In message ,
Phil wrote:

In articles (various) re apparent problem.

1/ As mentioned previously, most domestic appliances (TVs etc category)
are 2-wire connected.
These devices therefore have a 'chassis' of MID VOLTAGE ie 115V (230-0 /2)
which is not what most would consider 'earth' potential - as in the 3rd
pin.
(The 'chassis' is connected via high-impedance to the power supply, to
stop it floating in total isolation, uncontrollably.)

Where did you learn this?

In the UK, a "three wire" connected device, as you call it, does (or
should) indeed have any exposed metal parts of its chassis connected to
the earth conductor in the mains wiring. Therefore these parts will be
at "earth" potential.

[Aside: here's the bit which scares some people - in many installations
the earth will be connected directly to the neutral, either where they
come into your house (PME or TN-C-S system) or at the transformer
(TN-S).]

Here's where I think you are mistaken:

If a device has only a "two wire" connection then in order to be legal
in the UK and indeed in Europe it must be "double insulated". This means
that at all times there are *two* layers of insulation between any live
parts (live or neutral supplies) and any exposed metalwork. Such a
chassis is therefore not really at *any* voltage relative to the supply.

If it is, especially if you can measure it through a low-ish
resistance meter, there's a problem. Get it sorted.

Hwyl!

M.

--
Martin Angove: http://www.tridwr.demon.co.uk/
Two free issues: http://www.livtech.co.uk/ Living With Technology
.... Unable to load REALITY.SYS -- Invalid Parameter: /UTOPIA

"Martin Angove" wrote in message
...
In message ,
Phil wrote:

In articles (various) re apparent problem.

1/ As mentioned previously, most domestic appliances (TVs etc category)
are 2-wire connected.
These devices therefore have a 'chassis' of MID VOLTAGE ie 115V (230-0
/2)
which is not what most would consider 'earth' potential - as in the 3rd
pin.
(The 'chassis' is connected via high-impedance to the power supply, to
stop it floating in total isolation, uncontrollably.)

Where did you learn this?

I don't know, but it certainly doesn't follow by any teaching. Apologies if
I'm stating the obvious, and I'm addressing this to the post to which you
reply.

Firstly, the 230V which is quoted as UK mains voltage is RMS (root mean
square). If you just took the mean, you'd end up with zero, because we use
alternating current which fluctuates between about -330 and +330V. You'd
have a struggle getting 115V from anywhere.

Secondly, I'd be worried if you ever had anything other than earth connected
to a chassis - that indicates a serious miswiring.

[Aside: here's the bit which scares some people - in many installations
the earth will be connected directly to the neutral, either where they
come into your house (PME or TN-C-S system) or at the transformer
(TN-S).]

Are you sure about that? I thought some wiring systems would trip if earth
and neutral are connected together.

In article , altgrr
wrote:
"Martin Angove" wrote in message
...
In message , Phil
wrote:

In articles (various) re apparent problem.

1/ As mentioned previously, most domestic appliances (TVs etc
category) are 2-wire connected. These devices therefore have a
'chassis' of MID VOLTAGE ie 115V (230-0
/2)
which is not what most would consider 'earth' potential - as in the
3rd pin. (The 'chassis' is connected via high-impedance to the power
supply, to stop it floating in total isolation, uncontrollably.)

Where did you learn this?

I don't know, but it certainly doesn't follow by any teaching.
Apologies if I'm stating the obvious, and I'm addressing this to the
post to which you reply.
Phil: Please note that despite the wording above, this is not a reply by
me to the quote above.
Firstly, the 230V which is quoted as UK mains voltage is RMS (root mean
square). If you just took the mean, you'd end up with zero, because we
use alternating current which fluctuates between about -330 and +330V.
You'd have a struggle getting 115V from anywhere.
Phil: This is totally irrelevant to the point being made.
Of course I am well aware that 230v a.c. is the rms voltage (+/- 20%, of
nominal 50Hz +/- 2Hz uk / european electricity mains supply.

Secondly, I'd be worried if you ever had anything other than earth
connected to a chassis - that indicates a serious miswiring.

Phil: You are confusing 2-wire, and 3-wire devices!!!! amongst other
things.
3-wire devices, I would /expect/ the 'chassis' to be at the earth
potential (and this signified by an earth symbol, not simply chassis
symbol), but with a 2-wire device, it might be either at a live rail
potential (chassis potential as in a circuit diagram - not the externally
touchable part of a casing), or if it used a switchedmode power supply, to
be at the potential of 1 side of the secondary - BUT the point I was
making is as to what THAT potential might be - to avoid it floating freely.

