Best method for dropping supply voltage a volt or two for low power device?

[GregS]

What is the best method for dropping the voltage of a device that has a
range of 8 to 12V powered by a car battery(so max, say 14V) to bring it
within spec? The device uses from 100uA to 3mA depending on state.

Devices should generally be protected up to 16 volts in that invironment. Some
charging specs go up to 15. 5 volts, but just over 14 would be more common. I do use
diodes many times. How about diodes and a zener?

greg

 

[Jon Slaughter]

What is the best method for dropping the voltage of a device that has a
range of 8 to 12V powered by a car battery(so max, say 14V) to bring it
within spec? The device uses from 100uA to 3mA depending on state.

I was thinking I probably could just use a diode or two and that should be
fine but it would take about 4-6 to do it though and that assumes the
battery is fully charged.

Seems like a regulator is overkill though(as I do not need regulation). I
pretty much want the cheapest solution that will work.

I'm thinking of a simple zener regulator, say at 9V with a ~700 ohm current
limiting resistor but this doesn't seem to work all that well over the whole
supply range.


Basically I'd like something that will drop 2V when the supply is at about
14V and 0V when its at 8V. I imagine this is to much trouble and probably I
won't find a better solution than just two diodes in series? Although this
won't let me run the device if the voltage drops down past 10V ;/).

The biggest problem is that with low current draw(100uA) and high
voltage(14V) it will drop, say, 1V, but at high current draw(3mA) and low
voltage(8V) it will drop 2V ;/ So seems like I can't have my cake an eat it
too ;/

The zener regulator also has a similar problem unless I want to waste a lot
of power ;/

Any other methods I should look into that won't be to expensive? I thought
about DC-DC converters but I'd like to avoid the additional cost and all the
extra components.

(essentially the zener diode will work if I could somehow remove it from the
circuit as the supply voltage decreased. Maybe a transistor could accomplish
this?)

Thanks,
Jon

 

[Mark]

What is the best method for dropping the voltage of a device that has a
range of 8 to 12V powered by a car battery(so max, say 14V) to bring it
within spec? The device uses from 100uA to 3mA depending on state.

I was thinking I probably could just use a diode or two and that should be
fine but it would take about 4-6 to do it though and that assumes the
battery is fully charged.

Seems like a regulator is overkill though(as I do not need regulation). I
pretty much want the cheapest solution that will work.

I'm thinking of a simple zener regulator, say at 9V with a ~700 ohm current
limiting resistor but this doesn't seem to work all that well over the whole
supply range.

Basically I'd like something that will drop 2V when the supply is at about
14V and 0V when its at 8V. I imagine this is to much trouble and probably I
won't find a better solution than just two diodes in series? Although this
won't let me run the device if the voltage drops down past 10V ;/).

The biggest problem is that with low current draw(100uA) and high
voltage(14V) it will drop, say, 1V, but at high current draw(3mA) and low
voltage(8V) it will drop 2V ;/  So seems like I can't have my cake an eat it
too ;/

The zener regulator also has a similar problem unless I want to waste a lot
of power ;/

Any other methods I should look into that won't be to expensive? I thought
about DC-DC converters but I'd like to avoid the additional cost and all the
extra components.

(essentially the zener diode will work if I could somehow remove it from the
circuit as the supply voltage decreased. Maybe a transistor could accomplish
this?)

Thanks,
Jon

10K resistor to a zener diode and regular diode in series regulating
to 8.9V, that voltage feeding the base of an emmitter follower that
puts out 8.2 Volts.

Mark

 

[Paul E. Schoen]

What is the best method for dropping the voltage of a device that has a
range of 8 to 12V powered by a car battery(so max, say 14V) to bring it
within spec? The device uses from 100uA to 3mA depending on state.

---
View in Courier:

.9-14VDC>------+
. |
. [300]
. |
. +------>8-12V
. |
. [1N4742]
. |
.GND>----------+------>GND

The load has a maximum draw of 3 mA, so the 300 ohms will drop less than 1
volt under worst conditions, for 8 volts out with battery at 9 volts
(pretty much a dead battery). At the other end, with 14 volts input, the
zener clamps the output at 12 VDC maximum, with no more than 13 mW wasted
in the resistor. It's a good idea to add a capacitor across the zener.

Paul

 

[Bobo The Chimp]

.

call stuff like this "minor circuit design." So let's see what /your/
boy can do- your fan club is comprised of largely the lowest quality
dregs , vermin , and trash- could be wrong , but I f_cking doubt it...

