I have an issue with a very simple small board I designed in EasyEDA pro (I'm a novice so if the answer is easy, my apologies). I'm not sure I correctly connected the MOSFET in the schematic. Reference to Datasheet: https://atta.szlcsc.com/upload/public/pdf/source/20140314/1457707013919.pdf
Part Number:
IRLR7843TRPBF

I've spent a fair amount of time trying to understand this chip before using it, so I'm thinking I got it right, but it doesn't function as expected.

I designed a single channel MOSFET board for use in an LED-lit device that is being driven by an Arduino Uno. The Uno pulses a 5V output every second. The schematic below doesn't show the arduino, but both arduino and this board are on the same 0v. I am supplying 5v to the arduino with a buck transformer going from 24v to 5v. ardunio and power supply are checked and verified to function.

The board I designed gets a 5V input (from the arduino board) to Pin 2 of component P1. 24V is connected to Pin 4 of component P1, 0v is connected to Pin 3 of P1. 24v LED strip lighting (which only requires a 24vdc and 0vdc connection to light) is connected to Pins 1 and 2 of component U1.

Issue is, the LED illuminates with or without input signal. I probed pin 2 to ground and can see the alternating 5v/0v swing in the output from the arduino. I probed Pins 1 and 2 on component U1 when not powered to confirm I didn't just have a short on the MOSFET, and I was able to confirm.

 

Q1 source should go to 0V.
Led anodes should go to 24V.
Is the LED designed to take 24V DC?

 

Q1 source should go to 0V.
Led anodes should go to 24V.
Is the LED designed to take 24V DC?

So if Q1 should be connected to 0V, and LED anode is connected to 24V, would that mean that LED cathode is connected to drain?

or in other words, LED load is between 24v and MOSFET, and MOSFET is switching the ground?

The LED is designed to take 24vdc. I'm using a 50mm segment of this LED strip:
https://www.superbrightleds.com/cus...e-light-24v-ip20-250-lmft-stn-x-b6a-08c1m-24v

 

The board I designed gets a 5V input (from the arduino board) to Pin 2 of component P1. 24V is connected to Pin 4 of component P1, 0v is connected to Pin 3 of P1. 24v LED strip lighting (which only requires a 24vdc and 0vdc connection to light) is connected to Pins 1 and 2 of component U1.

It's difficult to understand what you're saying. It sounds like 24V going to the source of the MOSFET. It's unclear whether you have the Arduino 5V supply, or a signal, connected to the gate. How the LEDs are connected is also unclear because the source goes to some unspecified terminal on U1.

At any rate, you appear to have the MOSFET connected backwards and the parasitic diode is conducting.

 

It's difficult to understand what you're saying. It sounds like 24V going to the source of the MOSFET. It's unclear whether you have the Arduino 5V supply, or a signal, connected to the gate. How the LEDs are connected is also unclear because the source goes to some unspecified terminal on U1.

At any rate, you appear to have the MOSFET connected backwards and the parasitic diode is conducting.

Here's a better schematic showing 24v and 0v connected to what I'd term a bus on the board, and how I have the LED strip connected.

I'm trying to switch 24v to LED, but it sounds like I should be trying to switch 0v to ground.

 

I'm trying to switch 24v to LED, but it sounds like I should be trying to switch 0v to ground.

Yes, since you are using an N-MOSFET.
(Note that, in normal operation, the drain of an N-MOSFET must always be more positive than the source, and you have it connected opposite of that which forward biases the substrate drain-source diode).

So connect the MOSFET source to GND, its drain to the (-) of the LED strip, and the (+) of the LED strip to 24VDC.

 

I'm trying to switch 24v to LED, but it sounds like I should be trying to switch 0v to ground.

You need to use a P channel MOSFET to switch high side. You have the MOSFET source and drain reversed causing the parasitic diode to always be conducting and you should be switching low side.

Since you're a newbie.

1. You have the MOSFET drawn sideways.
2. If you shift P4 down, you can eliminate 3 wire jogs.
3. We don't draw wires over text.
4. We don't draw wires over components.
5. We don't like unnecessary wire jogs.
6. Schematics shouldn't use colors. Print to PDF in black and white and turn off the grid.

 

So if Q1 should be connected to 0V, and LED anode is connected to 24V, would that mean that LED cathode is connected to drain?

or in other words, LED load is between 24v and MOSFET, and MOSFET is switching the ground?

The LED is designed to take 24vdc. I'm using a 50mm segment of this LED strip:
https://www.superbrightleds.com/cus...e-light-24v-ip20-250-lmft-stn-x-b6a-08c1m-24v

You got it. Does it work now?
Just checking that the LEDs had built-in current limiting. That's OK.

 

You got it. Does it work now?
Just checking that the LEDs had built-in current limiting. That's OK.

This is a board that I am making using EasyEDA, so it'll be a week before I know. I spent $20 already to find out I connected the MOSFET incorrectly.

The LED Strips do have built in resistors every so often to limit current.

 

You need to use a P channel MOSFET to switch high side. You have the MOSFET source and drain reversed causing the parasitic diode to always be conducting and you should be switching low side.

Since you're a newbie.

1. You have the MOSFET drawn sideways.
2. If you shift P4 down, you can eliminate 3 wire jogs.
3. We don't draw wires over text.
4. We don't draw wires over components.
5. We don't like unnecessary wire jogs.
6. Schematics shouldn't use colors. Print to PDF in black and white and turn off the grid.

Thanks for the feedback and the patience, Dennis. Drawing updated and modified as mentioned and agree that this is much clearer. Please understand, my lack of MOSFET knowledge isn't due to lack of trying.

circuit revised to show that the drain is connected to the incoming voltage from the LED strip, and Source is grounded.

 

This is a board that I am making using EasyEDA, so it'll be a week before I know. I spent $20 already to find out I connected the MOSFET incorrectly.

Get your scalpel out and a few bits of wire, and modify it to make it work, like real engineers do.

 

Get your scalpel out and a few bits of wire, and modify it to make it work, like real engineers do.

I have no good reason not to, honestly. I hope to get a chance to do so later today.

 

I have no good reason not to, honestly. I hope to get a chance to do so later today.

After a bit of practice you'll soon be able to solder a wire on the pin of an 0.5mm pitch LQFP when you've missed a track off the board.

 

Drawing updated and modified as mentioned and agree that this is much clearer.

Agree. Below is a version that removes most of the wire crossings.

I messed up on the 2 pin connector. It's a custom component and I didn't notice until now that the terminal numbers need to be moved.

Normally the flow is left-to-right and top-to-bottom. I violated that rule because I thought it made the schematic more readable.

Please understand, my lack of MOSFET knowledge isn't due to lack of trying.

No problem. Being willing to listen and learn is about the only prerequisite for getting help.