# 431 shunt regulator help

J

#### Jim

Jan 1, 1970
0
trying to drop 12 vdc to 5vdc for an extended scale A/D, I would like
to use a TI431 to sense anything above 9 or 10 volts, feed it into my
uC for display on a digital meter. I am a software guy with limited HW
experience so here are my questions, they all pertain to a simple
shunt regulator circuit, eg; current resistor on Vin. Ref tapped into
voltage divider (R1, R2)

with a regulated input voltage that of 12v and a uC A/D pin that can
accept 5v, what resistors do I need for the current resistor and the
voltage dividers in order to capture the top 5 volts (lets say 9v and
above).

If you would rather "teach me to fish" I will accept a formula that is
written in 8th grade terms ; )

Any help
Thanks

B

#### [email protected]

Jan 1, 1970
0
trying to drop 12 vdc to 5vdc for an extended scale A/D, I would like
to use a TI431 to sense anything above 9 or 10 volts, feed it into my
uC for display on a digital meter. I am a software guy with limited HW
experience so here are my questions, they all pertain to a simple
shunt regulator circuit, eg; current resistor on Vin. Ref tapped into
voltage divider (R1, R2)

with a regulated input voltage that of 12v and a uC A/D pin  that can
accept 5v, what resistors do I need for the current resistor and the
voltage dividers in order to capture the top 5 volts (lets say 9v and
above).

If you would rather "teach me to fish" I will accept a formula that is
written in 8th grade terms ; )

The only useful help we can offer is the suggestion to go for a less
ambitious project.

If you want to drop the 12V to 5V to get a regulated 5V supply for
your A/D converter power supply, the TL431 is a poor choice of
regulator. The LM6805 will do a perfectly adequate job and leaves you
less room to screw up.

You can scale down the 0-12V input range to 0V to 5V with a simple
voltage divider. If you were to tell us which A/D converter you were
planning on using (or identify the processor of which it forms a part)
we could suggest specific resistor values.

If you want to expand the range 9V to 12V to fill the 0V to 5V range
of your A/D converter, life gets a little more complicated. If you
told us quite a lot more about what you think you are trying to do we
could certainly come up with something - but granting the expertise
you are exhibiting here, it is less than certain that you would be
able to get it to work.

J

#### Jim

Jan 1, 1970
0
The only useful help we can offer is the suggestion to go for a less
ambitious project.

If you want to drop the 12V to 5V to get a regulated 5V supply for
your A/D converter power supply, the TL431 is a poor choice of
regulator. The LM6805 will do a perfectly adequate job and leaves you
less room to screw up.

You can scale down the 0-12V input range to 0V to 5V with a simple
voltage divider. If you were to tell us which A/D converter you were
planning on using (or identify the processor of which it forms a part)
we could suggest specific resistor values.

If you want to expand the range 9V to 12V to fill the 0V to 5V range
of your A/D converter, life gets a little more complicated. If you
told us quite a lot more about what you think you are trying to do we
could certainly come up with something - but granting the expertise
you are exhibiting here, it is less than certain that you would be
able to get it to work.

Thanks for the reply, what I am doing is making a DVM to read the
12vdc output voltage of a power supply. Due to grounding issues, a DVM
cannot be powered by the supply it is measuring unless it is driven by
a microcontroller type device. My plan is to use an analog in pin on
an arduino board which can accept an input of up to 5 volts and
converts this input into 1023 digital steps (it can source or sink
25ma) Since the power supply output is regulated, I see no need in
monitoring voltage under, lets say, 10v. By using a zener, rather than
simply a voltage divider, this allows me to expand the DVM scale and
get a finer resolution on the voltage that matters.

My idea is to create a simple shunt regulator to get me there. I
realize that there is a minor non-linearity close to the zener
voltage, however, if I set the voltage around 10, by the time the
voltage reaches the range that I really care about, lets say
11.75-13.00 it should be linear again.

Now I suppose I could just go out and get a 10v zener, but thats not
the point. I want to play, I want to do something new and learn...I
mean I could have just bought a power supply, or a dvm. Im sure you
know how it is.

