1.5v src, LED needs 3v, now what?

[SklettTheNewb]

I had recently read and experimented w/ darlington pairs and thought I
could use them as a voltage increaser, but looking back, I realize they
only allow for a very small base current to open up the C/E.

My problem is that I have a 1.5V source and a circuit that needs 3v to
operate(I think) I have implemented a TI LM393P comparator and it
appears to need a MIN up 2v to operate (per the TI DataSheet) and my
LED would like to have 3v.

What is the standard solution to something like this? I'm a
newbie(duh) and I'm having a hard time finding the answer to this.

I hope someone will shed some light on this for me.

oh, adding another battery ins't an option and changing from a AAA alk.
isn't an option either

Thank you,
Steve

 

[Chris]

SklettTheNewb wrote:

I had recently read and experimented w/ darlington pairs and thought
I
could use them as a voltage increaser, but looking back, I realize
they
only allow for a very small base current to open up the C/E.

My problem is that I have a 1.5V source and a circuit that needs 3v
to
operate(I think) I have implemented a TI LM393P comparator and it
appears to need a MIN up 2v to operate (per the TI DataSheet) and my
LED would like to have 3v.

What is the standard solution to something like this? I'm a
newbie(duh) and I'm having a hard time finding the answer to this.

I hope someone will shed some light on this for me.

oh, adding another battery ins't an option and changing from a AAA
alk.
isn't an option either

Thank you,
Steve

Hi, Steve. You're might want to look at a step-up DC-to-DC converter
IC to solve your problem. I once used an LT1073 successfully for a one
AAA alkaline cell handheld tester for the factory floor, and would
personally recommend it, despite it's being a little pricey. One
inductor, one diode, two resistors for a voltage divider, and a small
output cap are all you need to finish the circuit. They're very
efficient, use little quiescent power, and efficiently give you the
20-40mA you need. As a big plus for newbies/hobbyists, the IC comes in
a standard 8-pin DIP package. The data sheet even tells you exactly
which inductor to buy -- can't be easier.

A couple of caveats. First, be sure to use the inductor specified.
Don't substitute. Second, get a GOOD low ESR aluminum electrolytic cap
designed for switchers. Third, if you're doing this on perfboard, be
very careful with layout. For a hobbyist, it might be best to use a
small piece of unetched FR-4 as a ground plane, and use "dead bug"
construction for minimum ground loop length.

Another option might be one of the newer single cell inductorless
charge pump ICs, like the TI TPS60301. I haven't used it myself, and
it only comes in 10-pin VSSOP package, which might be a bit of a
problem.

From a newbie/hobbyist standpoint, though, it might be be better to
reconsider your battery situation. Another AAA battery might be the

cheapest and easiest way to go. Many times, people say something is
"not an option", but when the alternatives are realistically presented,
"not an option" suddenly becomes negotiable.

Good luck
Chris

 

[SklettTheNewb]

"Many times, people say something is
"not an option", but when the alternatives are realistically presented,
"not an option" suddenly becomes negotiable. "

How true that is

I appreciate your detailed suggestions and I have to say that I'm a
little surprised that it is that involved, I was expecting some clever
2 component magical solution that would make everything OK.

BUT, I will look at the chip you mentioned, see if I can scam some
samples from somewhere and have a go with it.

I also found in some research last night that there are lower voltage
comparitors than the TI, I found one that stated it had a MIN Input of
1.0v (if using dual inputs, which I don't understand yet)
http://www.st.com/stonline/books/pdf/docs/2468.pdf

The description says:
"WIDE SINGLE SUPPLY VOLTAGE RANGE OR DUAL SUPPLIES FOR ALL DEVICES :
+2V TO +36V OR ą1V TO ą18V"

So I'm not sure what they mean by "Dual supplies" - if this means that
I can use this comparator w/ a single AAA cell, then I am a happy
person (assuming it's not too expensive)

Anyway, just thinking out loud now, thanks again for the post!

