With the "correct" choice of IXYS devices it looks like cascoding of the output fets might only reduce input capacitance by about 50% with an 8R load, even less with lower load impedances.
Last edited:
Keep going brother. 
Is this similar to the type of thermistor you are bolting to the heatsink?
http://www.vishay.com/docs/29092/ntcalug.pdf
Is this similar to the type of thermistor you are bolting to the heatsink?
http://www.vishay.com/docs/29092/ntcalug.pdf
Attachments
I have used these: Mouser part #594-NTCALUG02A472F http://www.mouser.com/ProductDetail...GAEpiMZZMuBd0%2bwiCVS29cFeDjED7IBg/nN8NMR3zg=
Attachments
yeah I got the 0.00X at least and first in Spice even uncascoded......
I remembered that the original XA25 is said to have 20dB gain.
I always worked with 26dB gain, coming from XA30.5.
So I stayed for the gain resistor with a value of around 110R and choose a lower value for the feedback resistor, around 1k2.
Now with 20dB gain only,
it is no more problem to get for 1W at 8Ohm the 0.00x value.
Let us see next days what happens in reality.....
Not to speak of the work to get possible neg k2 phase or use of CFB.
All in all playing is really fun.
More awful is of course the part to hear what really sounds better.... I tend to avoid this part and admire Nelson and his people doing this in a great scale.
I remembered that the original XA25 is said to have 20dB gain.
I always worked with 26dB gain, coming from XA30.5.
So I stayed for the gain resistor with a value of around 110R and choose a lower value for the feedback resistor, around 1k2.
Now with 20dB gain only,
it is no more problem to get for 1W at 8Ohm the 0.00x value.
Let us see next days what happens in reality.....
Not to speak of the work to get possible neg k2 phase or use of CFB.
All in all playing is really fun.
More awful is of course the part to hear what really sounds better.... I tend to avoid this part and admire Nelson and his people doing this in a great scale.
Attachments
yeah I got the 0.00X at least and first in Spice even uncascoded......
.
About bloody time you dumb bastard. Hahahahahaha
You were making me sound like a broken record. "you don't need cascode,,,,,,you don't need cascode...... you don;t need cascode"
a million bloody times.
Hahahahahaha
Good to see you happy.
Last edited:
you did not tell me how to do it, so it took me a time to find it out....!
Brocken record? long time ago this happened in my house.....
Excellent work brother.
Hello brother
As an alternative to careful part selection, and other more technical techniques to achieve your goal, why not do something dumber.
If you add 0.05 Ohms (50 mOhms) to the source of the N Channel Mosfet it will give you the effect you want. I don't think this small amount of degeneration would hurt the natural bias scaling of the output stage. For the cost of a dollar it's worth a try.
Even 25 mOhms will do the trick (2 in parallel).
I might play with this myself. I like dumb ideas. Hahahaha
https://www.digikey.com.au/product-...ctronics-inc/MPR3JB50L0/MPR3JB50L0-ND/1646064
Last edited:
Yes, i will try it.
First i have to train me again to measure the residual with Virtins software to know Rallye what k2 Phase comes out.
Lhquam already told that you have to overwhelm a lot of pos k2 phase to come in the neg region.
By the way I can live with pos phase as well, sometimes I like it even more.
First i have to train me again to measure the residual with Virtins software to know Rallye what k2 Phase comes out.
Lhquam already told that you have to overwhelm a lot of pos k2 phase to come in the neg region.
By the way I can live with pos phase as well, sometimes I like it even more.