"Is comprised of" doesn't mean anything:
http://www.google.com/search?hl=en&q=define:comprise

The expression is "is composed of" or "comprises".

And its incorrect usage is a double-whammy, because it's usually used
incorrectly by people who are trying to sound sophisticated, but are
merely pretentious.

Cheers!
Rich^H^H^H^HBobo

 

[Bobo The Chimp]

[/QUOTE]

Robert, Your present persona does not improve your reputation. You
are very close to earning a TURD rating ;-)

Who the heck is "Robert"?

 

[Eeyore]

I'm thinking of a simple zener regulator, say at 9V with a ~700 ohm current
limiting resistor but this doesn't seem to work all that well over the whole
supply range.

Zener with constant current drive plus an emitter follower.

Graham

 

[Eeyore]

He doesn't need precision voltage regulation, so something with a zener,
and some transistors with high enough rating to take a 60V load dump
will do, this one works down to 8.1V, a TVS at Vin will catch the HV
short duration transients:
View in a fixed-width font such as
Courier.

.
.
. Vin
. |
. .----+----------.
. | | |
. | | |
. | | |
. [10k] | [470
. | | |
. | |< |
. +--| |
. | |\ |
. | | |
. | '---+------|-------> out
. | | |
. | [4.7k] |
. \| | |
. |------|------+
. <| | |
. | | _/
. +--------' ^ 1N4734
. | | 5V6
. [3.9k] |
. | |
. '------+--------'
. |
. ---
. ///

Mmmmm ... negative feedback ! Nice.


Graham

 

[_]

On Tue, 22 Apr 2008 10:13:30 -0600, "Jon Slaughter"


What is the best method for dropping the voltage of a device that has a
range of 8 to 12V powered by a car battery(so max, say 14V) to bring it
within spec? The device uses from 100uA to 3mA depending on state.

---
View in Courier:

.9-14VDC>------+
. |
. [300]
. |
. +------>8-12V
. |
. [1N4742]
. |
.GND>----------+------>GND


The load has a maximum draw of 3 mA, so the 300 ohms will drop less than 1
volt under worst conditions, for 8 volts out with battery at 9 volts
(pretty much a dead battery). At the other end, with 14 volts input, the
zener clamps the output at 12 VDC maximum, with no more than 13 mW wasted
in the resistor. It's a good idea to add a capacitor across the zener.

Paul

I think 8V is the cranking voltage spec for most apps- definitely not
9V-has nothing to do with a dead battery.

A century ago, when I was designing ignition systems, "cold crank" was
4V ;-)

...Jim Thompson

Y'know, I'd bookmark a link to a page or site that gives tips on how to
build power supplies that'll work from typical auto voltage, and protect
the small stuff on the output side from any mess on the input side.

Anyone got a url?

 

[Paul E. Schoen]

Paul E. Schoen wrote:



On Tue, 22 Apr 2008 10:13:30 -0600, "Jon Slaughter"



What is the best method for dropping the voltage of a device that has
a
range of 8 to 12V powered by a car battery(so max, say 14V) to bring
it
within spec? The device uses from 100uA to 3mA depending on state.

---
View in Courier:

.9-14VDC>------+
. |
. [300]
. |
. +------>8-12V
. |
. [1N4742]
. |
.GND>----------+------>GND


The load has a maximum draw of 3 mA, so the 300 ohms will drop less
than 1 volt under worst conditions, for 8 volts out with battery at 9
volts (pretty much a dead battery). At the other end, with 14 volts
input, the zener clamps the output at 12 VDC maximum, with no more than
13 mW wasted in the resistor. It's a good idea to add a capacitor
across the zener.

Paul


I think 8V is the cranking voltage spec for most apps- definitely not
9V-has nothing to do with a dead battery.

A century ago, when I was designing ignition systems, "cold crank" was
4V ;-)

...Jim Thompson

The 1908 models? Had they even discovered lead-acid or was it carbon zinc
back then?

That must have been a 6 volt system, or possibly the voltage at the starter
terminals. Of course, in 1908 many (most/all?) cars had a manual crank, and
probably a magneto ignition, with no battery.

The lead-acid battery was invented in 1859
http://en.wikipedia.org/wiki/Lead-acid_battery. The first practical
electric starter was invented in 1911
http://en.wikipedia.org/wiki/Electric_starter. The first battery powered
ignition was introduced in 1910
http://en.wikipedia.org/wiki/Ignition_system.

