HOW TO MAKE A PROPER AMPLIFIER

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Here is an example of an amplifier that eventhough made almost 15 years ago preserve some rules that a lot of people scream about in our forum for many years ( for people that manufacture and construct ) and the amp it shelf sets a level of perfection in amplifier construction .

To my understanding what makes the construction perfect is that the designer had a perfect relation with the pcb and production engineers and set up rules that in no chance should be overlooked .

In the audio history is a common secret that things that designed from grate designers like NP for example when ended up in the production line results was seriously altered and final product was actually compromised by the production engineers or the accountant's office . Adcom i think is one of them but it should be a lot more ...

Important thing is to notice that amplifier achieves superior performance with use of simple material that can be found anywhere without the use of any exotic parts .


down under i will post a few pictures and a few comments on this superior amplifier

DSC00513.JPG


here is the general picture , middle 60.000 bank of caps , clever pcb even distribution , less wiring , very clean work with detail and order .

DSC00514.JPG


Here is the front board , switcable balanced inputs, ground lift switch , LTP placed a few cm from input RCA , on board you have the first stages of the amp up to the pro drivers area ( very clever since that area is the most clean and most far away from dirty stuff like the psu and output combo zobel /inductor/dumper ) loads of decoupling/bypass for this input stage 100mf +3.3mf+100nf per rail per channel , thermal junction on the LTP , arrays of input capacitors 2.2 mfd X3 times +100nf instead of any exotic capacitor of 6-10mfd less microphonic and much faster ( i think )

Notice also that the manufacturer had the balls to do what is proper IE add decoupling caps down under the board on the input IC hardwired standing on the air to benefit from them as close to IC is possible eventhough this doesn't look in order and neat .


DSC00515.JPG


Here is output detail while 8 mosfets form the output while the arrangement is 4 N and 4 P the implementation is N-P-N-P N-P-N-P resulting absolutely even distribution of power among the mosfets but also combined with a vbe multiplier in the middle an amplifier without any chance for any thermal run away

DSC00517.JPG


Enjoy the output board with all the dirty stuff on it zobel inductor dumper that of course features separated return to 0Volts in order to keep the amplifier's ground as clean is possible

notice also that all protections of the amp ( deferential overload , dc , thermal , and short circuit) are mostly around optocouplers .


ladies and gents you have enjoyed an audio masterpiece from PERREAUX
THE PERREAUX 350 proudly made in nuclear free New Zealand !!!!

kind regards
sakis
 
Last edited:
well if i understand the question correctly i can tell you that the pcb has drills through and its a dual layer but soldering is made on the top and if the amplifier needs to be serviced for any reason that will happen this way very easily .

Also some manufacturers think of that and make drills and pads and trough's with a bit of tolerance so replacing parts is easy .

For the specific amp all boards can be removed individually for service and without any complicated wiring which means that the designer thought of that also ....
 
Enjoy the output board with all the dirty stuff on it zobel inductor dumper that of course features separated return to 0Volts in order to keep the amplifier's ground as clean is possible

I hope the output current isn't "dirty" - large, yes

I would call AB rail currents that are "half wave rectified" "dirty" - they should be resolved on the power output/power supply connections - the interleaved output devices helps reduce any contaminating magnetic field radiation of these "dirty" currents

but the output wire and loudspeaker return gnd wire should have only the current demanded by the loudspeaker flowing in them - usually not characterized as "dirty" signals - most loudspeakers are only low order moderately "weak" nonlinear loads - the output coils could couple to the inputs so it is worth paying some attention to - but nothing like the evils of the PS rail currents

doing load and signal gnd "right" can be difficult in multichannel amps with shared power supply - "dual mono" is "best", differential amplification referencing the output terminal gnd, "gnd loop breaker" construction can work with less than "perfect" gnds - but lots of copper - applied intelligently in hierarchical gnd schemes can be practical, "good enough"

if you worry about external RF coupling to speaker cables the shown construction isn't particularly great
 
well i haven't finished the repair yet so i cannot tell ... obviously there is a question on how one will put next to each other a 2x60 BJT and CFP amplifier next to a mosfet 2x350W ...it will not be possible

Still my guess is that at low power and casual listening under the same speakers i always use ( near fields/ easy load ) the P3a will be warmer....Obviously i dont like mosfets and especially at class AB

That ... though is my personal opinion and if i had the chance or the money or the load ( proper speakers ) i would definitely get a Perreaux by far from so many i have seen in the market .

Since you are asking i will make a detailed presentation when i am done

Still i feel very privileged to be able to work with such a machine .....

( haven't seen you for a while chlorophile ...good to see you )

Kind regards
sakis
 
Have you ever worked on a Crown? My Com-tech 400 is laid out very well and only high quality components are used.

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


You might not be able to see this but there are two transformers used in this amplifier. The small one actually doubles as a heat-sensitive fan that turns on when the amplifier gets too hot. My comtechs run so cool though that this hasn't happened yet!

You also cannot forget the grounded bridge technology, large heat sinks, CAD-designed chassis that optimizes airflow across the components, interchangeable PIP cards, ODEP protection, and distortion-sensing technology. I don't mean to rain on your parade, but I believe that there are other contenders for the title of "proper amplifier". ;)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.