Reply by Martin Brown December 15, 20142014-12-15
On 12/12/2014 19:37, George Herold wrote:
> On Friday, December 12, 2014 12:44:23 PM UTC-5, jurb...@gmail.com wrote: >> I have an application where I need an LED to light on about 1.1 mA. >> >> It doesn't hav to be real bright, just indicate thaat a battery is good. Due to my wish to make it power switch free, the design is constrained. I am sure I could probably come up with more current, but at the cost of complexity as well as power drain. >> >> It is a pretty simple device, just an oscillator to put a signal to a DUT. What I want is when the test leads are open, the current drain is so low the battery life approaches shelf life. Even in operation I hope to keep the max drain less than like 4 mA. (this gives me more choices of batteries to use) >> >> It is a simple five transistor circuit comsisting of an astable multivibrator and a buffer. The buffer is just fine being bipolar, but I am wondering if I oculd use FETs for the multivibrator to reduce current drain even further because that part of it runs all the time. (depletion mode is looking pretty good fo that) The buffer will see no current unless it is in use and the current involved there is dictsated by other factors. >> >> Even in use though we only got a bit over 1 mA to light the thing. Got any suggestions ? (and green woulld be the preferred color I think) > > 1 mA is enough current to see an LED. Buy a few and see what you like. > You could flash a pulse every second or so to keep the drain lower.
Modern high efficiency white ones are visibly lit on around 10-100uA depending on the specimen. Water clear packaging and towards the higher end of that range if you want to have it visible in daylight. As someone else has said a flash every second or two can be a lot brighter and use less current on average. -- Regards, Martin Brown
Reply by Robert Baer December 13, 20142014-12-13
John Larkin wrote:
> On Fri, 12 Dec 2014 09:44:19 -0800 (PST), jurb6006@gmail.com wrote: > >> I have an application where I need an LED to light on about 1.1 mA. >> >> It doesn't hav to be real bright, just indicate thaat a battery is good. Due to my wish to make it power switch free, the design is constrained. I am sure I could probably come up with more current, but at the cost of complexity as well as power drain. >> >> It is a pretty simple device, just an oscillator to put a signal to a DUT. What I want is when the test leads are open, the current drain is so low the battery life approaches shelf life. Even in operation I hope to keep the max drain less than like 4 mA. (this gives me more choices of batteries to use) >> >> It is a simple five transistor circuit comsisting of an astable multivibrator and a buffer. The buffer is just fine being bipolar, but I am wondering if I oculd use FETs for the multivibrator to reduce current drain even further because that part of it runs all the time. (depletion mode is looking pretty good fo that) The buffer will see no current unless it is in use and the current involved there is dictsated by other factors. >> >> Even in use though we only got a bit over 1 mA to light the thing. Got any suggestions ? (and green woulld be the preferred color I think) > > 1 mA is blinding for a high-efficiency green LED. We run status LEDs > on PCBs at 200 uA, so we can see around them to probe. 1 uA is > typically visible in moderate office lighting. > > I have an Avago green LED that makes just detectable light, into a > dark-adapted eye, at about 800 pA. > > Here'a a Tadiran 3V lithium battery, 1M resistor, green LED. > > https://dl.dropboxusercontent.com/u/53724080/Optos/Tadiran.JPG > > It should look about the same in 20 years. > >
Check, that Rohm SML-P12YTT86 SMD Yellow 130 mcd LED seems similar in apparent brightness; probably better because it is yellow and fits the eye color sensitivity curve better.
Reply by Robert Baer December 13, 20142014-12-13
jurb6006@gmail.com wrote:
> I have an application where I need an LED to light on about 1.1 mA. > > It doesn't hav to be real bright, just indicate thaat a battery is good. Due to my wish to make it power switch free, the design is constrained. I am sure I could probably come up with more current, but at the cost of complexity as well as power drain. > > It is a pretty simple device, just an oscillator to put a signal to a DUT. What I want is when the test leads are open, the current drain is so low the battery life approaches shelf life. Even in operation I hope to keep the max drain less than like 4 mA. (this gives me more choices of batteries to use) > > It is a simple five transistor circuit comsisting of an astable multivibrator and a buffer. The buffer is just fine being bipolar, but I am wondering if I oculd use FETs for the multivibrator to reduce current drain even further because that part of it runs all the time. (depletion mode is looking pretty good fo that) The buffer will see no current unless it is in use and the current involved there is dictsated by other factors. > > Even in use though we only got a bit over 1 mA to light the thing. Got any suggestions ? (and green woulld be the preferred color I think)
I used the Rohm SML-P12YTT86 SMD Yellow 130 mcd LEDs with 10K in series driven from a microcontroller (5V); to the eye brighter that their red ones; that is roughly 300uA.
Reply by Jasen Betts December 13, 20142014-12-13
On 2014-12-12, meow2222@care2.com <meow2222@care2.com> wrote:
>> >> Here'a a Tadiran 3V lithium battery, 1M resistor, green LED. >> >> https://dl.dropboxusercontent.com/u/53724080/Optos/Tadiran.JPG >> >> It should look about the same in 20 years. > > Get an 'ultrabright'
why? ultrabright will likely have worse leakage than high-efficiency. -- umop apisdn
Reply by December 12, 20142014-12-12
On Friday, December 12, 2014 8:35:44 PM UTC, John Larkin wrote:
> On Fri, 12 Dec 2014 09:44:19 -0800 (PST), jurb6006@gmail.com wrote: >=20 > >I have an application where I need an LED to light on about 1.1 mA.=20 > > > >It doesn't hav to be real bright, just indicate thaat a battery is good.=
Due to my wish to make it power switch free, the design is constrained. I = am sure I could probably come up with more current, but at the cost of comp= lexity as well as power drain.=20
> > > >It is a pretty simple device, just an oscillator to put a signal to a DU=
T. What I want is when the test leads are open, the current drain is so low= the battery life approaches shelf life. Even in operation I hope to keep t= he max drain less than like 4 mA. (this gives me more choices of batteries = to use)
> > > >It is a simple five transistor circuit comsisting of an astable multivib=
rator and a buffer. The buffer is just fine being bipolar, but I am wonderi= ng if I oculd use FETs for the multivibrator to reduce current drain even f= urther because that part of it runs all the time. (depletion mode is lookin= g pretty good fo that) The buffer will see no current unless it is in use a= nd the current involved there is dictsated by other factors.
> > > >Even in use though we only got a bit over 1 mA to light the thing. Got a=
ny suggestions ? (and green woulld be the preferred color I think)
>=20 > 1 mA is blinding for a high-efficiency green LED. We run status LEDs > on PCBs at 200 uA, so we can see around them to probe. 1 uA is > typically visible in moderate office lighting. >=20 > I have an Avago green LED that makes just detectable light, into a > dark-adapted eye, at about 800 pA. >=20 > Here'a a Tadiran 3V lithium battery, 1M resistor, green LED.=20 >=20 > https://dl.dropboxusercontent.com/u/53724080/Optos/Tadiran.JPG >=20 > It should look about the same in 20 years.
Get an 'ultrabright' NT
Reply by John Larkin December 12, 20142014-12-12
On Fri, 12 Dec 2014 09:44:19 -0800 (PST), jurb6006@gmail.com wrote:

