Can someone take a gander at this diagram and see which, if either, is correct (Figure A or Figure B)?

I am trying to isolate my LEDs from the MCU so i can give them a little more voltage. I am doing both sides of the LED because this is part of a larger project. Also, after drawing the diagram i realized i am missing a resistor at the LED. i will add that in later.

Here is a diagram of the project i have completed and have working so far (prior to installing the MOSFETs):

I have PC0-5 set as input and rotate through them at 2ms each setting them to output (LOW). PD2-4 i control with PWM so i can control the brightness of any single LED. When i have the LEDs set to full brightness it is still dimmer than i would like. They need to be bright because they will be placed inside ping pong balls (it looks really cool) so the colors mix (RGB = Any Color).

And a follow-up question: Since i am going to have any single LEDs on for 2ms out of 12ms and the PWM will actually make it less than that, can i, technically speaking, give the LED more voltage than what it is rated for? For Example, if i have a LED rated for 2 volts can i safely give it 4 volts if i never take it above half brightness with PWM?

Thanks and any suggestions are welcome!

Bryan

Bryan,

It's late, so I didn't get a chance to really look at all of your thread (my apologies), but initially

1. Drawing 'A' is the more correct of the two
2. Both drawings are showing the BJT schematic symbol, not the FET symbol- you should fix that.

Something else to consider- put your LED in parallel, not series.

BM

BM,

Thank you for the response Boba. The main thing i was wondering is if Figure A is even a correct way of connecting devices. All the "tutorials" i looked at online show the load, the LED in this case, being on the Drain side of the FET. i was not sure if connecting the load to the source side would work properly. Since typing this question i have connected up a little test and the LED is about half as bright as LED's connected directly to the MCU even though i am trying to give it +5v through the FETs (should be very bright or even blow out).

here is the diagram of my "test":

The problem with A is that the top FET doesn't get a large enough gate-source voltage drop. B is better in that regard, but you have to pull that gate pin low to turn it on. I don't see the need for pull-down resistors either.

Bryan,

The only way to actually discuss your circuit is if we know what transistor you're actually trying to work with. FETs behave differently based on kind.

BM

Boba,

The N-Channel are the ones that came with my NerdKit. They are 2N7000. I do not currently have any P-Channel, this is why i was trying to use all N-Chan in this project if possible.

Bret,

Please help me see if i am understanding MOSFETs correctly. i am currently in "learn" mode. Using the bottom Diagram (the one with +5V). here are my assumptions:

(1)it takes 4.5 volts (Vgs) to turn "on" the FET.

(2)I am supplying 5v (i know it is more like 4.7 on the nerdkit, lets say 5 for my theory lesson)

(3)Only the PD2 and PC0 are currently connected.

(4)The LED has a 2 volt drop across it. I measured from Ground to the + side of the LED and got apx +3 volts.

If we just "say" my assumptions are correct would this be true; to turn on the LED i would have to give PC0 at least 4.5 volts and i would have to (theoretically) give PD2 at least 7.5 volts? as far as the MOSFET is concerned am i thinking down the right path?

My P-Channel MOSFETs came in the other day (DigiKey.com) and everything seems to work exactly like Figure B at the top of the thread. i did remove the "pull-down" resistors. it is also using 5v instead of 3v.

here is a link to a video of my project before adding the MOSFETs. it is wired up like the middle diagram (directly connected to MCU)

Chase Video