T

#### tomrei

Jan 1, 1970
0
Thanks for the reply, what I am doing is making a DVM to read the
12vdc output voltage of a power supply. Due to grounding issues, a DVM
cannot be powered by the supply it is measuring unless it is driven by
a microcontroller type device. My plan is to use an analog in pin on
an arduino board which can accept an input of up to 5 volts and
converts this input into 1023 digital steps (it can source or sink
25ma)  Since the power supply output is regulated, I see no need in
monitoring voltage under, lets say, 10v. By using a zener, rather than
simply a voltage divider, this allows me to expand the DVM scale and
get a finer resolution on the voltage that matters.

My idea is to create a simple shunt regulator to get me there. I
realize that there is a minor non-linearity close to the zener
voltage, however, if I set the voltage around 10, by the time the
voltage reaches the range that I really care about, lets say
11.75-13.00 it should be linear again.

Now I suppose I could just go out and get a 10v zener, but thats not
the point. I want to play, I want to do something  new and learn...I
mean I could have just bought a power supply, or a dvm. Im sure you
know how it is.

Hi Jim,

i can think of a few ways of doing this.

will sitting on top of 10v and sample 10-15V.

2, use a differential amplifier.
http://en.wikipedia.org/wiki/Differential_amplifier
use a resistor and zener in series from supply to get stable 10v, then
put both supply and zener 10v to the inputs of the differential
amplifier. the differential voltage will be amplified and be the input
to ADC. this is a good way, but trouble is you'll need +/-1515V for
the OP AMP supply.

3, quality resistor and high resolution ADC. personally i think this
i'll suggest a 12bits standard one, it's simple enough part. higher
resolution ones may be hard to setup.
you can use high quality resistors to divide the input voltage by say,
3. the voltage range you are interested are around 11-13V, so the band
is 2V, divide by 3, gives you 0.666V.
for a 12 bit ADC of input 0-5V, that gives you 2^12=4096 samples
5v/4096=1.22mV per sample.
0.666V/1.22mV=545samples.
log545/log2=9 bits

hope it helps.

yours Ren

J

#### James Arthur

Jan 1, 1970
0
Jim said:
Thanks for the reply, what I am doing is making a DVM to read the
12vdc output voltage of a power supply. Due to grounding issues, a DVM
cannot be powered by the supply it is measuring unless it is driven by
a microcontroller type device. My plan is to use an analog in pin on
an arduino board which can accept an input of up to 5 volts and
converts this input into 1023 digital steps (it can source or sink
25ma) Since the power supply output is regulated, I see no need in
monitoring voltage under, lets say, 10v. By using a zener, rather than
simply a voltage divider, this allows me to expand the DVM scale and
get a finer resolution on the voltage that matters.

My idea is to create a simple shunt regulator to get me there. I
realize that there is a minor non-linearity close to the zener
voltage, however, if I set the voltage around 10, by the time the
voltage reaches the range that I really care about, lets say
11.75-13.00 it should be linear again.

Now I suppose I could just go out and get a 10v zener, but thats not
the point. I want to play, I want to do something new and learn...I
mean I could have just bought a power supply, or a dvm. Im sure you
know how it is.

Is this what you're suggesting? ...

(view diagram in Courier font)

Vin >----------O-------.
9-15v | |
| .-.
| | | R1
| | |
| '-'
| |
,---' | R1-R2 set TL431 to desired
TL431 / \<-----O voltage drop (9 volts).
--- |
| |
| .-.
| | | R2
| | |
| '-'
| |
O-------'
|
|
.-.
| | R3 R3-R4 scale remaining 0-6volt signal to
| | 0-5v, AND, provide bias for TL431.
'-'
| ___
O--------------|___|----> Vout 0-5v
|
.-. R5
| | R4 (protects a/d input
| | against faults)
'-'
|
===
GND

That's a handful of 1% resistors--not terribly accurate, but
maybe good enough. Suitability depends on your needs. What are
they? (accuracy, power consumption, calibration...)

