volume meter and bar graph

E

electricked

Guest
Hi all,

I'm planning to make a 10 LED voltage/volume meter for my car. I'm planning
to tap the car speakers to acquire the voltage needed for this device. I
haven't measured the max voltage at the speaker but if it's not enough to
light up 10 LEDs. What's best way to raise voltage level in audio signal
(audio transformer or would a transistor work for this?)

So basically, I'm planning to rectify the current with a bridge diode, then
smoothen the output voltage with a cap, and then if needed raise the
voltage, then run it to the LEDs rail.

Here's what it would look like:

For 10 LEDs, I need 10V

Audio AC Audio DC
In o-------- |-----------------|
| | | CAP ---
rectifier|---|-- --- | | LED bar
bridge |---|-- --- | |
| | | ---
Out o-------- |-----------------|

I'm planning on making my own LED bar and it will be of the voltage divider
type. e.g.

+V Gnd
| |
| |
- |
| | R1 |
- |
| LED1 |
|--|>|---|
| |
- |
| | R2 |
- |
| LED2 |
|--|>|---|
| |
- |
| | R3 |
- |
| LED3 |
|--|>|---|
| |
- |
| | R4 |
- |
| |
| |
----------

and so on up to 10 LEDs

What do you think? Would I need some kind of regulator after the cap just to
make sure the voltage doesn't jump above the max? Anything else that I
should be concerned with this type of circuit? Would it work?

Thanks!

--Viktor
 
You might want to take a look at

http://electronickits.com/kit/complete/ligh/ck1005.htm

Click on the CK1005.pdf and there's a circuit diagram for the meter that
you're looking to create.

I'm trying to do something similar myself, but instead measuring the voltage
of a speaker I'm trying to measure the setting of a 1K pot to show the
current power setting of a pulse width modulator circuit that I've put
together. This doc might help me to alter the circuit to suit my needs :)

Regards

Colin.


"electricked" <no_emails_please> wrote in message
news:LoadnbHjbplo3v7d4p2dnA@comcast.com...
Hi all,

I'm planning to make a 10 LED voltage/volume meter for my car. I'm
planning
to tap the car speakers to acquire the voltage needed for this device. I
haven't measured the max voltage at the speaker but if it's not enough to
light up 10 LEDs. What's best way to raise voltage level in audio signal
(audio transformer or would a transistor work for this?)

So basically, I'm planning to rectify the current with a bridge diode,
then
smoothen the output voltage with a cap, and then if needed raise the
voltage, then run it to the LEDs rail.

Here's what it would look like:

For 10 LEDs, I need 10V

Audio AC Audio DC
In o-------- |-----------------|
| | | CAP ---
rectifier|---|-- --- | | LED bar
bridge |---|-- --- | |
| | | ---
Out o-------- |-----------------|

I'm planning on making my own LED bar and it will be of the voltage
divider
type. e.g.

+V Gnd
| |
| |
- |
| | R1 |
- |
| LED1 |
|--|>|---|
| |
- |
| | R2 |
- |
| LED2 |
|--|>|---|
| |
- |
| | R3 |
- |
| LED3 |
|--|>|---|
| |
- |
| | R4 |
- |
| |
| |
----------

and so on up to 10 LEDs

What do you think? Would I need some kind of regulator after the cap just
to
make sure the voltage doesn't jump above the max? Anything else that I
should be concerned with this type of circuit? Would it work?

Thanks!

--Viktor
 
"electricked" <no_emails_please> wrote in message
news:LoadnbHjbplo3v7d4p2dnA@comcast.com...
Hi all,

I'm planning to make a 10 LED voltage/volume meter for my car. I'm
planning
to tap the car speakers to acquire the voltage needed for this device. I
haven't measured the max voltage at the speaker but if it's not enough to
light up 10 LEDs. What's best way to raise voltage level in audio signal
(audio transformer or would a transistor work for this?)

So basically, I'm planning to rectify the current with a bridge diode,
then
smoothen the output voltage with a cap, and then if needed raise the
voltage, then run it to the LEDs rail.

Here's what it would look like:

For 10 LEDs, I need 10V

Audio AC Audio DC
In o-------- |-----------------|
| | | CAP ---
rectifier|---|-- --- | | LED bar
bridge |---|-- --- | |
| | | ---
Out o-------- |-----------------|

Sorry, Viktor, but this will not work. Automotive speakers have a low
impedance, and the voltage used generally does not rise very high. It will
hardly drive more than 1 LED reliably. Even for 1 LED, you would have the
volume set quite high. Test it. Connect a LED (with 10 Ohm res) to one of
the speakers and slowly raise the volume until it lights. Be careful however
not to overload the LED. Adding a bridge would take yet another volt before
there is even a small signal at CAP, so this likely would not do. The usual
controllers for such devices are ICs that check the AC signal against some
limits and light the LEDs if it exceeds. These controllers are not
expensive, but, since it looks like you want to build the circuit from
scratch to better understand it, see below.