>I have an application where I need an LED to light on about 1.1 mA. > >It doesn't hav to be real bright, just indicate thaat a battery is good. Due to my wish to make it power switch free, the design is constrained. I am sure I could probably come up with more current, but at the cost of complexity as well as power drain. > >It is a pretty simple device, just an oscillator to put a signal to a DUT. What I want is when the test leads are open, the current drain is so low the battery life approaches shelf life. Even in operation I hope to keep the max drain less than like 4 mA. (this gives me more choices of batteries to use) > >It is a simple five transistor circuit comsisting of an astable multivibrator and a buffer. The buffer is just fine being bipolar, but I am wondering if I oculd use FETs for the multivibrator to reduce current drain even further because that part of it runs all the time. (depletion mode is looking pretty good fo that) The buffer will see no current unless it is in use and the current involved there is dictsated by other factors. > >Even in use though we only got a bit over 1 mA to light the thing. Got any suggestions ? (and green woulld be the preferred color I think)
1 mA is blinding for a high-efficiency green LED. We run status LEDs on PCBs at 200 uA, so we can see around them to probe. 1 uA is typically visible in moderate office lighting. I have an Avago green LED that makes just detectable light, into a dark-adapted eye, at about 800 pA. Here'a a Tadiran 3V lithium battery, 1M resistor, green LED. https://dl.dropboxusercontent.com/u/53724080/Optos/Tadiran.JPG It should look about the same in 20 years. -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Reply by Artemus December 12, 20142014-12-12
<jurb6006@gmail.com> wrote in message news:b7847399-9e05-4056-9a7a-b0aa4f53d9e4@googlegroups.com...
I have an application where I need an LED to light on about 1.1 mA.

It doesn't hav to be real bright, just indicate thaat a battery is good. Due to my wish to make it 
power switch free, the design is constrained. I am sure I could probably come up with more current, 
but at the cost of complexity as well as power drain.

It is a pretty simple device, just an oscillator to put a signal to a DUT. What I want is when the 
test leads are open, the current drain is so low the battery life approaches shelf life. Even in 
operation I hope to keep the max drain less than like 4 mA. (this gives me more choices of batteries 
to use)

It is a simple five transistor circuit comsisting of an astable multivibrator and a buffer. The 
buffer is just fine being bipolar, but I am wondering if I oculd use FETs for the multivibrator to 
reduce current drain even further because that part of it runs all the time. (depletion mode is 
looking pretty good fo that) The buffer will see no current unless it is in use and the current 
involved there is dictsated by other factors.