Cheers,
James Arthur

J

#### Jim

Jan 1, 1970
0
Is this what you're suggesting? ...

(view diagram in Courier font)

Vin >----------O-------.
9-15v          |       |
|      .-.
|      | |  R1
|      | |
|      '-'
|       |
,---'     |      R1-R2 set TL431 todesired
TL431   / \<-----O      voltage drop (9 volts).
---      |
|       |
|      .-.
|      | | R2
|      | |
|      '-'
|       |
O-------'
|
|
.-.
| | R3  R3-R4 scale remaining 0-6volt signal to
| |     0-5v, AND, provide bias for TL431.
'-'
|               ___
O--------------|___|----> Vout 0-5v
|
.-.              R5
| | R4       (protects a/d input
| |           against faults)
'-'
|
===
GND

That's a handful of 1% resistors--not terribly accurate, but
maybe good enough.  Suitability depends on your needs.  What are
they? (accuracy, power consumption, calibration...)

Cheers,
James Arthur

yes, although I hadnt thought of R3, R4, R6, I was just going to
adjust the zener to around 9 or 10 and then just "assume" it would
never go 5v above that. I see that R3,4,5, supply a little insurance.
So how would I figure the resistor sizes.

Thank you to all of the rest of the solutions above, I was hoping to
avoid op-amps and the like only because my aim was to get the zener
thing dialed in. Also I have been on Usenet since it began, so I didnt
take offense to any perceived slight, I have a thick skin when asking

J

#### Jim

Jan 1, 1970
0
yes, although I hadnt thought of R3, R4, R6, I was just going to
adjust the zener to around 9 or 10 and then just "assume" it would
never go 5v above that. I see that R3,4,5, supply a little insurance.
So how would I figure the resistor sizes.

Thank you to all of the rest of the solutions above, I was hoping to
avoid op-amps and the like only because my aim was to get the zener
thing dialed in. Also I have been on Usenet since it began, so I didnt
take offense to any perceived slight, I have a thick skin when asking

By the way, as far as suitability, I dont care about current draw, its
going on a 30 amp supply, accuracy and calibration? It would be nice
if the display was +/- .01 volt, although I havent played with the a/d
on the arduino before.

B

#### [email protected]

Jan 1, 1970
0
John Fields seems to think that practical advice is nasty - if I give
it. I notice that he didn't bother to contribute his own particular
insight on the problem, so he may be less nasty than I am. but also

If you really want a suppressed zero scale, a zener isn't a
particularly nice way to do it; it's hard to get a tolerance of the
zener voltage that is is less than 2%, and the knee is never very
sharp.

There are are some cheap, close tolerance voltage references around -
the Linear Technology LT1009 2.500V reference offers +/-0.2% for a
couple of bucks, and Farnell is now offering a Texas Instrument second
source part for even less. Zetex has got into the act with an +/-0.5%
SOT-23 version of the LM4040 for little more than one dollar - the
LM4040C50FTA does 5.00V and the LM4040C25FTA does 2.50V.

You'd need a summing amplifier - probably two - and some precision
resistors to get from 0V to 12V in to a 0V to 5V output that covered
the 10V to 12V range on the power supply. The voltage rails available
make quite a difference to the ways in which you might do the job. and
you have to worry about what happens to the output to the A/D
converter input pin when the power supply is outside the range 10V to
12V (or whatever you want to look at).

Like I said in my first response, when we know more about the circuit
we can make more explicit suggestions.

B

#### [email protected]

Jan 1, 1970
0
Ah, the ever helpful and sociable Sloman.

Hasn't himself designed anything useful in decades, but professes
enough expertise to insult people who ask sincere questions.

Speaking of "work"...

John Larkin is not only thin-skinned in his own right, but also
amazingly tender about the sensitivities of unfamiliar posters. Like
John Fields, this sudden (and somewhat unexpected) burst of
consideration for the OP's fragile feelings didn't stretch far enough
to generate any advice on solving the technical problem, but John is a
busy man, desperately trying to keep up his cash flow while the
economy slides out from under him.

