Issue with N-channel MOSFET Transistor

[Leolicos]

http://www.digikey.com/product-sear...eywords=MOSFET+N-CH+250V+10A+TO-220FM&x=0&y=0

I just recently bought this N-channel MOSFET transistor and when I hook the first pin (gate) to 5V, the second pin (drain) is just an open wire, and the third pin (source) to 24V, everything works fine; I use the multimeter and check the open wire (drain) and it reads 24V..

The problem is when I remove the gate - I don't have it connected to anything - the source is still hooked up to 24V, yet the open wire (drain) still reads 24V. Is the source outputting too much current or something? It's 24V 6.5A. The drain-source breakdown voltage is 250V (I checked the data sheet. It's in the link)

Why am is the drain still reading 24V when the gate isn't hooked up to anything?

Thanks,
Leo

 

[Baluncore]

Because the huge gate capacitance is still charged and must be discharged before the MOSFET can switch off.
You cannot identify on/off state by high impedance voltage measurement, only by current flow.
The source of an N'chan MOSFET should be to zero volts, not 24V.

 

[Leolicos]

I hadn't tried that before, but I'm still getting the same thing. I even removed all the capacitors I had in the circuit.

Should the drain be the 24V? Maybe I got the wrong type of transistor.. I need to switch on/off 24V 6.5A with 5V 40mA (Arduino).

Here's my schematic if this helps.
https://www.circuitlab.com/circuit/fphdn3/automated-led-lighting-system/

For testing purposes, I'm bypassing the Arduino and hooking up the 5V in that normally goes to the Arduino directly to the Gate of the MOSFET.

Maybe I got the wrong transistor what I need it to do?

 

[Baluncore]

Your “MOSFET – Load” circuit is upside down...

Connect the source of the IRF530 to ground = 0V. Connect the gate through the 1k resistor as shown in your diagram. Connect the load between the drain and the +24V rail.

You must use an N-Channel MOSFET such as IRF530 as a low-side switch because control voltage on gate is switched between 1 and 4 volts only. A P-channel high side switch will not work because the control logic is close to 0V, not 24V.

The capacitance I referred to was the inherent base capacitance of a MOSFET. That must be charged and discharged through the 1k resistor every time the circuit switches.

 

[Baluncore]

The data sheet for your RCX100N25 shows a GATE THRESHOLD VOLTAGE VGS(th) range from 3.0V to 5.0V at VDS=10V for ID=1mA. That might be a problem if your logic level does not rise all the way to 5V. In that case you may need a pull-up resistor on the logic output.

What are you switching? LEDs are OK, but if the load is inductive such as a relay coil then you will need a flyback power diode reverse biased across the load to prevent the inductive kick destroying the MOSFET when it turns off.

 

[Leolicos]

ohh, so I did get the wrong transistor.. (Transistors are fairly new to me, so I'm still learning.)

I'm switching a 2d array of LEDs where each row has a resistor on it.

Do you happen to have any references I can look at to further my understanding of transistors? Most videos on youtube just go over my head..

 

[Baluncore]

If you jump in the deep end you will drown. It is probably safer to provide the required details here and then build your knowledge on that. I can give you a circuit diagram that solves your immediate design problem once I have enough details.

A 2D array of LEDs will need both row and column control.
How many rows? by how many columns?
LED matrix arrays may be multiplexed with cathode or anode common to the row or column.
Are the LEDs row cathode, or row anode?
Who is the manufacturer and what is the part number? Or a link to the LED data sheet.

24 volt is quite a high voltage to drive a LED array. But that can be sorted in the design later.

But in the meantime, play with the MOSFET you now have.
Connect it's drain through a couple of low wattage 12V automotive globes in series to +24V.
Connect the source to ground = 0V. Switch the gate between 0V and 12V to turn it on and off.
Remember that power is current times voltage. A switch likes only current or voltage, so avoid the half ON situation as that will have the highest power dissipation in the MOSFET.

 

[Leolicos]

That would be great, I really appreciate the help!

The only problem that I know of is the switch. I basically just need the Arduino to control the on/off state of the LED Panel.

