Circuit for 12V lights to come on each evening.

[Terryc]

discussing projects and some stated that he was having trouble finding
and economical 12V timer and that all the timers he could purchase
required mains supply.

The situation is solar panle charges a SLA battery that runs a few 12V
leds for a porch light each evening.

I redefined it to needing a circuit that responded to ambient light
dropping and then switching on some LED lights for a set period of time,
say 2-4 hours, thus removing any need for a "timer" and Rt syncing/drift.

Has anyone come across such a circuit?

 

[kreed]

On Feb 16, 7:59 pm, Terryc <newsfourspam-s...@woa.com.au> wrote:

discussing projects and some stated that he was having trouble finding
and economical 12V timer and that all the timers he could purchase
required mains supply.

The situation is solar panle charges a SLA battery that runs a few 12V
leds for a porch light each evening.

I redefined it to needing a circuit that responded to ambient light
dropping and then switching on some LED lights for a set period of time,
say 2-4 hours, thus removing any need for a "timer" and Rt syncing/drift.

Has anyone come across such a circuit?

Google will soon reveal plenty of suitable circuits.
along the lines of "light sensitive switch" "light activated switch".

Make sure that you dont use a circuit with a relay if you can avoid
it, if you are trying to extend battery life.
A relay will chew up power. A FET will make bugger all difference to
consumption.



This probably is the best way to do what you want to do, though
putting a switch on the porch is probably
better if you only need it on when you are there.

 

[Mr.T]

"kreed" <kenreed1999@gmail.com> wrote in message
news:3025d52b-5fbd-47de-8617-8a5159135b2c@k18g2000prf.googlegroups.com...

Make sure that you dont use a circuit with a relay if you can avoid
it, if you are trying to extend battery life.
A relay will chew up power. A FET will make bugger all difference to
consumption.

In fact a bi-stable relay that only needs to be powered for a few
milliseconds for it to turn on or off, will use far less power than a FET!

MrT.

 

[Bruce Varley]

"Mr.T" <MrT@home> wrote in message
news:4b7b436a$0$3003$afc38c87@news.optusnet.com.au...

"kreed" <kenreed1999@gmail.com> wrote in message
news:3025d52b-5fbd-47de-8617-8a5159135b2c@k18g2000prf.googlegroups.com...
Make sure that you dont use a circuit with a relay if you can avoid
it, if you are trying to extend battery life.
A relay will chew up power. A FET will make bugger all difference to
consumption.

In fact a bi-stable relay that only needs to be powered for a few
milliseconds for it to turn on or off, will use far less power than a FET!

MrT.

That's fine as long as the circuit can ensure that the switching doesn't get

reversed as a result of a short power dip or something else.

 

[Jasen Betts]

On 2010-02-17, Bruce Varley <bxvarley@weastnet.com.au> wrote:

"Mr.T" <MrT@home> wrote in message
news:4b7b436a$0$3003$afc38c87@news.optusnet.com.au...

"kreed" <kenreed1999@gmail.com> wrote in message
news:3025d52b-5fbd-47de-8617-8a5159135b2c@k18g2000prf.googlegroups.com...
Make sure that you dont use a circuit with a relay if you can avoid
it, if you are trying to extend battery life.
A relay will chew up power. A FET will make bugger all difference to
consumption.

In fact a bi-stable relay that only needs to be powered for a few
milliseconds for it to turn on or off, will use far less power than a FET!

MrT.

That's fine as long as the circuit can ensure that the switching doesn't get
reversed as a result of a short power dip or something else.

latching relays typically need different inputs to switch each way.
(different polarity or on a different pin)

I'm not convinced that a latching relay would use less power than a
mosfet driven by a nor-nor cmos bistable.




--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

 

[Mr.T]

"Jasen Betts" <jasen@xnet.co.nz> wrote in message
news:hlgfjn$2do$1@reversiblemaps.ath.cx...

In fact a bi-stable relay that only needs to be powered for a few
milliseconds for it to turn on or off, will use far less power than a
FET!

