Post by thread:[EE]:Turn off 3 pin regulator?

PICers, That thread about battery backup reminded me of something ... I have a 5V PIC cct and an 8v driver cct. (Coupling is by 4N25 optocouplers). To save power I'd want to switch off the 8V regulator, say a 7808 in this case. OK, Let's say I put an NPN transistor (say BD139) in the earth leg of the 3pin 7808 & switch it on/off with a 4N25. No worries, that kills the 7812 ... BUT the o/p voltage then floats to supply, 12V in this case. So what you need is some simple way to switch off the regulator and have its o/p volts go to zero when it's OFF = when the npn tarnsistor is OFF. Any ideas? Thanx - Ian __________________________________________________________________ Get your free Australian email account at http://www.start.com.au -- http://www.piclist.com hint: To leave the PICList spam_OUTpiclist-unsubscribe-requestTakeThisOuTmitvma.mit.edu

> >So what you need is some simple way to switch off the regulator and have >its o/p volts go to zero when it's OFF = when the npn tarnsistor is OFF. P Channel mosfet on the output, or use a reg with a shutdown pin. -- http://www.piclist.com hint: To leave the PICList .....piclist-unsubscribe-requestKILLspam@spam@mitvma.mit.edu

At 04:22 AM 2/1/02 +1000, you wrote: >So what you need is some simple way to switch off the regulator and >have its o/p volts go to zero when it's OFF = when the npn tarnsistor >is OFF. A P-channel logic MOSFET, a resistor, and an open-collector driver (2N4401 + a resistor), in series with the input of the regulator. +a zener and another resistor if the 12V might have transients that could kill the 20V MOSFET gate. Or use a PNP 'tarnsistor' instead of the MOSFET. But, really, it would be better to just use an LM2941 (in TO-220-5 or TO-263-5) that has shutdown built-in, is LDO and costs about 85 cents in 100's (adjustable output voltage). Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" speffKILLspaminterlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com 9/11 United we Stand -- http://www.piclist.com hint: To leave the PICList .....piclist-unsubscribe-requestKILLspam.....mitvma.mit.edu

Original message from: Spehro Pefhany <EraseMEspeffspam_OUTTakeThisOuTINTERLOG.COM> > >But, really, it would be better to just use an LM2941 (in TO-220-5 >or TO-263-5) that has shutdown built-in, is LDO and costs about 85 >cents in 100's (adjustable output voltage). > Thanks guys ... that sounds like thge way to go. Ian __________________________________________________________________ Get your free Australian email account at http://www.start.com.au -- http://www.piclist.com hint: To leave the PICList piclist-unsubscribe-requestspam_OUTmitvma.mit.edu

I've done this sort of thing several times. Use that strange "open drain" PIC output -- RA4, I think. Hook it to the base of a PNP transistor via a resistor (10K or so). Hook emitter to +8, collector to the + input of the regulator. When your PIC wakes up, the output is tri-state, so no current flows thru the transistor. Program the pin to output and set it low, then the PNP transistor turns on, with pretty low drop of Vcesat, so you only lose a couple hundred millivolts. Switch the output high, the transistor turns back off, and the regulator output goes to zero. Remember that the regulator will not switch on and off instantly, since it needs a cap on its output to be stable, and that cap will take at least a few microseconds to charge and discharge, and maybe quite a few milliseconds. Depending on the circuits it's feeding, it may take a very long time for the output to go to zero. On one design turning a transmitter on and off, it was necessary to add another transistor to short the output of the regulator to ground when it was supposed to be off, so that the transmitter's oscillator would really stop and not interfere with a nearby receiver. > {Original Message removed}

Original message from: Don Hyde <@spam@DonHKILLspamAXONN.COM> > >I've done this sort of thing several times. >..... Remember that the regulator will not switch on and >off instantly, since it needs a cap on its output to be stable, and that cap >will take at least a few microseconds to charge and discharge, and maybe >quite a few milliseconds. Hey this sounds like a good idea. will try it. Decay time won't be a problem - the drain is around 600mA Thanx - Ian __________________________________________________________________ Get your free Australian email account at http://www.start.com.au -- http://www.piclist.com hint: To leave the PICList KILLspampiclist-unsubscribe-requestKILLspammitvma.mit.edu

> >So what you need is some simple way to switch off the regulator and > >have its o/p volts go to zero when it's OFF = when the npn tarnsistor > >is OFF. > But, really, it would be better to just use an LM2941 (in TO-220-5 > or TO-263-5) that has shutdown built-in, is LDO and costs about 85 > cents in 100's (adjustable output voltage). The LM2941 has a regrettably high quiescent current in shut down mode. The LM2936 cannot be shut down but has a much lower quiescent current - around 10uA in practice AFAIR - still too high in many cases. RM -- http://www.piclist.com hint: To leave the PICList RemoveMEpiclist-unsubscribe-requestTakeThisOuTmitvma.mit.edu

