Neither. What you want the amp/speaker system to do is provide constant acoustical power. How that's done can take a range of forms- constant voltage amps with maximally flat to critically damped base is canonical, but not truly privileged. Once can get the same flat acoustic response from a constant current amp and an overdamped woofer.
The Mills and Hawksford paper is worth a close read.
The Mills and Hawksford paper is worth a close read.
Thanks a lot.
I found it here:
http://www.essex.ac.uk/dces/researc...bdocs/J14 Mills-Hawksford power amplifier.pdf
Studying....
I found it here:
http://www.essex.ac.uk/dces/researc...bdocs/J14 Mills-Hawksford power amplifier.pdf
Studying....
[url]http://www.essex.ac.uk/dces/research/audio_lab/malcolmspubdocs/J14%20Mills-Hawksford%20power%20amplifier.pdf[/url] said:
I would say that any power amplifier requires an output impedance that is linear and frequency independent. Overlooked non-linearity of output resistance is the cause of horrible sound of so called "voltage amplifiers" that are measured well loaded on resistors. The same true to "current output amplifiers" called in the article "Power Transconductance Amplifier". Changing from voltage to a current output can't solve the problem of distortions.
"Velocity feedback" does not belong to a "transconductance amplifier", it helps to any amp/woofer system.
I've experimented with feedback by an acceleration using an amp with corresponding EQ that boosted low frequencies, and the whole system was linearized by a piezo - crystal with a source follower glued to a cone of a speaker. There was an ordinary Soviet made bass guitar cabinet with Fs around 55 Hz.
As the result, it sounded very good on all available material (rock, jazz, symphonic orchestra), incomparable better than any stock Hi-Fi. I wanted to try it on lower frequencies and borrowed a vinyl disk with Bach's organ music... I put the disk, sat down preparing to enjoy.... and the amp spit out the cone from the basket: it tried to reproduce acceleration of a needle on a horrible stamped disk with shifted center...
Later I added however a 18 dB/Oct 16 Hz HPF, but used pair of smaller compression drivers; that time Fs was about 45 Hz. I linearized a frequency response by a positive feedback by current (i.e. output resistance was negative and of course less linear!).
You know, despite of absence of a truly mechanical feedback by acceleration, the system did not sound worse...
CLS said:Thanks.
But pentodes have too much intrinsic distortions, wouldn't they?
Altec 1568 is already made with output pentodes... It is a nice amp, though the output transformer is not "bassy", probably that's why you blame on your woofer.
Why you don't try to add a positive feedback by current (after trying a negative feedback by current) and compare results?
I have been researching an EL84 PP transconductance amp.
It looks like a good target output Z would be on the order of 64 ohms. (the F1 is about 80 Ohms)
Looking at the pentode tube curves, it looks like an EL84 has about a 40K ohm output Z.
Paralleling 2 would give a 20K. if I tried the same transformer as a voltage amp, that would be 8k:8 or 1000:1, so it would give me about a 20 ohm output Z. Am I figuring that correctly?
That would Imply I need closer to a 8K:32 to reach my output Z target?
How do I estimate Gain? Probably draw a load line, but can I figure it out?
Doug
It looks like a good target output Z would be on the order of 64 ohms. (the F1 is about 80 Ohms)
Looking at the pentode tube curves, it looks like an EL84 has about a 40K ohm output Z.
Paralleling 2 would give a 20K. if I tried the same transformer as a voltage amp, that would be 8k:8 or 1000:1, so it would give me about a 20 ohm output Z. Am I figuring that correctly?
That would Imply I need closer to a 8K:32 to reach my output Z target?
How do I estimate Gain? Probably draw a load line, but can I figure it out?
Doug
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.