• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Paralleling of MOSFET

Status
Not open for further replies.

Archit

Member
I need a high current carrying mosfet for some application. What will be the challenges if I use 2 or 3 mosfets in parallel,
 

rjenkinsgb

Well-Known Member
Most Helpful Member
Just make sure the current is equally divided between them.

If you are getting near the device limits, an equal low value source resistor with each FET will help - even if that's just a few inches (same length) of connecting wire for each.

Also ensure the rest of the connections are equal length to help current balance.
 

unclejed613

Well-Known Member
Most Helpful Member
http://www.irf.com/technical-info/appnotes/an-941.pdf

MOSFETS are more easily paralleled than BJTs, because they don't have a problem with "current hogging". they don't work well in parallel in self oscillating circuits like free running power inverters because of variations in gate turn-on thresholds.
 
Last edited:

ronsimpson

Well-Known Member
Most Helpful Member
an equal low value source resistor with each FET will help
Not much. That works with BJT but not with MOSFETS.
...With BJT the turn on voltage is 0.7V and a little Emitter resistance helps.
...With MOSFETs the Gate might be at 15 volts and jacking up the Source by 0.1 volt will do nothing.
I have seen application notes that want separate Gate resistors. (might oscillate)
 

crutschow

Well-Known Member
Most Helpful Member
Depends upon the use.
What is the "some application" that will use the MOSFET(s)?
 

Rich D.

Active Member
Not much. That works with BJT but not with MOSFETS.
Those MOSFETS should be driven on as hard as possible even in the worst case, so yes, a driving variation of .1 volt shouldn't make a bit of difference there.

However: (a fancy term for "but"):

It may not help the turn-on/turn-off thresholds, but if you are using a MOSFET that is 0.1Ω and you have it in series with a 10Ω resistor, they will balance out the Source-Drain currents, even if one of the MOSFETS is erroneously 0.2Ω.
Those are just example numbers, you probably want ohms to be much less than that.


...at least that's my understanding of it, but I didn't go to a college.
 

ronsimpson

Well-Known Member
Most Helpful Member
I think this project is about running great amounts of current. Any voltage across a Source resistor will heat. Efficiency is important.
I just pulled a data sheet for a large high voltage MOSFET. Its on resistance goes up when hot. If one FET takes most of the current, I think it will heat and increase resistance thus shifting the current more to a second part.
Data sheet: Max resistance = 0.25 ohms, Typical resistance = 0.19 ohms, Min resistance is unknown so I will pick a number (0.125 ohms).
Extreme example: Take a 0.25 in parallel with a 0.125 and one MOSFET has 10A while the other has only 5A. Sharing is not so good in the extreme example, but the 10A part will get hotter and shift some of the current to the colder part.
It is also hard to find the "best" and "worst" part in the same batch.
 

dknguyen

Well-Known Member
Most Helpful Member
I think this project is about running great amounts of current. Any voltage across a Source resistor will heat. Efficiency is important.
I just pulled a data sheet for a large high voltage MOSFET. Its on resistance goes up when hot. If one FET takes most of the current, I think it will heat and increase resistance thus shifting the current more to a second part.
Data sheet: Max resistance = 0.25 ohms, Typical resistance = 0.19 ohms, Min resistance is unknown so I will pick a number (0.125 ohms).
Extreme example: Take a 0.25 in parallel with a 0.125 and one MOSFET has 10A while the other has only 5A. Sharing is not so good in the extreme example, but the 10A part will get hotter and shift some of the current to the colder part.
It is also hard to find the "best" and "worst" part in the same batch.
Yeah, you need the small resistor with BJTs to current balance since BJTs conduct better as they heat up which causes the most efficient BJT to take up more and more of the shared load (positive feedback due to a negative temperature coefficient) cycle until it blows.

MOSFETs do not need this resistor because their internal resistance increase with temperature so if one MOSFET ends up conducting more than the others (negative feedback due to temperature coefficient), it heats up and its resistance increases causing it to conduct less forcing the under-utilized MOSFETs to take their share of the load.

Use gate resistors as close as possible to the gate of each parallel MOSFET to dampen ringing and overshoot due to parasitic inductances and the gate capacitances.

Use identical MOSFETs so that they try to share load equally and switch on and off at approximately the same time and speed.
 
Last edited:
Status
Not open for further replies.

EE World Online Articles

Loading
Top