I'm planning on making my own LED bar and it will be of the voltage
divider
type. e.g.

+V Gnd
| |
| |
- |
| | R1 |
- |
| LED1 |
|--|>|---|
| |
- |
| | R2 |
- |
| LED2 |
|--|>|---|
| <snip> |

and so on up to 10 LEDs

This will hardly work except for the first 2 LEDs. If you get the third one
to light, you'll fry the first one. The reason for this is the low
differential resistance of a LED when it lights. LED1 will always have an
almost constant voltage while the small difference will not be enough to
cause a sufficient current through R2 and allow LED2 to light up. If you add
a res. in series with each LED, it will help things a little, but more than
5 or 6 LEDs will still be difficult. Besides, +V would likely be more than a
car stereo can provide on highest volume even if you feed its final stage
with a square wave. There is another way to do this:

For a more workable solution, you would need an opamp and as many
comparators as you want to use LEDs. Let's assume 3 LEDs for the start, you
can always add more later.

Set up the opamp to work as a buffer amp. Since no -12V are available, set
up the reference point at half the supply voltage (voltage divider with
zener). The input signal should be applied to a potentiometer for adjusting
overall sensitivity and coupled through a capacitor (because of the 6V
offset). Set the voltage gain of the amp stage to something around 5, the
sensitivity pot will be used later for fine-tuning. Make sure the input is
protected from overload using a resistor. Then connect the opamp's output
(which is also 6V-biased) to the comparators' positive inputs (all). The
negative ones get connected to a multi-part voltage divider (just a bunch of
resistors in series) that defines the thresholds. The outputs of the
comparators can be used to (via diodes) charge up small caps and drive
transistors (or LEDs directly if the comparators you use can supply that
power).

Since the schematic is rather lengthy for ASCII, I'll only draw parts of it.

.---o---||---> to opamp in
input: | | C1
<---------\VR1 | /R2
in / R1 | \
\<---/\/\--' /
/ |
<---------\--------------o--------> gnd
gnd


opamp:
<-------o------------------.
vcc | |
12V '-/\/\/-.R3 |
o---------------------> cmp3
.-/\/\/-'R4 |
o-----------------------------> cmp2
'-/\/\/-.R5 |
o---------------------> cmp1
.-/\/\/-'R6 |
| |
| R7 |\amp
| ,--------------| \
<-------|--o--\/\/\-. +| \---o----> out
in | ,--------' -| / \
o--o--\/\/\---o---| / / R9
| R8 | |/ \
---, | | /
/ \|zener '----------'
--- 6V |
| |
<-------o------------------o----------> gnd
gnd

To interface the comparators to LEDs, 1 transistor for each should be
enough. Use base current limiting resistors (and put in caps to prevent
flickering).

You can also use opamps instead of comparators throughout this circuit.
Since they are often two in a package, the above can be implemented with two
ICs (plus one for each two extra LEDs used).

Dimitrij
 
"electricked" <no_emails_please> wrote in message
news:LoadnbHjbplo3v7d4p2dnA@comcast.com...
Hi all,

I'm planning to make a 10 LED voltage/volume meter for my car. I'm
planning
to tap the car speakers to acquire the voltage needed for this device. I
haven't measured the max voltage at the speaker but if it's not enough to
light up 10 LEDs. What's best way to raise voltage level in audio signal
(audio transformer or would a transistor work for this?)

So basically, I'm planning to rectify the current with a bridge diode,
then
smoothen the output voltage with a cap, and then if needed raise the
voltage, then run it to the LEDs rail.

Here's what it would look like:

For 10 LEDs, I need 10V
I meant to say 15-20V.

Audio AC Audio DC
In o-------- |-----------------|
| | | CAP ---
rectifier|---|-- --- | | LED bar
bridge |---|-- --- | |
| | | ---
Out o-------- |-----------------|

I'm planning on making my own LED bar and it will be of the voltage
divider
type. e.g.

+V Gnd
| |
| |
- |
| | R1 |
- |
| LED1 |
|--|>|---|
| |
- |
| | R2 |
- |
| LED2 |
|--|>|---|
| |
- |
| | R3 |
- |
| LED3 |
|--|>|---|
| |
- |
| | R4 |
- |
| |
| |
----------

and so on up to 10 LEDs

What do you think? Would I need some kind of regulator after the cap just
to
make sure the voltage doesn't jump above the max? Anything else that I
should be concerned with this type of circuit? Would it work?

Thanks!

