Topic 'LEDs not going fully off' in forum 'Electronics' - dslreports.com

Re: LEDs not going fully off

Fri, 18 Apr 2014 09:12:33 EDT

Jason posted : Ooh.. I'll have to remember that mnemonic. Its a hell of a lot more PC than the one I've known since youth ;-)
--
When cryptography is outlawed, bayl bhgynjf jvyy unir cevinpl.

Re: LEDs not going fully off

Thu, 17 Apr 2014 23:51:04 EDT

Thane_Bitter posted : Just remember:
Bad beer rots our young guts but vodka goes well – get some now. ;)

Re: LEDs not going fully off

Thu, 17 Apr 2014 22:03:51 EDT

Nick posted : So the goof is on me...It was definitely an issue with #1
I may have incorrectly read the resistor value since I had the same issue when I used an NPN transistor. It looks like my pull down resistor didn't fully close the gate. Once I switched to a different resistor the NPN transistors worked, as did the MOSFETs

Re: LEDs not going fully off

Wed, 09 Apr 2014 22:14:31 EDT

TheMG posted :

said by Nick:

It's hard to justify $6 shipping for $2 in parts...Of course if I had an idea of all the components I would potentially need for various projects I would just place one bigger order with Mouser.

Know anyone else that might need to order some parts?

Where I live, I'm a member of the local amateur radio club and what we do when we need parts is we'll ask around if anyone else needs anything, and put in a combined order split the shipping cost.

Re: LEDs not going fully off

Tue, 08 Apr 2014 11:40:17 EDT

Nick posted : I do realize dealing with eBay sometimes you get counterfeit or bad parts whether it's sunglasses or electronics components....

I guess the only thing that causes my hesitation for ordering from Mouser/DigiKey is the shipping charges...

It's hard to justify $6 shipping for $2 in parts...Of course if I had an idea of all the components I would potentially need for various projects I would just place one bigger order with Mouser.
--
-Stupidity, like hydrogen, is one of the basic building blocks of the Universe.
-Insanity: doing the same thing over and over again and expecting different results.

Misc

Re: LEDs not going fully off

Sat, 05 Apr 2014 01:21:00 EDT

TheMG posted :

said by Nick:

2. These are from a Chinese eBay seller so they could be poor quality rejects that exhibit this behavior

Inferior counterfeit parts are also a likely possibility.

Looking at the datasheet for the IRF830B shows the leakage current should be no greater than 10uA at 25C. That's at 500V drain-source voltage and you're at a fraction of that. Current should be negligible at 12V drain-source and not enough to light the LEDs at all.

Counterfeit components are a pretty big problem. This is why I stick to trusted distributors such as DigiKey and Mouser and avoid eBay as a source for components unless it's a rare obsolete part that can't be found anywhere else.

Another possibility could be the result of ESD damage. But if you've got several of them and they all behave similarly, it's probably counterfeit parts.

Re: LEDs not going fully off

Sat, 05 Apr 2014 01:13:48 EDT

leibold posted : The datasheets indicate a lower zero-gate-drain-current for the IRF830B (compared to the IRF510). Looks as if #2 is the more likely choice.
--
Got some spare cpu cycles ? Join Team Helix or Team Starfire!

Re: LEDs not going fully off

Fri, 04 Apr 2014 23:08:03 EDT

Nick posted : Ok, I stopped by Radio Shack today and picked up an IRF510 which works properly when directly swapped for the IRF830B....

Now, I narrowed it down to being the transistor I have 2 possible reasons for it:
1. I am using the wrong spec transistor for my circuit (a real possibility since I'm dumb when it comes to electronics and better at the programming aspect of it)
2. These are from a Chinese eBay seller so they could be poor quality rejects that exhibit this behavior
--
-Stupidity, like hydrogen, is one of the basic building blocks of the Universe.
-Insanity: doing the same thing over and over again and expecting different results.

Misc

Re: LEDs not going fully off

Thu, 03 Apr 2014 00:14:42 EDT

Nick posted : I am using an IRF830B and the resistor I'm using is 10k...

