how to separate ground potential

S

Saul Bernstein

Guest
Hi,

I have to design a printed circuit board for a high-voltage environment.
Hence, the "ground potential" of the board has to be on 4 kV. Du to the fact
that the absolute potential values don't matter, one could say that the 2 kV
are virtually my ground potential. The problem arises by reason of the power
supply. I have to power my board with 5 V and for the mains adapter ground
is really 0 V.

Can someone think of a possibility to supply my board with a mains adapter
(AC/DC 220V/5V) and translate between the two ground reference potentials?

Thank you

Saul
 
"Saul Bernstein" <jiffylube@freenet.de> wrote in message
news:48f6e1b8$0$6569$9b4e6d93@newsspool4.arcor-online.net...
Hi,

I have to design a printed circuit board for a high-voltage environment.
Hence, the "ground potential" of the board has to be on 4 kV. Du to the
fact that the absolute potential values don't matter, one could say that
the 2 kV are virtually my ground potential. The problem arises by reason
of the power supply. I have to power my board with 5 V and for the mains
adapter ground is really 0 V.

Can someone think of a possibility to supply my board with a mains adapter
(AC/DC 220V/5V) and translate between the two ground reference potentials?
Click to expand...
Sure. Feed it AC and use an isolation transformer. Might have
to get one special-made to withstand that sort of potential
difference. Watch for safety regs. Rectify/regulate on-board.

Bob M.
 
Saul Bernstein wrote:

Hi,

I have to design a printed circuit board for a high-voltage environment.
Hence, the "ground potential" of the board has to be on 4 kV. Du to the fact
that the absolute potential values don't matter, one could say that the 2 kV
are virtually my ground potential. The problem arises by reason of the power
supply. I have to power my board with 5 V and for the mains adapter ground
is really 0 V.

Can someone think of a possibility to supply my board with a mains adapter
(AC/DC 220V/5V) and translate between the two ground reference potentials?

Thank you

Saul


Nobody wants to touch this query because if you dont know HV stuff
Click to expand...
pretty well, you *will* be incapable of safely implementing any
suggestions.

The idea of using a pluggable adaptor is EXTREMELY DANGEROUS as any
fault could lead to having 4 KV on the plug pins.

Its in the category: Make your will, pay for a funeral plan but dont
bother with life insurance because they wont pay out for suicide.

BUY IN an isolated PSU rated for sufficiently more than your working
voltage, Ground the case to the same ground as the HV source with a
BOLTED on cable (not removable without tools). Have all the high voltage
wiring hooked up by a qualified HV technician and you *may* stay out of
that pine box. *DONT* order anything you haven't cleared with the
technician.

Alternatively why not consider battery power? A SLA battery and a
regulator could easily power most microcontroller circuits for a day's
running then you'd shut down, isolate your circuit, unplug the battery
and recharge it overnight with an off the shelf charger. No more risky
than the rest of your board if the battery is in the same enclosure.

If you need to transfer data, use fibre optic cable or any sort of
wireless link.

If this is an academic project you've just been given for ****s sake
consult with your tutor, the senior departmental technician and the
Prof. You *may* need another project if you are asking this sort of
question here (sci.electronics.basics) and you want to pass the course.
 
On Thu, 16 Oct 2008 08:40:17 +0200, "Saul Bernstein"
<jiffylube@freenet.de> wrote:

Hi,

I have to design a printed circuit board for a high-voltage environment.
Hence, the "ground potential" of the board has to be on 4 kV. Du to the fact
that the absolute potential values don't matter, one could say that the 2 kV
are virtually my ground potential. The problem arises by reason of the power
supply. I have to power my board with 5 V and for the mains adapter ground
is really 0 V.

Can someone think of a possibility to supply my board with a mains adapter
(AC/DC 220V/5V) and translate between the two ground reference potentials?
Click to expand...
---
Will the PCB be connected to anything other than the high voltage supply
positive and the 220V mains?

JF
 
On Thu, 16 Oct 2008 08:40:17 +0200, "Saul Bernstein"
<jiffylube@freenet.de> wrote:

Hi,

I have to design a printed circuit board for a high-voltage environment.
Hence, the "ground potential" of the board has to be on 4 kV. Du to the fact
that the absolute potential values don't matter, one could say that the 2 kV
are virtually my ground potential. The problem arises by reason of the power
supply. I have to power my board with 5 V and for the mains adapter ground
is really 0 V.