Even in use though we only got a bit over 1 mA to light the thing. Got any suggestions ? (and green 
woulld be the preferred color I think)

***********************
Does it have to on continuously or can you pulse it?
The LM3909 is no longer available but the idea is still valid.
Art 


Reply by Jan Panteltje December 12, 20142014-12-12
On a sunny day (Fri, 12 Dec 2014 09:44:19 -0800 (PST)) it happened
jurb6006@gmail.com wrote in
<b7847399-9e05-4056-9a7a-b0aa4f53d9e4@googlegroups.com>:

>I have an application where I need an LED to light on about 1.1 mA. > >It doesn't hav to be real bright, just indicate thaat a battery is good. Du= >e to my wish to make it power switch free, the design is constrained. I am = >sure I could probably come up with more current, but at the cost of complex= >ity as well as power drain. > >It is a pretty simple device, just an oscillator to put a signal to a DUT. = >What I want is when the test leads are open, the current drain is so low th= >e battery life approaches shelf life. Even in operation I hope to keep the = >max drain less than like 4 mA. (this gives me more choices of batteries to = >use) > >It is a simple five transistor circuit comsisting of an astable multivibrat= >or and a buffer. The buffer is just fine being bipolar, but I am wondering = >if I oculd use FETs for the multivibrator to reduce current drain even furt= >her because that part of it runs all the time. (depletion mode is looking p= >retty good fo that) The buffer will see no current unless it is in use and = >the current involved there is dictsated by other factors. > >Even in use though we only got a bit over 1 mA to light the thing. Got any = >suggestions ? (and green woulld be the preferred color I think)
1mA is more than enough with modern high efficiency LEDs, its blinding!
Reply by George Herold December 12, 20142014-12-12
On Friday, December 12, 2014 12:44:23 PM UTC-5, jurb...@gmail.com wrote:
> I have an application where I need an LED to light on about 1.1 mA.=20 >=20 > It doesn't hav to be real bright, just indicate thaat a battery is good. =
Due to my wish to make it power switch free, the design is constrained. I a= m sure I could probably come up with more current, but at the cost of compl= exity as well as power drain.=20
>=20 > It is a pretty simple device, just an oscillator to put a signal to a DUT=
. What I want is when the test leads are open, the current drain is so low = the battery life approaches shelf life. Even in operation I hope to keep th= e max drain less than like 4 mA. (this gives me more choices of batteries t= o use)
>=20 > It is a simple five transistor circuit comsisting of an astable multivibr=
ator and a buffer. The buffer is just fine being bipolar, but I am wonderin= g if I oculd use FETs for the multivibrator to reduce current drain even fu= rther because that part of it runs all the time. (depletion mode is looking= pretty good fo that) The buffer will see no current unless it is in use an= d the current involved there is dictsated by other factors.
>=20 > Even in use though we only got a bit over 1 mA to light the thing. Got an=
y suggestions ? (and green woulld be the preferred color I think) 1 mA is enough current to see an LED. Buy a few and see what you like. =20 You could flash a pulse every second or so to keep the drain lower. =20 George H.
Reply by mike December 12, 20142014-12-12
On 12/12/2014 9:50 AM, Klaus Kragelund wrote:
> On Friday, December 12, 2014 6:44:23 PM UTC+1, jurb...@gmail.com wrote: >> I have an application where I need an LED to light on about 1.1 mA. >> >> It doesn't hav to be real bright, just indicate thaat a battery is good. Due to my wish to make it power switch free, the design is constrained. I am sure I could probably come up with more current, but at the cost of complexity as well as power drain. >> >> It is a pretty simple device, just an oscillator to put a signal to a DUT. What I want is when the test leads are open, the current drain is so low the battery life approaches shelf life. Even in operation I hope to keep the max drain less than like 4 mA. (this gives me more choices of batteries to use) >> >> It is a simple five transistor circuit comsisting of an astable multivibrator and a buffer. The buffer is just fine being bipolar, but I am wondering if I oculd use FETs for the multivibrator to reduce current drain even further because that part of it runs all the time. (depletion mode is looking pretty good fo that) The buffer will see no current unless it is in use and the current involved there is dictsated by other factors. >> >> Even in use though we only got a bit over 1 mA to light the thing. Got any suggestions ? (and green woulld be the preferred color I think) > > Search digikey for a high brightness LED, whatever you want to pay for it > > If the batterypack has several cells, let one cell power the LED to keep losses down in the series resistor, but that creates unsymmetric loading of the cells, so that might not be a good idea. > > Add a LED driver (switch mode driven) to keep losses low, but that circuit will be more complex than your primary circuit > > Cheers > > Klaus >
Without knowing what voltage you have available, or much of anything about what you're doning, it's hard to go forward. Building a system around all the requirements is often superior to hanging blocks together. Spend a buck on a microcontroller that sleeps most of the time. Lots of options then.