-Steve

 

[Chris]

SklettTheNewb wrote:

"Many times, people say something is
"not an option", but when the alternatives are realistically
presented,
"not an option" suddenly becomes negotiable. "

How true that is

I appreciate your detailed suggestions and I have to say that I'm a
little surprised that it is that involved, I was expecting some
clever
2 component magical solution that would make everything OK.

BUT, I will look at the chip you mentioned, see if I can scam some
samples from somewhere and have a go with it.

I also found in some research last night that there are lower voltage
comparitors than the TI, I found one that stated it had a MIN Input
of
1.0v (if using dual inputs, which I don't understand yet)
http://www.st.com/stonline/books/pdf/docs/2468.pdf

The description says:
"WIDE SINGLE SUPPLY VOLTAGE RANGE OR DUAL SUPPLIES FOR ALL DEVICES :
+2V TO +36V OR ą1V TO ą18V"

So I'm not sure what they mean by "Dual supplies" - if this means
that
I can use this comparator w/ a single AAA cell, then I am a happy
person (assuming it's not too expensive)

Anyway, just thinking out loud now, thanks again for the post!

-Steve

Hi, Steve. Even if you use an ultra-low voltage comparator, you'll
still have to deal with the LED, so it looks like you'll need 3V
anyway.

A "dual supply" means having a plus supply and a minus supply, with
common/GND inbetween. For instance, you could use your 2 AAA batteries
as a +/-1.5V supply, by connecting the common to the junction of the
two batteries. The + of one battery would be +1.5V from common, and
the - of the other would be -1.5V from common. In short, this isn't
going to help you.

If you're space constrained, and you're not too concerned about expense
or battery power, there are a number of lithium cells that will give
you your 3V, and take up a lot less space than a AAA battery.

Since you have CircuitMaker, and are able to post .gifs, why don't you
just post the circuit you're trying to build? Or provide a description
of what you're trying to accomplish?

Good luck
Chris

 

[SklettTheNewb]

Hi Chris,

Thanks for the explanation of dual supply, makes sense now.

My power source is turning out to be a troublesome factor. I like the
idea of a smaller source(lithium) but I don't know what that would do
to my cost. I have also thought of using 2 hearing aid batteries to
get 3v, but I don't know if they will have enough mAh.

I like your suggestion that I post my circuit, here it is:
http://www.pmddirect.com/temp/comparator_c.gif

(BTW: If anyone has checked out www.pmddirect.com, that is my Dad's
company, I'm not an engineer for them, thank god!)

So I will explain this circuit, it's a basic sensing circuit that will
turn the LED on when the resistance at the sensor get's high enough. I
intend to have a flashing circuit that will cause the LED to flash w/
1500 ms off for each "flash" (to save power and get attention)

In no way do I think this is a correct circuit, it *works* for me, but
I'm sure it could be better/simpler and I would really value any
suggestions you may have in that regard as well.

In summary, I need a sensing circuit w/ a flashing LED(bright) in as
small a space as possible utilizing as little power as possible and I
would like it to be able to flash at the above mentioned rate for 45
days(if that is possible)

I am REALLY looking forward to any responses

Thanks!!
Steve

 

[Chris]

SklettTheNewb wrote:

Hi Chris,

Thanks for the explanation of dual supply, makes sense now.

My power source is turning out to be a troublesome factor. I like
the
idea of a smaller source(lithium) but I don't know what that would do
to my cost. I have also thought of using 2 hearing aid batteries to
get 3v, but I don't know if they will have enough mAh.

I like your suggestion that I post my circuit, here it is:
http://www.pmddirect.com/temp/comparator_c.gif

(BTW: If anyone has checked out www.pmddirect.com, that is my Dad's
company, I'm not an engineer for them, thank god!)

So I will explain this circuit, it's a basic sensing circuit that
will
turn the LED on when the resistance at the sensor get's high enough.
I
intend to have a flashing circuit that will cause the LED to flash w/
1500 ms off for each "flash" (to save power and get attention)

In no way do I think this is a correct circuit, it *works* for me,
but
I'm sure it could be better/simpler and I would really value any
suggestions you may have in that regard as well.