A simple way to avoid the effect of low voltage during cranking, in this
case, would be to add a diode and a fairly large storage capacitor. 10,000
uF should give you at least 10 seconds before the capacitor voltage dropped
below 8 volts (starting from 12 V).

Paul

 

[neon]

no matter what you add zeners diodes resistors there is going to be waste power. since you know the max load a current source is suitable solution.

 

[Rich Grise]

Is English even your first language, retard???


The hell if it is retard:
tr.v. com·prised, com·pris·ing, com·pris·es
1. To consist of; be composed of:
2. To include; contain:

Yes, exactly. If you say "is comprised of" you are saying "is contained
of", which doesn't mean anything.

Is English ANY of your languages?

Looks like you're another one of these high school diploma whizzes,
trying desperately to be learned but actually acting like just another
sophomoric 'tard... who can't even use the dictionary.

Yeah - get your English corrected, and rather than learn what the
word means, you attack those who wish to educate you.

Who's the nitwit now?

Sheesh!
Rich

 

[legg]

He doesn't need precision voltage regulation, so something with a zener,
and some transistors with high enough rating to take a 60V load dump
will do, this one works down to 8.1V, a TVS at Vin will catch the HV
short duration transients:
View in a fixed-width font such as
Courier.

.
.
. Vin
. |
. .----+----------.
. | | |
. | | |
. | | |
. [10k] | [470
. | | |
. | |< |
. +--| |
. | |\ |
. | | |
. | '---+------|-------> out
. | | |
. | [4.7k] |
. \| | |
. |------|------+
. <| | |
. | | _/
. +--------' ^ 1N4734
. | | 5V6
. [3.9k] |
. | |
. '------+--------'
. |
. ---
. ///

Mmmmm ... negative feedback ! Nice.

Almost full marks, but that bias current!!

Lowering the zener voltage and raising 470 to 4k7 would be a start.

Add another zener and bias current can be sub-milliamp.


Version 4
*LDO graham mod
SHEET 1 880 680
WIRE 0 -128 -144 -128
WIRE 112 -128 0 -128
WIRE 224 -128 112 -128
WIRE -144 -96 -144 -128
WIRE 0 -80 0 -128
WIRE 112 -32 112 -128
WIRE 0 16 0 0
WIRE 48 16 0 16
WIRE -144 64 -144 -16
WIRE 144 64 112 64
WIRE 352 64 144 64
WIRE 448 64 352 64
WIRE 0 80 0 16
WIRE 16 80 0 80
WIRE 224 80 224 -128
WIRE 16 112 16 80
WIRE 144 128 144 64
WIRE 224 160 80 160
WIRE 448 160 448 64
WIRE 352 176 352 64
WIRE 16 240 16 208
WIRE 144 240 144 192
WIRE 144 240 16 240
WIRE 16 256 16 240
WIRE 224 272 224 160
WIRE -144 384 -144 144
WIRE 16 384 16 336
WIRE 16 384 -144 384
WIRE 176 384 16 384
WIRE 224 384 224 336
WIRE 224 384 176 384
WIRE 352 384 352 240
WIRE 352 384 224 384
WIRE 448 384 448 240
WIRE 448 384 352 384
WIRE 176 416 176 384
FLAG 176 416 0
SYMBOL npn 80 112 M0
SYMATTR InstName Q1
SYMATTR Value 2N4401
SYMBOL pnp 48 64 M180
SYMATTR InstName Q2
SYMATTR Value 2N4403
SYMBOL res -16 -96 R0
SYMATTR InstName R1
SYMATTR Value 10K
SYMBOL res 208 64 R0
SYMATTR InstName R2
SYMATTR Value 47K
SYMBOL res 0 240 R0
SYMATTR InstName R3
SYMATTR Value 3K9
SYMBOL cap 368 240 R180
WINDOW 0 24 64 Left 0
WINDOW 3 24 8 Left 0
SYMATTR InstName C2
SYMATTR Value 1E-7
SYMBOL res 432 144 R0
SYMATTR InstName R4
SYMATTR Value 3K3
SYMBOL voltage -144 48 R0
WINDOW 3 -73 308 Left 0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V1
SYMATTR Value PULSE(0 14 1E-5 1E-3 1E-3 1E-4 2.2E-3 2)
SYMBOL zener 240 336 R180
WINDOW 0 24 72 Left 0
WINDOW 3 24 0 Left 0
SYMATTR InstName D1
SYMATTR Value CMPZ4625
SYMBOL res -128 0 R180
WINDOW 0 36 76 Left 0
WINDOW 3 36 40 Left 0
SYMATTR InstName R6
SYMATTR Value 1
SYMBOL zener 160 192 R180
WINDOW 0 24 72 Left 0
WINDOW 3 -57 -22 Left 0
SYMATTR InstName D2
SYMATTR Value BZX84C6V2L
TEXT 236 -4 Left 0 !.tran 0 4.4E-3 1E-5 1E-6