Can someone think of a possibility to supply my board with a mains adapter
(AC/DC 220V/5V) and translate between the two ground reference potentials?
Click to expand...
---
If you need your +5V to ride on the high voltage, you could do it like
this: (View in Courier)

+--------------------------------------+
| PCB |
| 7805 |
| +----+ +-----+ |
220AC>-----+-----+ +--|~ +|-----+-----| |--+5V |
| | P||S | | |+ +--+--+ |
| | R||E | | [BFC] | |
| | I||C | | | | |
220AC>---+-|-----+ +--|~ -|-----+--------+-----GND |
| | | +----+ | |
| | | | |
| | +----------------------------|---------+
| | |
| | |
| | |
| | +------+ |
| +------------------|~ +HV|-----+--->
| | |
+--------------------|~ -HV|--------->
+------+

You'll need a transformer with high isolation, and if your high voltage
is 4kV I'd recommend a dielectric strength of at least 5kV from
secondary to primary and from secondary to core. More would be better.

Be aware that this is a dangerous circuit and you could easily get hurt
or killed, or cause someone else to get hurt or killed



JF
 
On Thu, 16 Oct 2008 08:40:17 +0200, "Saul Bernstein"
<jiffylube@freenet.de> wrote:

Hi,

I have to design a printed circuit board for a high-voltage environment.
Hence, the "ground potential" of the board has to be on 4 kV. Du to the fact
that the absolute potential values don't matter, one could say that the 2 kV
are virtually my ground potential. The problem arises by reason of the power
supply. I have to power my board with 5 V and for the mains adapter ground
is really 0 V.

Can someone think of a possibility to supply my board with a mains adapter
(AC/DC 220V/5V) and translate between the two ground reference potentials?
Click to expand...
You haven't indicated the voltage and power requirement for this
housekeeping supply.

At lower power levels, some simple self-constructed solutions are
possible.

RL
 
"Wim Lewis"
The easy way to do this is probably to use an isolated DC-DC converter.
A few kV of isolation is typical; I'm sure you can find ones with
higher ratings if needed.
Click to expand...

** One of these might be suitable, if all you need is 1 watt.

http://www.xppower.com/pdfs/SF_IV.pdf

Farnell have them for around $10.

Ground the 5 volt input side ( for human safety ) and install it INSIDE
the HV device.



If you do not have any circuitry whose ground
potential needs to be at 0V then you can use an isolated switching power
supply to go from mains current to a floating 5V output (this would
be cheaper than a non-isolated supply followed by an isolated
converter). The transformer in many power supply designs provides
this kind of isolation "for free".
Click to expand...

** Even if you find such a well insulated PSU, the problem is in having 3
or 4 kV to ground on the cable coming out from it - that must not be left
around for folk to touch !!!!

So, the PSU must be fitted INSIDE the device to be powered and the AC power
fed in.



...... Phil
 
In article <48f6e1b8$0$6569$9b4e6d93@newsspool4.arcor-online.net>,
Saul Bernstein <jiffylube@freenet.de> wrote:
I have to design a printed circuit board for a high-voltage environment.
Hence, the "ground potential" of the board has to be on 4 kV. [....]

Can someone think of a possibility to supply my board with a mains adapter
(AC/DC 220V/5V) and translate between the two ground reference potentials?
Click to expand...
The easy way to do this is probably to use an isolated DC-DC converter.
A few kV of isolation is typical; I'm sure you can find ones with
higher ratings if needed. If you do not have any circuitry whose ground
potential needs to be at 0V then you can use an isolated switching power
supply to go from mains current to a floating 5V output (this would
be cheaper than a non-isolated supply followed by an isolated
converter). The transformer in many power supply designs provides
this kind of isolation "for free".

--
Wim Lewis <wiml@hhhh.org>, Seattle, WA, USA. PGP keyID 27F772C1
"We learn from history that we do not learn from history." -Hegel
 
On Thu, 16 Oct 2008 17:49:25 -0400, legg <legg@nospam.magma.ca> wrote:

You haven't indicated the voltage and power requirement for this
housekeeping supply.

At lower power levels, some simple self-constructed solutions are
possible.
Click to expand...
At very low power levels, just send optical power up on an optical
fiber and the data down on an other fiber.

This system is used on some high voltage measurement instruments
hanging on an overhead high voltage line.

Paul
 
Paul Keinanen wrote:
On Thu, 16 Oct 2008 17:49:25 -0400, legg <legg@nospam.magma.ca> wrote:


You haven't indicated the voltage and power requirement for this
housekeeping supply.

At lower power levels, some simple self-constructed solutions are
possible.


At very low power levels, just send optical power up on an optical
fiber and the data down on an other fiber.

This system is used on some high voltage measurement instruments
hanging on an overhead high voltage line.

Paul

Thats an interesting technique I've been hearing a little about lately.
Click to expand...

Whats state of the art for power over fiber? How big a laser does one
need for how many mW at the remote end? Got any good links?

Thanks.

Ian.
 