--Viktor
 
"Colin Dawson" <news@cjdawson.com> wrote in message
news:c3vhlt$5sk$1@hercules.btinternet.com...
You might want to take a look at

http://electronickits.com/kit/complete/ligh/ck1005.htm

Click on the CK1005.pdf and there's a circuit diagram for the meter that
you're looking to create.

I'm trying to do something similar myself, but instead measuring the
voltage
of a speaker I'm trying to measure the setting of a 1K pot to show the
current power setting of a pulse width modulator circuit that I've put
together. This doc might help me to alter the circuit to suit my needs
:)

Regards

Colin.
Yeah, that's pretty much what I want to do. My VU meter will detect mono
signals only, which is not exactly accurate since I'll run it from a single
speaker (either left or right) but it's close enough. I might bridge it, so
this way I might not need to boost the voltage.

In any case, I don't want to use LM3915 since I'm doing this to learn
electronics and I'd like to design and build my own circuit for driving the
LEDs. Plus, a nicely designed VU meter looks great while the audio is
playing :)

I can use comparators for each LED but I'm looking for a simpler way at
first, even if the circuit is not as stable as a comparator. I think the
LM3915 uses a bunch of comparators, right?

In any case, good luck with your project, and I'd like to see the final
product.

--Viktor
 
On Thu, 25 Mar 2004 14:42:04 -0800, "electricked" <no_emails_please>
wrote:

I'm planning to make a 10 LED voltage/volume meter for my car.
see my Led Peak meter on my site under electronics ... :)

-- Regards, SPAJKY ÂŽ
& visit my site @ http://www.spajky.vze.com
"Tualatin OC-ed / BX-Slot1 / inaudible setup!"
E-mail AntiSpam: remove ##
 
"electricked" <no_emails_please> wrote in message
news:Kcmdnf8ne8B64f7d4p2dnA@comcast.com...

snip
You are right, the ones I have are 4ohms so the voltage required to drive
them is minimal. Can't the voltage be boosted somehow without causing much
interference with the working of the speakers?
Well, it can, but everything has limits. The main problem is the 12V supply.
It's the only voltage available and it's available with only one polarity.
This means the maximal AC voltage you can have with a rail-to-rail opamp
will be 6/sqrt(2) ~= 4.2VRMS, biased by an offset of +6VDC. With 2 amps in
reverse-phase, you'll get 2 times this voltage (between amp outputs), but
that's all, no further increase possible unless you put a transformer in.

+V Gnd
| |
| |
- |
| | R1 |
- |
| LED1 |
|--|>|---|
| |
- |
| | R2 |
- |
| LED2 |
|--|>|---|
| <snip> |

This will hardly work except for the first 2 LEDs. If you get the third
one
to light, you'll fry the first one. The reason for this is the low
differential resistance of a LED when it lights. LED1 will always have
an
almost constant voltage while the small difference will not be enough to
cause a sufficient current through R2 and allow LED2 to light up. If you
add
a res. in series with each LED, it will help things a little, but more
than
5 or 6 LEDs will still be difficult. Besides, +V would likely be more
than
a
car stereo can provide on highest volume even if you feed its final
stage
with a square wave. There is another way to do this:

Can you explain this a little bit? Say you have 15V to start with, and
each
resistor drops 1/10 of the voltage. For 10 LEDs, that's 1.5V a piece, just
enough to light up each LED. Why wouldn't work? Also, what happens if the
voltage is about 3V? Would there be no current passing through the bottom
8
LEDs?
A LED is basically a diode, and it behaves like one. Only a diode has a
forward drop of about 0.7V, while an LED has about 2V. It will not rise
significantly no matter what current you force through. This means the first
LED will "short out" the rest and there will not be enough voltage for the
others, that is if it does not have a series resistor to allow for voltage
increase. If you increase the total V, the first LED will get more and more
current, eventually it will burn out, but its voltage drop will still hardly
exceed 2V, so the voltage at R2 and the rest will still be 2V and not rise
with the input. It may (or may not) be enough to light the second LED, but
it will surely not be enough for all others. Imagine boiling water in some
container when there is another smaller container, also with water, inside
it. If you increase the power, more water from the outer container will boil
out, but not from the inner container. There the water will not boil at all.
This is because the temperature in the outer containder will not rise. It's
almost the same with an LED's voltage - it will not rise enough. To allow
for this rising, the LEDs' resistances must be adjusted using series
resistors. Besides, you don't have that 15V to start with (see above) and
the 4 (maximal 8) volts you have are not enough for this.

Set up the opamp to work as a buffer amp. Since no -12V are available,
set
up the reference point at half the supply voltage.

