Power amp under development

Jump to Latest
#3,261
qsa said:
against 2SD669 is a driver

ok its true, but I have different BJT PA Amplifier with this device as VBE and its working very stable?
Click to expand...
the 2sb649/d669 hits maxFt@~120mA.
and hits maxhFE@~100mA.
a 2sc3423 does the maxfT @~14mA and max hFE from 0.3mA to 14mA.
@4mA is manages ~190Mhz and 200.
Whereas the d669 manages ~<70MHz and 205, But I don't believe the hFE graph applies to the majority of available d669 and certainly not at 4mA.
 
#3,263
look at how a CFP is protected.
you need half of that plus half of the protection for a EF stage.

The bottom of the quasi is effectively CFP and the top of the quasi is EF.

Both halves of the protection do not need to connect to each other. They monitor the Voltage across the device and the current through the device and link the drive signal to the adjacent rail.
The EF drive gets linked to the output line.
The CFP drive is linked to the negative rail.

Look at the protection on the Cyrus one. It's a Thyristor triggered by excess current. The Thyristor mutes the amp to remove drive. However, I think it resets on every half cycle. It would be nice if it muted for a fixed period or latched.
 
#3,265
The crowbar won't work the way shown either, the triac is the wrong way round and the 0.47 ohm resistor would need to be very big not to explode if the crowbar needs to activate. Also, did you make sure the triac fusing I^2t product is greater (with a margin) than the let-through I^2t rating of the rail fuse?


AndrewT said:

Both halves of the protection do not need to connect to each other. They monitor the Voltage across the device and the current through the device and link the drive signal to the adjacent rail.
The EF drive gets linked to the output line.
The CFP drive is linked to the negative rail.
Click to expand...

But if done that way (after the driver) the driver and the protection will fight and one or both will blow up and the output stage is likely to follow.

A common way to do this in quasi amps is to insert a collector/drain resistor into the collector/drain of one of the devices (used in some Crown amps among others) or a common resistor for all of them (see the BGW schematic) and shunt drive to the output before the driver.

A latching protection muting the amp might be a better idéa though and won't need the extra resistor.
 
#3,267
triac protection

hi qsa,

What is the purpose of crowbar triac? since original quasi design use allready a relay protection?

for my understanding, f u put the triac protection, fusing must be required, because if the triac is activated due to unwanted signal or overswing signal, there is possible that your trannies will be damage.

rlg,🙂
 
#3,269
Hi

have changed T8 VBE Transistor from 2SD669a to BD139, MJE 340/350 to MJE15034/15035 because this new devices are faster and modern

I have add Short circuit protection, Clip and Signal LEDs circuit, crowbar protection is removed


RAIL +/-130 V

Maximum Output power about 3000W@2OHM EIAJ / IHF......?

I dont know....

I think the time is right to produce sample PCB and build this amplifier.

Does somebody find major faults or ideas to improvement ?

Have a look to pdf
 

Attachments

#3,270
Hi,
2915W into 2r0 is an impossible target for an 8pair output stage when asked to drive a reactive speaker load reliably.

The peak dissipation per device when the 2ohm load has a 45phase angle is ~586W.
How long will a 460 last, trying to achieve that on a transient?
and if the load is 60degree phase angle the peak dissipation per device jumps to 720W.
 
You must log in or register to reply here.
Top Bottom