On Fri, 17 Oct 2008 07:36:36 +0100, IanM <Invalid@totally.invalid>
wrote:

Paul Keinanen wrote:
On Thu, 16 Oct 2008 17:49:25 -0400, legg <legg@nospam.magma.ca> wrote:


You haven't indicated the voltage and power requirement for this
housekeeping supply.

At lower power levels, some simple self-constructed solutions are
possible.


At very low power levels, just send optical power up on an optical
fiber and the data down on an other fiber.

This system is used on some high voltage measurement instruments
hanging on an overhead high voltage line.

Paul

Thats an interesting technique I've been hearing a little about lately.

Whats state of the art for power over fiber? How big a laser does one
need for how many mW at the remote end? Got any good links?
Click to expand...
Do a Google search such as

"optically powered" "current transformer"

should give some useful links.

For instance
http://homepage.ufp.pt/alobo/page8/page5/files/IEEE%20PPT%20Conf2001_2.pdf
describe a 300 mW laser at 810 nm and a photocell efficiency of 40 %,
so you would get about 100 mW of usable DC power for the sensor and
the optical transmitter to send back the information.

Paul
 
Hi,

thanks for your kind answers so far. To answer your question: The PCB won't
be connected to anything other than the 4kV and the 220V mains.

Saul


"John Fields" <jfields@austininstruments.com> schrieb im Newsbeitrag
news:3ftef4hjuggm6qgerbrq9snn6i4b7fvem7@4ax.com...
On Thu, 16 Oct 2008 08:40:17 +0200, "Saul Bernstein"
jiffylube@freenet.de> wrote:

Hi,

I have to design a printed circuit board for a high-voltage environment.
Hence, the "ground potential" of the board has to be on 4 kV. Du to the
fact
that the absolute potential values don't matter, one could say that the 2
kV
are virtually my ground potential. The problem arises by reason of the
power
supply. I have to power my board with 5 V and for the mains adapter ground
is really 0 V.

Can someone think of a possibility to supply my board with a mains adapter
(AC/DC 220V/5V) and translate between the two ground reference potentials?

---
Will the PCB be connected to anything other than the high voltage supply
positive and the 220V mains?

JF
Click to expand...
 
Paul Keinanen wrote:

On Fri, 17 Oct 2008 07:36:36 +0100, IanM <Invalid@totally.invalid
wrote:


Paul Keinanen wrote:

On Thu, 16 Oct 2008 17:49:25 -0400, legg <legg@nospam.magma.ca> wrote:



You haven't indicated the voltage and power requirement for this
housekeeping supply.

At lower power levels, some simple self-constructed solutions are
possible.


At very low power levels, just send optical power up on an optical
fiber and the data down on an other fiber.

This system is used on some high voltage measurement instruments
hanging on an overhead high voltage line.

Paul


Thats an interesting technique I've been hearing a little about lately.

Whats state of the art for power over fiber? How big a laser does one
need for how many mW at the remote end? Got any good links?


Do a Google search such as

"optically powered" "current transformer"

should give some useful links.

For instance
http://homepage.ufp.pt/alobo/page8/page5/files/IEEE%20PPT%20Conf2001_2.pdf
describe a 300 mW laser at 810 nm and a photocell efficiency of 40 %,
so you would get about 100 mW of usable DC power for the sensor and
the optical transmitter to send back the information.

Paul
Click to expand...
Thanks Paul.

6V (open circuit) GaAs photovoltaic diode. It looks like there would be
no particular problem getting 15 mA at 3.3 V for a microcontroller or
whatever. I suspect that it would be out of the price range of the
average hobbyist but if one is building Van Der Graph generators or
similar it might be justifiable.

Its a far cry from two minature PM DC motors coupled shaft to shaft with
a couple of feet of Perspex rod in between :)

Ian.
 
Saul Bernstein wrote:
thanks for your kind answers so far. To answer your question:
The PCB won't be connected to anything other than the 4kV and
the 220V mains.
Click to expand...
Please do not top-post. Your answer belongs after (or intermixed
with) the quoted material to which you reply, after snipping all
irrelevant material. See the following links:

<http://www.catb.org/~esr/faqs/smart-questions.html>
<http://www.caliburn.nl/topposting.html>
<http://www.netmeister.org/news/learn2quote.html>
<http://cfaj.freeshell.org/google/> (taming google)
<http://members.fortunecity.com/nnqweb/> (newusers)

--
[mail]: Chuck F (cbfalconer at maineline dot net)
[page]: <http://cbfalconer.home.att.net>
Try the download section.
 
On Fri, 17 Oct 2008 14:24:43 +0100, IanM <Invalid@totally.invalid>
wrote:

Its a far cry from two minature PM DC motors coupled shaft to shaft with
a couple of feet of Perspex rod in between :)
Click to expand...
At least this mechanical system can be scaled to quite large power
levels :)
http://en.wikipedia.org/wiki/Lyon-Moutiers_DC_transmission_scheme

Other low power isolated power transmission systems include feeding
acoustic vibrations into an isolating rod or using radio frequency
power transmission (after all the isolation distance required is only
a few meters).

