Forums

PNP for soft switch and reverse battery protection

Started by Peabody February 9, 2017
I'm working on a small circuit that will be powered by a 9V battery.  I would 
like to combine the functions of a soft power switch and reverse battery 
protection in a single device.  I think the answer may be to use a PNP 
transistor on the high side, with a base resistor to ground - actually to 
ground through a physical momentary switch, or indirectly to ground via an 
NPN transistor controlled by a microcontroller output pin.  But for purposes 
of this post, just think of a base resistor to ground.  I only need a few mA 
of current, and the base resistor needed will be 100K.

My question concerns reverse polarity protection.  I believe I understand 
correctly that current will not flow backwards from collector to emitter of 
a PNP so long as the transistor's voltage rating is not exceeded.  So the 
"main" power flow will be protected.

But the absolute maximum base-emitter reverse voltage for this transistor 
(BC560C) is 5V. So if 9V were to be applied directly to the base, with the 
emitter at ground, I think the transistor would be blown.  The question is 
whether the 100K resistor limits current enough so that the transistor is not 
damaged.  I just don't have any experience with this situation, and don't 
knpw what actually works.  There is also a separate question as to whether 
the limited reverse current flowing through the resistor would damage the 
microcontroller (max 3.6V Vcc) since all of that current would actually flow 
through it via its protection diodes.

I guess if it's clear that, at 9V, 100K is gonna keep anything from losing 
its smoke, I would just go with that.  But I'm more than a little goosey 
about that.  The easiest alternative I can think of is to just insert a diode 
between the base and resistor.  That would prevent any reverse current 
flowing into the PNP base up to the rating of the diode, which will be way 
above 9V, while not really affecting how the transistor functions.  In 
particular, it would have no effect on the main E-C voltage drop in normal 
operation.

So what do you think?  Am I right at least about the theoretical risk?  If 
so, does the 100K resistor cover me, or do I need the diode?  Or is there a 
better way?  Well, I guess there's always a better way.  As I said, I want to 
combine a soft switch with polarity protection.  A P-channel MOSFET can't do 
both because the body diode has to be one way to function as a switch, and 
the opposite way to function as polarity protection.  The PNP looks like the 
right solution.

Thanks for any suggestions.

On Thu, 09 Feb 2017 13:16:44 -0600, Peabody
<waybackNO584SPAM44@yahoo.com> wrote:

>I'm working on a small circuit that will be powered by a 9V battery. I would >like to combine the functions of a soft power switch and reverse battery >protection in a single device. I think the answer may be to use a PNP >transistor on the high side, with a base resistor to ground - actually to >ground through a physical momentary switch, or indirectly to ground via an >NPN transistor controlled by a microcontroller output pin. But for purposes >of this post, just think of a base resistor to ground. I only need a few mA >of current, and the base resistor needed will be 100K. > >My question concerns reverse polarity protection. I believe I understand >correctly that current will not flow backwards from collector to emitter of >a PNP so long as the transistor's voltage rating is not exceeded. So the >"main" power flow will be protected. > >But the absolute maximum base-emitter reverse voltage for this transistor >(BC560C) is 5V. So if 9V were to be applied directly to the base, with the >emitter at ground, I think the transistor would be blown. The question is >whether the 100K resistor limits current enough so that the transistor is not >damaged. I just don't have any experience with this situation, and don't >knpw what actually works. There is also a separate question as to whether >the limited reverse current flowing through the resistor would damage the >microcontroller (max 3.6V Vcc) since all of that current would actually flow >through it via its protection diodes. > >I guess if it's clear that, at 9V, 100K is gonna keep anything from losing >its smoke, I would just go with that. But I'm more than a little goosey >about that. The easiest alternative I can think of is to just insert a diode >between the base and resistor. That would prevent any reverse current >flowing into the PNP base up to the rating of the diode, which will be way >above 9V, while not really affecting how the transistor functions. In >particular, it would have no effect on the main E-C voltage drop in normal >operation. > >So what do you think? Am I right at least about the theoretical risk? If >so, does the 100K resistor cover me, or do I need the diode? Or is there a >better way? Well, I guess there's always a better way. As I said, I want to >combine a soft switch with polarity protection. A P-channel MOSFET can't do >both because the body diode has to be one way to function as a switch, and >the opposite way to function as polarity protection. The PNP looks like the >right solution. > >Thanks for any suggestions.
There are some ideas here: http://www.ti.com/lit/an/slva139/slva139.pdf -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
On Thu, 09 Feb 2017 13:16:44 -0600, Peabody
<waybackNO584SPAM44@yahoo.com> wrote:

[snip]

>The easiest alternative I can think of is to just insert a diode >between the base and resistor. That would prevent any reverse current >flowing into the PNP base up to the rating of the diode, which will be way >above 9V, while not really affecting how the transistor functions. In >particular, it would have no effect on the main E-C voltage drop in normal >operation. >
[snip] That won't keep the load from seeing a reverse voltage. If you could find a small P-Channel device that would be your best bet. ...Jim Thompson -- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | STV, Queen Creek, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 | Thinking outside the box... producing elegant solutions.
Jim Thompson says...

