Message-ID: <3FE526020D404BB7B87FA52527E5DD7D@gnat>
From: "Alan Melia" <alan.melia@btinternet.com>
To: <rsgb_lf_group@blacksheep.org>
References: <56D6D9BC.2040308@abelian.org> <CAA8k23R-AaOP2-BMcfz3iHMua_-eaUmTf1B0dy+3kL_pyDKUCA@mail.gmail.com>
Date: Wed, 2 Mar 2016 14:12:47 -0000
MIME-Version: 1.0
Subject: Re: LF: A divider with no power supply
Content-Type: multipart/alternative;
	boundary="----=_NextPart_000_0068_01D1748D.98D98CE0"
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Status: O

This is a multi-part message in MIME format.

------=_NextPart_000_0068_01D1748D.98D98CE0
Content-Type: text/plain;
	charset="utf-8"
Content-Transfer-Encoding: quoted-printable

Yes used to be a well known "phantom supply" effect. Another effect =
which I dont think effects CMOS because I dont think there are any =
layers that used to be called "substrates". In some MOS devices like the =
AMD7910 (PMOS ??) used in a lot of packet modems and also the Texas Bell =
Modem chip,.......if you left audio going into the modem chip from the =
radio with the TNC power off the internal substrate would charge up =
though the isolation diodes until the input was biassed completely to =
cut off the input signal. It was often refered to as the "deaf modem" =
effect. It could be recovered by heating the chip to about 120deg C and =
leaving it unpowered for several days to allow the excess charge to leak =
away as the reverse diode leakage was increased with temperature.

A variant of this signal-powering was used by several ADCs (Pico =
Technology, and I think Andy used it on a PIC) of toggling the RS-232 =
modem control lines and rectifying the signal to supply power in the =
days before powered USB.=20

The only bad effect I came across with CMOS was where the ground rail =
(or input) was taken more than 1v above the Vdd in a short pulse. This =
caused an internal 4-layer structure mimicking a thyristor to switch on =
shorting the supply rail to ground. This usually produced a crater =
effect over the chip in plastic devices, or if the supply was beefy =
enought blowing the encapsulation to pieces.=20

Alan
G3NYK
  ----- Original Message -----=20
  From: Andy Talbot=20
  To: rsgb_lf_group@blacksheep.org=20
  Sent: Wednesday, March 02, 2016 12:22 PM
  Subject: Re: LF: A divider with no power supply


  Yes, I had that once.
  Surreal, isn't it ;-)


  'jnt


  On 2 March 2016 at 12:17, Paul Nicholson <vlf0403@abelian.org> wrote:


    I just discovered something as I turned off the 12V supply to
    the CD4040 I've been using to divide 477.69kHz by 16.

    With the supply off - it still works - with a nice square
    29855.625 Hz output!

    It seems that the input to the divider, 477.69kHz sine at
    4.5V peak biased at about 6V is passing through the input
    protection diodes of the CD4040 and supplying the Vdd rail
    quite sufficiently for the divider to operator.  The rail
    voltage is 7.76V and the output is a square with
    amplitude 7.4V.

    --
    Paul Nicholson
    --


------=_NextPart_000_0068_01D1748D.98D98CE0
Content-Type: text/html;
	charset="utf-8"
Content-Transfer-Encoding: quoted-printable

