Is a MOSFET floating drain OK?

Thread Starter

browner87

Joined Nov 15, 2008
10
What I'm trying to do:

I have a bunch of RGB LED strips and a controller for them. Obviously to get different colors the controller uses PWM to dim the R, G, or B lines.

Now, I have several strips of LEDs that I want to be able to turn on/off separately and I'd rather use a MOSFET instead of a mechanical switch.

The problem:

The lines are presumably floating between PWM pulses and when a color channel is completely off (i.e. if I set it to red, the blue and green LEDs will be always off).

My thoughts so far:

I'd love to just put an P-channel MOSFET on the common anode for the line. But, I'm not sure if this will cause a problem because likely the drain will be in a floating state any time a color is off.

I know a floating gate causes problems, and a floating source can cause chip damage, but I can't find anything to suggest the effects of a floating drain. Can I presume that since the gate is referenced against the source it should be safe?
 

#12

Joined Nov 30, 2010
18,224
I think you'll be OK treating open drain MOSFETs like open collector transistors.
Of course, if you showed a schematic, I could be sure.
 

Thread Starter

browner87

Joined Nov 15, 2008
10
I guess this would be the essence of it. Repeated once for every R, G and B channel (and again for every strip I have).

(I figured there's no point trying to draw up the whole schematic because I don't know what's actually in the controller, but this is the best I can figure)
 

Attachments

Last edited:

#12

Joined Nov 30, 2010
18,224
Is the thing labeled, "1" a ground, and the PWM goes to the gate of the MOSFET?
If so, the drain is not floating. It is connected to the LEDs.

However, your LEDS must be equipped with resistors. Probably included in the manufactured strings, right?

Just a note, you're thinking upside-down to the way I think. For your circuit, the PWM controller must have a condition where the output matches whatever the positive voltage is. Most people use an N-channel MOSFET and ground the negative side of the controller to the negative side of the LED strings. However. you are allowed to think upside-down. If you get it wrong, the LEDs will tell you.
 

Thread Starter

browner87

Joined Nov 15, 2008
10
No, the "1" (labelled PWM_Circuit) is the ground for the LEDs, and it's those ground lines that do the PWM. The "arrow" is Vcc (common anode for all LEDs).

I do think very upside down, but I'm usually pretty normal with my circuit diagrams ;)
 

Thread Starter

browner87

Joined Nov 15, 2008
10
So there shouldn't be any negative effects of the fast switching of current flow on and off? I was slightly worried about that.

I'm afraid my brain refused to retain 99% of my formal education on MOSFETs and I've only spent a day now trying to relearn them :)

Thanks for all the help!
 

#12

Joined Nov 30, 2010
18,224
As long as the gate to source voltage is under control it should be OK. It's the gate to source that has a low voltage limit, no matter what MOSFET you buy. Usually 20 volts or less. I assume you are using a MOSFET rated for more than the voltage you are using. That's the drain to source limit. Voltage applied, voltage not applied, the MOSFET won't care as long as you don't get an inductive reaction that over-volts the drain. If that happens, a MOV is often used to protect a MOSFET.
 
Top