[Aside: here's the bit which scares some people - in many
installations the earth will be connected directly to the neutral,
either where they come into your house (PME or TN-C-S system) or at
the transformer (TN-S).]

Are you sure about that? I thought some wiring systems would trip if
earth and neutral are connected together.

Phil: I'm not sure if either of you, in quoting my reply to an earlier
item, have read it or digested its content appropriately!

Discussion of rms or otherwise voltage is totally irrelevant!
I also gave a specific, industry example, of a public situation where the
'otherwise floating' rails are ties VIA HIGH IMPEDANCE to the source side.
Perhaps that the underground uses d.c confused you here.

If a 'double insulated' device is left TOTALLY FLOATING, yet has an
external set of CONNECTIONS as does a TV set with VIDEO/AUDIO sockets,
then there is NO CONTROL of the potential to which the connectors may
float (and with just a 1 volt difference between the 2 of them).
THAT IS WHY there is a HIGH IMPEDANCE link between the live chassis side
and the secondary side - to stop the 'floating' and this therefore could
set it 'either side' of the live (2 wire reversible connectors) or a mid
average level.
It IS in the regulations - and I find it 'shocking' - but the current
permitted is too small to cause injury by electric shock (unless / even
with wet hands across a heart?) - hence my comment about the 'surprise'
shock element which may give rise to injury indirectly.
You might regard this level of 'shock' as a 'static' discharge as you
provide a path to connect the 'floating'-almost side to another potential.
(The impedance is over 1M, possibly 10M, - that I can't recall - probably
because I think it was referred to in the regulations more by the maximum
leakage current in terms of its maximum safe limit)

--
Phil Spiegelhalter:
==== Technical Training for Broadcasters =====
*RE CUE Mobile DV Multi-Camera Production and Non-Linear Editing*

"Phil" wrote in message
...
In article , altgrr
wrote:
"Martin Angove" wrote in message
...
In message , Phil
wrote:

In articles (various) re apparent problem.

1/ As mentioned previously, most domestic appliances (TVs etc
category) are 2-wire connected. These devices therefore have a
'chassis' of MID VOLTAGE ie 115V (230-0
/2)
Firstly, the 230V which is quoted as UK mains voltage is RMS (root mean
square). If you just took the mean, you'd end up with zero, because we
use alternating current which fluctuates between about -330 and +330V.
You'd have a struggle getting 115V from anywhere.
Phil: This is totally irrelevant to the point being made.
Of course I am well aware that 230v a.c. is the rms voltage (+/- 20%, of
nominal 50Hz +/- 2Hz uk / european electricity mains supply.

So where do you get your 115V from?

Secondly, I'd be worried if you ever had anything other than earth
connected to a chassis - that indicates a serious miswiring.

Phil: You are confusing 2-wire, and 3-wire devices!!!! amongst other
things.

No, not at all. Regardless of whether a device has two or three wires
coming into it, the only thing I'd expect to be connected to a chassis
(which I inferred as casing from your post and the reply to it) is the
earth.

3-wire devices, I would /expect/ the 'chassis' to be at the earth
potential (and this signified by an earth symbol, not simply chassis
symbol), but with a 2-wire device, it might be either at a live rail
potential (chassis potential as in a circuit diagram - not the externally
touchable part of a casing), or if it used a switchedmode power supply, to
be at the potential of 1 side of the secondary - BUT the point I was
making is as to what THAT potential might be - to avoid it floating
freely.

[Aside: here's the bit which scares some people - in many
installations the earth will be connected directly to the neutral,
either where they come into your house (PME or TN-C-S system) or at
the transformer (TN-S).]

Are you sure about that? I thought some wiring systems would trip if
earth and neutral are connected together.

Phil: I'm not sure if either of you, in quoting my reply to an earlier
item, have read it or digested its content appropriately!

I'm not sure why you raise this here - I seem to be replying directly to
your point.

Discussion of rms or otherwise voltage is totally irrelevant!
I also gave a specific, industry example, of a public situation where the
'otherwise floating' rails are ties VIA HIGH IMPEDANCE to the source side.
Perhaps that the underground uses d.c confused you here.

Where, for this paragraph, "you" refers to the author of the post I replied
to, not me, presumably?

If a 'double insulated' device is left TOTALLY FLOATING, yet has an
external set of CONNECTIONS as does a TV set with VIDEO/AUDIO sockets,
then there is NO CONTROL of the potential to which the connectors may
float (and with just a 1 volt difference between the 2 of them).

Was something missing from your parenthesis? I'm not entirely sure what you
mean here.