B

#### [email protected]

Jan 1, 1970
0
If I understand you correctly, you simply want to map...

+9V to +14V into 0V to +5V

Is that correct?

Simplest solution:

OpAmp whose output range includes negative rail (or you have a
negative supply).

OpAmp powered +12V/0V

Plus input of OpAmp to +7V (divider from +12V supply?)

R from signal to minus input of OpAmp

R (same value) from minus input to OpAmp output

This maps +9V to +14V "signal" into +5V to 0V OpAmp output

In other words a phase reversal, but I'm sure your uC can cope ;-)

Using the +12V supply (that you are measuring) as the the source of
the +7V reference (via a voltage divider) used to be the kind of "drop-
off" that excited a certain measure of hilarity in the places where I
used to work (all those years ago).

Since Jim is even older than I am, it's less funny than it used to be.
Maybe it is as well that John Fileds and John Larkin didn't run the
risk of exposing their own decaying neurones.

E

#### ehsjr

Jan 1, 1970
0
Jim said:
yes, although I hadnt thought of R3, R4, R6, I was just going to
adjust the zener to around 9 or 10 and then just "assume" it would
never go 5v above that. I see that R3,4,5, supply a little insurance.
So how would I figure the resistor sizes.

V = Vref * (1 + R1/R2) Vref is a characteristic of the
431, and is about 2.5 volts. So, if you make R1 2.6K and
R2 1K, you get ~9V.

Alternatively, you can use a 5K 10 turn pot in place
of R1 and R2 and adjust it to get your 9 volts.

Ed

J

#### James Arthur

Jan 1, 1970
0
Pick R1-R2 for, say, a 9v drop, per the data sheet formula.

(R1 + R2)
V = Vref * ---------
R2

Then you'd make the impedance of that R1-R2 low enough to
drive the TL431 input's bias current without significant
error. I(bias) = 4uA. That gives you R1 and R2.

Next you'd pick a series combination of R3-R4 that provides the
TL431 with at least 1mA of idle current wherever you want it
to work accurately. The LMV431 is better--it'll do this job on
80uA and has just 0.5uA input current (Vref = 1.24v for that
part though--don't forget this when calculating R1-R2).

R5 isn't critical.

Note that this circuit craps out at low Vin since, ultimately,
R3-R4 won't provide the TL431 with enough bias current for the
IC to regulate accurately.

You could improve that by dropping less voltage across the
TL431, leaving some drop across R3-R4, and removing the
offset voltage in software, and/or by using the LMV431 instead
of the TL431.

We'd usually just use an op-amp difference amplifier for
this job:

R1 R2
Vin >----\/\/\/----O----\/\/\/---.
| |
| ===
| |\ GND
'---|+\
| >---------O--\/\/\/--> Vout
.---|-/ | R5
| |/ | (a/d protection)
| |
+9v >---/\/\/\----O----/\/\/\------'
R3 R4

R1=R2=R3=R4, precision resistors.

By the way, as far as suitability, I dont care about current draw, its
going on a 30 amp supply, accuracy and calibration? It would be nice
if the display was +/- .01 volt, although I havent played with the a/d
on the arduino before.

If you want a floating DVM supply for one of those little LCD DVM
modules, just make a little floating supply, like the cap-coupled
charge-pump thingie at the bottom of this web page:

http://www.techlib.com/electronics/regulators.html

Best,
James Arthur

J

#### Jim

Jan 1, 1970
0
V = Vref * (1 + R1/R2)   Vref is a characteristic of the
431, and is about 2.5 volts.  So, if you make R1 2.6K and
R2 1K, you get ~9V.

Alternatively, you can use a 5K 10 turn pot in place
of R1 and R2 and adjust it to get your 9 volts.

Ed

Thank you very much

J

#### Jim

Jan 1, 1970
0
Pick R1-R2 for, say, a 9v drop, per the data sheet formula.

(R1 + R2)
V = Vref * ---------
R2

Then you'd make the impedance of that R1-R2 low enough to
drive the TL431 input's bias current without significant
error.  I(bias) = 4uA.  That gives you R1 and R2.

