Krill - The Next Generation

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Calling "various voices that mired the threads in the past" is a safe one. They'll be back, on the double.

Windmills can't win; they only spin.

I was not the one that said that.

he charged at Rocinante's
fullest gallop and fell upon the first mill that stood in front of
him; but as he drove his lance-point into the sail the wind whirled it
round with such force that it shivered the lance to pieces, sweeping
with it horse and rider, who went rolling over on the plain, in a
sorry condition.
 
Member
Joined 2006
Paid Member
Don Quixote, his horse Rocinante and his squire Sancho Panza after an unsuccessful attack on a windmill. By Gustave Doré.
 

Attachments

  • 220px-Quixo-panza.jpg
    220px-Quixo-panza.jpg
    26 KB · Views: 359
Steve:

'Progress can only depend on the unreasonable man' - GB Shaw, 1904

Fashions and trends abound in science and technology, just like the fashion industry. The conventionalists control proceedings with high priesthoods and ridicule. Galileo and Copernicus come to mind. But it's hard to forget that Nikola Tesla was regarded as an eccentric fringe dweller, yet he gave us many high voltage techniques, the AC power grid, the AC induction motor, and many other inventions.
 

GK

Disabled Account
Joined 2006
Perhaps your Dick Miller quote has given them pause.


Yes, I'm sure that was it. As far as the operation of this circuit goes there really is nothing to add to what Andy said in post 7 and showed quite clearly in his subsequent simulation plots.

And can't believe those who think this scheme works are still sticking that big cap between the driver bases, which absolutely guarantees reverse bias of the driver+output device b-e junctions at all but the lowest audio frequencies. Unreal! :rolleyes:

I just tried to download that Intusoft demo package, but it is 48Mbyte - far too large for my 48000bps dial-up connection. I'll try it on a computer at work some time next week.
 
andy_c said:

It will eventually turn off, given time. Try it for 20 Hz for example.

GK said:

And can't believe those who think this scheme works are still sticking that big cap between the driver bases, which absolutely guarantees reverse bias of the driver+output device b-e junctions at all but the lowest audio frequencies.

So, which one is correct?
 
I just tried to download that Intusoft demo package, but it is 48Mbyte - far too large for my 48000bps dial-up connection. I'll try it on a computer at work some time next week.

I just downloaded and installed it, and loaded Steve's file into it. Steve's file is a netlist and not a "drawing file" as Intusoft calls it, so it doesn't allow viewing the circuit in schematic form. The demo won't do a simulation either, as Steve's circuit exceeds its maximum allowed device count. On the plus side, the netlist has all the models embedded in it, allowing them to be ported to LTspice. But reverse engineering a schematic from a netlist is quite time-consuming.

Steve, would it be possible to post a graphic of the schematic used to generate that netlist? The schematic I saw in the demo had the node numbers clearly marked, which would allow quick reconstruction of the sim in LTspice, using the exact same models as are in the Intusoft netlist you posted.
 

GK

Disabled Account
Joined 2006
andy_c said:

So, which one is correct?

No contradiction there.

Here is a run of the sim file posted by Andy in post 11:

http://www.diyaudio.com/forums/solid-state/151295-krill-next-generation-2.html#post1918262

I'm showing something different this time - the voltage at the base of the NPN driver along with the voltage at the emitter of the NPN output device, at 2kHz with a 4 ohm load.

On negative swings the B-E junctions are reverse biased, exactly as theory predicts.
 

Attachments

  • krill.jpg
    krill.jpg
    128 KB · Views: 353
Last edited:
Here is a run of the sim file posted by Andy in post 11:

If you look at the Vbe of just the output device by itself though, it ends up being forward-biased when one might expect reverse bias. But its base when in that state is essentially open-circuited (except for strays), since the Vbe of its corresponding driver has a hard reverse bias. So the output device is kind of hanging in the breeze there. Its Vbe is forward biased, but only due to stored charge.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.