Re: LEDs not going fully off

Thu, 03 Apr 2014 00:03:23 EDT

TheMG posted : You might be able to divert enough of that current away from the LEDs with a resistor in parallel with the LEDs. Experiment with a few different values. Don't put one of too small value though as you don't want the resistor to dissipate excessive power when the LEDs are full on.

I'd start with a 10k resistor. Theoretically I think 10k should be more than sufficient but I haven't done the math on it.

What are you using for a MOSFET anyways (part number)?

Re: LEDs not going fully off

Wed, 02 Apr 2014 21:37:49 EDT

Nick posted : Can you recommend a MOSFET i can pick up from RS that I can use to drive 6 LEDs without leakage?

Re: LEDs not going fully off

Wed, 02 Apr 2014 21:31:34 EDT

Nick posted : Here's a photo of my breadboard

Re: LEDs not going fully off

Wed, 02 Apr 2014 20:45:09 EDT

TheMG posted : If it's still happening when you ground the gate of the MOSFET, then the most likely culprit is the leakage current of the MOSFET itself.

Modern high-efficiency LEDs, IIRC, will produce visible light at currents of only a few microamps. Large power MOSFETs can have leakage currents of several tens or hundreds of microamps.

Re: LEDs not going fully off

Wed, 02 Apr 2014 19:27:42 EDT

Nick posted : I'll try to rewire it again and see what happens

Re: LEDs not going fully off

Wed, 02 Apr 2014 19:13:04 EDT

Jason posted : I saw this in the comments at the site you posted..

He's right, the pull-down resistor is on a different mosfet pin in the first 2 colors than on the 3rd, in the diagram.

"This was helpful with my own project but I did have a problem wiring it up as shown. In the first two instances the resistor is in the wrong place, the last FET has the resistor on the gate, where it needs to be or it won't turn fully off when it's supposed to."

edit: I just saw you tried this.

Hmmm. I dunno then.
--
When cryptography is outlawed, bayl bhgynjf jvyy unir cevinpl.

Re: LEDs not going fully off

Wed, 02 Apr 2014 09:24:11 EDT

Nick posted : I don't have an Oscilloscope but when I took the gate cable from the arduino pin and grounded it the strip stayed dimly lit.

It might be that my strip is very short (I cut only 1 segment off the whole strip to play with) but I tried a long section (full strip minus 1 segment) and instead of having all the LEDs stay dimly lit only a few stayed on. This is confusing me quite a bit...

Re: LEDs not going fully off

Wed, 02 Apr 2014 00:47:40 EDT

TheMG posted : What happens if you disconnect and ground the gate of the MOSFET? Do the LEDs turn off completely?

If so, then it is output from the microcontroller that is causing it. An oscilloscope should be able to confirm that.

Re: LEDs not going fully off

Tue, 01 Apr 2014 23:54:11 EDT

Nick posted : i wired mine like the red color but same behavior...No change when I removed the resistor.

Re: LEDs not going fully off

Tue, 01 Apr 2014 21:38:00 EDT

shdesigns posted : Yes, that makes sense. They make no sense on the drain pins of the MOSFETs.

Re: LEDs not going fully off

Tue, 01 Apr 2014 18:45:54 EDT

mackey posted :

said by shdesigns:

Those resistors are in wrong and i see no reason for them.

I think it's to make sure they stay off during power-up as the I/O pins start tri-state.

And the Red is wired right, it's just the Grn and Blu which are wrong.

/M

Re: LEDs not going fully off

Tue, 01 Apr 2014 18:29:09 EDT

shdesigns posted : Those resistors are in wrong and i see no reason for them.

Try removing them, they will keep a small current through teh LEDs. The may have been intended to go to +12V instead of GND.

LEDs not going fully off

Tue, 01 Apr 2014 17:53:11 EDT

Nick posted : I am playing with an analog Rgb strip from which I cut s small section to play with. I am trying to build a simple circuit using an arduino but when I turn the pin off the LEDs don't go fully off they just dim... I am following this wiring:
»www.jerome-bernard.com/b ··· arduino/

I am only wiring up one color and using the basic blink sketch...
--
-Stupidity, like hydrogen, is one of the basic building blocks of the Universe.
-Insanity: doing the same thing over and over again and expecting different results.

Misc

trim.058FBF2···9B4E.MOV
5,375,458 bytes