THAT IS WHY there is a HIGH IMPEDANCE link between the live chassis side
and the secondary side - to stop the 'floating' and this therefore could
set it 'either side' of the live (2 wire reversible connectors) or a mid
average level.

Calm down, there's no need to shout!

Apologies for stirring up an argument :-) Just a couple of things,
heavily snipped...

In message ,
Phil wrote:


3-wire devices, I would /expect/ the 'chassis' to be at the earth
potential (and this signified by an earth symbol, not simply chassis
symbol), but with a 2-wire device, it might be either at a live rail
potential (chassis potential as in a circuit diagram - not the externally
touchable part of a casing), or if it used a switchedmode power supply, to
be at the potential of 1 side of the secondary - BUT the point I was
making is as to what THAT potential might be - to avoid it floating freely.


Ok, fine. You are using "chassis" in what seems to me to be an
old-fashioned sense, but I will concede the point that you are referring
to the "ground" or "0V" potential of the *circuitry*, not to the actual
potential of the casing of the device. In that case, what I wrote
probably doesn't make much sense!

If a 'double insulated' device is left TOTALLY FLOATING, yet has an
external set of CONNECTIONS as does a TV set with VIDEO/AUDIO sockets,
then there is NO CONTROL of the potential to which the connectors may
float (and with just a 1 volt difference between the 2 of them).
THAT IS WHY there is a HIGH IMPEDANCE link between the live chassis side
and the secondary side - to stop the 'floating' and this therefore could
set it 'either side' of the live (2 wire reversible connectors) or a mid
average level.

This is where we need to hear from someone who designs double-insulated
HiFi gear. I am more electrically-biased these days, though I have an
electronics background. Ignoring switched-mode PSUs for the moment, if
you have a conventional isolating transformer - rectifier - capacitor -
regulator powersupply, then the thing is so totally floating that it
couldn't care less what you connect it to. As soon as you connect it,
it simply and without fuss becomes referenced to the thing you have
connected it to.

One side of the transformer winding becomes the circuit ground (what you
refer to as the chassis) and the other becomes the power rail. An
(non-isolated) external interface will cause the ground of some other power
supply to be electrically connected to the ground of the power supply in
question, but *it doesn't matter*. Even if the second supply wasn't truly
floating, the first just floats up to meet it. How on earth else is it possible
to interface battery powered devices to mains powered devices?

The danger would come if the circuit ground was indeed connected to one
or other of the live supplies - ancient radios used to do this. If all
equipment is used on a single phase and everything is referenced to
neutral then this might just work, but imagine the case where one device
references to neutral, and another to *earth*. In practically every
electrical installation with anything more than a clock connected, the
neutral will be at a significantly different potential to the earth.
This is how, even in a good installation, connecting neutral and earth
together (such as might be done when you cut a cable), even where live
is not involved, can cause an RCD to trip. A normal RCD will trip if
something above 15mA flows. [but see my reply to "altgrr"]

It IS in the regulations - and I find it 'shocking' - but the current
permitted is too small to cause injury by electric shock (unless / even
with wet hands across a heart?) - hence my comment about the 'surprise'
shock element which may give rise to injury indirectly.

Which "regulations"? The "wiring regulations" (BS7671:2001 r713-04)
state that for a 240V circuit there should be at least 500kOhms
resistance between each conductor and each other, the test voltage being
500V d.c. The guidance suggests that a resistance reading under 2MOhms
should be taken as a sign that all is not well. I don't think this
applies to the user interfaces of consumer equipment though :-)

You might regard this level of 'shock' as a 'static' discharge as you
provide a path to connect the 'floating'-almost side to another potential.
(The impedance is over 1M, possibly 10M, - that I can't recall - probably
because I think it was referred to in the regulations more by the maximum
leakage current in terms of its maximum safe limit)


BS7671 attempts to limit discharge of energy mainly in order to protect
cables and flexes. This is done through the use of fuses and circuit
breakers. Additional precautions are required on certain socket outlets,
which usually results in the installation of an RCD. An RCD limits the
discharge of energy through a person and one can imply what the powers
reckon to be safe limits by the following information:

A (normal domestic use) RCD must

NOT trip if the leakage current is 15mA or less
TRIP within 300mS if the leakage current is 30mA
TRIP within 40mS if the leakage current is 5*30mA (150mA) or higher.

I'm not sure where this argument is leading. Perhaps I'd better stop
there :-)

Hwyl!

M.

--
Martin Angove: http://www.tridwr.demon.co.uk/
Two free issues: http://www.livtech.co.uk/ Living With Technology
.... Motel mattresses are better on the side away from the phone.