 >> The easiest alternative I can think of is to just
 >> insert a diode between the base and resistor.  That
 >> would prevent any reverse current flowing into the PNP
 >> base up to the rating of the diode, which will be way
 >> above 9V, while not really affecting how the transistor
 >> functions.  In particular, it would have no effect on
 >> the main E-C voltage drop in normal operation.

 > That won't keep the load from seeing a reverse voltage.

 > If you could find a small P-Channel device that would be
 > your best bet.

I don't understand.  It seems to me the PNP C-E does exactly the
same thing as a p-channel mosfet, which is to block current
from flowing when the battery is connected backward.  So
it's the same for the main load with either device.  It's
just the pesky PNP base that I'n worried about.  The mosfet
doesn't present that problem, but then it can't both switch
and protect in a single device.

John Larkin says...

>There are some ideas here: > >http://www.ti.com/lit/an/slva139/slva139.pdf
Yes, I saw that. But it doesn't address using a PNP at all.
Den torsdag den 9. februar 2017 kl. 20.18.05 UTC+1 skrev Peabody:
> I'm working on a small circuit that will be powered by a 9V battery. I would > like to combine the functions of a soft power switch and reverse battery > protection in a single device. I think the answer may be to use a PNP > transistor on the high side, with a base resistor to ground - actually to > ground through a physical momentary switch, or indirectly to ground via an > NPN transistor controlled by a microcontroller output pin. But for purposes > of this post, just think of a base resistor to ground. I only need a few mA > of current, and the base resistor needed will be 100K. > > My question concerns reverse polarity protection. I believe I understand > correctly that current will not flow backwards from collector to emitter of > a PNP so long as the transistor's voltage rating is not exceeded. So the > "main" power flow will be protected. > > But the absolute maximum base-emitter reverse voltage for this transistor > (BC560C) is 5V. So if 9V were to be applied directly to the base, with the > emitter at ground, I think the transistor would be blown. The question is > whether the 100K resistor limits current enough so that the transistor is not > damaged. I just don't have any experience with this situation, and don't > knpw what actually works. There is also a separate question as to whether > the limited reverse current flowing through the resistor would damage the > microcontroller (max 3.6V Vcc) since all of that current would actually flow > through it via its protection diodes. > > I guess if it's clear that, at 9V, 100K is gonna keep anything from losing > its smoke, I would just go with that. But I'm more than a little goosey > about that. The easiest alternative I can think of is to just insert a diode > between the base and resistor. That would prevent any reverse current > flowing into the PNP base up to the rating of the diode, which will be way > above 9V, while not really affecting how the transistor functions. In > particular, it would have no effect on the main E-C voltage drop in normal > operation. > > So what do you think? Am I right at least about the theoretical risk? If > so, does the 100K resistor cover me, or do I need the diode? Or is there a > better way? Well, I guess there's always a better way. As I said, I want to > combine a soft switch with polarity protection. A P-channel MOSFET can't do > both because the body diode has to be one way to function as a switch, and > the opposite way to function as polarity protection. The PNP looks like the > right solution. > > Thanks for any suggestions.
use two pfets back to back or maybe a diode and a voltage regulator with enable
"Peabody" <waybackNO584SPAM44@yahoo.com> wrote in message 
news:20170209-224630.33.0@Peabody.ssl.astraweb.com...
> I don't understand. It seems to me the PNP C-E does exactly the > same thing as a p-channel mosfet, which is to block current > from flowing when the battery is connected backward. So > it's the same for the main load with either device. It's > just the pesky PNP base that I'n worried about. The mosfet > doesn't present that problem, but then it can't both switch > and protect in a single device. >
If your battery is single cell (i.e., up to 4V), the PNP blocks current in both directions (because 4V is below Vebo). Low-Vce(sat) types usually have quite high hFE in both forward and reverse directions. All you need to do, to make it "turn on" (doesn't matter which direction current flows), is draw base current. The base current goes in from whichever remaining terminal is at a higher voltage, emitter or collector (traditionally, emitter). With high hFE, the efficiency can be quite good (~99%), but if you aren't controlling base current to keep it proportional to load current, your average efficiency on a variable load will suck. If your load is highly variable, back-to-back PMOS starts to look really good -- and as a bonus, low voltage MOS has great specs: low Qg, high gain (1.8V logic level!) and low Rds(on). Tim -- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website: http://seventransistorlabs.com
On Thu, 09 Feb 2017 16:47:38 -0600, Peabody
<waybackNO584SPAM44@yahoo.com> wrote:

