Favorite High Power Output Transistor

Status
Not open for further replies.
Jump to Latest
#22
My impression is the "speed" (as in fT?) of a device becomes more important as power and rail voltage increases, i.e., slew limit.

However, just upgrading to a newer device may not be a slam dunk. I've read from a Phase Linear tech that substituting the certain newer MJ15xxx devices sets off oscillations in the old PL700s and PL400s. This sounds to me like combining slow drivers with faster output devices presents a problem. I could also believe that the gain and feedback network might create problems as well.

I may be off base on this, so don't take as truth.
 
#23
Hi sam9,
Right you are as far as oscillation is concerned. But, you may end up with oscillations in a low voltage amp just as easily. So, the voltage & current only makes the problem more spectacular. Any make of amplifier may suffer from this if your original outputs were much slower. The same thing used to happen when a Japanese output set where older JAN types were used.
If you blow the outputs, it's always a good idea to change the drivers as well.
-Chris
 
#24
"But, you may end up with oscillations in a low voltage amp just as easily."

<Sigh> I'm olny too painfully aware of this.


"makes the problem more spectacular."

I'm afraid I can help thinking about the old sci-fi TV shows where the contol pannels on the "bridge" would go up in smoke and sparks every other episode. Imagine a rack of PL700's going poof all at once!
 
#26
I had calculated about the same figures via a slightly different route. Walter Jung published some articles in the mid 70's that were based on the required v/uS at 80khz in order to set a safety factor of 4 and I think Nelson Pass may have recommended slew rates adequate for 200kHz as desirable.

I can understand designing for 2 to 4 times beyond what is needed for 20kHz, but I don't know the reasons why even higher might be thought beneficial.
 
#31
sam9 said:
... and I think Nelson Pass may have recommended slew rates adequate for 200kHz as desirable.
Click to expand...

:scratch:

Slew-rate distortion (triangular look) of a sine wave occurs when the initial slop of the sine wave is greater than the slew rate. And, guide of power bandwidth (maximum frequency for the undistorted large signal operation) is suggested by the formula:

f(max) = Slew-rate/(2 x pi x Vp)

If an amp has a slew-rate big enough to cover the top critical frequency of 20 kHz at its peak output voltage and above this fc the output peak voltages are reduced at abt. -20dB/decade rate, isn't the slew rate adequate for 200kHz overpaid?
 
#33
Easyamp said:
My fav's for high power in a to-3 pac are the MJ11033/32 pair and in the to-264 pac the highly underated flj4215/4315 fairchild pair. The new on semi mjl's are nice but don't sound as good as the fairchilds.
Click to expand...


I just compare the 21193/21194 pair vs. the 4213/4313 pair. No audible difference.
 
#34
Commentable Thoughts

A Simple Truth "BJT's can't beat Vertical Mosfets in any aspect" .
If u don't believe ask Sir Nelson Pass or Anthony holton about the greatness of Mosfets over Bjt's in the output stage.

Regards
Ampman
 
#35
Re: Commentable Thoughts

amp_man_1 said:
A Simple Truth "BJT's can't beat Vertical Mosfets in any aspect" .
If u don't believe ask Sir Nelson Pass or Anthony holton about the greatness of Mosfets over Bjt's in the output stage.

Regards
Ampman
Click to expand...


single blanket statements like the above are doomed to be wrong. As a big fan of mosfets, I can tell you with personal experience that there are tons of cases where a BJT is far superior than a MOSFET.

the reserve is also true in other cases. it is naive to make a sweeping statement with regards to any device's superiority in general.

It is always wise NOT to put your words in other's mouth.
 
#37
just compare the 21193/21194 pair vs. the 4213/4313 pair. No audible difference.
Click to expand...

Here's the schem of the amp I used, it was on a proto board so I just swapped them and I meant I liked Fairchilds FJL**** over onsemi's MJL**** I haven't done any listening test between the MJ11033/94 and the MJL's

An externally hosted image should be here but it was not working when we last tested it.


The resevoir caps were only a few inches away if you're wondering about the 2 100n filter cap's only.

Personal taste's are very subjective on many levels, (especially in audio) for me to say one sounds better than the other could be affected by many things other then how they actually sound so I guess it's not correct to say that one sounded better. Maybe I should just say "I like the fairchilds better" but who-cares??
 
Status
Not open for further replies.
Top Bottom