Maximum channel temperature - RF Cafe Forums

shireesha

Post subject: Maximum channel temperature Posted: Tue Oct 20, 2009 12:58 am

Captain

Joined: Wed Jun 04, 2008 3:23 am

Posts: 11

My question is on the maximum channel temperature specified in the power amplifier data sheets. How is this value determined? What does it signify?

Top

nubbage

Post subject: Re: Maximum channel temperaturePosted: Wed Oct 21, 2009 4:16 am

General

Joined: Fri Feb 17, 2006 12:07 pm

Posts: 218

Location: London UK

Hi shireesha

This is a basic parameter for the channel and substrate, above which the device performance will degrade, and much above it the device will suffer thermal runaway and be irreversibly damaged.

It is used to input the value into thermal design software in order to arrive at the design of heatsink needed to keep the device within the stated performance parameters.

_________________

At bottom, life is all about

Sucking in and blowing out.

Top

shireesha

Post subject: Re: Maximum channel temperaturePosted: Wed Oct 21, 2009 6:24 am

Captain

Joined: Wed Jun 04, 2008 3:23 am

Posts: 11

Sir,

Thanks for your reply. Could you explain how this value (of maximum channel temperature) is arrived at? Is it generally provided by the foundry or is there a way of determining it ourselves?

Top

nubbage

Post subject: Re: Maximum channel temperaturePosted: Fri Oct 23, 2009 4:04 am

General

Joined: Fri Feb 17, 2006 12:07 pm

Posts: 218

Location: London UK

Hi shireesha

This is most accurately measured by the manufacturer of the device, because they have access to the open die of a test sample during pre-production. With the device in a test fixture they are able to measure device performance parameters and measure the channel temperature by a non-contact method such as a calibrated infra-red scanner.

None of these facilities are normally available to the end user.

_________________

At bottom, life is all about

Sucking in and blowing out.

Top

biff44

Post subject: Re: Maximum channel temperaturePosted: Tue Nov 10, 2009 5:47 am

Colonel

Joined: Wed Apr 08, 2009 11:07 am

Posts: 33

A semiconductor device is made a little like how you bake a chocolate chip cookie. Once it is made, it is delicious to eat. However, if you leave it on the dashboard of your car in the summer, the chocolate chips melt and run all over the place--making a big mess. The same happens with your semiconductor. If you keep it nice and cool after it is made, all the various dopants and metals stay where they are supposed to, and you can enjoy a good semiconductor operating life. Let it get too hot, and those dopants and metals start to diffuse around. If they move around too much, then a catastrophic failure can occur. Some devices are more prone to this (such as silicon junction devices), while others are less prone to this (GaAs mesfets), so there are different allowable operating temperatures.

If you wanted to determine the allowable operating temperature, you would do an "acelerated Life" test, where you take 100 or so devices (set up as amplifiers, for example), and measure an important parameter (like gain or output power). You put them in an oven at some very high temperature, and take them out every once in a while and mesure the electrical parameter. Once it is seriously degraded, you extrapolate when the average device would fail. From that you can determine the maximum operating temperature for a given "mean" failure time. A complicated test--so usually only the device manufacturer does it.

_________________

Rich

Maguffin Microwave Consulting

www.MaguffinMicrowave.com

Posted  11/12/2012