That's fine as long as the circuit can ensure that the switching doesn't
get
reversed as a result of a short power dip or something else.

latching relays typically need different inputs to switch each way.
(different polarity or on a different pin)

Easy enough for the circuit to check the state first in any case.

I'm not convinced that a latching relay would use less power than a
mosfet driven by a nor-nor cmos bistable.

You know of a mosfet with no losses at all, or you think the relay will
consume more in a few milliseconds than the mosfet will in a few hours?

MrT.

 

[kreed]

On Feb 18, 10:22 am, "Mr.T" <MrT@home> wrote:

"Jasen Betts" <ja...@xnet.co.nz> wrote in message

news:hlgfjn$2do$1@reversiblemaps.ath.cx...

In fact a bi-stable relay that only needs to be powered for a few
milliseconds for it to turn on or off, will use far less power than a
FET!

That's fine as long as the circuit can ensure that the switching doesn't
get
reversed as a result of a short power dip or something else.

latching relays typically need different inputs to switch each way.
(different polarity or on a different pin)

Easy enough for the circuit to check the state first in any case.

I'm not convinced that a latching relay would use less power than a
mosfet driven by a nor-nor cmos bistable.

You know of a mosfet with no losses at all, or you think the relay will
consume more in a few milliseconds than the mosfet will in a few hours?

MrT.

In this application, the Mosfet losses should be negligible, depending
on the load.

Also what happens with the latching relay if the battery discharges
beyond the point of it being able to
turn itself off ? (especially as the battery ages). This could easily
happen as the LED's will be on overnight (due to the light sensor) and
even if the battery has sufficient capacity (and sufficient charging
current) to power them until dawn, (when switched off by the latching
relay) it would only need an overcast day to lower the battery's
charge rate to less than normal, and run the system out of juice
before dawn when it normally turns itself off.

Having the LED load locked "on" while charging may also prevent the
battery from ever reaching a decent charge (to unlatch the relay) for
a lot longer than usual. Every night that the charging stops before
it can get far enough, the LEDS will still be latched on killing
whatever charge may have been accumulated.

This would result in the load staying connected 24/7, probably
destroying the battery over time.

You can add battery level monitoring and auto-off, or a timer to do
the same thing, but these extra circuits being on all the time will
probably consume similar power than the FET would have in losses.

 

[Terryc]

kreed wrote:

Also what happens with the latching relay if the battery discharges
beyond the point of it being able to
turn itself off ? (especially as the battery ages). This could easily
happen as the LED's will be on overnight

The idea was that they would only be one for a few hours each evening.
simple timer after the light sensor

 

[Jasen Betts]

On 2010-02-18, Mr.T <MrT@home> wrote:

"Jasen Betts" <jasen@xnet.co.nz> wrote in message
news:hlgfjn$2do$1@reversiblemaps.ath.cx...
In fact a bi-stable relay that only needs to be powered for a few
milliseconds for it to turn on or off, will use far less power than a
FET!

That's fine as long as the circuit can ensure that the switching doesn't
get
reversed as a result of a short power dip or something else.

latching relays typically need different inputs to switch each way.
(different polarity or on a different pin)

Easy enough for the circuit to check the state first in any case.

I'm not convinced that a latching relay would use less power than a
mosfet driven by a nor-nor cmos bistable.

You know of a mosfet with no losses at all, or you think the relay will
consume more in a few milliseconds than the mosfet will in a few hours?

hmm it seems they cost about 100nA (ballpark) to run.
that adds up quickly.


--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

 

[Harry Pfeifer]

A suggestion for a starting point for a light operated control is the solar
powered lights that can be bought for a few dollars for use in a garden.
They come with a LED light at the top, a plastic tube about 400 mm long,
and a spike to stick in the ground. If you get one of these, remove the LED
and extend the connections that were on the LED to connect to the base
circuit of a transistor or FET which connects power to turn on the lights at
night


OldZ80





"Terryc" <newsfourspam-spam@woa.com.au> wrote in message
news:hldqa4$avk$1@speranza.aioe.org...

discussing projects and some stated that he was having trouble finding and
economical 12V timer and that all the timers he could purchase required
mains supply.

