Discrete Darlingtons vs. Packaged Darlingtons

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
This is probably a newbie quetion, but I never understood why:

How come many class AB or Class B solid state amps that use BJTs for their output stages use discrete-built Darlingtons (sometimes upwards to three - two driver pairs and then the output pair) as opposed to fabricated Darlington transistors (the ones that have two transistors in one package)?

I know that fabricated Darlingtons have lower transition frequencies, but I would *think* that making a discrete pair would also have a low ft...
 
Discrete components allow Class A drivers and more robust (SOA) parts in general; packaged Darlingtons are usually hard-wired so one is locked into the classic configuration and that limits design choices and
flexibility.

Some designs such as the Leach amplifier use a triple Darlington configuration. I don't know of any packaged devices like that. I've never seen a Sziklai/compound-complementary device either.

Ft varies, but there are plenty of audio power devices that go up to
30 Mhz or so, and drivers that go even higher.

I think there have been some hybrid packages that contain a full
output stage, but that locks one into the choice of devices and usually a considerable limit on output power.

Onsemi may have a good idea in packaging a thermal diode with an output device, even though the device is pricey and requires a 5 pin
package with a non-standard pinout array.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.