Pass "DeLite" Amp from BAF

[stefanobilliani]

Originally Posted by vdi_nenna

If not using light bulbs, can we call it the Zen Ten?

Jajaja , nice name

[Skip Pack]

In standard form, the ouput capacitor is sized at 10k uf, presumably
yielding a high-pass frequency in the single digits when used with
a nominal 8 ohm speaker. I'm considering assembling a stereo Delite
for some compression tweeters that are used with a passive
high-pass spec'ed to 1700hz. The first element in these 2nd
order high-pass filters is a 5 uf cap.

Is there any reason I could not put that cap in the amp in place
of the big electrolytic and connect the output to the coil that is
the parallel element in the crossover? I realize this would make it a
system-specific amp, but I am comfortable with this sort of thing.
Am I missing some impedance related issues that make this a
bad idea?

Thanks,

Skip

[Zen Mod]

just go for it , as long you remember that's "system specific amp "

[Skorpio]

Originally Posted by stefanobilliani

And it works . Thanks a lot . So I set R5 1,47 for now , if it is correct probably we can also rise the rail Voltage to 40 volts or so

This looks familiar (my ZEN4). Would it be a worthwhile upgrade with the IXTH6N50D2 or would it be better to go for the power JFET (SJDP120R085)?

My idea is to drop the input buffer and biasing part completely and go for a Gate resistor of 47k and a source resistor...

Seems like nearly plug and play...

[Nelson Pass]

The Source resistor will be a major player, since it is likely to be
several ohms to give you the bias you want. Of course you can
bypass it with a large cap if you need to.

[Roscoe Primrose]

I tried the self biased R085 with a 1:1 OPT rather than the bulbs for a load. Needed about 2.5ohms source resistor with a +24V supply. Preferred the IXTH6N50D2 with 1ohm source resistor. Haven't tried bypassing the source resistor with the R085, that's a lot of uF ;-)

May try the R100 in there though....

Roscoe

[Pano]

Hey guys,
I've been putting together a dual rail power supply version of the De-Lite. Same basic idea, same MOSFET even, but a +/- supply to eliminate the DC on the output. No cap or transformer needed on the output.

It's works pretty well so far using +/-12V and a 12V-27W automotive taillight.

Just wondering if anyone has a spice model of the IXTH20N50 MOSFET that would work in Microcap. It's the spice flavor I'm familiar with, so it would be handy for me. So far, I've haven't simulated it, just built it - but the sims might get me closer to my goals faster.

I think Nelson uses Microcap, right? Any chance for a IXTH20N50 model? I've searched but have not found. Thanks!

[Pano]

Nobody, eh? Too bad.....

[ra7]

Pano, I'm interested. I had suggested earlier in this thread that a no-cap, dual rail version could be interesting. Do you have a schematic to share?

Although, I may not be able to help you on the simulation side.

[Pano]

Yep, there ain't much to it. See below. 12V and 12R is what I've been working with to start, but getting about 2A current flow may be the sweet spot. My 12V 25W bulbs can do that, but they are bright!

The idea is simple - as long as the voltage drop is the same across the MOSFET and the resistor (or bulb) then the node between them is at 0V compared to ground. Apply an AC signal and that 0V will fluctuate up and down. Voila - you've got an amp. To get the depletion mode MOSFET to drop the same voltage as the resistor/lamp, you have to turn it off a little. You want it to be the same resistance as the lamp. About 1V does it in this case.

There are some problems with this, but will go into that later, if anyone cares.