Ultima Thule said:
You're taking this quite far afield.
The issue is simple. Charles claimed that the Sziklai pair has more feedback than the Darlington pair.
I'm simply trying to find a definitive answer as to whether or not this is true. And so far I haven't really seen anything definitive. And I don't consider "You're wrong. Go read a book" to be a definitive answer.
Since Charles made the original claim, then I guess it's Charles' definition of feedback that you'll want to get.
Perhaps Charles can give an example using a specific transistor pair and describe the process of determining how much feedback a Sziklai and Darlington pair each have and state the amount of feedback in dB.
se
AKSA said:
And it's in a single-ended amp that I'm using the CFP (though I'm using a JFET/bipolar combination) which I found to work best for me in this particular situation compared to half a dozen or so different follower combinations.
But still there has been no definitive answer as to whether the CFP employs more feedback than the Darlington.
se
john curl said:
There isn't any argument that there is a feedback loop. But there's negative feedback in the Darlington pair as well as the simple emitter follower.
But there has yet to be a definitive answer as to whether the CFP employs more feedback overall than the Darlington.
The CFP tends to be a bit more linear than the Darlington, but I'm wondering whether that's due to the CFP employing more feedback or because of the use of complimentary devices.
se
Re: Which has more feedback?
Yeah. And the same would seem to hold true even if you look at the devices as transconductance devices.
But my confidence in that notion isn't much greater than my confidence in Charles' notion. Which is why I'd really like to see a truly definitive answer rather than all this "gut level" stuff. And while I've been looking for some time, I've yet to come across a truly definitive answer.
se
SY said:
Yeah. And the same would seem to hold true even if you look at the devices as transconductance devices.
But my confidence in that notion isn't much greater than my confidence in Charles' notion. Which is why I'd really like to see a truly definitive answer rather than all this "gut level" stuff. And while I've been looking for some time, I've yet to come across a truly definitive answer.
se
Re: Re: Which has more feedback?
Hi guys,
I've got some Middlebrook loop gain stuff for the CFP ready to post in the other thread. I don't know about a "definitive answer", since there's a huge variety of different operating points, etc for the CFP. But I was surprised at how low the loop gain of this particular CFP was.
Steve Eddy said:
Hi guys,
I've got some Middlebrook loop gain stuff for the CFP ready to post in the other thread. I don't know about a "definitive answer", since there's a huge variety of different operating points, etc for the CFP. But I was surprised at how low the loop gain of this particular CFP was.
Re: Re: Which has more feedback?
Hey Steve,
Didja ever look at Douglas Self's web site:
http://www.dself.dsl.pipex.com/ampins/discrete/cfp.htm
The previous page in his discourse discusses the emitter follower. The page linked above discusses the CFP. He says:
"The simple emitter-follower is lacking both in linearity and load-driving ability. The first shortcoming can be addressed by adding a second transistor to increase the negative feedback factor by increasing the open-loop-gain. This also allows the stage to be configured to give voltage-gain, as the output and feedback point are no longer inherently the same. This arrangement is usually called the Complementary Feedback Pair (hereafter CFP) though sometimes known as the Szilaki configuration." (emphasis mine)
But hey, maybe you think Douglas Self is some sort of tweaky low-life too... who knows?
Charles Hansen
Steve Eddy said:
Hey Steve,
Didja ever look at Douglas Self's web site:
http://www.dself.dsl.pipex.com/ampins/discrete/cfp.htm
The previous page in his discourse discusses the emitter follower. The page linked above discusses the CFP. He says:
"The simple emitter-follower is lacking both in linearity and load-driving ability. The first shortcoming can be addressed by adding a second transistor to increase the negative feedback factor by increasing the open-loop-gain. This also allows the stage to be configured to give voltage-gain, as the output and feedback point are no longer inherently the same. This arrangement is usually called the Complementary Feedback Pair (hereafter CFP) though sometimes known as the Szilaki configuration." (emphasis mine)
But hey, maybe you think Douglas Self is some sort of tweaky low-life too... who knows?
Charles Hansen
Re: Re: Re: Which has more feedback?
Yes, I've read that page.
But did you read this page?
http://www.dself.dsl.pipex.com/ampins/discrete/ef.htm
When Self says "The simple emitter-follower" he's referring to an emitter-follower using a SINGLE transistor.
To wit:
This page deals only with simple emitter-followers, ie those with one transistor as the actual follower.
Since this discussion has to do not with the Sziklai versus a single-device "simple emitter-follower" but a Darlington pair, what you quote above is comparing apples to oranges.
So how 'bout we compare the Darlington to the "simple emitter-follower"?
The Darlington also adds a second transistor. The Darlington also is more linear than the "simple emitter-follower." Does the Darlington not achieve this improved linearity in the same fashion that the Sziklai does? I.e. by "increasing the open-loop gain"?
If not by increasing the open-loop gain, then by what mechanism DOES the Darlington achieve its greater linearity over the "simple emitter-follower"?
Grow up, Charles.
se
Charles Hansen said:
Yes, I've read that page.
But did you read this page?
http://www.dself.dsl.pipex.com/ampins/discrete/ef.htm
When Self says "The simple emitter-follower" he's referring to an emitter-follower using a SINGLE transistor.
To wit:
This page deals only with simple emitter-followers, ie those with one transistor as the actual follower.
Since this discussion has to do not with the Sziklai versus a single-device "simple emitter-follower" but a Darlington pair, what you quote above is comparing apples to oranges.
So how 'bout we compare the Darlington to the "simple emitter-follower"?
The Darlington also adds a second transistor. The Darlington also is more linear than the "simple emitter-follower." Does the Darlington not achieve this improved linearity in the same fashion that the Sziklai does? I.e. by "increasing the open-loop gain"?
If not by increasing the open-loop gain, then by what mechanism DOES the Darlington achieve its greater linearity over the "simple emitter-follower"?
Grow up, Charles.
se
Charles Hansen said:
What would you like me to drink, Charles? The apple juice you're selling as orange juice?
Just answer the question. How does the Darlington manage to have increased linearity over the simple emitter-follower? Is it via the same mechanism as the CFP (i.e. increasing the open-loop gain) or is it some other mechanism?
se
Guys (CH and SE), cool the personal stuff. There's an interesting technical discussion buried in the invective that we don't want to lose sight of.- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.