What are these NOT gates for?

[hg_man]

I'm looking at this schematic of PAIA's starved-plate tube mic pre:

http://www.paia.com/fantusch.gif

(Yes, it's a bad graphic!)

I can figure out most of what's what here, but what is this arrangement of 5 NOT gates with two resistors and a cap at the upper center? They feed into a voltage multiplier at upper right, that much I figured out. But what do they do? I'm betting that the resistors control the rate at which the cap charges and discharges, but I can't figure out how the NOT gates fit in.

Also, at center left there's an LED, which is the phantom power indicator. It's turned on by the phantom power bringing the NOT gate input high, which brings the output low, forming a voltage differential between Vcc and the output. But why bother? Why not just have the LED tapped into the phantom power so that when it's on, the LED turns on? Isolation? because the designer had an extra NOT gate handy?

Thanks for the help,

Alden

 

[bcarso]

The inverters etc. are an oscillator etc. and drivers for the cap-diode voltage multiplier.

 

[SSLtech]

Drat you bcarso. Always first with your right answers... -well, I 'll have no more of it, -you hear?

Harrumph...

To put a different slant on things, the NOT gate is NOT for keeping the sheep in the meadow.

Go ahead... try and prove me wrong! :wink:

Keef

 

[hg_man]

Thanks for the informative answers.

On the way home on the bus I figured it out, starting with the premise of "OK, what if the input of this one is low.", and going round and round until everything repeated itself.

NOT gates for NOT keeping the sheep in, huh? AND what else are they good for? :wink:

 

[Mark Burnley]

Further mind expansion available here:

http://www.web-ee.com/primers/files/WebEE/AN-118.pdf

It's an XORbitant use of gates if you ask me

:wink:

Mark

 

[bcarso]

I like the schmitt-trigger inverters for a single-gate surefire oscillator with one R and one C, although the delta V of the hysteresis is spec'd all over the map and with it the oscillation frequency.

Ah--I see it is on page three of the app note.

Funny story about using CMOS inverters of the non-schmitt variety for linear uses I have. But no time to relate.

 

[SSLtech]

[quote author="hg_man"]AND what else are they good for? :wink:[/quote]
.. it's what they're NOT good for that matters! :wink:

Keef

 

[PRR]

> Why not just have the LED tapped into the phantom power

OK, we start with 15V, jack it up toward 75V, clamp it at 51V, feed a tube, a mike, and an LED that only needs 1.7V with significant current. That's pretty simple and efficient..... NOT!

If this was a proper 48V power supply, another few mA for the LED would not be a big deal. But this is the output of two CMOS gates, levered-up by 5. I bet they don't even make proper 48V 10mA power as the recent P48 specs call for. It may be straining to do 2mA, the old P48 spec. A similar mini-mike-amp in my bag sags badly with just 2*2mA, without the tube or LED. PAIA realized that, didn't want to add another CMOS chip just to feed an LED the hard way. While they could have used a double-pole switch to connect the 48V Phanton .AND. feed 15V to the LED, if the switch half-fails it could be misleading. Their trick of slamming a CMOS input with 48V through 470K seems rude, but does not pretend that switches never go wonky.

 

[josan]

I did this as one of my first DIY and it now sits on the shelf unusable. With all the info gained from the Lab one look at the pcb with wires poking out from every place on the board and a switched power phantom power supply on the same pcb. ....................Then I did a JLM board............... "it's not the design but the details." a PRR quote sticks constantly in my mind now.

 

[hg_man]

The more I've been looking at it, the more I realize how many places they cheated... but there it is. I'm having a Basic Electronics class build one as a class project, because they've been asking to do something "real" in addition to the little $6 preamp kits. I'm hoping that it will work out as a class project - perhaps I can get some mileage out of the things by upgrading and replacing bits and pieces (starting with the power supply, I think).

I'd welcome suggestions on a better kit or project to use that could be finished in a 10 hours or less and doesn't cost tons of money. (yeah, I know... and can I pick up a perpetual-motion machine too?)

 

[clintrubber]

from Brad:

The inverters etc. are an oscillator etc. and drivers for the cap-diode voltage multiplier.

Be sure to check the output-voltage this supply-arrangement. I once build the PAiA SIAB which uses an alike supply but didn't deliver the expected amount of Volts. Starves the tube even further...