In message ,
"altgrr" wrote:

"Martin Angove" wrote in message
...

[Aside: here's the bit which scares some people - in many installations
the earth will be connected directly to the neutral, either where they
come into your house (PME or TN-C-S system) or at the transformer
(TN-S).]

Are you sure about that? I thought some wiring systems would trip if earth
and neutral are connected together.



Earth and neutral will be connected together somewhere in all but very
specialised installations. In this country (uk) there are basically
three sorts of domestic installation:

TT (terre-terre): The neutral is physically connected to the planet
Earth at the transformer. Your supply is live and neutral, and you make
your own earth by connecting to an earth rod or plate or similar. In
this case a live-earth fault in your installation causes a current to
flow through your earth rod, through the bulk of the planet Earth and
back to the neutral at the transformer. Done properly, this can work
very well, but it can also lead to an impedance too high to safely blow
a fuse or trip an MCB. For this reason it is required that TT systems
have an RCD protecting *all* circuits (note: this differs from a
standard 30mA RCD). An RCD will trip at much lower levels of fault
current than an MCB or fuse.

TT systems are now mainly seen in rural areas with overhead supply,
although these are being converted to TN-C-S (see below).

TN-S (Terre Neutral - Separated): The neutral is physically connected to
the Earth at the transformer. Your supply is live, neutral *and* earth -
usually the metal sheath of the supply cable. In this case a live-earth
fault in your installation causes a current to flow through the
supplier's earth back to the neutral at the transformer. This path is
usually good enough to blow a fuse under such a fault so an overall RCD
is not required, though RCD protection is required for certain outlets.

TN-S systems are commonly found in urban areas, though they are being
supplanted by

TN-C-S (Terre Neutral - Combined [for the supply] - Separated [in the
installation]), also known as PME (Protective Multiple Earthing): The
neutral is physically connected to the Earth at the transformer, and at
several points along the route from transformer to your house. Your
supply is live and neutral, and your installation's earth is connected
directly to the incoming neutral at the supplier's "cut off" (main
fuse). As you may surmise, a live-earth fault in this situation also
causes plenty of current to flow to blow a fuse. The problem with a PME
system is that if someone puts a digger through the cable, severing the
neutral you also lose your safety earth, and if the live is intact this
can be dangerous. There are very strict regulations concerning "bonding"
exposed metalwork and electrical earths together in these circumstances.
RCD requirements are similar to TN-S.

Regarding your point about systems "tripping" if neutral and earth are
connected, in any of these cases (particularly in the TT system) there
will be a difference in potential between the neutral in the
installation and the earth whenever something is switched on. This is
described by Ohm's Law regarding resistances, flow of current and
voltages. If you have an RCD fitted, it may only take 15mA flowing down
the earth instead of the neutral for the thing to trip.

As I said however, the earth will (at some point) be connected to the
neutral quite deliberately.

Hwyl!

M.

--
Martin Angove: http://www.tridwr.demon.co.uk/
Two free issues: http://www.livtech.co.uk/ Living With Technology
.... See that slate, That's your keyboard that is.

"Martin Angove" wrote in message
...
Here's where I think you are mistaken:

If a device has only a "two wire" connection then in order to be legal
in the UK and indeed in Europe it must be "double insulated". This
means
that at all times there are *two* layers of insulation between any
live
parts (live or neutral supplies) and any exposed metalwork. Such a
chassis is therefore not really at *any* voltage relative to the
supply.

These days a TV usually has a switching power supply so that if nothing
was done to make it otherwise the circuitry including various external
connectors (chassis) would be floating.

However this would be dangerous - in a TV there are parts of the TV with
are anything up to 25000 volts or so positive with respect of the
external connectors. If everything was just left to float the connectors
could easily end up with a massive voltage on them. You would get a
nasty jolt if this discharged through you. To stop this happening two
high value resistors in series are including between one of the mains
inputs and the chassis. The resistors value is high enough to count as
an insulator for double insulated.

--

Brian Gregory (In the UK).

To email me remove the letter vee.

So where do you get your 115V from?

Phil: As per the original query - about blowing things up when items were
CONNECTED via SCARTs: - and to which my original reply was directed....


If, albeit via High Impedance, one side of the electrical (not physical
case) chassis is connected to a power rail, to stop it floating, and the
next device is similarly connected, then the potential of the 'shield'
connector (and by definition, almost the same for the signal connectors)
might be at 'Neutral', 'Live' OR (50:50 chance) 115v (mid voltage) since
there is no identification externally for the orientation of a telefunken
or similar 2 wire connector: therefore if device A is connected / tied to
live via say 10M, and the 2nd device is tied to neutral via 10M: result of
both devices is now a compromised 115v value (simple potential divider) -
and if another connector is touched / made to a third device, this will be
experienced at this potential.

