simplest push-button power switch

[Pawel Paron]

I want to make the simplest power on/off switch, operated with single push
button, using one MOSFET and some kind of a flip-flop. My question concerns
this flip-flop, it has to be something really simple and small, most
desirably a single SMD chip with a minimal number of pins, is there anything
like that available? I know how to do it with discrete components, or with
standard CMOS gates, but these are all too big. I'm looking for something,
that I can solder directly to the BUZ11 pins, and it will not protrude
significantly, so the volume of the complete switch is not much bigger than
the transistor itself.

Or, can I build something like a D flip-flop, using ne555? I have some of
these chips in SMD version. Just wonder if I can get it to work as a
bistable flip-flop, without many external components.

Regards
Pawel

 

[Rheilly Phoull]

I want to make the simplest power on/off switch, operated with single push
button, using one MOSFET and some kind of a flip-flop. My question concerns
this flip-flop, it has to be something really simple and small, most
desirably a single SMD chip with a minimal number of pins, is there anything
like that available? I know how to do it with discrete components, or with
standard CMOS gates, but these are all too big. I'm looking for something,
that I can solder directly to the BUZ11 pins, and it will not protrude
significantly, so the volume of the complete switch is not much bigger than
the transistor itself.

Or, can I build something like a D flip-flop, using ne555? I have some of
these chips in SMD version. Just wonder if I can get it to work as a
bistable flip-flop, without many external components.

Regards
Pawel

Surely the components needed for this must take more space than a miniature
switch ??

 

[Pawel Paron]

Surely the components needed for this must take more space than a miniature
switch ??

Well, it's not a must, but If I can do it really simply, like soldering some
tiny chip, then why not. There must be that MOSFET for another reason
(another circuit, that protects the power supplying battery from complete
discharge), so using a small switch instead of the whole thing is not an
option, it would replace just that flip-flop. In my opinion push buttons are
more reliable, than bi-stable switches, and they are usually smaller (or at
least flatter)

Pawel

 

[Fred Bloggs]

I want to make the simplest power on/off switch, operated with single push
button, using one MOSFET and some kind of a flip-flop. My question concerns
this flip-flop, it has to be something really simple and small, most
desirably a single SMD chip with a minimal number of pins, is there anything
like that available? I know how to do it with discrete components, or with
standard CMOS gates, but these are all too big. I'm looking for something,
that I can solder directly to the BUZ11 pins, and it will not protrude
significantly, so the volume of the complete switch is not much bigger than
the transistor itself.

Or, can I build something like a D flip-flop, using ne555? I have some of
these chips in SMD version. Just wonder if I can get it to work as a
bistable flip-flop, without many external components.

Regards
Pawel

If you look hard enough, you can find a PUSH-ON/PUSH-OFF switch. These
have the same form factor as momentary switches.

 

[Active8]

I want to make the simplest power on/off switch, operated with single push
button, using one MOSFET and some kind of a flip-flop. My question concerns
this flip-flop, it has to be something really simple and small, most
desirably a single SMD chip with a minimal number of pins, is there anything
like that available? I know how to do it with discrete components, or with
standard CMOS gates, but these are all too big. I'm looking for something,
that I can solder directly to the BUZ11 pins, and it will not protrude
significantly, so the volume of the complete switch is not much bigger than
the transistor itself.

Or, can I build something like a D flip-flop, using ne555? I have some of
these chips in SMD version. Just wonder if I can get it to work as a
bistable flip-flop, without many external components.

Regards
Pawel

google search this group for mini-gate? i haven't checked 'em but they
sound like something that will do the job.

give it some time and JT might chime in to confirm this.2

brs,
mike

 

[Pawel Paron]

google search this group for mini-gate? i haven't checked 'em but they
sound like something that will do the job.

Thanks, this is the solution, there is a single D flip-flop NL17SZ374.
Although I think these are the "hard-to-find" components, at least around my
place.

Pawel

 

[Active8]

Thanks, this is the solution, there is a single D flip-flop NL17SZ374.
Although I think these are the "hard-to-find" components, at least around my
place.