At 08:49 AM 2/1/02 +1300, you wrote: > >The LM2941 has a regrettably high quiescent current in shut down mode. I don't see it specified. It isn't the horrible regular Iq is it? 8-( It's a lateral-PNP type that draws a LOT of current when Vin-Vout gets small and at high output currents. >The LM2936 cannot be shut down but has a much lower quiescent current - >around 10uA in practice AFAIR - still too high in many cases. More expensive, but there is the LP3964 which uses an internal MOSFET pass transistor. 15uA in shutdown., but doesn't have the automotive resistant specs of the LM2941. Too high a shutdown current to eliminate the on/off switch in many applications, but a heck of a lot better than the few-mA Iq of a 78xx. Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" spamBeGonespeffspamBeGoneinterlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com 9/11 United we Stand -- http://www.piclist.com hint: To leave the PICList TakeThisOuTpiclist-unsubscribe-requestEraseMEspam_OUTmitvma.mit.edu

Original message from: Spehro Pefhany <RemoveMEspeffTakeThisOuTINTERLOG.COM> > >At 08:49 AM 2/1/02 +1300, you wrote: >> >>The LM2941 has a regrettably high quiescent current in shut down mode. > >>More expensive, but there is the LP3964 which uses an internal MOSFET >pass transistor. 15uA in shutdown., but doesn't have the automotive >resistant specs of the LM2941. Too high a shutdown current to eliminate >the on/off switch in many applications, but a heck of a lot better than >the few-mA Iq of a 78xx. Hey guys - I just tried that configuration with the PNP transistor in the i/p of the 3T regulator. Works like a charm! (transistor base low, o/p = 8V, base high, o/p= 0V) Uh .. didn't measure total Iq. Better do that ... :) Regards - Ian __________________________________________________________________ Get your free Australian email account at http://www.start.com.au -- http://www.piclist.com hint: To leave the PICList piclist-unsubscribe-requestEraseME.....mitvma.mit.edu

> >The LM2941 has a regrettably high quiescent current in shut down mode. > > I don't see it specified. It isn't the horrible regular Iq is it? 8-( > It's a lateral-PNP type that draws a LOT of current when Vin-Vout gets > small and at high output currents. The datasheet in the 1993 NatSemi Power IC's databook that I have has a diagram at the very end showing a pullup resistor on the on/off enable line (low = on, high = off). A note says "To assure shutdown, select resistor R3 [the pullup] to guarantee at least 300 uA of pullup current when S1 is open. Assume 2v at the on/off pin" The data sheet proper has a value for On/Off threshold current of 50 uA typical and 00 uA max at Von/off = 2v and Iout <= 1 amp. (Clearly Iout = 0ish when it's off but this is preeumably the curremt one is switching off with this Ion/off current. ) So the answer is - at least 300 uA (appnote) - typically 50 uA (data sheet) - worst case (data sheet) Choose one :-) This is a MUCH worse current drain than the LM2936's always on quiesecent current of 9 uA tyoical, 15 uA max (at load of 100 uA - lower presumably at no load current. Turning the supply off with a FET or even a PNP transistor you should be able to get off state quiescent current down to leakage levels - certainly far far far below 1 uA. Cost is 2 transistors max (1 NPN, 1 PNP and 2 resistors)(1 transistor if you use the PIC RA4 pin trick to drive the PNP). Russell McMahon -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

On Thu, 31 Jan 2002, David VanHorn wrote: > > > >So what you need is some simple way to switch off the regulator and have > >its o/p volts go to zero when it's OFF = when the npn tarnsistor is OFF. > > P Channel mosfet on the output, or use a reg with a shutdown pin. like LM317 for positive voltage, or a simple npn in series with the output reg. Vasile -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

That should have said > > >The LM2941 has a regrettably high quiescent current in shut down mode. > - at least 300 uA (appnote) > - typically 50 uA (data sheet) > - worst case 100 uA (data sheet) <====================== changed > > Choose one :-) -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