The safety regulations are satisfied because the leakage current may be
measured in micro amps - but this could still damage components, and cause
'surprise' to someone touching the connector - causing a subsequent
accident.

We have experienced also, cases where individual items of equipment show a
minimal leakage (eg 3microA) - but the combined effect of several devices
becomes enough to trip a sensitive RCCB

--
Phil Spiegelhalter:
==== Technical Training for Broadcasters =====
*RE CUE Mobile DV Multi-Camera Production and Non-Linear Editing*

In message ,
Phil wrote:


If, albeit via High Impedance, one side of the electrical (not physical
case) chassis is connected to a power rail, to stop it floating, and the
next device is similarly connected, then the potential of the 'shield'
connector (and by definition, almost the same for the signal connectors)
might be at 'Neutral', 'Live' OR (50:50 chance) 115v (mid voltage) since
there is no identification externally for the orientation of a telefunken
or similar 2 wire connector: therefore if device A is connected / tied to
live via say 10M, and the 2nd device is tied to neutral via 10M: result of
both devices is now a compromised 115v value (simple potential divider) -
and if another connector is touched / made to a third device, this will be
experienced at this potential.

I understand your reasoning now. I'm still not convinced it's a good
idea, but if it is true I can see that what you're saying might happen.


The safety regulations are satisfied because the leakage current may be
measured in micro amps - but this could still damage components, and cause
'surprise' to someone touching the connector - causing a subsequent
accident.

We have experienced also, cases where individual items of equipment show a
minimal leakage (eg 3microA) - but the combined effect of several devices
becomes enough to trip a sensitive RCCB


Surely though this will only happen if there is an earth connection
somewhere? For 2-wire (no earth) devices the only place current can flow
is within the L/N circuit.

It is a well known fact that some 3-wire devices, particularly those
with filters on their mains inputs (computers for example) leak quite
badly to earth as a matter of course. There is even a section in
BS7671:2001 dedicated to allowing for these "high protective conductor
currents". Basically it involves using high integrity earth conductors
and splitting things onto several RCDs, if indeed an RCD is necessary at
all. As I said somewhere else, a normal RCD should not trip below 15mA
leakage, but anything above this and it is quite likely that it will,
though 30mA is the official design current.

Hwyl!

M.

--
Martin Angove: http://www.tridwr.demon.co.uk/
Two free issues: http://www.livtech.co.uk/ Living With Technology
.... Help! I'm parked diagonally in a parallel universe.

"Martin Angove" wrote in message
...
In message ,
Phil wrote:

In articles (various) re apparent problem.

1/ As mentioned previously, most domestic appliances (TVs etc category)
are 2-wire connected.
These devices therefore have a 'chassis' of MID VOLTAGE ie 115V (230-0
/2)
which is not what most would consider 'earth' potential - as in the 3rd
pin.
(The 'chassis' is connected via high-impedance to the power supply, to
stop it floating in total isolation, uncontrollably.)

Where did you learn this?


What Martin said here is correct.

In the UK, a "three wire" connected device, as you call it, does (or
should) indeed have any exposed metal parts of its chassis connected to
the earth conductor in the mains wiring. Therefore these parts will be
at "earth" potential.

Ok, but I thought we were talking about Class 2 devices.

[Aside: here's the bit which scares some people - in many installations
the earth will be connected directly to the neutral, either where they
come into your house (PME or TN-C-S system) or at the transformer
(TN-S).]

Here's where I think you are mistaken:

If a device has only a "two wire" connection then in order to be legal
in the UK and indeed in Europe it must be "double insulated". This means
that at all times there are *two* layers of insulation between any live
parts (live or neutral supplies) and any exposed metalwork. Such a
chassis is therefore not really at *any* voltage relative to the supply.


It can well be. I regularly get (unharmful) shocks from TV and STBs when
I'm connected them together due to the different chassis voltage.

The double insulted thing is not meant to be taken literally at all times,
for instance sufficient air gaps are acceptable, also components can be
connected across primary and secondard without 'double insulation' however
these components are covered by strict contruction and quality assurance
rules.

If it is, especially if you can measure it through a low-ish
resistance meter, there's a problem. Get it sorted.



Not correct, there can be up to 4Mohm between primary and exposed metal
parts in a class 2 device.

--
Tony Stanley ++Always Learning++