What is wrong with Darlingtons?

Member
Joined 2009
Paid Member
Hi to Everyone ! at the cost of sounding tedious i would like to understand more about why darlingtons are so neglected in audio power amps
apart from a few small-scale British designers it is almost impossible to find integrateds or power amplifiers that use darlingtons as output devices
If nothing else they allow you to skip a driver stage, right? thus simplifying the project as a whole
From a completely ignorant point of view they should be a godsend for lovers of minimalism
Not only that I know at least one medium power amp that uses darlington and it sounds pretty good, definitely good actually
These are the Albarry m408 monos which at least in my country Italy had become quite famous after a highly rated reviewer had praised them
Why does everyone avoid them? do they stink maybe? 😵
😉
 
Bebo Moroni? His reviews should have be taken with a grain of salt as many of the stuff he reviewed seemed to be like the holy grail of audio.
Hi ! yes ! he was him Nevertheless people who have been exposed to the amps in question had all a very positive impression They are mono
It is using a darlington output pair but also a very big power supply built with a 300VA transformer (20VAC-0-20VAC)
A lot for a 40W/8 ohm amp
 
  • Like
Reactions: shadowplay62
Hi thanks a lot for the very valuable information I read somewhere that the main issue with the SAP15N could be the embedded power resistor
For sure the more embedded things are the easier the design comes out (just think of power packs) But could it be that the resistor heats the chip up ?
Instead usual darlingtons do not have that power resistor inside If i am not wrong
 
Hi thanks a lot for the very valuable information I read somewhere that the main issue with the SAP15N could be the embedded power resistor
For sure the more embedded things are the easier the design comes out (just think of power packs) But could it be that the resistor heats the chip up ?
Instead usual darlingtons do not have that power resistor inside If i am not wrong
Sanken STD03 had no 0,22R embedded emitter resistor, all are obsolete afaik.
 
  • Thank You
Reactions: ginetto61
never did make any mistakes
I am sure he would not have published this amp if it was a mistake.
after all he had plenty of time to think over it:

I am inclined to demonstrate my goodwill in this manner and
do in fact have a 13 year old piece laying around unpublished,

my comment wasn't to be taken that seriously, by the way ... 😊
 
  • Thank You
Reactions: ginetto61
  • Thank You
Reactions: ginetto61
Apart from biasing, thermal drift, double Vbe loss and charge storage troubles they are fine. One should look at the discrete Sziklai/CFP pair.
https://en.wikipedia.org/wiki/Sziklai_pair
hi thank you for the very valuable advice Still if i am not wrong at the output of common amps i see a bjt driver driving an output bjt of the same gender
i.e. like in a darlington
If they develop more heat they can be designed to dissipate it
I think that in some cases the idea of putting together on the same chip more similar parts is very nice Like in the Sanken multiple power bjts I love them
Here where i live they are called savonette
 
Hi dear Mr Pass thank you so much for your kind and very precious advice
so in the end there are reasons for their not so great popularity I am sure and understand that any designer likes to have free hands in designing his/her creations
Heat is also a problem Maybe embedding more parts on a same chip can work when the parts are identical like in some Sanken power devices ?
and they are quite more popular actually
 
Last edited:
Another reason I was never a fan of integrated darlingtons was that (at least back in the day) their data sheets were often a bit sparse for data that you might want to consider for audio application decisions. fT and hFE flatness vs current come to mind as a couple of examples of missing info - I had a college instructor once show a massive frequency response difference between similarly biased emitter followers built with incompletely specified integrated darlingtons vs darlingtons composed of discrete parts that were more thoroughly characterized. He was trying to make the point that what's not spec'd can bite your a$$ if you don't pay attention.
 
  • Thank You
Reactions: ginetto61