=EF=BB=BF<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content=3D"text/html; charset=3Dutf-8" http-equiv=3DContent-Type>
<META name=3DGENERATOR content=3D"MSHTML 8.00.6001.23588">
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT size=3D2 face=3DArial>Yes used to be a well known "phantom =
supply"=20
effect. Another effect which I dont think effects CMOS because I dont =
think=20
there are any layers that used to be called "substrates". In some MOS =
devices=20
like the AMD7910 (PMOS ??) used in a lot of packet modems and also the =
Texas=20
Bell Modem chip,.......if you left audio going into the modem chip from =
the=20
radio&nbsp;with the TNC power off the internal substrate would charge up =
though=20
the isolation diodes until the input was biassed completely to cut off =
the input=20
signal. It was often refered to as the "deaf modem" effect. It could be=20
recovered by heating the chip to about 120deg C and leaving it unpowered =
for=20
several days to allow the excess charge to leak away as the reverse =
diode=20
leakage was&nbsp;increased with temperature.</FONT></DIV>
<DIV><FONT size=3D2 face=3DArial></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2 face=3DArial>A variant of this signal-powering was =
used by=20
several ADCs (Pico Technology, and&nbsp;I think Andy used it on a PIC) =
of=20
toggling the RS-232 modem control lines and rectifying the signal to =
supply=20
power in the days before powered USB. </FONT></DIV>
<DIV><FONT size=3D2 face=3DArial></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2 face=3DArial>The only bad effect I came across with =
CMOS was=20
where the ground rail (or input) was taken more than 1v above the Vdd in =
a short=20
pulse. This caused an internal 4-layer structure mimicking a thyristor =
to switch=20
on shorting the supply rail to ground. This usually produced a crater =
effect=20
over the chip in plastic devices, or if the supply was beefy enought =
blowing the=20
encapsulation to pieces.&nbsp;</FONT></DIV>
<DIV><FONT size=3D2 face=3DArial></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2 face=3DArial>Alan</FONT></DIV>
<DIV><FONT size=3D2 face=3DArial>G3NYK</FONT></DIV>
<BLOCKQUOTE=20
style=3D"BORDER-LEFT: #000000 2px solid; PADDING-LEFT: 5px; =
PADDING-RIGHT: 0px; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px"=20
dir=3Dltr>
  <DIV style=3D"FONT: 10pt arial">----- Original Message ----- </DIV>
  <DIV=20
  style=3D"FONT: 10pt arial; BACKGROUND: #e4e4e4; font-color: =
black"><B>From:</B>=20
  <A title=3Dandy.g4jnt@gmail.com =
href=3D"mailto:andy.g4jnt@gmail.com">Andy=20
  Talbot</A> </DIV>
  <DIV style=3D"FONT: 10pt arial"><B>To:</B> <A =
title=3Drsgb_lf_group@blacksheep.org=20
  =
href=3D"mailto:rsgb_lf_group@blacksheep.org">rsgb_lf_group@blacksheep.org=
</A>=20
  </DIV>
  <DIV style=3D"FONT: 10pt arial"><B>Sent:</B> Wednesday, March 02, 2016 =
12:22=20
  PM</DIV>
  <DIV style=3D"FONT: 10pt arial"><B>Subject:</B> Re: LF: A divider with =
no power=20
  supply</DIV>
  <DIV><BR></DIV>
  <DIV dir=3Dltr>
  <DIV>Yes, I had that once.</DIV>
  <DIV>Surreal, isn't it ;-)</DIV>
  <DIV><BR></DIV>
  <DIV>'jnt</DIV>
  <DIV><BR></DIV>
  <DIV><BR></DIV>
  <DIV class=3Dgmail_extra><BR>
  <DIV class=3Dgmail_quote>On 2 March 2016 at 12:17, Paul Nicholson =
<SPAN=20
  dir=3Dltr>&lt;<A href=3D"mailto:vlf0403@abelian.org"=20
  target=3D_blank>vlf0403@abelian.org</A>&gt;</SPAN> wrote:<BR>
  <BLOCKQUOTE=20
  style=3D"BORDER-LEFT: rgb(204,204,204) 1px solid; MARGIN: 0px 0px 0px =
0.8ex; PADDING-LEFT: 1ex"=20
  class=3Dgmail_quote><BR>I just discovered something as I turned off =
the 12V=20
    supply to<BR>the CD4040 I've been using to divide 477.69kHz by=20
    16.<BR><BR>With the supply off - it still works - with a nice=20
    square<BR>29855.625 Hz output!<BR><BR>It seems that the input to the =

    divider, 477.69kHz sine at<BR>4.5V peak biased at about 6V is =
passing=20
    through the input<BR>protection diodes of the CD4040 and supplying =
the Vdd=20
    rail<BR>quite sufficiently for the divider to operator.&nbsp; The=20
    rail<BR>voltage is 7.76V and the output is a square =
with<BR>amplitude=20
    7.4V.<BR><BR>--<BR>Paul=20
  =
Nicholson<BR>--<BR><BR></BLOCKQUOTE></DIV><BR></DIV></DIV></BLOCKQUOTE></=
BODY></HTML>

------=_NextPart_000_0068_01D1748D.98D98CE0--