low voltage LED

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Post Reply
KC0MRZ
Posts: 4
Joined: Sat Jan 11, 2003 1:01 am
Location: Mn.
Contact:

low voltage LED

Post by KC0MRZ » Thu Feb 13, 2003 5:41 pm

Hi,<p> I need help!!! I'm a student in college for electronics. I'm trying to make a a bobber that will light up when it goes under water. It will run on a 3 volt battery and will have a couple of wires hangin out of it. When it hits the water that will complete the circuit and a LED will light. Now my problem is this.....
I tried using a LM6511 comparotor that when you touch the two wires the LED will light. No such luck. It works on other op amps and comparitors but I'm dealing with a small area and low voltage. The idea is that you have a ref. voltage and when you touch the wires your negative input voltage goes higher then the ref. voltage and your ouput will go up in voltage and the LED will light. Now for voltage I'm using a 3volt coin battery, and using a voltage divider circuit. I hope you guys get the general idea on this and I didn't confuse you more. My instructor and I have been bucking heads on this and get it to work on other IC's but there too big. I need to fit this inside a bobber thats about 1/1/2" dia. Any info would be appreciated. Would like to use the LM6511 still but will try others if I could.

User avatar
dacflyer
Posts: 4503
Joined: Fri Feb 08, 2002 1:01 am
Location: USA / North Carolina / Fayetteville
Contact:

Re: low voltage LED

Post by dacflyer » Thu Feb 13, 2003 6:58 pm

hey.. K-Mart sold them,,and i think they still do sell them,, i got 2-3 of them,,, they twist on / off and use 2-3 watch batteries... they work great... go look...they do exhist..

Ron H
Posts: 360
Joined: Tue Dec 18, 2001 1:01 am
Location: Boise, ID
Contact:

Re: low voltage LED

Post by Ron H » Thu Feb 13, 2003 7:07 pm

The LM6511 is has an open collector output. It can only sink current. You need to connect the cathode of the LED to the output, and connect your current limiting resistor from the anode to +3v. When water is sensed, the (-) input needs to go more positive than the (+) input, so that the output goes low. You need to set up your inputs so that neither goes more positive than 1.2 volt below the positive supply (1.8 volts in this case), nor more negative than 0.3 volts. These are the common mode limits of this part.
Also take a look at LMV331. It has better common mode range. The worst-case output current is marginal, but unless you are building hundreds, this probably wouldn't be a problem.<p>Ron

KC0MRZ
Posts: 4
Joined: Sat Jan 11, 2003 1:01 am
Location: Mn.
Contact:

Re: low voltage LED

Post by KC0MRZ » Fri Feb 14, 2003 7:53 am

I know Kmart has them but they stay on all the time I want the LED to light only when under water. I'll try the Rons idea and pleace give me some more ideas if you have them. I'll let you guys know if Rons idea works. thanks Ron!

L. Daniel Rosa
Posts: 400
Joined: Tue Dec 25, 2001 1:01 am
Location: Bellingham, WA
Contact:

Re: low voltage LED

Post by L. Daniel Rosa » Fri Feb 14, 2003 12:37 pm

Does it need to be done with a chip? A few transistors can do what you ask.

Ron H
Posts: 360
Joined: Tue Dec 18, 2001 1:01 am
Location: Boise, ID
Contact:

Re: low voltage LED

Post by Ron H » Fri Feb 14, 2003 2:15 pm

Good point, L.
A low threshold MOSFET with ESD protection (series gate resistor (10k?) and back-biased diodes from gate to both battery terminals), a BFR (10meg) from the other end of the 10k to V-, water probes from V+ and the junction of the 10k/10meg, source to V-, LED cathode to drain, and current-limiting resistor from LED anode to V+ should do it.<p>Ron

josmith
Posts: 340
Joined: Tue Dec 04, 2001 1:01 am
Contact:

Re: low voltage LED

Post by josmith » Fri Feb 14, 2003 4:28 pm

YOu could use a cmos logic gate like a 4000 or 74C series. They are not intended for this purpose but it's common to use them in experimental circuits.
Just use a big pullup reisistor for the inputs and a current limiter on the output. Use a few gates in parallel if you need to.

gadgeteer
Posts: 102
Joined: Fri Mar 07, 2003 1:01 am
Location: Ok!
Contact:

