|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
160.62.4.10
In Reply to: RE: Solid State Power: MOSFET vs. Bipolar? posted by Duster on October 4, 2010 at 13:26:32
I have found that JFETS and MoSFETS can be nearly invisible sounding in lower current applications (as input and driver stages), especially when used like a triode: single ended, Class A and no negative feedback. Somehow they still don't sound quite as good as a properly done tube amp but much better than bipolar amps.
When used in the typical Class AB, high feedback and complementary designs then I would say that MOSFETS seem to underperform compared to the best bipolar output stages but both sound bad.
The FET is a closer to tube substitute and when used in that way it can sound very good.
Follow Ups:
You know, I felt the same way until I recently made some changes to the low frequency EQ/buffer unit on my Infinity RSIIb speakers. I wanted to get rid of the output coupling caps so I changed the single-ended, class A MOSFET output unity gain buffer to an OPA827 opamp unity gain buffer (opamps of all things!). The extremely low DC offset of the opamp allowed me to remove the output coupling caps - the result was clearer, less congested sound, without introducing any harshness that I could hear. I'm starting to think that a properly implemented opamp (with enough PS decoupling/bypassing) can sound damn good. Perhaps it might be a different story in a gain stage.
The opamp buffer stage is on the tan board in the picture below (just below the board you can see the MOSFETs still in place, I left them in just in case the experiment was a flop).
I don't think so.
How can you compare the performance you get out of modern, ultra-fast bipolar/FET devices to what you get out of an output transformer, esp. SE? Not to mention the sound?
I like my frequency response flat, not looking like a bell curve, but that's just me.
I realize alot of people like the euphonic 2nd-order harmonic distortion, the reactive nature of the OPT that causes phase shifts and the high output impedance that is another problem.
However, that's not what we're discussing here.
Never attribute to malice that which can be adequately explained by stupidity.
"How can you compare the performance you get out of modern, ultra-fast bipolar/FET devices to what you get out of an output transformer, esp. SE? Not to mention the sound?"
This depends on what you define as performance. I define it as how the amp sounds not how it measures because, as I am sure you are aware, there is not a good correlation between percevied sound quality and test measurements like THD. It seems that how linearity is achieved has as much or more of an impact than the raw number in terms of % THD.
"I like my frequency response flat, not looking like a bell curve, but that's just me"
As long as your damping factor is more than 5 there will not be a large amplitude swing in the FR curve. The flatness of most speakers will be far worse.
"I realize alot of people like the euphonic 2nd-order harmonic distortion"
Oh this old chestnut. Well seeing as it has been demonstrated that it takes at least 1% of 2nd order distortion to be readily audible I don't think that there is much in the way of "Euphonizing" the sound. The real problem is that we are very sensitive to high order harmonics that are present in high feedback and Class AB designs. These destroy the sound, whereas little old 2nd order is essentially sonically invisible.
The real problem with many tube amps is the output transformer but not at the high frequencies (A good one is quite linear up to 20Khz). The real problem is at low frequencies and core saturation. This is one of the main reasons for "tube" sound. If you have ever heard good OTL amps you will note that they sound distinctly non-tubey in that golden way but also distinctly non-transistory.
"However, that's not what we're discussing here"
What exactly are we discussing then if not perceived sound quality as THE benchmark for amplifier performance. Clearly the numbers, at least on the surface, don't support subjective findings. See papers by Geddes and CHeever and you will see what I mean.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: