Does this MOSFET makes sense for this project? Should I identify the continuous drain current value?

Hi amazing people. I'm planning to build an aeroponic watering system. So I have a DC 12V water pump (15L, 4.2bar) LightEU DC 12V pressure pump, pressure water pump liquid pump for marine, boats, yacht, caravan, camping, outdoor, garden. : Amazon.de: Automotive

I'm powering it from mains and using an AC adapter to 12V at 8A.

I want to automatically switch the power of the pump by using a MOSFET, and I want to just see if I am using the right MOSFET. I have been reading the MOSFET datasheet but I cannot successfully determine if I have the right MOSFET. Also, to makes things easier (hopefully), I've found a MOSFET module:

component number is FR120N. I found a datasheet:

So, I was a bit concerned using other MOSFET modules which advised to run use more than 5A of current.

THe FR120N markets its product as "100V 9.4A". I guess the mosfet and the mosfet modules can have different max current ratings. But, anyhow should I be looking at the continuous drain current on the data sheet? It is 13A. Also, do I need a heat sink?

Thanks :slight_smile:

That MOSFET is not a logic level MOSFET, and may not work at all with Arduino. It is also very poor choice. The Rds(on) is over 0.1 Ohms at Vgs = 10V, 15A.

You can do much better than that. Also, make SURE you know the start/stall current for the pump and that the pump power supply can provide it. It is much higher than the typical running current.

1 Like

The RDS(ON) is given for Vgs @10v, this is not a good choice for your application i.e. when direct connection to an Arduino output pin.


You could add an intermediary NPN transistor to allow control from the Arduino though.

1 Like

I looked closer at the data sheet, and indeed that MOSFET won't work at all with Arduino. This plot is the key, and shows that the drain to source current is typically less than 2A maximum when 5V is applied to the gate.

Capture

Something like this logic level MOSFET would be a better choice.

1 Like

The module mentioned by the author has a PC817 optocoupler on board.
What circuit after the optocoupler is unknown.
FR120N Module

1 Like

Well this changes things, would be nice to see a schematic showing all the components :frowning: .

I'm not sure if this is exactly his scheme, but it is very similar.


It will probably work with Arduino.

Just for a 12V and 8A pump, I would choose a similar module with D4184 and a slightly more powerful power supply.

1 Like

No, you look at the Rds(on) keeping heat sinking in mind. The dissipation is I-squared-R for a MOSFET when on, so for instance 8A and 0.1ohm would dissipate 6.4W, far too much unless you have a medium sized heatsink.

Choose a MOSFET with 0.005 ohms on-resistance and that 8A dissipates 0.3W, so you wouldn't even need a small heatsink.

So long as you do these power calculations and provide appropriate heatsinking for the power being dissipated you won't have to worry about current ratings (which are only usually applicable to using heavy water-cooling thermal management, not practical in 99.9% of applications).

And if you want to switch from 5V, use a logic-level MOSFET.

For high currents sometimes you'd parallel a few MOSFETs, using separate gate resistors.

1 Like

No, that is clearly non-functional. Should be this one:


OK, there was one wire missing in yours!

This one.

2 Likes

Thanks for the schematic.

What do you mean with a slightly more powerful power supply? Do you mean supply power with more than 12V? or more current to compensate for the possible high stall current?

I found a module with a D4184 MOSFET. Hopefully, it has the same resistors and components as the schematic:

It is exactly the same module as I just cited.

Just more expensive on Amazon and you would probably get it faster if it were available as your link says it currently is not. :face_with_raised_eyebrow:

1 Like

Ok, thanks! It's seem to be available for my location.
I see that there is a flyback diode shown on the load on the schematic you have attached. Do you know if I need one, or it would be safe to use one on the 12V pump that I'm using? And do you have a reason for not using an AC adapter to supply 12V DC to the pump and module? Thanks.

It is always safe, and highly advisable.

If you do not propose to use PWM, then any diode adequately rated for at least the voltage and current that the pump uses, will be adequate.

I don't believe I advised against using an AC adapter. That would be the proper way to do it, as long as it is rated adequately for the stall current of the pump.

1 Like

The module has an optocoupler for isolation. I think that is providing the switching for the MOSFET so the Gate voltage would be same as the supply voltage. As long as the supply voltage is over 10 and below the limit for the device it should switch OK.

1 Like

Yes, I meant that the power supply is rated at 8A and the pump consumes 8A. The power supply has no reserve. If the stall current is more than 8A, the protection of the power supply can be triggered during startup and overload.
I would take a 12V and 10-15A power supply.

1 Like

Thanks everyone for your responses. I asked the manufacture for the motor specs so I can calculate the stall current of the motor. However, they haven't responded.
I also noticed that the motor stops running at a threshold current, as seen in the figure below (in this case it stops at 13.48A)

Does this mean, I should ignore the stall current, since the pump will not run at a higher current than the stall current?

So I am planning to use a 12v 12.5A AC adapter as a power supply.

Here are more specs and instructions for a fuse. I guess I should install a fuse at the positive terminal.
image

image

Hello again. This is about reading MOSFET datasheets and understanding current ratings.

If I understand correctly, I should probably use 5V as Vgs because that way most likely 12.5 A, as the drain-to-source current, can be delivered through the MOSFET and therefore to the motor, correct? In other words, the curve for using 3.3V as Vgs might be too flat and reach its saturation limit below 12.5A.
image

Here is another figure, in case.
image

and the datasheet is here

That device has 165milliohm on resistance at 5V drive, so will dissipate 26W at 12.5A - that'll mean a lot of heatsinking is required.

I'd go for a much beefier MOSFET, something with less than 30 milliohm on-resistance perhaps (harder to find for p-channel I know). Heatsinking can be smaller and easier then.

Usually the graphs in such a datasheet are marked "typical", which pretty much means you should ignore them. Only the guaranteed on-resistance spec is of use to you. "Typical" is no guarantee of anything - specs vary significantly with temperature and between devices (manufacturing spread is large for FETs).

1 Like

I know (from previous comments I get) that my thoughts on such things are sometimes unpopular.
But when switching a "large" load like this, why do people automatically turn to mosfets?
I know it looks like an elegant solution. But mosfets for power switching require heatsinks, and a beefy flyback diode.

A simple relay here will be easier, smaller and cheaper. And it doesn't get hot.
Yes, you do need a small mosfet and a small flyback diode (e.g. IN4004) to operate the relay coil but that mosfet is tiny and cheap (e.g. a DIL mosfet), and no heatsink is needed.

The application described by the OP is a perfect match to the designed purpose of a relay.
I have built a number of watering systems and use a relay every time.

I hope this helps.

1 Like

Hi there. This is actually what I've been thinking the past few days, why not use relays.

So, thanks for the comment.