Next you'd pick a series combination of R3-R4 that provides the
TL431 with at least 1mA of idle current wherever you want it
to work accurately.  The LMV431 is better--it'll do this job on
80uA and has just 0.5uA input current (Vref = 1.24v for that
part though--don't forget this when calculating R1-R2).

R5 isn't critical.

Note that this circuit craps out at low Vin since, ultimately,
R3-R4 won't provide the TL431 with enough bias current for the
IC to regulate accurately.

You could improve that by dropping less voltage across the
TL431, leaving some drop across R3-R4, and removing the
offset voltage in software, and/or by using the LMV431 instead
of the TL431.

We'd usually just use an op-amp difference amplifier for
this job:

R1             R2
Vin >----\/\/\/----O----\/\/\/---.
|             |
|            ===
|   |\       GND
'---|+\
|  >---------O--\/\/\/--> Vout
.---|-/          |    R5
|   |/           | (a/d protection)
|                |
+9v >---/\/\/\----O----/\/\/\------'
R3             R4

R1=R2=R3=R4, precision resistors.

If you want a floating DVM supply for one of those little LCD DVM
modules, just make a little floating supply, like the cap-coupled
charge-pump thingie at the bottom of this web page:

http://www.techlib.com/electronics/regulators.html

Best,
James Arthur

thank you very much

J

#### Jim

Jan 1, 1970
0
Thank you all for the advice and entertainment. This place as a
certain neighborhood bar feeling about it.

K

#### krw

Jan 1, 1970
0
As in, 'not capabale of'.

Right. It's not "work" Slowman's a piece of.

J

#### Jim

Jan 1, 1970
0
well, thank you all again...last call

B

#### [email protected]

Jan 1, 1970
0
He did post a simulation of a rather weird version of a Baxandall
resonant oscillator. Barbaric newbie biasing. And he didn't understand
why it didn't squegg in simulation.

But then again. neither does John, on the evidence available.
Better than nothing, I suppose.

Since it's worth nothing to anybody else, there's not a lot of
motivation.

I've spent some time (and a little money) trying to find Peter
Baxandall's original paper, on "Class-D oscillators",

Baxandall, P.J, Proc I.E.E 106, B, 748 (1959)

without any success. My little brother (who has convenient accesss to
university libraries that ought to have copies of the journal) reports
" it wasn't in their complete 1959 Part B collection. From some
checking of citations, I've worked out it is in a supplement of
presumably the May 1959 Part B issue."

The late Tony Williams, who was the source of the reference, isn't

B

#### [email protected]

Jan 1, 1970
0
John Fields seems to think that practical advice is nasty - if I give
it.

---
Not at all, when you so infrequently do.

In this case, however, the "practical advice" you offered was to try to
get the OP to shut down by demeaning him.
---
I notice that he didn't bother to contribute his own particular
insight on the problem, so he may be less nasty than I am. but also

---
My intent, at the time, wasn't meant to clue in the OP, but rather to
show you up for the hateful spirit you are.
---

If you really want a suppressed zero scale, a zener isn't a
particularly nice way to do it; it's hard to get a tolerance of the
zener voltage that is is less than 2%, and the knee is never very
sharp.
There are are some cheap, close tolerance voltage references around -
the Linear Technology  LT1009 2.500V reference offers +/-0.2% for a
couple of bucks, and Farnell is now offering a Texas Instrument second
source part for even less. Zetex has got into the act with an +/-0.5%
SOT-23 version of the LM4040 for little more than one dollar - the
LM4040C50FTA does 5.00V and the LM4040C25FTA does 2.50V.
You'd need a summing amplifier - probably two - and some precision
resistors to get from 0V to 12V in to a 0V to 5V output that covered
the 10V to 12V range on the power supply. The voltage rails available
make quite a difference to the ways in which you might do the job. and
you have to worry about what happens to the output to the A/D
converter input pin when the power supply is outside the range 10V to
12V (or whatever you want to look at).
Like I said in my first response, when we know more about the circuit
we can make more explicit suggestions.