In summary, I need a sensing circuit w/ a flashing LED(bright) in as
small a space as possible utilizing as little power as possible and I
would like it to be able to flash at the above mentioned rate for 45
days(if that is possible)

I am REALLY looking forward to any responses

Thanks!!
Steve

You're several years too late, Steve. National Semiconductor used to
make a chip called the LM3909, which did exactly what you're talking
about -- long life flashing of an LED from one 1.5V battery. You could
even set it up so it would trigger on a voltage input.

Take a look at this website, and also take a look at the old LM3909
datasheet (.pdf). You should be able to get some ideas from this, and
also construct a circuit practically identical to the LM3909 out of
discrete components.

If you set your LED flasher up for a short enough flash time with long
enough space between flashes, you might be able to get 6 months out of
a AAA battery. However, with the voltage divider for your sensor, I'm
not sure. There are other ways to do that part, though.

Also, even though the LM3909 is obsolete and has been for years, you
still might be able to find a couple or scrounge them from something
else.

Look at this stuff and see. You might also just want to build the
"de-integrated" circuit for yourself for experimenting.

http://home.cogeco.ca/~rpaisley4/LM3909.html
http://www1.jaycar.com.au/images_uploaded/LM3909.PDF

Good luck
Chris

 

[SklettTheNewb]

Bummer, it does sounds like it would fit the bill. I will check out
the data sheet and see what I can gain from it.

You mentioned that there are other ways to handle the voltage divider,
do you have any suggestions?

I'm off to replicate the LM3909

Thanks again!

 

[Chris]

SklettTheNewb wrote:
<snip>

You mentioned that there are other ways to handle the voltage
divider,
do you have any suggestions?

Can you characterize your sensor? From your schematic, apparently it's
a variable resistance, like a photocell. Can you describe minimum,
maximum, trigger resistance, and whether you want the flasher to be on
if the resistance is greater or lesser than the trigger resistance?

Good luck
Chris

 

[SklettTheNewb]

min 2k ohm
max 64k ohm

if trigger resistance is the "reference", then that would be 4K, but I
suspect you are asking me something else.

When the sensor resistance > reference resistance
OR when input V < ref. V = LED on

 

[SklettTheNewb]

I'm gonna start a new thread with a more appropriate subject. Thanks
again for the help!

 

[Chris]

SklettTheNewb wrote:

min 2k ohm
max 64k ohm

if trigger resistance is the "reference", then that would be 4K, but
I
suspect you are asking me something else.

When the sensor resistance > reference resistance
OR when input V < ref. V = LED on

Hi, Steve. Sorry about the delay. Let's recap the bidding. You've
got a sensor that exhibits a variable resistance between 2K ohms and
64K ohms. When the sensor resistance is greater than 4K, you want an
LED to flash about once a second. The catch here is that you've got a
fresh Duracell AAA battery, and need to be able to have this thing
flash for 45 days with that battery as its only power source.

We talked before about your power requirements, and how a standard
oscillator won't do the job because of excess current which will eat up
the battery. The National Semiconductor LM3909 was made exactly for
flashing a standard LED off a single 1.5V cell with very low power
drain. Unfortunately, they're not making it anymore. However, there
are some parts still available if you look or scrounge, and there's
always the possibility of building one out of discrete components, as
linked in an earlier post.