 

[legg]

. Vin
. |
. .----+----------.
. | | |
. | | |
. | | |
. [10k] | [470
. | | |
. | |< |
. +--| |
. | |\ |
. | | |
. | '---+------|-------> out
. | | |
. | [4.7k] |
. \| | |
. |------|------+
. <| | |
. | | _/
. +--------' ^ 1N4734
. | | 5V6
. [3.9k] |
. | |
. '------+--------'
. |
. ---
. ///
.

good one, but high bias. Swapping in a zener and lowering 5V6 to 2V7
gives submilliamp bias and 'no' headroom.

. Vin
. |
. .----+----------.
. | | |
. | | |
. | | |
. [10K] | [47K]
. | | |
. | |< |
. +--| |
. | |\ |
. | | |
. | '---+------|-----+-> out
. | | | |
. | _ | |
. \| ^ 6V2 | [2K7] (3mA load)
. |------|------+ |
. <| | | |
. | | _/ |
. +--------' ^ 2V7|
. | | |
. [3.9K] | |
. | | |
. '------+--------+-----'
. |
. ---
. ///

 

[legg]

I don't think it's a good idea to operate those zeners at such low
currents, you're at the knee and lucky to be within 20% of nominal
voltage, and SPICE doesn't model zeners in that region of operation well
enough to be trusted...If you don't like the crummy 20mA bias, go with a
'431, at least you will know what you're getting and you can derive the
majority of the bias from the regulated output, using the unregulated
input for start-up only, ending up with enormous ripple rejection. I
went with the 5V6 1W because it's popular, readily available, cheap,
takes the 60V load dump, and good enough...

I think the CMP or CMH zeners are characterized at 250uA, with a
dynamic impedance around 1K3. The spice model showed the biasing
kicking in around 4V with the 2K7 load resistor.

My own version uses a low power 431 and a steering diode instead of
the two zeners, but I like the simplicity of your arrangement for this
lower power app, with apparently reduced compensation issues. The OP
only wanted 3mA out. Interesting to see effects of capacitive loading.

I don't think there's a load dump issue, except possibly for the
transistors.

Taking all models with a grain of salt, this circuit shouldn't raise
breadboarding issues with the novice.

RL

 

[Rich Grise]

On Tue, 22 Apr 2008 21:39:52 -0400, Fred Bloggs <[email protected]>

Bloggs is....
[snip]

a brain damaged drug addict, needs
help...

[snip]

Yeah, I should have known better. Since I reinstalled everything
recently, I haven't gotten around to re-filtering him; I guess
it's time for me to clue up. ;-)

Thanks!
Rich

 

[Rich Grise]

"Usage
Although it has been in use since the late 18th century, sense 3 is
still attacked as wrong. Why it has been singled out is not clear,...

Well, it's perfectly clear to one who speaks English correctly.

Sigh, I guess it's time to clue up again, and quit trying to insert
abstract thoughts into concrete heads.

Thanks,
Rich

 

[Rich Grise]

You're another one, you don't know anything about electronics, and
you're too stupid to learn, so why do you even post here at all?

Bloggs added to master-baiter file ;-)[/QUOTE]

No, no, no! He's not a baiter - he's the troll that _you're_ baiting! ;-)

The only reason I responded at all was my own pedantry - I just had a
lapse of judgement there and for a moment I actually believed it was
possible to teach a brick to sing. ;-)

Cheers!
Rich

 

[Rich Grise]

Despite the evidence of existing usage, you insist on the circular
argument about what is "correct" according to your preconception.
How's that for concrete?

Well, just because millions of people have been using it wrong, doesn't
make it right.

And yes, that many people can be wrong, and many more - just look at who
got elected president! ;-)

Cheers!
Rich