---
We?

mind?

Version 4
SHEET 1 880 680
WIRE 192 16 -288 16
WIRE 368 16 192 16
WIRE 192 64 192 16
WIRE 368 144 368 16
WIRE 192 160 192 144
WIRE 336 160 192 160
WIRE 544 176 400 176
WIRE -288 192 -288 16
WIRE -128 192 -288 192
WIRE 336 192 64 192
WIRE -288 256 -288 192
WIRE 192 272 192 160
WIRE 256 272 192 272
WIRE 400 272 336 272
WIRE 544 272 544 176
WIRE 544 272 480 272
WIRE -128 288 -176 288
WIRE 192 384 192 272
WIRE 416 384 192 384
WIRE 544 384 544 272
WIRE 544 384 480 384
WIRE 192 448 192 384
WIRE -288 560 -288 336
WIRE -176 560 -176 288
WIRE -176 560 -288 560
WIRE 192 560 192 528
WIRE 192 560 -176 560
WIRE 368 560 368 208
WIRE 368 560 192 560
WIRE -288 592 -288 560
FLAG -288 592 0
SYMBOL References\\LT1021-5 -32 240 R0
SYMATTR InstName U1
SYMBOL Opamps\\LT1006A 368 112 R0
SYMATTR InstName U2
SYMBOL voltage -288 240 R0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V1
SYMATTR Value 12
SYMBOL res 176 48 R0
SYMATTR InstName R1
SYMATTR Value 10k
SYMATTR SpiceLine tol=0,1
SYMBOL res 176 432 R0
SYMATTR InstName R2
SYMATTR Value 10k
SYMATTR SpiceLine tol=0.1
SYMBOL res 496 256 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R3
SYMATTR Value 124k
SYMATTR SpiceLine tol=0.1%
SYMBOL res 352 256 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R4
SYMATTR Value 1k
SYMATTR SpiceLine tol=1
SYMBOL cap 480 368 R90
WINDOW 0 0 32 VBottom 0
WINDOW 3 32 32 VTop 0
SYMATTR InstName C1
SYMATTR Value 0.1µ

The LT1021-5 is a ridiculous over-kill for the application, but it is
in the LTSpice library while the cheaper LM4040C50FTA I mentioned
earlier isn't.

I've also left out the 100nF bypass capacitor that even the LT1006
single supply op amp should have.

Almost any other low off-set single supply op amp that could survive
the peak supply voltage would do as well as the (elderly) LT1006.

I've followed Jim Thompson in figuring that the OP wants to digitise
the supply voltage range from 10V (which would give 5V at the output
of the LT1006 in this circuit) to 14V (which would give 0V at the
output of the LT1006, if the LT1006 could pull right down to the 0V
rail, which it can't, since it can only pull 40uA down to about 50mV).

The 0.1% resistors are E96 values and available ex-stock from Farnell,
some of them as single parts (though for many you have to buy five at
once, typically for more than a buck apiece).

B

#### [email protected]

Jan 1, 1970
0
Of course I know why it doesn't squegg. Squegging requires a
squegging-frequency modulator/demodulator loop with the usual
oscillation criteria, namely gain >1 at 0 degrees phase shift around
the loop. The classic tube and transistor oscillators often had such

Except that the classic circuit does squegg in real life, if you make
the inductor too big. The first one I ever built back on 1968 squegged
continuously, until I took off a lot of the turns that that I'd
tediously handwound onto the inductor.

Control theory. You're self-declared to be an expert here.

Don't remember ever making that claim. I can tune a DPI control loop,
but state-variable control theory and self-tuning control loops I've
so far been able to palm off onto people who do think they have
mastered control theory.

You seem to share their enthusiam for applying their expertise to
situations that they didn't really comprehend.

J
Replies
4
Views
2K
Palindrome
P
B
Replies
8
Views
2K
John Woodgate
J
J
Replies
4
Views
2K
Jasen Betts
J
Replies
5
Views
730
Replies
1
Views
321