National published an app note on the LM3909, called AN-154: "1.3V IC
Flasher, Oscillator, Trigger or Alarm"

http://www.fulcrum.ru/Read/CDROMs/NS/AN-154.pdf

On the second and third pages of the app note, it describes the circuit
operation for the basic 1.5V flasher circuit shown in Fig. 2. From
looking at their description and examining the feedback paths, it seems
that the flasher will not oscillate if only we can keep pin 2 (the +
end of the 300uF cap) from exceeding 1V. That shouldn't be too hard,
if we can find a comparator to do the rest of the job. Here's the
circuit (view in fixed font or M$ Notepad):


~~
~~
.--------o-----|<------. .---------.
| | RED LED | | |
| | | | |
| | | | |
| | | | |
| .--o------o------o------o--. |
| | 8 7 6 5 | |
| | | |
| | | +|
| | | ---
| | LM3909 | -
| | | 1.5V|
| | | |
| | | |
| | 1 2 3 4 | |
| '--o------o------o------o--' |
| | | | |
| | 330 | | |
| | uF | | |
| | ||+ | | |
'--------o--||--o '---------'
|| |
|
Mystery Signal |
From Control Ckt |
|

---------------'
created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de

The only question left is figuring out a way to turn the LM3909 (or the
discrete equivalent) on and off depending on the sensor resistance.
National Semiconductor came to your rescue again with the LM10 (which
is still in production).

http://cache.national.com/ds/LM/LM10.pdf

It's a low power combination op amp and 200mV reference that will work
well on power supplies down to 1.1V. Life is good. Here's the mystery
control circuit:

___
.----|___|---o----------------------------|<------->
| 330K | .-----------------. 1N4001 To
| | | | Flasher
| | | |
| .--------o------o------------. |
| | |6 |8 7| |
| |OpAmp.--|------|-----o------o----|--------.
| | | | | | | | |
| |2 |\| | | |\| Ref | | |
.--|---o---|-\ | '---|-\ Amp 1| |200mV +|
| | | | >-' LM10 | >-----o----o ---
| o---o---|+/ .---|+/ | | -
| | |3 |/| +| |/| | | |1.5V
| | | | --- | | | |
| | | | 200mV - | | | |
| | | | Ref | | | | |
| | | '---------o-----o------o----|--------'
| | | 4| o----.
| | '----------------------------' | |
| | .-. |
| | 10K| | |
| | | | |
| | '-' |20K
| | | .-.
| '-------------------------------------o | |<---.
| | | | |
| .-. '-' |
| Sensor| | | |
| 2K to 64K| | | |
| '-' | |
| | | |
| GND GND |
'---------------------------------------------------'
created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de


And here's how it works. The 200mV reference part of the LM10 provides
a reference voltage for a voltage divider between the 10K resistor and
your sensor. When your sensor goes above 4K, the VDVR goes above about
57mV. That's the value you've set with the 20K pot. Both of these
voltages are compared by the op amp section of the LM10. The output of
the LM10 goes high when that happens, which means it isn't interfering
with the charging of the 300uF cap, and the flasher does its thing.
When the sensor gets below 4K, the output of the op amp goes low. With
the series diode drop, it will keep the voltage at pin 2 of the LM3909
at around 0.8V, which will prevent oscillation/flashing. The 330K
resistor is added as positive feedback at the non-inverting input to
provide hysteresis, which prevents the in-betweens" from messing things
up.

I guess you're about set here. Even though both ICs are "micropower"
(at least by early 80s standards), you're going to be cutting it a
little close with the 45 day part on one AAA battery. I suppose it
could be tweaked a bit, but I think it will probably be good enough.
As long as you use a Duracell alkaline, you should make it.

Good luck with your project, Steve.
Chris

 

[Chris]

Slight error -- you need a series resistor with the diode to avoid
overcurrent at the LM10 output:

___ ___
.----|___|---o-|___|----------------------|<------->
| 330K | 100 .-----------------. 1N4001 To
| | | | Flasher
| | | |
| .--------o------o------------. |
| | |6 |8 7| |
| |OpAmp.--|------|-----o------o----|--------.
| | | | | | | | |
| |2 |\| | | |\| Ref | | |
.--|---o---|-\ | '---|-\ Amp 1| |200mV +|
| | | | >-' LM10 | >-----o----o ---
| o---o---|+/ .---|+/ | | -
| | |3 |/| +| |/| | | |1.5V
| | | | --- | | | |
| | | | 200mV - | | | |
| | | | Ref | | | | |
| | | '---------o-----o------o----|--------'
| | | 4| o----.
| | '----------------------------' | |
| | .-. |
| | 10K| | |
| | | | |
| | '-' |20K
| | | .-.
| '-------------------------------------o | |<---.
| | | | |
| .-. '-' |
| Sensor| | | |
| 2K to 64K| | | |
| '-' | |
| | | |
| GND GND |
'---------------------------------------------------'
created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de

Good luck
Chris

 

[Chris]

Lord Garth wrote:

Chris, Are those connection dots between the ref amp output and the
power rails? Looks like a short across the battery to me. The 7 and
4
pin numbers are probably misplaced as well.

