tek

Ok, i know i know...

But its in the modding section..

Ok so here goes, i have been working on some thing.. I want to make 3 leds light up at various voltages, say 4 5 6 or close to it.. Because of this i decided to use a lm3914n http://www.alldatasheet.com/datashee...C/LM3914N.html to get it to work the way i wanted.. At first i made a diode cascade using the 1n9xx diode, and that did not work as intended, the leds would not snap on at all, but would come on almost 2 volts apart, and i had no idea when full brightness was achived so that did not work, any way on to this.. Its a common led driver, and the national semi conductor pages says its easy.. WTF. I dont get it, i wired it perfectly <i think> to the spec on the data sheet under the common applications, used the 0-5 volt signal divider, and was able to i thought go....

I want 4 5 6 volt settings for leds to turn on

why cant i get this to work....

any one who is good with this sort of electronic prototyping or rocket scientists are welcomed, and any one who can... arg..

Steampunk

I think your problem is with the V(ref) of 1.25V. This diagram http://html.alldatasheet.com/html-pd...2/LM3914N.html shows a "typical application", with a calculation for the resistors. This page http://html.alldatasheet.com/html-pd...7/LM3914N.html goes into more detail on how to calculate the resistor values for the particular application of the circuit.

This page gives some hints, too: http://html.alldatasheet.com/html-pd...4/LM3914N.html

Basically, it says to keep the wire runs as short as possible to avoid oscillations. Perhaps part of the problem is you're running the wires too close together?

tek

I thank you very much steam for the help, ill reconfig it in the morning.. I am beat lol.. But i was working on a bread board with 10 leds <just to see it work right> and used short jumpers all the way. I think i might be getting in to trouble using the same source for power for all of it, but yet idk. Im deffinatly going to look in to more of it in the morning, myby my 1.25 ref was too low, and i need to change that to about 2.2 buy why would they tell you to run leds on 1.25? idk.. to many question for the night so time to pass the f out..

Steampunk

The 1.25V is an internal reference voltage generated by the input voltage. From the block diagram it looks like there is a 1.25V regulator built into the chip, and the resistors look like a voltage divider, but probably also provide current limiting to the reference. Hopefully you didn't fry the internal reference!

Also, if you look at the block diagram http://html.alldatasheet.com/html-pd...6/LM3914N.html Pin 9 determines whether you're using a bar display or discrete diodes, but I didn't read it enough to see what determines the mode.

Um...wait....look at the "Typical Applications" page... http://html.alldatasheet.com/html-pd...2/LM3914N.html

The circuit is looking for ~6.8-18V on Pin 9 and Pin 3. Are you supplying that kind of voltage to the ckt? That makes this ~12V ckt...

berger

something like this? http://www.ebay.com/itm/1pc-10-LED-B...item1c22b2766b

RY4

I remember those

Steampunk

I would say $7 is worth not getting a headache...

berger

I doubt I could get all the components for 7 bucks...I understand wanting to build one but without the board it gets a little unruly

tek

working with 8.4 volts on two 18650s, for the purpose of wanting three leds to light up, it should be simple, i may have fryed two chips.. i am going to try again today, since i have every thing, the right resistors, caps and 3 more of the lm3914

i may end up with one of them yet...

Ok i know its a head ake, but, its on a proto board, and there is some thing i am missing.. Also i want to learn this a nother reason why i am trying it out lol..

berger

take a look at this

tek

i wonder if i can substutie the 1n4001 for a 1n9001 hmm....

Steampunk

You're going to make me think, eh?

I think a 4001 is a signal diode while a 9001 is a power diode.Um,it's been a while.

Rocketman

Tek,
I'm too lazy to look it up but Google conduction "Knee" for silicon diode, and the spec graphs for both part numbers. Until you exceed the current on either the 9001 or the 4001 the functionality will be about the same.

The short answer: It'll be fine

JohnnyJames

How about something like this?

http://www.hobbyking.com/hobbycity/s...idProduct=2038

tek

the main issue is i only want 3 leds, and it to be small enough to fit in a space about 36 x10mm and that is approx for the room i have left in the mod, hence why i am building it lol.. i will figure it out eventually.. I think i might need to get a couple more in of the lm chip..

Steampunk

What's the fun in that?!?!

tek

i did it, turns out i needed the full voltage from the two li ion cells to the positive of the leds in order for it to work, pics to follow..

Steampunk

Cool! Bring it! What was the solution?

tek

Ok, so let me explain the pics, and try to do this right



Ok this above is my bread board, using the www.ti.com lay out for the components, and leds in slots (led positions, 6 7 and 9) This gave me round about voltages that are real close to 4 5 6 respectively, and that for some reason is how it worked out when tested with my multi meter. The voltage jump between leds is supposed to be .5 volt steps along a range of 1.25 volts, threw the ic, but turns out to be .74 volts, due to me using 2 18650 batts in series.

Now, this is not precision but round about, and that is fine for me, cause i know i am within my target voltage this way, and dont need fine meter tuning but by taste..



Tonight i added a hole for the pot, now that i have all the parts assembled, and 3 holes for leds, 3mm blue with holders.. Also cut in a main on off slider switch, because i was having to remove a batt and flip it to power down the mod while i went to sleep <i dont know why, but i sleep vape..>



Lowish voltage, about 4 volts..



Med, about 5 voltish



6 voltish, high setting..

The leds, need a positive off of the battery supply to the positive of the led, at 8.4 volts. I suspect that this opens the negative gates in the lm3914 and allow them to work. One of the original problems i was having was with trying to supply the 4 volts to the reference and to the load, for some reason that did not work for me.. But i followed the instructions again, and this time hooked the leds to the positive rail of the bread board.. It worked.. You tube to come later of actual use..

berger

pretty cool tek...nice job

slimest

Looks nice
As far as I remember, I made approx the same 4-step indicator based on LM339. This is not special chip, just 4 low voltage comparators in a single body. Used some external resistors and one zener...

tek

The lm3914 has ten compairtors, and that is way more then what i need, but at less then 3 dollars us each and the ability to not use a zener diode, makes it a lil better for me. Also it can be used with more then one lm3914 to make scales of 100, 200 leds possible, yes 200 are easily done with 20 lm3914 scaling function lol..