Re: low voltage LED

Post by gadgeteer » Fri Mar 07, 2003 10:30 pm

<blockquote><font size="1" face="Verdana, Helvetica, sans-serif">quote:</font><hr>Does it need to be done with a chip? A few transistors can do what you ask.<hr></blockquote>Heck ONE darlington would do it. At the currents required for a modern super-bright LED, such as 10-15 milliams, the darlington common-emitter would work fine. On a 3 volt supply, red LED, your resistor would be: (3Vbat - 0.2Vce - 1.8Vled)/.01mA = 100 ohms.<p>Piece of cake. Bias the base to gnd (we're assuming NPN) with a couple hundred K, and use a 1K input (in case of shorts). The bias resistor subject to experimentation---it keeps the thing OFF until the water current is sufficient to light the LED. Hfe on a darlington is easily 1k to 10k, which means the DRIVE CURRENT (0.01mA) is divided by the transistor gain, so that the BASE current is in the range of 10µA max. Easily provided through your average stream or pond...

Ron H
Posts: 360
Joined: Tue Dec 18, 2001 1:01 am
Location: Boise, ID
Contact:

Re: low voltage LED

Post by Ron H » Sun Mar 09, 2003 7:47 pm

[ <blockquote><font size="1" face="Verdana, Helvetica, sans-serif">quote:</font><hr>Originally posted by gadgeteer:
Heck ONE darlington would do it. At the currents required for a modern super-bright LED, such as 10-15 milliams, the darlington common-emitter would work fine. On a 3 volt supply, red LED, your resistor would be: (3Vbat - 0.2Vce - 1.8Vled)/.01mA = 100 ohms.<p>Piece of cake. Bias the base to gnd (we're assuming NPN) with a couple hundred K, and use a 1K input (in case of shorts). The bias resistor subject to experimentation---it keeps the thing OFF until the water current is sufficient to light the LED. Hfe on a darlington is easily 1k to 10k, which means the DRIVE CURRENT (0.01mA) is divided by the transistor gain, so that the BASE current is in the range of 10µA max. Easily provided through your average stream or pond...<hr></blockquote><p>A saturated Darlington's Vce is more like 0.7 volts, so if you want 10ma, you'll need 50 ohms (use 47). However, as battery voltage reaches end-of-life, the LED output will drop significantly. The CMOS gate solution mentioned by Josmith will work. A 74HC14 hex Schmitt trigger with all 6 gates in parallel should give enough drive to provide 10ma to a red LED through 100 ohms to +V on the 3 volt supply. Connect a 1k resistor in series with the paralleled inputs, 1meg to -V on the other end of the 1k. One water probe would go to this node, the other to +V.<p>Ron

gadgeteer
Posts: 102
Joined: Fri Mar 07, 2003 1:01 am
Location: Ok!
Contact:

Re: low voltage LED

Post by gadgeteer » Tue Mar 11, 2003 1:37 am

I had a "Dicken's Christmas Village Santa"---open frame plastic display, 71 LEDs, with all its LEDs tied in parallel (very poor "design"; all reds hog the current)---once I separated the colors and added proper ballast resistors, the current consumption dropped from 1.5 amps to less than ½ amp. And THEN (and only then) his yellow cuffs lit up. But I wanted his EYES to be BLUE---impossible for the 3 volt supply.<p>I simply wired two gates of a 4069 as a multivibrator, and tied the other gates in parallel for current---and added two capacitors and two diodes (schotky of course---1n5817). Presto! The 3 volt supply became 5.5 volts! Works so well, that as the batteries die, the BLUE EYES are the last lights to go out.<p>It occurs to me you could do a similar configuration, using the high-impedance inputs to gate the multivibrator with your WATER-SENSE-ELECTRODES. Maybe the feedback resistor can couple through the water?<p>OR, use a 4011 QUAD NAND, and use the extra inputs as the multivibrator gate (via your "water-sense-electrodes"). <p>Using ONE BATTERY (say an "N" alkaline), you should be able to run a red, orange, green or yellow LED for a very long time. Use a superbright (2000mcd-10,000mcd) and lower current, and you should be able to light it for weeks. Efficiency is quite high. With more diodes and capacitors, it should be possible to charge-pump high enough for even BLUE or WHITE LEDs from one battery.<p>You hafta use schotky diodes, 'cause a SILICON diode will drop 0.7 volts each; the 5817 will only drop 0.2 volts...

Post Reply

Who is online

Users browsing this forum: No registered users and 14 guests