Pawel

great! i remembered Win Hill or someone mentioning finding D-FF in
single gate.

but i don't think they're that hard to find, it's just that they're new.
google this group on Mini-Gate and Jim Thompson and Active8. he replied
to me that if you go to whatever site he mentioned, and click refresh a
few times or so, the ad will come up for sample kits. 500 pieces IIRC.

i gotta get a pile of 'em.

brs,
mike

 

[Tom Bruhns]

Not too long ago I played with a simple on/off toggle using one power
mosfet which switched a 12V load, and the pushbutton switch and a
resistor and a capacitor. It's only useful in an application where you
will switch at least once a week or so, and where it doesn't get too
hot. For reliable operation, it should switch a load powered by at
least a couple times the gate turnon voltage, and if the supply is
more than the max gate voltage, you have to add another resistor. The
memory is simply charge on the gate of the FET, and that limits how
long it can go between toggles, as each toggle resets the gate
voltage. It requires a switch with low leakage in the "off" state.
Might this work for you?

Cheers,
Tom

 

[Pawel Paron]

Not too long ago I played with a simple on/off toggle using one power
mosfet which switched a 12V load, and the pushbutton switch and a
resistor and a capacitor.

But how do you turn it on and off with a single button? Or you use two
buttons, for charging and discharging the gate?

It's only useful in an application where you
will switch at least once a week or so, and where it doesn't get too
hot. For reliable operation, it should switch a load powered by at
least a couple times the gate turnon voltage, and if the supply is
more than the max gate voltage, you have to add another resistor. The
memory is simply charge on the gate of the FET, and that limits how
long it can go between toggles, as each toggle resets the gate
voltage.

As I understand it, I would expect a problem when the gate slowly
discharges, and at some point the transistor becomes a "resistor", heating
and finally burning?

Pawel

 

[Tom Del Rosso]

In Active8 typed:

but i don't think they're that hard to find, it's just that they're
new. google this group on Mini-Gate and Jim Thompson and Active8. he
replied
to me that if you go to whatever site he mentioned, and click refresh
a
few times or so, the ad will come up for sample kits. 500 pieces IIRC.

I found the message and hit refresh many times, but I think the ad
expired. If they still have any kits left, they might have pulled the
ad for it.

 

[Tom Del Rosso]

In Active8 typed:

but i don't think they're that hard to find, it's just that they're
new. google this group on Mini-Gate and Jim Thompson and Active8. he
replied
to me that if you go to whatever site he mentioned, and click refresh
a
few times or so, the ad will come up for sample kits. 500 pieces IIRC.

It appears they have none left, since it is was "backordered" but it's
not even that now. Google for "minigate sample kit" on the onsemi site
and they have a cached page that said it was backordered, but the
current page doesn't list it at all.

This guy claims to be selling them, but who knows if he even has them.
http://www.resellerratings.com/forum/t78845.html

 

[Active8]

but i don't think they're that hard to find, it's just that they're
new. google this group on Mini-Gate and Jim Thompson and Active8. he
replied
to me that if you go to whatever site he mentioned, and click refresh
a
few times or so, the ad will come up for sample kits. 500 pieces IIRC.

It appears they have none left, since it is was "backordered" but it's
not even that now. Google for "minigate sample kit" on the onsemi site
and they have a cached page that said it was backordered, but the
current page doesn't list it at all.[/QUOTE]

i take it that means you found JT's article where he gives the link. if
so, save me some time looking and repost it, please.

aside from that, I usually pick up the phone for more info.

This guy claims to be selling them, but who knows if he even has them.
http://www.resellerratings.com/forum/t78845.html

i don't think i'd order from him. money order.., he's never heard of
PayPal.

brs,
mike

 

[Active8]

i forgot the reason i came back here, duh. it was to say:

http://www.onsemi.com/site/support/literature/list/0,4858,databooks_
0,00.html

the PDF databook page. they have a mini-gate data book *and* product
selector guide.

mike

 

[Tom Del Rosso]

In Active8 typed:

i take it that means you found JT's article where he gives the link.
if so, save me some time looking and repost it, please.

Actually it was John Larkin. I couldn't find a similar one from Jim.

The link and the whole message:

http://www.onsemi.com/site/home/0,4370,0,00.html

http://groups.google.com/groups?hl=...ki61n3m2ffq49cpkojpkcv3pq1av%404ax.com&rnum=2

 

[Tom Del Rosso]

In Active8 typed:

i forgot the reason i came back here, duh. it was to say:

http://www.onsemi.com/site/support/literature/list/0,4858,databooks_
0,00.html

the PDF databook page. they have a mini-gate data book *and* product
selector guide.

But no samples in sight, right?

 

[Tom Bruhns]

Hi Pawel,

Yes, there are potential problems with the simplest circuit. If the
load is heavy, then it has the potential to destroy the FET. But it
IS simple! Consider a 12V 1A light bulb as the load. Connect the
load to +12V, the other end to the FET drain. FET source to the minus
side of the 12V supply (call it ground). Connect a 10 megohm resistor
from the drain to an 0.1uF capacitor. Other end of the capacitor to
ground. Switch from the capacitor/resistor to the FET gate. So when
the FET is "off" the capacitor charges through the load and resistor
to 12V. Push the switch for a short time: the time must be short
compared with R*C, or in this case, let's say not more than 1/10
second. The FET gate is charged to 12V, a little less because the
resistor starts to discharge the cap when the FET drain voltage drops.
The FET is now on, and the capacitor discharges to the FET drain
voltage, which is nearly zero. The next time you press the switch
button the FET gate goes to that new voltage. So the limitations are
that you must not press the button too long (t<<RC), that you not
press it too often (time between presses >> RC), and that you press it
more often than the leakage current drains off the gate charge. My
tests with an IRF540 (I think it was) amazed me: it took TWO WEEKS
for the gate charge to decay enough to significantly affect the
conduction. BUT this will be affected strongly by the FET and by the
cleanliness and insulation around it.

You can modify the circuit fairly easily by adding another PNP
transistor which will keep the FET gate solidly ON indefinetly, and a
resistor from gate to source to keep it OFF when it's in that state.
All this is easiest to implement when the power source is on the order
of 12V, but higher voltages can be accomodated using resistive
dividers, adding a couple resistors to the circuit. Because the
current required to hold the gate in either state is extremely small,
the resistors can be high values and not dissipate much power. Let me
know if the addition of the PNP isn't obvious...basically the base is
driven through a resistance from the FET drain and the emitter ties
through a high resistance to about +12V. The collector goes to the
FET gate. You can use the same scheme for toggling that I mentioned
above. Be sure the cap can't charge to a voltage that can destroy the
FET gate!

Cheers,
Tom

 

[Active8]

i take it that means you found JT's article where he gives the link.
if so, save me some time looking and repost it, please.

Actually it was John Larkin. I couldn't find a similar one from Jim.

The link and the whole message:

http://www.onsemi.com/site/home/0,4370,0,00.html

http://groups.google.com/groups?hl=...ki61n3m2ffq49cpkojpkcv3pq1av%404ax.com&rnum=2
[/QUOTE]

ok, john, then.

i'm not having any luck. i should have jumped on it asap.

that little gator on the PCB ad is nice, though.

brs,
mike

 

[Pawel Paron]

Thanks for this brilliant explanation. I got the whole idea. But I see the
problem, that I mentioned earlier: relying on the gate charge may lead to
destroying the FET, when the transistor enters some state between on and
off. As I observed, this can happen even with a static electricity, when you
just approach your hand or something near the gate, sometimes the transistor
may switch from off state to high resistance and possibly burn. So leaving
the gate open and "floating" is not an option, I want to use it for my
bicycle lights (12V/20W hallogen bulb), imagine I'm stuck in the forest,
because my lights got damaged, and all the wolves are around Actually
I've provided a simple solution against such an accident, all the
electronics between the bulb and the battery is connected via plugs, I can
unplug it and connect the battery directly to the lamp. Normally there is a
circuit, that warns me with sound when the voltage drops to 10.5V, and
another one, that cuts the power off at 8V, this one has a MOSFET and I
wanted to use it as an on/off switch too. And this has to be robust, in all
conditions (wet weather, static electricity - I carry battery with this
electronic stuff on my belt, and lamp on the helmet). But your solution is
really interesting and tricky, perharps I'll make some other use for it
(this version with an additional PNP transistor)

Regards
Pawel