 Semiconductor Components Industries, LLC, 2005

January, 2005 − Rev. 0

1

Publication Order Number:

AND8199/D

AND8199

Thermal Stability of

MOSFETs

Prepared by: Alan Ball

ON Semiconductor

Application Engineering Manager

A variety of applications use hot−swap controllers, often

to increase the reliability of a system. However, a failure in

the hot−swap circuit would defeat that purpose. When you

use MOSFETs in their active region to control current, such

as you would for a controller that operates in a

constant−current mode of operation, they have an inherent

failure mechanism. In this mode, the MOSFET can get

hot spots and fail, long before the device exceeds its

Safe Operating Area (SOA) ratings.

Engineers have long understood that MOSFETs are

positive temperature coefficient devices. Therefore, as the

temperature of the device increases, the resistance increases.

In other words, higher temperatures result in lower currents.

This fact is important if you want to operate MOSFETs in

parallel. With a good thermal path between devices, the

positive temperature coefficient reduces the current in the

hottest device and forces more of it to flow in the cooler

device, thereby avoiding thermal runaway.

Engineers often think of a MOSFET as a single power

transistor, but it is a collection of thousands of tiny power

FET cells connected in parallel. In terms of sharing current,

the same application of the positive temperature coefficient

applies. In this case, the thermal path between the cells is

better than that of separate packaged devices, because the

cells are all on the same die.

As the current density of a small group of cells increases,

those cells heat up, increasing the resistivity of those cells

and forcing current to flow in neighboring cells, which

minimizes the thermal gradient and avoids hot spots. This

process is an essential physical tenet that allows the parallel

array of cells to function reliably.

If the MOSFET exhibits a negative thermal coefficient,

today’s parallel cell structure would cause serious reliability

issues. In fact, in some modes of operation, the thermal

coefficient goes negative. You can easily understand this

phenomenon by looking at the transconductance curves for

a FET device (refer to Reference 1).

A typical set of transconductance curves clearly

demonstrates this effect as shown by Figure 1. Below are

curves

from

three

typical

devices

used

in

hot swap applications.

24

12

8

4

7

6

5

4

3

2

1

0

0

89

16

20

0

Figure 1. Transfer Characteristics for NTD12N10

Figure 2. International Rectifier IRF530

100

10

4.0

5.0

6.0

7.0

8.0

20 ms = Pulse Width

http://onsemi.com