[R-390] Re: BallasTubes (was inrush current limiters)

[Drew Papanek]

> >As far as regulator-induced noise is concerned, filtering would help but 
>I
> >am not sure that this is necessary.

Jim Shorney wrote:

>Manufacturer's recommended filter/bypass caps are always necessary,
>IMHO.  I remember the time a CBer brought me a home-built 12v power
>supply that would spike to >18 volts when he unkeyed his radio.  The
>solution was to get out the data book and install the caps for the 317T
>that National Semi said shouldbe there for stability and transient
>response.  There was also the blurb in QST years ago from the ham who
>tossed a bunch of bypass caps at the regulators and zeners in, IIRC, a
>TR7 and IC551 and saw a noticable improvement in the noise floor.

The maufacturer-recommended filter/bypass caps go without saying!  Three 
terminal regulators can make good oscillators without them.  For LM317 
certain values of output capacitance will cause excessive ringing: a 
too-close cousin of oscillation.  I believe the evil values lie within 
500-5000pF.  This range is swamped out by the recommended value.
Additional filtering beyond that needed for stability may not be necessary, 
but wouldn't hurt.
>
> >Another solid state regulator approach is to half wave rectify the 
>25.2VAC,
> >filter and apply to a 3 terminal regulator configured as a 300 mA DC
> >regulator.
>
>
>The same without filtering should give a clipped half-sine wave as
>above.  Since we start out with an RMS of 17.8 volts, there should be
>enough headroom.  Another variation I've been thinking about.

Interesting method I hadn't thought of.  Dr. Jerry's clipped sinewave 
circuit regulates on only part of the waveform; when instantaneous value 
drops low enough, regulator saturates.  This requires a peak current of 
about 360 mA to achieve 300 mA RMS.  Dr. Jerry verified this value with a 
fair amount of effort, and my computer simulation agreed.  With the clipped 
half-sine wave circuit, that peak would have to be somewhat greater for the 
same RMS, so more fiddling about with true RMS current measurement 
techniques (can sometimes be a real pain) would be required for 
verification.  This would make for another interesting computer simulation.

These circuits would generate considerable heat; the advent of the coveted 
cool-running 3DW7 Tubester makes all of our regulator musings sound trivial 
(sigh).


Merry Christmas to all; and to all a 3DW7 Tubester in the stocking!

Drew

_________________________________________________________________
MSN 8: advanced junk mail protection and 3 months FREE*. 
http://join.msn.com/?page=features/junkmail&xAPID=42&PS=47575&PI=7324&DI=7474&SU= 
http://www.hotmail.msn.com/cgi-bin/getmsg&HL=1216hotmailtaglines_advancedjmf_3mf