I had a working circuit that turned on and off properly that looked like this: https://www.facebook.com/photo.php?...56910701019121_1032331824_n.png&size=1074,282
Everything is the same except I'm using 2 relays as a switch. The Arduino's 5V signal essentially turns on 24V 6.5A to power the LED panel. Though, this caused a different problem where - as people told me based on what I said was happening - the ground of the 24V relay would leak into the ground from the 12V relay and the Arduino's power would shut off if the relays were on for more than a minute.

Someone suggested eliminating the relays and using a transistor, so that's where I got the current schematic; I just need to find the right transistor to be able to switch on/off 24V 6.5A to power the LED panel.

The LED panel was constructed using this site that puts together a 2d led array; it's not that I want to control each one individually, it's they're either all on or all off.
http://led.linear1.org/led.wiz

 

[Leolicos]

I'm using 2 different LEDs where each row is the same color, but the adjacent rows is the opposite color.

Red: http://www.digikey.com/product-search/en/optoelectronics/led-indication-discrete/524729?k=P408-ND

Blue: http://www.digikey.com/product-sear...&lang=en&site=us&keywords=365-1184-ND&x=0&y=0

All together I have 20 rows (10 red and 10 blue) with 10 leds in each row

 

[Baluncore]

I am surprised that the Arduino digital pin was happy to drive a 12V relay coil directly with only 5 volts. It must be close to marginal and certainly is not a reliable design. The Arduino's digital pin must have some flyback protection to survive the negative spike when turning off the inductive load.

Your diagram shows the relay coils and LED array connected to ground with high side switches. Conceptually, switching power connects one side of the load to 0V ground. But N-channel MOSFETs are much cheaper than the slower P-channel versions so low side switching is much preferred, the load must then be connected between the MOSFET drain and the +supply.

“the ground of the 24V relay would leak into the ground from the 12V relay” This does not really seem to make sense as those grounds are, and should be, connected on your diagram.

“the Arduino's power would shut off if the relays were on for more than a minute.” That is probably because the 24V power supply is unable to support the current required by the LEDs and so is shutting down due to a thermal overload. The 7812 and 7805 regulators also have internal thermal shutdown, if they get hot, (maybe because they are not mounted on a heatsink), then they could turn it all off while they recover. The progressive elimination of chunks of the circuit is a good way to isolate a problem. Did you test it without the LEDs connected to the 24V relay, or without the 24V relay coil connected to the 12V relay contacts. That may help identify the problem. What gets hot when it runs for a minute? 24V power supply? 12V reg? 5Vreg?

Early power MOSFETs used to need between 4 and 8 volts on the gate to turn them on, so they were controlled by 0V to 12V gate voltages. They have since been refined to switch cleanly on logic levels between 0V and 5V. Unfortunately the MOSFET you selected is not guaranteed to switch fully on at 5V. But it might work for testing so try it.

Connect your LED array(–) terminal to the MOSFET drain and the LED array(+) to the +24 volt supply rail. Connect the MOSFET source to 0V=Ground.
Connect your 1k gate resistor and check you can switch the LED array on and off by manually switching the free end of the 1k between 0V=OFF and 12V=ON.
If you can do that then you are well on your way to sorting out this design.

If that does not work then it may be because the MOSFET has died or you have the pinout interpreted wrongly. Let me know how it goes.

 

[schip666!]

I'm not sure your question has been answered, so, to beat a dead horse, here's a schematic for a MOSFET output driver board using a PIC rather than an ATmega controller:
http://www.etantdonnes.com/DOC/USBPIC/doc/USBPIC_trans_schematic.png
See the right side where there are 8 (enhanced) N-channel fets in an open drain configuration (with the source pin at ground, pulling down a load when switched on). When I started out on this I found that I needed "enhanced" N-channel devices in order to make them act just like good-old-NPN transistor switches, but I no longer find the "enhanced" word on specsheets so I'm not sure if there is a distinction now.

I think your IRF530 (or was that what someone else suggested?) is the right transistor for the application when used in the pull-down config. You should be able to swap the load and transistor in your schematic but may have to invert the drive signal from the controller to get ON and OFF in the right places.

I have a version of that board with lower-power (circa .5A max) drivers for an 8x8 LED matrix. If you rummage through the full doc at: http://www.etantdonnes.com/DOC/USBPIC/USBPICindex.html, maybe you'll find some useful tidbits.