Paul
 
On Oct 15, 11:40 pm, "Saul Bernstein" <jiffyl...@freenet.de> wrote:
Hi,

I have to design a printed circuit board for a high-voltage environment.
Hence, the "ground potential" of the board has to be on 4 kV. Du to the fact
that the absolute potential values don't matter, one could say that the 2 kV
are virtually my ground potential. The problem arises by reason of the power
supply. I have to power my board with 5 V and for the mains adapter ground
is really 0 V.

Can someone think of a possibility to supply my board with a mains adapter
(AC/DC 220V/5V) and translate between the two ground reference potentials?
Click to expand...

How about a battery?

At low powers and bad efficiencies, transformers to handle a 4KV
isolation is fairly easy. At bigger power levels or efficiencies
above 50% it gets harder fast.

What does this PCB have to do? How much power etc?

Does the PCB live inside a shield?

At low powers, the simplest design is to use an "air core" transformer
with the primary and secondaries tuned and loose coupling between then
because of the large clearance. The down side is that this tends to
radiate a lot of RF at your operating frequency.
 
Paul Keinanen wrote:
IanM <Invalid@totally.invalid> wrote:

Its a far cry from two minature PM DC motors coupled shaft to
shaft with a couple of feet of Perspex rod in between :)

At least this mechanical system can be scaled to quite large
power levels :)

http://en.wikipedia.org/wiki/Lyon-Moutiers_DC_transmission_scheme
Click to expand...
Highly sensitive to dust, thus needing continuous maintenance, IMO.

--
[mail]: Chuck F (cbfalconer at maineline dot net)
[page]: <http://cbfalconer.home.att.net>
Try the download section.
 
On Wed, 15 Oct 2008 23:40:17 -0700, Saul Bernstein wrote
(in article <48f6e1b8$0$6569$9b4e6d93@newsspool4.arcor-online.net>):

Hi,

I have to design a printed circuit board for a high-voltage environment.
Hence, the "ground potential" of the board has to be on 4 kV. Du to the fact
that the absolute potential values don't matter, one could say that the 2 kV
are virtually my ground potential. The problem arises by reason of the power
supply. I have to power my board with 5 V and for the mains adapter ground
is really 0 V.

Can someone think of a possibility to supply my board with a mains adapter
(AC/DC 220V/5V) and translate between the two ground reference potentials?

Thank you

Saul
Click to expand...
How much power do you need? How about a couple of solar cells and a halogen
work light?

-- Charlie Springer
 
Saul Bernstein wrote:

Hi,

I have to design a printed circuit board for a high-voltage
environment. Hence, the "ground potential" of the board has to be on 4
kV. Du to the fact that the absolute potential values don't matter, one
could say that the 2 kV are virtually my ground potential. The problem
arises by reason of the power supply. I have to power my board with 5 V
and for the mains adapter ground is really 0 V.

Can someone think of a possibility to supply my board with a mains
adapter (AC/DC 220V/5V) and translate between the two ground reference
potentials?

Thank you

Saul
Click to expand...
Yes this sort of thing can be done but you really should know what you
are doing in the high voltage areas when doing projects like this. Some
of the systems I work with have rails of 132kV DC (in the MW region) and
this takes very special handling and checking from the design through
the implementation to maintenance planning before we build such systems.

As many of the professional engineers here who would know this stuff and
know of the dangers it is hardly surprising you are not getting much
help without knowing your level of experience. It is an area that is at
such a level that if you have to ask then you shouldn't be doing this
without adequate oversight and supervision. The professionals will want
to know that you have that in place and by asking here it is obvious
that you probably haven't and it would be criminal for them to offer a
solution.

--
********************************************************************
Paul E. Bennett...............<email://Paul_E.Bennett@topmail.co.uk>
Forth based HIDECS Consultancy
Mob: +44 (0)7811-639972
Tel: +44 (0)1235-811095
Going Forth Safely ..... EBA. www.electric-boat-association.org.uk..
********************************************************************
 
On Oct 15, 11:40 pm, "Saul Bernstein" <jiffyl...@freenet.de> wrote:
I have to design a printed circuit board for a high-voltage environment.
Hence, the "ground potential" of the board has to be on 4 kV.
Click to expand...
This is the same sort of problem as electron microscopes (and
even microwave ovens) have in driving the filament. The easy
solution is to box up a 6V battery and regulate down to 5V.
The commercial solution is a custom-made, tested, transformer
that safely holds off 4 kV. It's likely to be potted (embedded
in a blob of tar or silicone).

You don't want any of the 4kV parts outside a shielded and
interlocked enclosure, of course...
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top