discrete vs. IC opamps? has anyone tried the pass labs design?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
FETs and BJTs combined

To the combination of FETs and BJTs

Using FETs for input and BJTs for output is exacty what is done in the legendary KANEDA preamp design, so praised by the fanatics some (?) years ago....
Have had a Kaneda on my bench for some time now, just for the kicks...although I have not tested it with real audio yet, as I had problems in balancing the line stage due to too few FETs to do a proper matching.
BTW - the Kaneda design is still available on the net, and the parts won't cost more than your shirt... yu still got th e pants left...
 
I like JFET's for the input, as was pointed out BJT's Have hoght Input Bias Currents and the DC off set will Change with Source Impedance. I am not to fond of MOSFET's for use in any part of The Gain Stage and that includes the Output. Contrary to Popular Belief Mosfets are rather Nonlinear, have Lots of Capacitence and that capacitence when combined with the High gate Impedance can leed to Stability Problems, Moreover The Sound quality of Mosfets to Me is Not Good in the Mids and High's. The Audiophile Press Termed "MOSFET MIST" as a problem years ago when mosfets first arived. I think thy still have this problem. So my perfered Topology is JFETS on INPUT Jfets on the Second gain stage. BJT's on the Driver and output stage. Also i think Casscoding all stages is The best aproach with any device Tube, BJT, MOSFET, JFET, and yes you can cascode opamps. Casscoding opamps is most often done to allow Increced Voltage swing by allowing Higher voltage Supply rails.
 
Mosfet Mist

ppl,

I must say that your comments on mosfets are rather harsh, but I shared your sentiments years ago.

First a little history, when mosfets first came out most designers tried to use them like bipolars (big mistake). For one they have higher distortion than bipolars (maybe five times greater )at low bias currents so they need to be biased at higher currents.

Even though they were easy to drive, their higher capacitance could cause problems at higher frequencies, and most designers did not take this into account.

Lead inductance was an issue which could cause stability problems, which was seldom addressed.

Early mosfet devices were relatively high impedance( lateral) devices but now we have low impedance (vertical) mosfets like the IR's.

Mosfet mist, to me, is a myth for most part propagated by bad design and lack of understanding of the device.
Some of the worlds best designs use mosfets, Pass Labs could not exist if mosfets were as bad as you say. It is easy to proclaim that all mosfets amplifiers are bad having been exposed to few bad designs or ideas from designers like Douglas Self.

Used correctly mosfets are more reliable than bipolars, last longer and are generally easier to design with. (Try building a Zen with a bipolar). They might be more expensive to design with but this no indication of their sonic potential.

Regards,
Jam

[Edited by jam on 06-28-2001 at 09:46 PM]
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.