> > OK, Let's say I put an NPN transistor (say BD139) in the earth leg of > > the 3pin 7808 & switch it on/off with a 4N25. No worries, that kills > > the 7812 ... BUT the o/p voltage then floats to supply, 12V in this > > case. > > > > So what you need is some simple way to switch off the regulator and > > have its o/p volts go to zero when it's OFF = when the npn tarnsistor > > is OFF. > > > > Any ideas? Hi, there's a circuit here: centauri.ezy.net.au/~fastvid/self_swi.htm that might do what you want. :o) -Roman -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Original message from: Roman Black <EraseMEfastvidEZY.NET.AU> > > >Hi, there's a circuit here: >centauri.ezy.net.au/~fastvid/self_swi.htm >that might do what you want. :o) >-Roman > Yeah that cct looks cool. If the PIC did draw less than 5mA, though, you might want a resistor from 12V rail down to the SCR's anode to keep its current >Ido when it's in the on state? Just a thought ... Ian __________________________________________________________________ Get your free Australian email account at http://www.start.com.au -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Ian Hynes wrote: > > Original message from: Roman Black <RemoveMEfastvidEraseMEEraseMEEZY.NET.AU> > > > > > >Hi, there's a circuit here: > >centauri.ezy.net.au/~fastvid/self_swi.htm > >that might do what you want. :o) > >-Roman > > > Yeah that cct looks cool. If the PIC did draw less than 5mA, though, > you might want a resistor from 12V rail down to the SCR's anode to > keep its current >Ido when it's in the on state? Just a thought ... > > Ian Actually it's better to ensure the PIC (and other 5v stuff) draw more than 5mA when you choose, allowing the PIC to turn everything off by reducing the current below Ih. One interesting method is to put a 330 ohm resistor from a PIC output pin to gnd, when the PIC sets the pin high it will draw about 10mA. Turning the pin low will drop the 10mA and may alone be enough to break Ih and turn the entire circuit off. Also, a decoupling cap may be needed across the A and K of the scr to avoid breaking hold with rapid current changes. :o) -Roman -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Original message from: Roman Black <RemoveMEfastvidspam_OUTKILLspamEZY.NET.AU> > >>Actually it's better to ensure the PIC (and >other 5v stuff) draw more than 5mA when you >choose, allowing the PIC to turn everything off >by reducing the current below Ih. >One interesting method is to put a 330 ohm >resistor from a PIC output pin to gnd, when >the PIC sets the pin high it will draw about >10mA. Turning the pin low will drop the 10mA >and may alone be enough to break Ih and turn >the entire circuit off. >Also, a decoupling cap may be needed across >the A and K of the scr to avoid breaking hold >with rapid current changes. :o) >-Roman Hmm .. all very true. Getting back to the transistor turn on/off method, I had a diode from the 12V battery to the regulator i/p to protect against polarity reversal. With the emitter of the transistor seeing the high side & collector to regulator i/p, I guess you could dispense with the protecting diode? But a transistor's reverse Vbe doesn't want to be much more than 5V so should I keep the diode? With diode forward volts of, maybe 0.5V & VceSAT of maybe 0.4V, supply=12 (nominal) & regulator o/p=8V you're cutting it fine with the regulator's dropout level. To keep the diode or not to keep it ... that is the question :o -Ian __________________________________________________________________ Get your free Australian email account at http://www.start.com.au -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Ian Hynes wrote: {Quote hidden}> > Original message from: Roman Black <RemoveMEfastvidTakeThisOuTspamEZY.NET.AU> > > > >>Actually it's better to ensure the PIC (and > >other 5v stuff) draw more than 5mA when you > >choose, allowing the PIC to turn everything off > >by reducing the current below Ih. > >One interesting method is to put a 330 ohm > >resistor from a PIC output pin to gnd, when > >the PIC sets the pin high it will draw about > >10mA. Turning the pin low will drop the 10mA > >and may alone be enough to break Ih and turn > >the entire circuit off. > >Also, a decoupling cap may be needed across > >the A and K of the scr to avoid breaking hold > >with rapid current changes. :o) > >-Roman > > Hmm .. all very true. > Getting back to the transistor turn on/off method, I had a diode from > the 12V battery to the regulator i/p to protect against polarity > reversal. With the emitter of the transistor seeing the high side & > collector to regulator i/p, I guess you could dispense with the > protecting diode? > > But a transistor's reverse Vbe doesn't want to be much more than 5V so > should I keep the diode? With diode forward volts of, maybe 0.5V & > VceSAT of maybe 0.4V, supply=12 (nominal) & regulator o/p=8V you're > cutting it fine with the regulator's dropout level. To keep the diode > or not to keep it ... that is the question :o I can't remember the circuit Ian, can you re-post it? -Roman -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

part 1 1224 bytes content-type:text/plain; charset="us-ascii" (decoded 7bit) Original message from: Roman Black <EraseMEfastvidspamspamBeGoneEZY.NET.AU> > >Ian Hynes wrote: >> >> >> Getting back to the transistor turn on/off method, I had a diode from >> the 12V battery to the regulator i/p to protect against polarity >> reversal. With the emitter of the transistor seeing the high side & >> collector to regulator i/p, I guess you could dispense with the >> protecting diode? >> >> But a transistor's reverse Vbe doesn't want to be much more than 5V so >> should I keep the diode? With diode forward volts of, maybe 0.5V & >> VceSAT of maybe 0.4V, supply=12 (nominal) & regulator o/p=8V you're >> cutting it fine with the regulator's dropout level. To keep the diode >> or not to keep it ... that is the question :o > >I can't remember the circuit Ian, can you re-post >it? >-Roman OK - it's an attached 2k gif. Hope there's enough res left to make it out. I reckon you could dispense with the blocking diode? Thanks - Ian. > __________________________________________________________________ Get your free Australian email account at http://www.start.com.au part 2 1926 bytes content-type:image/gif; name="PWR01.gif" (decode) part 3 131 bytes -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.