Will a true class B have thermal runaway? - Page 3 - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 11th January 2004, 05:03 PM   #21
maylar is offline maylar  United States
diyAudio Member
 
Join Date: Jun 2003
Location: Connecticut
There are a lot of factors that influence thermal stability, one being the impedance of the drive or biasing circuitry. Icbo doubles with each 6*C of junction temperature and that current can cause forward biasing of the transistor and lead to thermal runaway even in Class B.
__________________
dave
  Reply With Quote
Old 12th January 2004, 03:17 AM   #22
azira is offline azira  United States
diyAudio Member
 
Join Date: Dec 2003
Location: Near Seattle
Ok. So... my reading suggested that thermal runaway was something along the lines of: A fixed bias accross a BE junction causes current to flow. As the temperature rises, the amount of current that is allowed to flow increases which causes more heat...

w/rt to distortion, here's a spice sim of a class B in a feedback amplifier and a fft of the output, what is it that spice doesn't know about?
Attached Images
File Type: jpg cir.jpg (21.6 KB, 162 views)
  Reply With Quote
Old 12th January 2004, 03:18 AM   #23
azira is offline azira  United States
diyAudio Member
 
Join Date: Dec 2003
Location: Near Seattle
FFT...
Attached Images
File Type: jpg fft.jpg (61.3 KB, 157 views)
  Reply With Quote
Old 12th January 2004, 07:05 AM   #24
Steven is offline Steven  Netherlands
diyAudio Member
 
Steven's Avatar
 
Join Date: Oct 2003
Location: The Netherlands
I think you should change your Y axis scale. Now it is linear and higher harmonics will not be visible, because the amplitude may be 1/1000 of the fundamental or less.
First change the scale to logarithmic or dB, then look again.

Steven
  Reply With Quote
Old 12th January 2004, 07:29 AM   #25
sreten is offline sreten  United Kingdom
diyAudio Member
 
Join Date: Nov 2003
Location: Brighton UK
I'm really surprised some people will find something to argue
about even with the the most basic concepts of electronics.

I wish they would read a decent textbook and stop
telling people what they think a concept really is.

Thermal runaway is caused by instability of DC biasing of a
transistor, whatever prompts the instability is not relevant,
destruction must be due to a runaway increase in DC bias
current because of thermal instability to classed be as
thermal runaway.

Class C by definition cannot suffer thermal runaway as it no DC bias.

Class C also by definition exhibits crossover distortion that cannot be
removed by any real world application of negsative feedback.

With standard single pole compensation feedback falls at 6dB/octave,
consequently the crossover distortion rises with frequency.

Not only that : At ~1k crossover distortion components are
liberally produced right up to 20K and beyond. Standard 1 pole
compensation above the turnover frequency spectrally skews
these components increasing higher harmonics at 6dB/octave.
(lees effectively reducing the harmonics at 6dB/octave)

sreten.
  Reply With Quote
Old 12th January 2004, 01:28 PM   #26
diyAudio Member
 
JFETglare's Avatar
 
Join Date: Oct 2003
Location: Raleigh/Cary NC
Send a message via AIM to JFETglare
I'd like to cast a vote too! A true class B amp, where the drivers and outputs are "off", will NOT runaway thermally.

A true class B amp sounds like dookie compared to one that is moderately biased, even with gobs of feedback. It would not be a hi-fi amplifier, that's for sure. In a TV or clock radio, you probably wouldn't notice it enough to care.
__________________
Oh frack!!
  Reply With Quote
Old 12th January 2004, 01:55 PM   #27
Account Disabled
 
Join Date: Apr 2003
Location: US
Quote:
Originally posted by azira
w/rt to distortion, here's a spice sim of a class B in a feedback amplifier and a fft of the output, what is it that spice doesn't know about?
Not sure about real components but if you go through reasoning, you will see that feedback will take out cross-over distortion. It is, however, arguable (or unlikely depending on your perspective) that the same is true with real life circuits or simulations using reasonably good models.

a much interesting plot would be to chart Vin, Vo and Pin 6 of the Opamp. You will see that the voltage on Pin6 crosses zero very fast (to reduce cross-over distortion at Vo).

Quote:
Originally posted by sreten
Thermal runaway is caused by instability of DC biasing of a
transistor,
sreten.

i have always thought that thermal run-away is caused by hot spotting which may or may no thave anything to do with biasing.
  Reply With Quote
Old 12th January 2004, 03:46 PM   #28
azira is offline azira  United States
diyAudio Member
 
Join Date: Dec 2003
Location: Near Seattle
Quote:
Originally posted by sreten
I'm really surprised some people will find something to argue
about even with the the most basic concepts of electronics.

