended for
> standard computers only. (use LCG3/4/5 or SCG4/5 for Apple, IBM & Dell
> computers).
>
> The LCG3/4/5 & SCG4/5 can be adjusted to turn on with any load 7 watts
> to 1000 watts. Turn off is always set to 10% lower then turn on.
>
> ...Jim Thompson
> --

Great to know that info!!! Thanks for asking the question to the vendor...
I had wondered about that spec, but never enough to follow up on it.
Thanks for posting it.

Fred Bloggs

unread,
Aug 22, 2005, 9:53:26 PM8/22/05
to

Jim Thompson wrote:

>
>
> Got this back from the vendor...
>
> The adjustment range is limited on the LCG1/2 models & recommended for
> standard computers only. (use LCG3/4/5 or SCG4/5 for Apple, IBM & Dell
> computers).
>
> The LCG3/4/5 & SCG4/5 can be adjusted to turn on with any load 7 watts
> to 1000 watts. Turn off is always set to 10% lower then turn on.
>
> ...Jim Thompson

150:1 range? May be hard to set and with that 10% hysteresis. That's a
dying niche market for you, now that everything is green.

Rich The Newsgroup Wacko

unread,
Aug 24, 2005, 6:52:58 PM8/24/05
to
On Wed, 17 Aug 2005 18:55:59 +0100, John Woodgate wrote:

> I read in sci.electronics.design that Jim Thompson
> <thegr...@example.com> wrote (in
> <0os6g1d8hv5l1qn39...@4ax.com>) about 'Sense Power Line
> Cord, Turn-on Fan', on Wed, 17 Aug 2005:


>>I need to sense the line current of a TV set and turn-on a fan.

>>Probably need adjustable sense current since set has your typical
>>standby mode.
>>
>>Ideas?
>

> Pass *one* conductor of the mains lead through a high-permeability (low
> frequency) ferrite toroid and wind say 50 turns of thin wire on the
> toroid.

It'd probably be easier to wind the 50 turns of thin wire on the toroid
_beore_ you pass the mains lead through the center. ;-P
--
Cheers!
Rich
------
"tear leather: To become excited, as in the sentence "Robin Hood tore his
leather jerkin' off.""

Ken Moffett

unread,
Aug 24, 2005, 7:16:44 PM8/24/05
to
Rich The Newsgroup Wacko <wa...@example.com> wrote in
news:pan.2005.08.24....@example.com:

> On Wed, 17 Aug 2005 18:55:59 +0100, John Woodgate wrote:
>
>> I read in sci.electronics.design that Jim Thompson
>> <thegr...@example.com> wrote (in
>> <0os6g1d8hv5l1qn39...@4ax.com>) about 'Sense Power Line
>> Cord, Turn-on Fan', on Wed, 17 Aug 2005:
>>>I need to sense the line current of a TV set and turn-on a fan.
>>>Probably need adjustable sense current since set has your typical
>>>standby mode.
>>>
>>>Ideas?
>>
>> Pass *one* conductor of the mains lead through a high-permeability (low
>> frequency) ferrite toroid and wind say 50 turns of thin wire on the
>> toroid.
>
> It'd probably be easier to wind the 50 turns of thin wire on the toroid
> _beore_ you pass the mains lead through the center. ;-P

I've made several "cheap & dirty" current transformers from low voltage E-I
laminated step-down transformers. I picked the kind that had the primary
and secondary windings on separate halves of the bobbin. Cut out the
secondary windings and wind one side of your power cord in it's place. One
or more turns, depending how many will fit. Use the original primary
(120vac) as the output. Add the appropiate load resistor.

Jim Thompson

unread,
Aug 24, 2005, 7:24:53 PM8/24/05
to
On Wed, 24 Aug 2005 23:16:44 GMT, Ken Moffett <KMo...@mn.rr.com>
wrote:

[snip]


>
>I've made several "cheap & dirty" current transformers from low voltage E-I
>laminated step-down transformers. I picked the kind that had the primary
>and secondary windings on separate halves of the bobbin. Cut out the
>secondary windings and wind one side of your power cord in it's place. One
>or more turns, depending how many will fit. Use the original primary
>(120vac) as the output. Add the appropiate load resistor.

As of this morning the LCG3 Smart Power Strip from BITS Limited is up
and running perfectly. The kind of solution I always prefer... $39
and plug it in... time required of me, a few minutes at their web site
;-)

DaveM

unread,
Aug 25, 2005, 8:37:31 AM8/25/05
to
"Jim Thompson" <thegr...@example.com> wrote in message
news:f80qg1tss0c5vg7vb...@4ax.com...

> On Wed, 24 Aug 2005 23:16:44 GMT, Ken Moffett <KMo...@mn.rr.com>
> wrote:
>
> [snip]
>>
>>I've made several "cheap & dirty" current transformers from low voltage
>>E-I
>>laminated step-down transformers. I picked the kind that had the primary
>>and secondary windings on separate halves of the bobbin. Cut out the
>>secondary windings and wind one side of your power cord in it's place. One
>>or more turns, depending how many will fit. Use the original primary
>>(120vac) as the output. Add the appropiate load resistor.
>
> As of this morning the LCG3 Smart Power Strip from BITS Limited is up
> and running perfectly. The kind of solution I always prefer... $39
> and plug it in... time required of me, a few minutes at their web site
> ;-)
>
> ...Jim Thompson
> --


Glad to hear that the Smart Strip is working for you Jim. Just curious..
did you have to adjust the sensivity pot on it to make it work with your TV?

Cheers!!!!