I am trying to understand how to properly select a MOSFET for my application. When reading the datasheet. In the attached sheet, in the 'Product Summary', the 'Vds' is 20 V, and the drain current is 2.8 at 4.5 V and 2.4 at 2.5 V gate voltage. Additionally, the 'Continuous Drain Current' is 2.8 A at 25oC and 2.2 A at 70oC. I have several questions about this:
Why is this different than the Continuous Source Current (Diode Conduction)? I would think the source and drain current would be the same since the current flows from the source to the drain, doesn't it?
The 'Power Dissipation' is 1.25 W at 25oC and 0.8 W at 70oC. Based on the 'Continuous Drain Current', and Id @ 20 V in the Product Summary, I would have though the power dissipation capability should be 56W (20 V* 2.8 A). Why is the Power Dissipation different than the specs in the 'Product Summary'.
When selecting a MOSFET, I would imagine the maximum current, maximum voltage, and maximum power all need to be considered. How should I properly get this information from a datasheet (such as the 1 attached).
The diode conduction does not use the drain to source channel: it conducts
outside the channel, so that current is not the same as Id.
Dissipation is how much (power) heat can get out of the package. When the
surroundings are hot, it is harder for heat to escape from the device. It starts
out being hot so it reaches maximum internal temperature sooner.
You have to understand that the maximum (or continuous) ratings are measured
separately and under different conditions. Many times the temperature is the
limiting factor. How large a heatsink can you tolerate or afford?
Herb
I do understand that the dissipation is how much heat can leave the unit, and that this value will change based on ambient temperature. I was more curious about why the dissipation value didn't seem to match with the maximum values listed for current and voltage. From what you're saying, it sounds like the maximum voltage and maximum current are measured separately from each other. Thus, if I have an application that requires 250 mA at 6 V at ~ 25oC, I can expect the transistor to overheat (1.5 W vs. a rated value of 1.25 W) despite the 'maximum' values for current and voltage being far higher than this.
TwoChain:
I am trying to understand how to properly select a MOSFET for my application.
So what is your application.
Mosfets are usually used here as a switches, where the only two important parameters are 'on' resistance (closed switch) and max S/D voltage (open switch).
Typical 'on' resistance of the fet you linked to is about 0.08 ohm with a Vgs of ~3volt.
From there you can see if temp rise is of any concern.
Make sure Drain-Source voltage stays under 20volt, and Gate-Source voltage stays under ±8volt.
Leo..
Concerning mosfet datasheets. You need to look at the details and not mindlessly accept their claims which is based on ideal conditions and maximum values.
Gate voltage is what creates the channel so that electrons can flow between the drain and source. The relationship between this current and gate is ID=K(Vgs-Vth)^2 where k is the conduction parameter. Since this channel is created by the gate voltage your current flow will be affected by it as you can see from the equation.
Concerning 2. You are not dissipating 56 watts across D and S. The actual power dissipated is based on Rds. If you had 20V across drain and source you would not be passing any current. You must be missing something about the parameters of a mosfet and how it actually develops voltage across it. Unfortunately it is not an ideal conductor when it’s switched on and due to the internal resistance of the channel there is heat generated. How much heat it can tolerate is based on its size or package and heat sink. This goes for just about all electronic components. Electron flow, jumping the valence bonds, diffusion and impedance generate heat. You have to mitigate and design for it. Are you sure you don’t need a relay? You could easily pass current.
Selecting a mosfet is a lot of work. There are so many these days. It will really depend on what you’re doing. As with many things it’s a trade off. But do your homework.
My question was meant to be more general in terms of how to read the datasheet, but for a specific application, I have attached a quick drawing. In this example, I would be using the MOSFET to switch on a water pump operating at 6V and with a current draw of 180 mA.
I have made this circuit using the Si2302DS MOSFET that I referenced in my original post, and it works as expected. However, when looking at the datasheet for this MOSFET, I became a bit confused regarding why the heat rejection didn't seem to match the maximum values.