Thanks. Possibly this will get through unmunged:
' ___ ___
' .----|___|---o-|___|----------------------|<------->
' | 330K | 100 .-----------------. 1N4001 To
' | | | | Flasher
' | | | |
' | .--------o------o------------. |
' | | |6 |8 7| |
' | |OpAmp.--|------|-----o------o----|--------.
' | | | | | | | | |
' | |2 |\| | | |\| Ref | | |
' .--|---o---|-\ | '---|-\ Amp 1| |200mV +|
' | | | | >-' LM10 | >-----o----o ---
' | o---o---|+/ .---|+/ | | -
' | | |3 |/| +| |/| | | |1.5V
' | | | | --- | | | |
' | | | | 200mV - | | | |
' | | | | Ref | | | | |
' | | | '---------o-----o------o----|--------'
' | | | 4| o----.
' | | '----------------------------' | |
' | | .-. |
' | | 10K| | |
' | | | | |
' | | '-' |20K
' | | | .-.
' | '-------------------------------------o | |<---.
' | | | | |
' | .-. '-' |
' | Sensor| | | |
' | 2K to 64K| | | |
' | '-' | |
' | | | |
' | GND GND |
' '---------------------------------------------------'
'
created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de

The LM10 has a bit of an unusual pinout:
Pin 1 Reference output
Pin 2 OpAmp - Input
Pin 3 OpAmp + Input
Pin 4 (Batt -)
Pin 5 Balance (unused)
Pin 6 OpAmp Output
Pin 7 (Batt +)
Pin 8 Reference Amp Feedback

In the diagram, Pin 8 is tied to Pin 1 for a ref amp voltage follower,
which would give 200mV Vref for the VDVR and the pot. Pin 7 is
connected to the battery +. I showed the internal "power connections"
of the LM10 like on the appnote, instead of just pinouts. Just focus
on what's going on outside the LM10 box (bubbles indicate pins).
Battery + (Pin 7) and Reference Output (Pin 8) are not connected
together here.

Thanks for the note. It seems Google munges ASCII art sometimes no
matter what. I don't mind it apart from that, but I used to post from
AOL groups. If it were a choice between AOL groups and a jammed
teletype machine with a spool of torn tape spewed all over the floor,
I'd have to think 'er over. Possibly I should chip in for a
store-bought newsreader and write off the monthly fees ;-)

Chris

 

[Lord Garth]

"Chris" <cfoley1064@yahoo.com> wrote in message news:1113024996.784822.203580@o13g2000cwo.googlegroups.com...

Lord Garth wrote:

Chris, Are those connection dots between the ref amp output and the
power rails? Looks like a short across the battery to me. The 7 and
4
pin numbers are probably misplaced as well.

Thanks. Possibly this will get through unmunged:
' ___ ___
' .----|___|---o-|___|----------------------|<------->
' | 330K | 100 .-----------------. 1N4001 To
' | | | | Flasher
' | | | |
' | .--------o------o------------. |
' | | |6 |8 7| |
' | |OpAmp.--|------|-----o------o----|--------.
' | | | | | | | | |
' | |2 |\| | | |\| Ref | | |
' .--|---o---|-\ | '---|-\ Amp 1| |200mV +|
' | | | | >-' LM10 | >-----o----o ---
' | o---o---|+/ .---|+/ | | -
' | | |3 |/| +| |/| | | |1.5V
' | | | | --- | | | |
' | | | | 200mV - | | | |
' | | | | Ref | | | | |
' | | | '---------o-----o------o----|--------'
' | | | 4| o----.
' | | '----------------------------' | |
' | | .-. |
' | | 10K| | |
' | | | | |
' | | '-' |20K
' | | | .-.
' | '-------------------------------------o | |<---.
' | | | | |
' | .-. '-' |
' | Sensor| | | |
' | 2K to 64K| | | |
' | '-' | |
' | | | |
' | GND GND |
' '---------------------------------------------------'
'
created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de

