[AMPS] Screen regulators circuits
John Fielding
John Fielding" <johnf@futurenet.co.za
Wed, 18 Nov 1998 23:38:33 +0200
This is a multi-part message in MIME format.
------=_NextPart_000_0006_01BE134C.A4E89220
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
This discussion seems to be going around in circles and not really =
getting anywhere.
May I be permitted to add my $10 worth on what in fact is a straight =
forward problem.
1. The problem regarding the break-down voltage of transistors or =
fets is really very simple to get around. In the off-line switch mode =
power supply industry this problem was solved many years ago by the =
simple expedient of connecting 2 transistors in series to double the =
break-down rating, exactly the same as putting several zeners in series =
to achieve a higher voltage. By utilising a low cost IC known as a =
"High-side / Low-side" driver the top transistor base / gate is operated =
at the required higher voltage (half-rail)and the bottom device is =
simply treated as a single transistor shunt stabiliser. The two =
transistors are connected in series and the bases / gates are =
effectively controlled in parallel, except the top transistor has the =
higher required voltage to ensure correct operation. High-side driver =
IC's are available with up to 600V operating capabilities and cost a =
couple of $'s. The circuit operates as the classic shunt stabiliser, in =
my opinion the only suitable solution for a tetrode screen supply.
2. On the question of the required dissipation rated transistor for a =
shunt high voltage stabiliser. Take a look at TV line timebase =
transistors, here operating voltages of 1500V are common with power =
dissipations of 50 to 125W. They are low cost and freely available. =
They only need the appropriate heatsink and control circuitry to operate =
safely.
3. Much has been said on the transient effects at switch-on before =
the control loop takes over the control of the output voltage. A low =
cost MOV across the output catches the initial over-voltage surge and =
clamps it to a safe level. In my opinion anyone who builds a screen =
supply for an expensive tube and omits to fit an MOV or similar device =
to catch spikes and surges, which happen more often than you think, =
especially during high negative screen current modes, is nuts and should =
be lead gently away to a padded cell! Why risk a couple of hundred $'s =
of tube for the sake of a couple of $'s worth of protection.
4. Series-pass screen stabilisers are for the birds! I have after =
many years of using them finally come to the conclusion that they cause =
more problems than they solve. Play with a couple of 4CX250's with the =
maximum anode rating and you soon realise that no matter how much static =
bleed you provide across the output of the regulator you are only =
transferring the dissipation from one part of the regulator to another, =
and you will pay the price dearly the first time the tubes flash-over! =
I have destroyed several good tubes using series stabilisers under these =
fault conditions. Take a little time to read the excellent section in =
Ian Whites book on tetrode screen supplies and you will learn a thing or =
two!
5. Finally let me say that I have learnt a lot from this discussion. =
Some really bad ways to make a screen stabiliser and some promising =
ideas that need some more thought and experimentation. To all who have =
participated, I thank you. =20
Now what is the next topic for the long winter nights?
John ZS5JF
=20
------=_NextPart_000_0006_01BE134C.A4E89220
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN">
<HTML>
<HEAD>
<META content=3Dtext/html;charset=3Diso-8859-1 =
http-equiv=3DContent-Type>
<META content=3D'"MSHTML 4.71.1712.3"' name=3DGENERATOR>
</HEAD>
<BODY bgColor=3D#fff8f0>
<DIV><FONT color=3D#000000 size=3D2>This discussion seems to be going =
around in=20
circles and not really getting anywhere.</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>May I be permitted to add my $10 =
worth on what=20
in fact is a straight forward problem.</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>1. The problem =
regarding the=20
break-down voltage of transistors or fets is really very simple to get=20
around. In the off-line switch mode power supply industry this =
problem was=20
solved many years ago by the simple expedient of connecting 2 =
transistors in=20
series to double the break-down rating, exactly the same as putting =
several=20
zeners in series to achieve a higher voltage. By utilising a low =
cost IC=20
known as a "High-side / Low-side" driver the top transistor =
base /=20
gate is operated at the required higher voltage (half-rail)and the =
bottom device=20
is simply treated as a single transistor shunt stabiliser. The two =
transistors are connected in series and the bases / gates are =
effectively=20
controlled in parallel, except the top transistor has the higher =
required=20
voltage to ensure correct operation. High-side driver IC's are =
available=20
with up to 600V operating capabilities and cost a couple of $'s. =
The=20
circuit operates as the classic shunt stabiliser, in my opinion the only =
suitable solution for a tetrode screen supply.</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>2. On the question =
of the=20
required dissipation rated transistor for a shunt high voltage =
stabiliser. =20
Take a look at TV line timebase transistors, here operating voltages of =
1500V=20
are common with power dissipations of 50 to 125W. They are low =
cost and=20
freely available. They only need the appropriate heatsink and =
control=20
circuitry to operate safely.</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>3. Much has been =
said on the=20
transient effects at switch-on before the control loop takes over the =
control of=20
the output voltage. A low cost MOV across the output catches the =
initial=20
over-voltage surge and clamps it to a safe level. In my opinion =
anyone who=20
builds a screen supply for an expensive tube and omits to fit an MOV or =
similar=20
device to catch spikes and surges, which happen more often than you =
think,=20
especially during high negative screen current modes, is nuts and should =
be lead=20
gently away to a padded cell! Why risk a couple of hundred $'s =
of =20
tube for the sake of a couple of $'s worth of protection.</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>4. Series-pass =
screen=20
stabilisers are for the birds! I have after many years of using =
them=20
finally come to the conclusion that they cause more problems than they=20
solve. Play with a couple of 4CX250's with the maximum anode =
rating and=20
you soon realise that no matter how much static bleed you provide across =
the=20
output of the regulator you are only transferring the dissipation from =
one part=20
of the regulator to another, and you will pay the price dearly the first =
time=20
the tubes flash-over! I have destroyed several good tubes using =
series=20
stabilisers under these fault conditions. Take a little time to =
read the=20
excellent section in Ian Whites book on tetrode screen supplies and you =
will=20
learn a thing or two!</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>5. Finally let me =
say that I=20
have learnt a lot from this discussion. Some really bad ways to =
make a=20
screen stabiliser and some promising ideas that need some more thought =
and=20
experimentation. To all who have participated, I thank you. =20
</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>Now what is the next topic for the =
long winter=20
nights?</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>John =
ZS5JF</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2> </FONT></DIV></BODY></HTML>
------=_NextPart_000_0006_01BE134C.A4E89220--
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm