Is my MOSFET broken or my circuit is defective?

[dzem]

I've made a simple circuit (see attached image) with mosfet IRFZ44N and some part of the LED strip (12V, 0.3A). Both parts I've found when making some orders, so they are not new. The circuit is supposed to control the LED strip (on and off) with a small voltage input. I'm having a problem with running this - the strip is not shining as bright as it does when directly connected to the 12V power source. So, I've added a voltmeter to find out the problem but it only confused me.

There are two switches - SW1 providing 3.3V input to the transistor gate and SW2 connecting LED strip "-" with transistor drain. So, depending on the switch's position the circuit has four states. I've made voltmeter readings for all of them. Here it is:

SW1: OFF, SW2: OFF, expected: 0V  - yes, this is what voltmeter shows
SW1: OFF, SW2: ON,  expected: 0V  - yes
SW1: ON,  SW2: OFF, expected: 12V - yes
SW1: ON,  SW2: ON,  expected: 12V - ...voltmeter shows 7.5V... and LED strip is barely shining.

Is there anything that I'm doing wrong or the mosfet is partially broken? I'm still a beginner and I do not know where this weird reading 7.5V can come from. Any thoughts?

PS. As a secondary, by the way, question - does my drawing in Kicad meet the standards of such drawings, or will you change/correct something?

Thank you

[MarkF]

Look at the datasheet.
The gate-threshold-voltage is between 2-4V.
With a 3.3V power supply, you're probably not able to turn the MOSFET completely 'on' or at all.

[langwadt]

3.3V is not enough to turn the mosfet fully on

[oPossum]

IRFZ44N requires 5 to 10 volts Vgs. See figure 3 in the datasheet.

https://www.infineon.com/dgdl/Infineon-IRFZ44N-DataSheet-v01_01-EN.pdf?fileId=5546d462533600a40153563b3a9f220d

[wraper]

IRFZ44N requires 5 to 10 volts Vgs. See figure 3 in the datasheet.

https://www.infineon.com/dgdl/Infineon-IRFZ44N-DataSheet-v01_01-EN.pdf?fileId=5546d462533600a40153563b3a9f220d

Even 5V is a bit too low for guaranteed operation.

[Kim Christensen]

IRFZ44N requires 5 to 10 volts Vgs. See figure 3 in the datasheet.

https://www.infineon.com/dgdl/Infineon-IRFZ44N-DataSheet-v01_01-EN.pdf?fileId=5546d462533600a40153563b3a9f220d

Even 5V is a bit too low for guaranteed operation.

Perhaps he was confusing it with the IRLZ44N? Though even that one would be a bit marginal at 3.3V Vgs since RDSon is spec'd at 4, 5, and 10 Vgs but it would be fine at 0.3A Id.

[BrianHG]

As everyone said above, if you want this circuit to work correctly, you need a mosfet which guarantee full turn-on way down at 2.5v if you expect to guarantee clean operation at 3.3v without residue voltage loss and heat.

Look for a replacement logic-level mosfet, one which runs on the really low voltage side, not the ones designed for 5v.

Your other choice is to use a gate-driver IC with your current mosfet.  A gate driver IC will have a connection to your 12v supply so that when you drive that IC's input above 0.7v, the output of that gate driver IC will send the 12v to the gate of the mosfet.

[BrianHG]

Only the Hexfets type mosfets will operate with such a low gate voltage reliably.

Example: https://www.infineon.com/dgdl/irlhm620pbf.pdf?fileId=5546d462533600a4015356638be325a3

[wraper]

Only the Hexfets type mosfets will operate with such a low gate voltage reliably.

Example: https://www.infineon.com/dgdl/irlhm620pbf.pdf?fileId=5546d462533600a4015356638be325a3

HEXFET (technology/trademark) has nothing to do with this whatsoever. In fact IRFZ44N is made with HEXFET technology too and it's a way how to achieve power density by making a MOSFET with many identical small cells in parallel.

[Kim Christensen]

I assume that your end goal is to control the LED strip with a microcontroller and you'd like to use parts you already have on hand.
If so, you could use an NPN transistor, with pullup resistor to 12V, to drive the IRFZ44N's gate. Assuming this is just for low speed on/off control this should be fast enough. There'd be a bit of current flowing in the pullup resistor when the LEDs are off and the logic would be inverted. (MCU high = off, MCU low = on) Don't know if that'd be an issue.

[Apollo1234]

if you need a low gate opening voltage use a irl... mosfet

[CaptainBucko]

In general, when you are starting out, new and experimenting, and if you are not an expert, and not driven by a low cost requirement, I would avoid non isolated designs. A non isolated design means that if an electrical component fails, you could theoretically dump for example, 12v into 3v3, and blow stuff up.

For example, assuming the OPs finds a FET that works in this design, if the FET is blown up dead short drain to gate (ie: A slip of a ground problem shorts out the FET, for example), then there is only a 330R resistor between 12v and 3v3 supplies.

During development, its normal to connect a PC via USB to your micro controller, so now you potentially are feeding 12v onto the USB. That is a dangerous outcome. I know there are quite a few what-ifs in this argument, but shit happens, like what happened to this guy: https://www.reddit.com/r/AskElectronics/comments/1c8f27a/power_serge_from_arduino_laptop_not_booting_back/

My recommendation is that if you want to control a 12v device from a 3v3 microcontroller, use an isolated design (relay or opto-isolated circuit). The uC can pull down a LED in the opto-coupler, and then you can do whatever (mostly) on the BJT side to control the 12v stuff (or use it as a FET driver).

I know this is not the OPs question, but I would really hate to see anyone blow up a PC USB port because they did not understand the risks they were taken. Obviously, this is totally different if you are building to a cost, you know exactly what risk you are taking, and its unlikely a slip of a probe will ever occur in the real world (like a packaged consumer device).

[dzem]

Thank you all for your responses. I've modified my circuit by adding a TC4420 driver. Could you please verify, if
- it is supposed to work correctly
- it is safe, assuming that 3.3V input may be the microcontroller output

[wraper]

Thank you all for your responses. I've modified my circuit by adding a TC4420 driver. Could you please verify, if
- it is supposed to work correctly
- it is safe, assuming that 3.3V input may be the microcontroller output
(Attachment Link)

You cannot just leave TC4420 input floating, there is not pull down. Also just get a proper MOSFET, you don't need a gate driver for this unless you will PWM those LED strips.