IRF530N mosfet doesn't cut off

Other Hardware General Electronics
1

Hi everyone,
Kinda new to the forum.

I have a couple of IRF530N mosfets with wich I try to hook my RGB strip to my Arduino.
I've wired everything according to this explaination: http://bildr.org/2012/03/rfp30n06le-arduino/ (only thing I changed was the fact that the IRF530N has a different gate-drain-source order).
First I'm trying just to get one colour working.

My problem is that when I try a very simple blink sketch (only HIGH and LOW, in order to avoid any possible issues with analogWrite values not being in range in order for the RGB strip to visibly light up), the RGB-strip doesn't go all the way out. It seems to properly fully light up when HIGH, and it also seems to dim a bit when pulled to LOW, but it won't go all the way out.

I've installed a troubleshooting-led on my breadboard between ground and the hypothetical PWM-pin, (which now only writes HIGH or LOW, as I explained) in order to see if the Arduino seems to give a good signal and also to see if the Arduino writes HIGH or LOW in the fully-on and a bit dimmed state of the RGB-strip. I'm sure now the Arduino writes HIGH when the RGB-strip is fully on, and it writting LOW when the RGB-strip is a bit dimmed.

The blink sketch has a 2 full seconds delay, to make sure the resistors are able to pull the PWM-pin fully to ground, in case the Arduino for some reason doesn't do that properly. I've tried putting a second pull-down resistor in parrallel in order to have it being pulled down faster, but it doesn't seem to help.

Has anyone seen this phenomenon before? I've tried to search the web for it, couldn't find anything.
I have very little experience in electronics, so please know you're dealing with a noob.

2

IRF530N is not a logic MOSFET.

Show us a good image of your wiring.
Posting images:
https://forum.arduino.cc/index.php?topic=519037.0

3

My error Arduinius -- sorry. Didn't follow your link.

If you disconnect your MOSFET gate from your Arduino and connect it to GND, does your strip go off?

Data sheet for your mosfet here: datasheet IRF530N. The pin designations (gate/drain/source) are shown on the last page.

4

the RGB-strip doesn't go all the way out. It seems to properly fully light up when HIGH, and it also seems to dim a bit when pulled to LOW, but it won't go all the way out.

It looks like your Arduino is driving the gate high enough, at least high enough to turn it on somewhat.

The fact that the lights are not turning off fully suggests that the FET is not wired up correctly and you should check your pin out for this part again. The FET should turn off when the gate is low, not when the gate is high.

When you have the pinout correct then you will need a logic level FET to turn the LEDs on fully. You might think they are fully in at the moment but wait and see how bright they are with a proper FET. They will also get not as hot as they will with the FET you have

5

First of all: thank you all for your quick reactions!

larryd:
IRF530N is not a logic MOSFET.

Looking at the datasheet the gate-to-source voltage shows at 5 volt a 20 amps drain-to-source amperage right? Should be enough right? I see that you are right about the IRF series not being a logic level mosfet, but still, it should work right? Many people seem to use non-logic level mosfets, like the IRFZ44N. Anyway, that isn't the problem at hand.

CurtCarpenter:
My error Arduinius -- sorry. Didn't follow your link.

If you disconnect your MOSFET gate from your Arduino and connect it to GND, does your strip go off?

Data sheet for your mosfet here: DigChip IC database. The pin designations (gate/drain/source) are shown on the last page.

Nope, it doesn't go off. I've checked the datasheet multiple times, checked my setup multiple times, rebuild it from scretch 4 times. Nothing seems to solve it.

Grumpy_Mike:
It looks like your Arduino is driving the gate high enough, at least high enough to turn it on somewhat.

The fact that the lights are not turning off fully suggests that the FET is not wired up correctly and you should check your pin out for this part again. The FET should turn off when the gate is low, not when the gate is high.

When you have the pinout correct then you will need a logic level FET to turn the LEDs on fully. You might think they are fully in at the moment but wait and see how bright they are with a proper FET. They will also get not as hot as they will with the FET you have

Couldn't agree more with you. Just so we're clear, this: "The FET should turn off when the gate is low, not when the gate is high. " is already happing right now, only is it not turning off entirely...

I guess what I'm trying to ask is if there's anyone who's seen this kind of behaviour with what maybe turned out to be faulty mosfets..?

Later on today I'll send a picture of my wiring diagram.

6

A IRF530N is rated at 10 volts, don’t use this with the Arduino.

RDS(on) Static Drain-to-Source On-Resistance ––– ––– 90mΩ VGS =10V,ID =9.0A „

7

"The FET should turn off when the gate is low, not when the gate is high. "

100% correct.

Looking at the datasheet the gate-to-source voltage shows at 5 volt a 20 amps drain-to-source amperage right? Should be enough right?

I don't think you can make that conclusion because that graph is only for a 20uS pulse width of 50V between the drain and source.
The one you want to see is the RDS(on) value in the table which shows 90mR for a VGS of 10V and a drain current of 9 amps. At 5V is is barely turning on, but it will be a bit on. So while it might appear to function the on resistance will be high and so it will dissipate a lot of heat.

I have not known a faulty FET to behave like this but there is always a first time. But just to be on the safe side show a photo of your wiring.

8

Do you have another IRF530N you can try? I think it should be working, but that you may have a bad MOSFET.

You can't be sure from the data sheet, but it shows a Drain-to-Source Leakage Current (IDSS) of 25 microamps with VDS = 100V and VGS=0 (that is, with the gate connected to ground). It would be odd if that leakage current increased as you lowered VDS with the gate grounded.

The fact that your strip is still lit with VGS=0 suggests to me that you have more than a few microamps of leakage current, either through the MOSFET, or somehow around it. You don't by any chance have the mounting tab of the MOSFET connected to anything do you? Remember, it's connected to the drain (pin 2).

9

Hi,
Welcome to the forum.

Please read the first post in any forum entitled how to use this forum.
http://forum.arduino.cc/index.php/topic,148850.0.html.

I've wired everything according to this explaination: http://bildr.org/2012/03/rfp30n06le-arduino/ (only thing I changed was the fact that the IRF530N has a different gate-drain-source order).

What makes you think the RFP30N06LE and the IRF530N have different pinouts?

They are the SAME. Gate-Drain-Source, left to right looking from the front with the leads pointing down.

Cab you post a picture of your project so we can see your component layout.

Have you included the 10K resistor from gate to source?

Thanks.. Tom.. :slight_smile:

10

Arduinius:
First of all: thank you all for your quick reactions!

Looking at the datasheet the gate-to-source voltage shows at 5 volt a 20 amps drain-to-source amperage right? Should be enough right?

Listen to us. The device is not a logic level MOSFET.

You've seen a graph labelled "typical characteristics", and assumed it is guaranteed - it is not, MOSFET variability can be +/-1V on any gate voltage for a start...

And secondly you've misinterpreted the graph completely - this is not a graph of the device as a switch,
its a graph of the device as an analog amplifier with 50V between source and drain, which at 20A means
1000W is being dissipated in the die (typically) with 5V gate drive. The die is in a state which if maintained
for more than a few microseconds will lead to it exploding... In fact due to the variability between devices
it might be anywhere from an amp or two to 35. The on-resistance is somewhere in the range 1.5 to 30 ohms,
not <= 0.09 ohms as would be the case for the device properly on. 5V gate drive is not enough.

The graph is actually there to estimate the switching losses during the few microseconds or less that
it is halfway between being on and off. It can be used to estimate transconductance too.

The only specification in any MOSFET datasheet you can use to determine its use as a switch is the
value of Rds(on) quoted for a particular Vgs value. The "maximum" column for that specification is
the only guarantee of its losses as a switch when turned on. With 10V (or more) gate drive, and 9A (or less) of drain
current, these devices are guaranteed to be 90milliohm or less.

11

But the problem here, as I understand it, isn't turning the MOSFET ON, but turning it OFF.

I don't think anything magical or dramatically non-linear is going to happen to the transconductance at VGS below 4.5V, VDS < 10V, or to IDSS with VGS=0 and VDS < 10V. It would be interesting to see for sure though. Does anyone have a power MOSFET data sheet that would prove otherwise? It would be a good to know!