[sci.electronics] high voltage Power Supplies

levy@ttrdc.UUCP (Daniel R. Levy) (07/12/87)

In article <622@neoucom.UUCP>, wtm@neoucom.UUCP (Bill Mayhew) writes:
> I like the idea of using a switcher to supply the B+.  By using a
> switcher, you eliminate bulk and the higher frequency means that
> smaller capacitors can be used to provide equal ripple removal.

Switcher or no switcher, you still have to take care of power line ripple.
It doesn't magically go away with a switching power supply.  Think about it.
If the switcher is powered from unfiltered full-wave rectifier output, the
high-frequency output will also be thusly "modulated."  Sufficient filtering
(and possibly some regulation, which will also have the desirable side effect
of reducing ripple) on the rectified output must be provided.
-- 
|------------dan levy------------|  path: ..!{akgua,homxb,ihnp4,ltuxa,mvuxa,
|         an engihacker @        |		vax135}!ttrdc!ttrda!levy
|    at&t data systems division  |  disclaimer:  i am not a yvel nad
|--------skokie, illinois--------|

dje@datacube.UUCP (07/14/87)

The  cleanest,  lowest  noise  power  supplies  are linear regulated.
Switchers are *mostly* for high  power applications  since their main
asset is efficiency.   Their  other assets  are light  weight and the
ability to convert DC to DC.  If you want DC to  DC conversion that's
clean, one possibility is a switcher followed by a linear supply.  

I missed the original posting, but I heard  a few  references to tube
type preamps.  That means relatively low power and  I assume throwing
efficiency and weight to the wind.   If I  was building  one, I would
use a standard  60 Hz  power transformer  followed by  either tube or
silicon diodes and a reasonable amount of filtering.  Then a tube for
a  series  pass regulator  (cathode follower  configuration) or maybe
even a high voltage amplifier transistor.  

Tubes  as  diodes  have  more  inherent resistance  than silicon thus
causing  lower  charging  currents and  lower 120Hz  harmonics on the
power supply.  The linear stage then has an easier job.  

Tubes as  high voltage  series pass  elements take  abuse better than
transistors.  If  you're really  into nostalgia,  use an  0A2 (or the
like) as the power supply reference. They have a great glow.

 				Dave Erickson
----------------------------------------------
  ------      Datacube Inc. 
 /    /|      4 Dearborn Rd. 
------ |      Peabody, Ma 01960
||  \| |      ihnp4!datacube!dje
||  /|/       Human:(617)535-6644
------        Fax:  (617)535-5643

wtm@neoucom.UUCP (Bill Mayhew) (07/19/87)

OK, I'll conceede that if you want to perform linear regulation at
tube type voltages, a tube for the pass element is a good idea. If
you accidentaly short the output cathode to ground, you've a few
seconds to rectify (*a pun *) your mistake, rather than the
microseconds you'd get before you send a transistor off to stick in
the ceiling.  A 6L6 is an easy to get tube that makes a good pass
element.  A pentode is would be a good idea to isolate your 0A2,
0D2, or whatever reference you use.  There are some VR circuits in
the old RCA receiving tube catalogs and probably old ARRL ham radio
books.

As far as switchers go, they aren't too bad if you design carefully
so that it can operate in the MHz region, rather than KHz.  Any
noise wouldn't be audible.  With a switcher, you can use something
like a .1 uF fancy "audio grade" capacitor to get all the filtering
you need.  It's getting downright difficult to get decent n hundred
uF capacitors* that are required for 60 (120) Hz operation.  60 Hz
magnetics weigh a lot.  Of course, building RF switchers calls for
very careful design, and care at bypassing leads, etc.  I think I'd
put a linear VR circuit after the switcher for extra insurance.

* at least, not at tube voltages!

We have RF switchers in medical equipment that amplifies uV signals
and don't have any noise trouble.  Using an RF power supply is
required so that ~5 cm. air gap can separate the primary and
secondary of the *air core* power transformer.  This baloney is
necesary for "patient safety".-- although our patients are lab
rats.

Bill Mayhew
Division of Basic Medical Sciences
Northeastern Ohio Universities College of Medicine
Rootstown (what a name), OH  44272-9989  USA  phone: 216-325-2511
(wtm@neoucom.UUCP  ...!cbatt!neoucom!wtm)

ken@rochester.arpa (Ken Yap) (07/21/87)

Isn't this all getting a little of overkill, regulators and all that?
As I recall the original poster only wanted to feed a couple of voltage
amplifying triodes in his audio preamp.

A centre tap 250-0-250 transformer @ 20 mA, a pair of 1N4007 diodes, a
pi section R-C filter, or maybe two, should do the job just nicely. And
a fuse on the primary of course.

Oh well, back to the nostalgia discussion... :-)

	Ken

wtm@neoucom.UUCP (Bill Mayhew) (07/25/87)

The high voltage power supply over kill issue was brought about by
the use of the word, "audio".  From reading net.rec.aduio, I've
learnt that circuits for audio applications must be over designed
by a factor of 5 to 10 in order to convince audiophiles that some
horrible bastardization of the sound is not taking place.

For circuits that require a few mA of current, a switcher supply
isn't too terrible.  Recently, I wanted ~200v at about 25 mA.  I
was able to use a ferrite ring and a couple of power transistors
from my junk box to build a "ringing choke".  The output of this
went into about a 1 uF mylar capacitor after being recitified.  I
used two 0B2 tubes to shunt-regulate the output. The whole mess,
except for the tubes, fit inside of one of those little Radio Shark
aluminum boxes and only weighed about 100 grams.  I resorted to
that since I didn't have any old 60 Hz transformers and big filter
caps laying around.  I followed the design (not really copied) from
one of those big "we publish every basic circuit you can imagine"
type reference books.

Bill