Not my words...
In page 83, app note an47 from linear technology, it goes:
"Some poorly designed amplifiers exhibit a substantial
thermal tail after responding to an input step. This phenomenon,
due to die heating, can cause the output to
wander outside desired limits long after settling has apparently
occurred..."
In page 83, app note an47 from linear technology, it goes:
"Some poorly designed amplifiers exhibit a substantial
thermal tail after responding to an input step. This phenomenon,
due to die heating, can cause the output to
wander outside desired limits long after settling has apparently
occurred..."
ppl said:
On big ICs (5x5 to 10x10 mm, which actually isn't big as we make circuits up to 50x50mm) you can easily find hot spots, and they aren't too hard to track down and see their influence on their neighbouring components.
Some figures for thermal resistance
-silicon: 100 W/mK (about 600um thick)
-glue: 2 W/mK (30um)
-ceramic package: 20 W/mK
Don't know about plastic packages, we never use them. Presumably directly moulded, hence glueless, but with a lower thermal conductivity for the package itself.
Dear Eric,
input pair:
Nonlinearities in the gain of the input pair will be corrected by global feedback, just like any other gain in the circuit!
Sorry, that statement is wrong when you consider temperature difference between emitter junctions. The correspondent error voltage will be added to feedback signal.
input pair:
Nonlinearities in the gain of the input pair will be corrected by global feedback, just like any other gain in the circuit!
Sorry, that statement is wrong when you consider temperature difference between emitter junctions. The correspondent error voltage will be added to feedback signal.
mikek said:this reminds me a lot about the homeopathic assertion that water has memory...
It does. According to
http://www.newscientist.com/news/news.jsp?id=ns99993817
Maybe switching amplifiers sidestep several of the mechanisms giving rise to this kind of distortion. I have read that some Single Ended Triode enthusiasts have been known to warm to the sound of switching amps!
Another observation is that power amplifier design officianados tend to have fairly undistinguished loudspeakers to evaluate their latest creations. This could explain why the likes of Linkwitz seems so attracted to switching amps, having addressed most of the acoustic issues with his speaker designs.
The icing on the cake!
Keith
Another observation is that power amplifier design officianados tend to have fairly undistinguished loudspeakers to evaluate their latest creations. This could explain why the likes of Linkwitz seems so attracted to switching amps, having addressed most of the acoustic issues with his speaker designs.
The icing on the cake!
Keith
-input stage transistors clamped together for thermal tracking - now common
-cascode input - now common
-Hawksford cascode VAS - now common
----------
I wonder.. if using ThermakTrak output transistors, plus exceptional Vbe temperature compensation has a larger thermal memory effect than some of the other specialized cascodes etc. in the Lavaradin Amp paper?
Andrew C Russell
"Ideas on Temperature Compensation for Emitter Follower Triples in Class AB Audio Amplifiers"
Andrew uses opamps with precise multi-slope gain to maintain the lowest distortion temperature independent output bias points.
-cascode input - now common
-Hawksford cascode VAS - now common
----------
I wonder.. if using ThermakTrak output transistors, plus exceptional Vbe temperature compensation has a larger thermal memory effect than some of the other specialized cascodes etc. in the Lavaradin Amp paper?
Andrew C Russell
"Ideas on Temperature Compensation for Emitter Follower Triples in Class AB Audio Amplifiers"
Andrew uses opamps with precise multi-slope gain to maintain the lowest distortion temperature independent output bias points.
Member
Joined 2009
Paid Member
Clearly unacceptable behaviour that won't be tolerated around here. We'll need their names. They will be struck-off.
For one branch of the differential input stage, the first two transistors are a kind of Sziklai pair with the load of the first being a Constant Current Source instead of the usual resistor. The emitters of second and the third transistors (PNP and NPN) are coupled so they effectively form a Rush circuit.
Hello
There is also loading the VAS transistor with a resistor (20 k to 50 k) from the collector to ground, it's ad very little thd and it should be good because it's made it more load tolerant and others reasons, but I think it also because that by reducing the OLG it's also reduce the memory distortion in the vas transistor, because that the memory distortion in the VAS are partially link to the OLG.
Thank
Bye
Gaetan
There is also loading the VAS transistor with a resistor (20 k to 50 k) from the collector to ground, it's ad very little thd and it should be good because it's made it more load tolerant and others reasons, but I think it also because that by reducing the OLG it's also reduce the memory distortion in the vas transistor, because that the memory distortion in the VAS are partially link to the OLG.
Thank
Bye
Gaetan
Last edited:
Not perse.
It would even less linear than a conventionnal differential-long tail pair circuit but it is nevertheless an interesting circuit.
A Rush circuit uses two devices in a series emitter-coupled differential configuration. Having two Vbe in series is more sensible to temperature than one.
There is some thermal compensation in a standard parallel emitter-coupled differential configuration also known as a long tail pair.
Perrot's circuit belongs to this last category. The fact that the cascoding makes two devices per branch forming a Rush circuit is anecdotic.
Member
Joined 2009
Paid Member
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.