The situation is solar panle charges a SLA battery that runs a few 12V
leds for a porch light each evening.

I redefined it to needing a circuit that responded to ambient light
dropping and then switching on some LED lights for a set period of time,
say 2-4 hours, thus removing any need for a "timer" and Rt syncing/drift.

Has anyone come across such a circuit?

 

[Ban]

Harry Pfeifer wrote:

A suggestion for a starting point for a light operated control is the
solar powered lights that can be bought for a few dollars for use in
a garden. They come with a LED light at the top, a plastic tube
about 400 mm long, and a spike to stick in the ground. If you get
one of these, remove the LED and extend the connections that were on
the LED to connect to the base circuit of a transistor or FET which
connects power to turn on the lights at night


OldZ80

But that is completely unnessessary, as he has already a solar panel in his

setup. He could use it as a dawnswitch by measuring the voltage across the
diode feeding the battery, so he would only need a timer for the light to go
off.
ciao Ban

 

[Mr.T]

"Jasen Betts" <jasen@xnet.co.nz> wrote in message
news:hliv89$454$1@reversiblemaps.ath.cx...

You know of a mosfet with no losses at all, or you think the relay will
consume more in a few milliseconds than the mosfet will in a few hours?

hmm it seems they cost about 100nA (ballpark) to run.
that adds up quickly.

Since when is "100nA" a "power" loss? Seems you still fail to consider the
FET's on resistance and actual load current.
If it works for you, then I sure don't care But your claim that "a latching
relay would use less *power* than a mosfet driven by a nor-nor cmos
bistable." is still a little shaky IMO though.

MrT.

 

[Mr.T]

"kreed" <kenreed1999@gmail.com> wrote in message
news:c7aa4463-01b2-457e-a82b-600210200acc@t32g2000pre.googlegroups.com...

In this application, the Mosfet losses should be negligible, depending
on the load.

Well "negligable" is purely subjective and the load is unknown, so that
statement is pointless.

Also what happens with the latching relay if the battery discharges
beyond the point of it being able to
turn itself off ?

Best to monitor the battery to prevent total discharge in any case.

Having the LED load locked "on" while charging

Why on earth would you ever do that? If there is sufficient light for the
solar cells to charge the battery, wouldn't you have the circuit turn the
LED's OFF?

You can add battery level monitoring and auto-off, or a timer to do
the same thing,

Of course.

but these extra circuits being on all the time will
probably consume similar power than the FET would have in losses.

Only if the design were bad, or the load current *extremely* small.

MrT.

 

[Jasen Betts]

On 2010-02-19, Mr.T <MrT@home> wrote:

"Jasen Betts" <jasen@xnet.co.nz> wrote in message
news:hliv89$454$1@reversiblemaps.ath.cx...
You know of a mosfet with no losses at all, or you think the relay will
consume more in a few milliseconds than the mosfet will in a few hours?

hmm it seems they cost about 100nA (ballpark) to run.
that adds up quickly.

Since when is "100nA" a "power" loss? Seems you still fail to consider the
FET's on resistance and actual load current.

have a look at some figures:

OMRON - G6BK-1114P-1-US-DC12 : 30 milliohms

FAIRCHILD SEMICONDUCTOR - FDU8580 : 9 milliohms

also, the relay costs twenty times the price of the mosfet, and can
only handle half the current.

If it works for you, then I sure don't care But your claim that "a latching
relay would use less *power* than a mosfet driven by a nor-nor cmos
bistable." is still a little shaky IMO though.

What lets the mosfet down is the gate leakage when on and the D-S leakage when
off.

I think there's more power to be lost in choosing an inefficient dark detector
and/or timer circuit.


--- news://freenews.netfront.net/ - complaints: news@netfront.net ---