>John Larkin says... > >>There are some ideas here: >> >>http://www.ti.com/lit/an/slva139/slva139.pdf > >Yes, I saw that. But it doesn't address using a PNP at all.
Then use a mosfet! Or TPS2660, if your budget allows. -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
On Friday, February 10, 2017 at 6:18:05 AM UTC+11, Peabody wrote:
> I'm working on a small circuit that will be powered by a 9V battery. I would > like to combine the functions of a soft power switch and reverse battery > protection in a single device. I think the answer may be to use a PNP > transistor on the high side, with a base resistor to ground - actually to > ground through a physical momentary switch, or indirectly to ground via an > NPN transistor controlled by a microcontroller output pin. But for purposes > of this post, just think of a base resistor to ground. I only need a few mA > of current, and the base resistor needed will be 100K. > > My question concerns reverse polarity protection. I believe I understand > correctly that current will not flow backwards from collector to emitter of > a PNP so long as the transistor's voltage rating is not exceeded. So the > "main" power flow will be protected. > > But the absolute maximum base-emitter reverse voltage for this transistor > (BC560C) is 5V. So if 9V were to be applied directly to the base, with the > emitter at ground, I think the transistor would be blown. The question is > whether the 100K resistor limits current enough so that the transistor is not > damaged.
The resistor will prevent gross damage, but it is known that even a small reverse current through the base emitter junction of a bipolar transistor progressively degrades the junction and the forward current gain of the transistor. The damage can be annealed out, and there was a Hewlett-Packard function generator where one of the routine service procedures involved the service engineer stubbing out his cigarette on a particular output transistor to do just that. The story dates back to the days when most people smoked ...
> I just don't have any experience with this situation, and don't > knpw what actually works. There is also a separate question as to whether > the limited reverse current flowing through the resistor would damage the > microcontroller (max 3.6V Vcc) since all of that current would actually flow > through it via its protection diodes.
The protection diodes will be fine. The microcontroller might not be. The protection diodes inject charge carriers into the substrate of the device, and they can do unexpected things. The microcontroller can't be relied to perform as specified when the protection diodes are conducting. -- Bill Sloman, Sydney
On Thu, 9 Feb 2017 17:01:03 -0600, "Tim Williams"
<tiwill@seventransistorlabs.com> wrote:

>"Peabody" <waybackNO584SPAM44@yahoo.com> wrote in message >news:20170209-224630.33.0@Peabody.ssl.astraweb.com... >> I don't understand. It seems to me the PNP C-E does exactly the >> same thing as a p-channel mosfet, which is to block current >> from flowing when the battery is connected backward. So >> it's the same for the main load with either device. It's >> just the pesky PNP base that I'n worried about. The mosfet >> doesn't present that problem, but then it can't both switch >> and protect in a single device. >> > >If your battery is single cell (i.e., up to 4V), the PNP blocks current in >both directions (because 4V is below Vebo). > >Low-Vce(sat) types usually have quite high hFE in both forward and reverse >directions. All you need to do, to make it "turn on" (doesn't matter which >direction current flows), is draw base current. The base current goes in >from whichever remaining terminal is at a higher voltage, emitter or >collector (traditionally, emitter). > >With high hFE, the efficiency can be quite good (~99%), but if you aren't >controlling base current to keep it proportional to load current, your >average efficiency on a variable load will suck. > >If your load is highly variable, back-to-back PMOS starts to look really >good -- and as a bonus, low voltage MOS has great specs: low Qg, high gain >(1.8V logic level!) and low Rds(on). > >Tim
Bipolar devices are bilateral except that forward and reverse Beta usually differ. Leakage will do you in if you float the base. PMOS device are not bilateral, unless you swap the body, which I sometimes do ;-) ...Jim Thompson -- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | STV, Queen Creek, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 | Thinking outside the box... producing elegant solutions.