If I have 12V, how do I get an output of -12V and still have ground that's
0V? I've been wondering about this and I'm thinking it can be done using a
diode to redirect the current in the opposite direction from the current
produced by the 12V.
You don't get -12V at all. See above for maximal AC RMS output.

If there is still a desperate need for a -12V supply, it can be created from
the +12V one using some capacitors, a lot of transistors and some sort of HF
oscillator circuit to switch them. To increase efficiency, using additional
inductors may help, but the circuit won't be very simple. :) For such a
simple circuit as an LED gauge driver, one is better off not to start with
switching power supplies in the first place.

Since the schematic is rather lengthy for ASCII, I'll only draw parts of
it.

.---o---||---> to opamp in
input: | | C1
---------\VR1 | /R2
in / R1 | \
\<---/\/\--' /
/ |
---------\--------------o--------> gnd
gnd


opamp:
-------o------------------.
vcc | |
12V '-/\/\/-.R3 |
o---------------------> cmp3
.-/\/\/-'R4 |
o-----------------------------> cmp2
'-/\/\/-.R5 |
o---------------------> cmp1
.-/\/\/-'R6 |
| |
| R7 |\amp
| ,--------------| \
-------|--o--\/\/\-. +| \---o----> out
in | ,--------' -| / \
o--o--\/\/\---o---| / / R9
| R8 | |/ \
---, | | /
/ \|zener '----------'
--- 6V |
| |
-------o------------------o----------> gnd
gnd

To interface the comparators to LEDs, 1 transistor for each should be
enough. Use base current limiting resistors (and put in caps to prevent
flickering).

This is a little more complex than I expected. I mean I understand what's
going on, but it's more complex than I want it to be. I might get the
LM3915
IC and work with that. I was hoping that I would get it to work with a
variation of the circuit I provided above, even if I have to boost the
voltage to about 1.5V, if that's possible (which I think it is with the
help
of a few caps or transformer?)
You don't need to boots the V to 1.5. The max output on full volume will be
more, so you will probably even have to attenuate it. Also caps alone are
useless when it comes to amplifying a voltage.

aren't comparators basically an opamp connected in a specific way?
Not really. A comparator has a higher gain and is meant to switch the output
voltage between 0 and max. It is not meant to work in a linear way and may
start to oscillate if used that way. An opamp has a lower maximal gain, is
more stable and meant to amplify signals with little distortion.

Dimitrij
 
"Spajky" <Spajky##@volja.net> wrote in message
news:ijv76095ceuom1c9to3f1lttda3hcl6crj@4ax.com...
On Thu, 25 Mar 2004 14:42:04 -0800, "electricked" <no_emails_please
wrote:

I'm planning to make a 10 LED voltage/volume meter for my car.

see my Led Peak meter on my site under electronics ... :)
I'll have a look, Thanks Spajky!

--Viktor

-- Regards, SPAJKY Ž
& visit my site @ http://www.spajky.vze.com
"Tualatin OC-ed / BX-Slot1 / inaudible setup!"
E-mail AntiSpam: remove ##
 
Just an ordinary transistor will make a fair to middlin' quasi-
comparator - when vbe exceeds .7v, it conducts, roughly. Just
from a quick glance, I don't think your R/LED ladder will do
what you want, but it might be instructive to slap one together
on the bench with a variable supply and see what it does.

Good Luck!
Rich

"electricked" <no_emails_please> wrote in message
news:NOGdneuovPXHwP7dRVn-sw@comcast.com...
"Colin Dawson" <news@cjdawson.com> wrote in message
news:c3vhlt$5sk$1@hercules.btinternet.com...
You might want to take a look at

http://electronickits.com/kit/complete/ligh/ck1005.htm

Click on the CK1005.pdf and there's a circuit diagram for the meter that
you're looking to create.

I'm trying to do something similar myself, but instead measuring the
voltage
of a speaker I'm trying to measure the setting of a 1K pot to show the
current power setting of a pulse width modulator circuit that I've put
together. This doc might help me to alter the circuit to suit my needs
:)

Regards

Colin.

Yeah, that's pretty much what I want to do. My VU meter will detect mono
signals only, which is not exactly accurate since I'll run it from a
single
speaker (either left or right) but it's close enough. I might bridge it,
so
this way I might not need to boost the voltage.

In any case, I don't want to use LM3915 since I'm doing this to learn
electronics and I'd like to design and build my own circuit for driving
the
LEDs. Plus, a nicely designed VU meter looks great while the audio is
playing :)

I can use comparators for each LED but I'm looking for a simpler way at
first, even if the circuit is not as stable as a comparator. I think the
LM3915 uses a bunch of comparators, right?

In any case, good luck with your project, and I'd like to see the final
product.

--Viktor
 

Welcome to EDABoard.com

Sponsor

Back
Top