I need help with this amp

[wonder_b]

I am trying to change this amp a bit but I need some help to make sure I do it right not twice .I have changed the output transisotrs from npn on both sides(npn.jpg) to npn on - side and pnp on + side and what do the cf diodes do can i get rid of them with no problems and are the transistors i have marked the ones that control thermal runaway.and also more volts but I dont think this will be a problem can anyone tell me if I have it correct or not will it work

[wonder_b]

the new schimatic with outputs npn - side pnp + side and no cf diode

[destroyer X]

Regards,

Carlos

[wonder_b]

carlos do you want the spice file

[destroyer X]

I have to stop and go out..... in a matter of some ours i will try more...not working!

I did not think, just put it in Multisim....this afternoon i will concentrate my last neuronium on it.

regards,

Carlos

[MikeB]

Hi !

Are you sure, that you want increase the voltagesupply from
94v to 120v ? This will change all biasing, unless you adapt
some resistors. (like the 30k to 38.3k)
And you will push the mje15032/33 to its absolute limit of 250v.
You could use mje15034/35 instead.
But the mj15024 output still has a 250v limit, and theres no
substitute in this family with higher voltageratings.

And i think you should skip the 12e-resistor at collector of q3.

When building the amp, you should match the mje340/350 pairs,
these are rated with a hFe 30 - 240. If you use that different
transistors in the VAS or Diffamp, you will get big problems.

http://www.onsemi.com/site/products/...MJE340,00.html

Michael

[ChocoHolic]

Hi wonder!
Yes, the marked transistors are for thermal compensation.


But if you want to go for a symmetrical npn/pnp outputstage, then I would strongly recommend the following:
-Towards the positive rail keep npn for q5 and CF as it was in the original schematic. CF diode provides a current feedback at high output currents and will limit the output current in case of overload or short circuit.
-Towards the negative supply use a pnp type in the symmetrical way and also but the diode.
- If you go for a higher rail, you should keep some reasonable margin against max. rating. About 100V would be enough for my taste.
Already then you should increase the 0.1 Ohms to 0.12 Ohms as
with higher currents you must limit at lower current values.
And also take care about the changed biasing of all stages at higher supply as Michael already stated.

... I will modify your schematic to show what I mean....

Cheers Markus

[ChocoHolic]

....uhps, while modifying the schematic I noticed that diode CF cannot provide a current limit, if it is put with the polarity as your first schematic shows....
sorry, have to think a little bit more about that.....

[K-amps]

Originally posted by MikeB


But the mj15024 output still has a 250v limit, and theres no
substitute in this family with higher voltageratings.

Michael

Ofcourse Micheal, but the Mj21195/96 will give much better SOAR therefore better reliability at high voltages... right?

As for family of devices, the MJL4302 and MJL4381 are same family (albeit in T-O3PL pack) i.e. Powerbase power transistors and have VCE ratings of 350v. Then again you will need many in paralell...

[ChocoHolic]

OK, it looks like that the diode CF is put in order to enable higher output voltage swing.
Without diode CF the 12 ohms emitter resistor of q3 would limit
the max. output swing at a remarkable value below the rails, if high output current is drawn. If CF is placed then q3 can draw it's current
also through the diode CF and will be able to allow an output swing close to the rails.

Nevertheless I would recommend to arrange the npn at the positive rail and the pnp at the negative rail for this circuit as shown in my attached image.