Similar Transistors swap changes everything....why

quadklipsh

Super Member
Hi all folks.
Heres a quicky.
Id like to forward a question here that my tech would like to be answered by experts only.
He recently built two amps . One is a copy/ clone of a bang olufsen sounds fabulous.
Other is an amp based on the 3055 design but not using 3055 . Instead we used 3442 and this one sounds great.
The question is when i replace the 3442s with a set of 5200 transistors he says the sound completely loses its beautiful character .ill send the two transistors details for you.plz tell us why this is so pronounced
 
Last edited:
I don't understand the question. Are you inquiring about differences between transistors utilized in a given circuit/application?
 
The question is when i replace the 3442s with a set of 5200 transistors he says the sound completely loses its beautiful character

If you like the sound of the 3N3442's in your design then don't change them, what are you trying to achieve by doing this?

For clarity with these types of questions please state the full transistor type number - then there can be no confusion, otherwise you leave us guessing.
 
The schematic of those amps is gonna be sent to me soon. Im waiting from my techs responce .....
Hyperion ...... The main reason my tech asks of this from you experts is that when he used less than quality transistors in his clone B&O it made no significant difference and the amp performed like a charm. while putting poor quality caps in his supreme 600 made a night n day difference. And the supreme sounded like crap.

He has some Transistors for various projects ahead. . Original caps are costly here while cheap ones are handy and v cheap but sound crap..... So if a design sounds good with cheap caps why use the rare and expensive ones there ....
 
The speed of transistors considerably varies. Poor design choices can cause amplifiers to oscillate in the MHz when higher-speed devices are used. This burns power, overheats the devices, and degrades the sound. It ultimately results in failure.

The two transistors you cited have bandwidths differing by a factor of about 400.

Randomly substituting components without an understanding of circuits is doomed to fail.
 
IMO, neither one of those circuits is very good for all sorts of reasons. If you want to know why, read the Self and the Cordell amplifier books. The result is differing transistor properties cause differing performance. One of the hallmarks of a good circuit is lack of sensitivity to component variations.
 
Thanks for the input guys ,

My tech added ;

D669 and B649 were not used anywhere.
D313 were used as drivers.

Binary Mike....yes he used 3 darlingtons as specified by original circuit diagram. But he just did not draw the large symbols of the darlingtons for his own use at his desk.

Ok i discussed with him what he wants to ask in particular.....
He wants to know why these 4 transistors differ in sound quality when their specs are comparable and closer to each other

Group1 .....metal Transistors.
(3055 and 3442)
THESE SOUND SIMILAR.

GROUP 2) plastic bodied flat ones .( 1047 and D5200)
THESE SOUND SIMILAR too but are poor in quality than the group 1 .

Why are plastic transistors generally lesser in mid range and musical details than metallized steel ones .....

Its a common belief here that amps that use flat plastic body transistors sound tinny and V-Shaped with lack of lustre and detail and aaflat responce while metal Transistors have a better midrange character .
Hope i have conveyed his point .
 
Well, I think the whole premise of metal vs. plastic is nonsense, so probably the wrong person to ask. Hint- if you blindfolded me and hooked various transistors to my curve tracer, there's no way I could tell you if they were metal or plastic.
 
I tend to be one of those people that believes individual component choices, provided the components meet required design criteria, should not affect how a well designed circuit behaves. Its why I don't get into tube rolling, fancy capacitor or resistor selection, etc.

Plastic transistor vs metal, assuming the datasheet specs are comparable, should make no difference at all. I'm more of a numbers guy though, just because its impossible for me to hear exactly what you hear. Maybe when we meet the Vulcans and they teach us mind melds that will be different.
 
Plastic transistor vs metal, assuming the datasheet specs are comparable, should make no difference at all.
I believe there are some transistors that use exactly the same die, but are mounted in either metal or plastic, presumably for customer choice. (i.e. very likely MJ3055/MJ2955 & MJE3055/MJE2955)
 
Last edited:
Well, I think the whole premise of metal vs. plastic is nonsense, so probably the wrong person to ask. Hint- if you blindfolded me and hooked various transistors to my curve tracer, there's no way I could tell you if they were metal or plastic.

Absolutely true.

The issue in actual equipment arises because plastic transistors are often poorly heatsinked or overdriven, or both, which overheats the device and gradually damages it. So the difference between in vivo and in vitro is a consequence of poor implementation, not a difference in specifications or performance when the part is operated within manufacturer guidelines and limits.
 
Well, I think the whole premise of metal vs. plastic is nonsense, so probably the wrong person to ask. Hint- if you blindfolded me and hooked various transistors to my curve tracer, there's no way I could tell you if they were metal or plastic.

Of course I couldn't see the curve tracer either. I'm blindfolded. Ba dum tss!
 
This is why you need a small hole in the blindfold.

Isn't that how audiophile experiments are conducted?
 
Back
Top Bottom