The LM10 has a bit of an unusual pinout:
Pin 1 Reference output
Pin 2 OpAmp - Input
Pin 3 OpAmp + Input
Pin 4 (Batt -)
Pin 5 Balance (unused)
Pin 6 OpAmp Output
Pin 7 (Batt +)
Pin 8 Reference Amp Feedback

In the diagram, Pin 8 is tied to Pin 1 for a ref amp voltage follower,
which would give 200mV Vref for the VDVR and the pot. Pin 7 is
connected to the battery +. I showed the internal "power connections"
of the LM10 like on the appnote, instead of just pinouts. Just focus
on what's going on outside the LM10 box (bubbles indicate pins).
Battery + (Pin 7) and Reference Output (Pin 8) are not connected
together here.

Thanks for the note. It seems Google munges ASCII art sometimes no
matter what. I don't mind it apart from that, but I used to post from
AOL groups. If it were a choice between AOL groups and a jammed
teletype machine with a spool of torn tape spewed all over the floor,
I'd have to think 'er over. Possibly I should chip in for a
store-bought newsreader and write off the monthly fees ;-)

Chris

Here's my edit on the circuit

' ___ ___
' .----|___|---o-|___|----------------------|<------->
' | 330K | 100 1N4001 To
' | | Flasher
' | |
' | .--------o------o------------.
' | | |6 |8 7 |
' | |OpAmp.--|------|-----o------|-------------.
' | | | | | | | |
' | |2 |\| | | |\| Ref | |
' .--|---o---|-\ | '---|-\ Amp 1| 200mV +|
' | | | >-' LM10 | >-----o----o ---
' | o-------|+/ .---|+/ | -
' | |3 |/| +| |/| | |1.5V
' | | | --- | | |
' | | | 200mV - | | |
' | | | Ref | | | |
' | | '---------o-----o------|-------------'
' | | 4 o----.
' | | | |
' | | .-. |
' | | 10K| | |
' | | | | |
' | | '-' |20K
' | | | .-.
' | '--------------------------------o | |<---.
' | | | | |
' | .-. '-' |
' | Sensor| | | |
' | 2K to 64K| | | |
' | '-' | |
' | | | |
' | GND GND |
' '----------------------------------------------'

 

[Chris]

Lord Garth wrote:

Here's my edit on the circuit

' ___ ___
' .----|___|---o-|___|----------------------|<-------
' | 330K | 100 1N4001 To
' | | Flasher
' | |
' | .--------o------o------------.
' | | |6 |8 7 |
' | |OpAmp.--|------|-----o------|-------------.
' | | | | | | | |
' | |2 |\| | | |\| Ref | |
' .--|---o---|-\ | '---|-\ Amp 1| 200mV +|
' | | | >-' LM10 | >-----o----o ---
' | o-------|+/ .---|+/ | -
' | |3 |/| +| |/| | |1.5V
' | | | --- | | |
' | | | 200mV - | | |
' | | | Ref | | | |
' | | '---------o-----o------|-------------'
' | | 4 o----.
' | | | |
' | | .-. |
' | | 10K| | |
' | | | | |
' | | '-' |20K
' | | | .-.
' | '--------------------------------o | |<---.
' | | | | |
' | .-. '-' |
' | Sensor| | | |
' | 2K to 64K| | | |
' | '-' | |
' | | | |
' | GND GND |
' '----------------------------------------------'

Same circuit, _much_ cleaner look. Thanks.

Chris