I wish they would read a decent textbook and stop
telling people what they think a concept really is.

Thermal runaway is caused by instability of DC biasing of a
transistor, whatever prompts the instability is not relevant,
destruction must be due to a runaway increase in DC bias
current because of thermal instability to classed be as
thermal runaway.

1) Is 0-volts at the base considered DC biasing if the load is referenced to ground?

2)

Sedra/Smith, Microelectronics 4th ed. pg. 768-9

"... if Vbe is held constant and the temperature increases, the collector current increases. The increase in collector current increases the power dissipation which in turn increases the collector current. Thus a positive feedback mechanism exists that can result in thermal runaway."

Can you please suggest a 'decent textbook' that I should use instead?


3) Thank you Steven for the tip on the crossover distortion, feedback won't remove it. Here is the sim output with a log plot for anyone who's interested.
Attached Images
File Type: jpg fft.jpg (57.3 KB, 88 views)
  Reply With Quote
Old 12th January 2004, 08:23 PM   #29
sreten is offline sreten  United Kingdom
diyAudio Member
 
Join Date: Nov 2003
Location: Brighton UK
First a correct :

Quote:
Class B by definition cannot suffer thermal runaway as it no DC bias.

Class B also by definition exhibits crossover distortion that cannot
be removed by any real world application of negative feedback.

(I'd just got up , class C slipped in instead of class B)
Quote:
1) Is 0-volts at the base considered DC biasing if the load is referenced to ground?
No. Base (and x beta Collector) current must flow.

Quote:
"... if Vbe is held constant and the temperature increases, the collector current increases. The increase in collector current increases the power dissipation which in turn increases the collector current. Thus a positive feedback mechanism exists that can result in thermal runaway."
This is one form of thermal instability, there are others.
It also implies a base current (collector current / beta)
which implies the base voltage > 0.6V.

QUOTE]Can you please suggest a 'decent textbook' that I should use instead?

Well the above book isn't wrong, as such.
Its referring to trying to set a collector current by holding
the base voltage without an emitter resistor - this is not
a good idea for lots of reasons, but a dead transistor is
certainly the the most cogent arguement against it.

Quote:
Here is the sim output with a log plot for anyone who's interested.
Shows some of the limitations of the simulator.
No even harmonics -> transistor models are perfectly symmetrical.


Quote:
It is, however, arguable (or unlikely depending on your perspective) that the same is true with real life circuits or simulations using reasonably good models.
This is pure conjecture born from ignorance, reality
has nothing to do with personal perspective.
Crossover distortion cannot be removed from real world Class
B amplifiers because of fundamental electrical principles.
(As outlined in my previous post)

Quote:
i have always thought that thermal run-away is caused by hot spotting which may or may no thave anything to do with biasing.
No. Hot spotting causes the maximum allowable current to be
lower than the current implied by thermal limits at high currents.

Whilst hotspotting is a form of local thermal runaway (not the
other way round) it applies to any transistor in any circuit, if
it is operated beyond its current limits.

When discussing thermal runaway in circuit topologies the
reference is always to unstable DC biasing arrangements.

Quote:
a much interesting plot would be to chart Vin, Vo and Pin 6 of the Opamp. You will see that the voltage on Pin6 crosses zero very fast (to reduce cross-over distortion at Vo).
Not fast enough, because of fundamental electrical principles.
(Unless the simulator assumes the op-amp has infinite gain
and infinite speed, in this case its not much of a simulator)

sreten.
  Reply With Quote
Old 12th January 2004, 08:31 PM   #30
Account Disabled
 
Join Date: Apr 2003
Location: US
Quote:
Originally posted by sreten
fundamental electrical principles.

sreten.

just curious. What are those "foundamental electricl principles" and how would they work in this case?
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Thermal runaway question ricsmuts Car Audio 39 22nd May 2008 04:03 PM
thermal runaway problem Dan2 Solid State 6 25th March 2008 08:11 PM
Krell KSA-250 not true class-A? K-amps Solid State 11 27th May 2004 09:54 PM
thermal runaway diodes Wagener Solid State 19 22nd May 2004 11:44 AM
Transistor Thoery Help. Thermal Runaway hugeli60 Pass Labs 1 1st June 2003 10:29 AM


New To Site? Need Help?

All times are GMT. The time now is 04:55 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2