Best 100-W high fidelity audio NPN Transistor ?

Status
Not open for further replies.
Jump to Latest
#61
>You could actually improve performance by choosing a transistor (such as the "C" version of a BC550) with a higher beta and a lower Early voltage for the common-emitter portion of the VAS cascode. The improvement would come from increased current gain of the VAS.

Andy, this occurred to me as well - the super-beta, low Early voltage transistors like the 2sc3112/3113/3807 would work well as the lower-limb of a cascode, and there's no need to worry about the Vceo either. The upper, common-base section can take care of Vceo as well as linearity, even if it has lousy beta; 2sc3953 and other NEC/Sanyo CRT chroma/video amplifier transistors come to mind as good candidates.
 
#62
Krokotam

Too bad I don't understand a word of bulgarian, I would have a lilttle
more comprehension of the details you guys discussed there about the amplifier. Bulgarian is not one of the languages I speak.

However it looks like that amp is a little on the classic side of an AB
class amplifier with protection, which is a good thing. Mine (I wish it
was mine) is a class A (lots of heat...) designed by an engineer named
John Lindsay Hood who made his name by designing this kind of amplifier, among other very valuable audio projects.

The only thing I did was to implement his design in a more powerful version with japanese transistors. This amplifier requires heafty
heatsinks to dissipate all the heat from 4 output transistors (each side)

This the schematic of a similar amp I built a few years ago. It's not that different and it has a very good quality.
 

Attachments

#65
Glen I think you still got it wrong, the vas is a single darlington transistor, a mpsa13 in fact. This is cascoded, and then followed by a buffer.

If you have his book, its a combination of figures d and e. In d the vas transistor is a darlington and cdom connected between c b of it. This is followed by buffer as in figure e. There is no emiter follower like in the blameless.

Any comments on this arrangement.
 
#68
homemodder said:
Glen I think you still got it wrong, the vas is a single darlington transistor, a mpsa13 in fact. This is cascoded, and then followed by a buffer.
Click to expand...



By "buffer" I was referring to the first transistor making the darlington.

If the VAS is not cascoded then the collector of this transistor should be connected to AC ground so that the VAS performance isn't ruined by signal feedback via the non-linear b-c capacitance.
 
#69
homemodder said:
but the scheme I mean is completely different and is not in his books. Lately he has used a single transistor darlington vas , cascoded and followed by a buffer.
Click to expand...
pages 85 to 92 of audio power amplifier design handbook by D Self ed2 discusses VAS with and without cascode and with and without Buffer and EF.
I think fig4.20 shows the improvement between a cascoded Vas and a cascoded VAS with output buffer.
It would appear from your description that Self is now using the slightly more expensive VAS+Cascode+Buffer in the Cambridge if I have understood your description.
 
#70
Best 100W ....

Douglas Self has also made another change. The attached is from a reply to me in another earlier thread.

SandyK


...... the input stage and VAS are fed from a separate supply a few volts below the main V- rail; this gives more output and allows enough voltage headroom for the VAS circuitry. This aux supply is RC filtered but not regulated in any way.....
 
#71
Adrew you got it right exept that the vas is a single darlington transistor with very high gain. I would rather call it cascoded darlington + buffer. Look at mpsa13 datasheet. The collector of this transistor is connected to the current source, there is no connection to ground.

Sandyk,besides other benefits to having seperate supplies for the LTP and vas i think in this case he is addressing the unsymmetrical slewrate deficiency of blameless design.

Which thread is this going on??
 
#74
Can anyone provide Spice models for the 100 Watt Sanken

PNP 2sa1186
NPN 2sc2837

discussed in this thread? LT Spice would be best, but models from any Spice simulator would be very helpful. A Google and Sanken site search found nothing.

The Sanken datasheet shows ft=60Mhz and Cob=100pf, which makes these interesting $3 outputs.

Thank you,
Thank you,
Thank you.
 
Status
Not open for further replies.
Top Bottom