I was playing around with the MOSFET to try and understand it better, and found 1 observation that confuses me:
When I measure the resistance between the source and drain I was expecting the MOSFET to show < 1 Ohm on my Ohmmeter. However, when I actually made the measurement (in the attached circuit labelled 'Si2302DS 82 Ohm'), I am getting a reading of ~ 12 Ohms. Also, if I measure the current passing through this circuit, and knowing the voltage (5V) and resistance of the resistor (82 Ohm), I also calculate that the 'MOSFET resistance' is ~ 12 Ohms. I considered it was coming from poor connections in my test cables, but when I measure them separate from the MOSFET, I also get << 1 Ohm. Is there a reason for this?
Wolframore, thanks for your comments. I'm no expert on MOSFETs (clearly). However, I am fully aware I need to turn them on with the gate voltage. In my original post, I had assumed that Rds would be close enough to 0 Ohm (when the MOSFET was fully on) that I could consider the full 20 V and 2.8 A to be experienced between the drain and source. When you say "If you had 20V across drain and source you would not be passing any current", I assume you are referring to when the gate voltage is 0, or when the MOSFET is 'off'.
You should have a volt drop of 0.08 * 0.18 = 0.0144volt across the fet when 'on'.
That should be a dissipation of ~1mW. Not worth calculating a heatsink for that.
Low voltage SOT-23 mosfets are very sensitive to ESD. You blow the gate with more than 8volt.
Not wise to measure the transistor with a DMM, solder with an iron, or handle without ESD safe tools/mat.
The pump must also have a back-emf diode across.
Leo..
Ad 2):
The MOSFET survives 20V when it is off (no current).
It also survives 2.8A when fully on (minimal voltage drop). Probably limited by power dissipation in ideal conditions.
If the conditions are worse than ideal you need to do the math based on clues in the datasheet. You cannot have both maxima at once: if you have large DS voltage (low GS voltage; MOSFET in linear region) the current must be small enough to keep power dissipation low.
Cooked the hell out of the poor little MOSFET with a baking hot soldering iron
Did it all while fumbling around with the impossibly small MOSFET as it bounced and crashed around on my table due to its tiny size and my clumsy hands.
Thus, I guess I should just be happy that it's functioning at all (for now)...
Please post the schematics and exactly how you’re measuring Rds. You should be measuring Vds and Ids and calculating for Rds. Hopefully you understand why you can’t just probe DMM probes across it and expect the reading to be correct. Think about how voltage is dropped across a circuit using KVL.
Here is a schematic of how I measured Rds. I used an ammeter to measure current, then knowing the resistance of my 82 Ohm resistor, calculated what the resistance of Rds should be. I continually calculated ~ 12 Ohms, which as I stated, seemed odd.
No need to break the circuit. Easier to measure voltage across the fet.
Then you can calculate 'on' restance from the ratio of the voltage across the fet and across the resistor.
There must have been 12/(12+82)*5= 0.38volt across that fet. Way too high according to the datasheet.
Leo..
Page 2-2, 6th spec.
Rds(on) > the resistance between drain and source when the fet is 'on' is listed as 0.07 ohm.
That should give a drain voltage of 0.07/(82+0.07)*5= 4.3 millivolt.
Leo..
FYI, I played around a bit and found my biggest issue with these readings...
For the 5 V power supply, I was using the 5V pin from the Arduino. When the MOSFET is turned on, the voltage dropped to ~ 4.85 V. Accounting for this, things are making a lot more sense. I have read before that I should not rely on the Arduino for stable voltage when pulling current, but.... ya... I did anyways.
Thanks again everyone for your help!
Dustin
p.s. Despite what I've been reading about the MOSFETs, I have found them amazingly durable! I've been abusing them horribly yet they all seem to have been able to withstand my bumbling approach.
MOSFETs are extremely rugged unless you exceed the gate-source breakdown voltage (for even a nanosecond) at which point they break immediately. This means handling precautions are very important to avoid static electricity damage (no protection diodes in most power MOSFETs). Even mains pickup voltage is enough to destroy a device.