1. Home
  2. Electronics Design
  3. Galvanic isolation without transformers ??
  4. Page 3

Galvanic isolation without transformers ??

:

te:

rote:

rote:

m

for

ors

ce

oad

ut 55?

ng

with

give full isolation. Touch the load and you can still get a shock. Cs on bo th L&N give a load floating at half mains V, ground the load and i doubles.

de

h current depends on the capacitance. A cap that passed 1A to a load would also pass about 1A to a human, frying them. Its only safe when, among other things, the cap is small enough to limit current to a touch-safe value.

th

o

e,

ching the mains directly.

ower supplies for sub-mA [touchable] loads, and could be allowed quite safe ly..

f

urrent control for safety from either C or R in the psu, you get something needing multiple failures for any risk to occur, which is perfectly satisfa ctory safety-wise. At that point there is no need for any insulation from t he user.

od

o a

e

ne

ng

red

l

e,

have

tching supplies at 120V via the Y capacitors

,

to a TV

correct, I'm in Denmark

The chassis is not hot, but the capacitive divider formed by the input filt er puts the chassis at half the mains voltage, very low current but still e nough to give you a buzz or kill an IC input

formatting link
$file/faq8.jpg

-Lasse

-Lasse

Reply to
Lasse Langwadt Christensen
Loading thread data ...

--
Since everything I wrote referred to a floating load, was geared to 
respond to the OP's query, and nowhere mentioned providing access to 
either side of the load for humans or other machines to touch, It 
seems you're setting up straw men in order to try to shoot down I 
don't know what... 

Do you?
Reply to
John Fields
[about ungrounded TV power connection]

Maybe not for modern TVs (with touchable inputs for sound, VGA video, HDMI, composite video, component video, etc.), but for simpler items (like a clock-radio) that's still a viable design strategy.

Reply to
whit3rd

If you need a low cost, low power supply, the high frequency low capacity coupling idea is good.

The limiting factor is, beyond EMC, the approval agency maximum "leakage" current over the barrier (for domestic is is often 0.5mA at 230V ac, so the capacitance must be below 6nF)

Cheers

Klaus

Reply to
risskovboligrenovering

Surely you risk RF burns doing this, besides all the cost and complexity disadvantages. You might not die from a bad RF burn the way you might from 50/60 Hz, but you might wish you had. :(

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net
Reply to
Phil Hobbs

RF burn?

Where might that come from?

The HF loop is closed. Square wave generator/buffer -> Two plate capacitors -> bridge rectifier -> Load. No HF escapes the loop, only low frequency mains leakage is running across the barrier

Cheers

Klaus

Reply to
risskovboligrenovering

:

te:

te:

rent control for safety from either C or R in the psu, you get something ne eding multiple failures for any risk to occur, which is perfectly satisfact ory safety-wise. At that point there is no need for any insulation from the user.

a

there are:

Its a big improvement on nearly all devices in such a condition, as used to be the case [until the 1970s in the UK].

d

I certainly prefer earthed power tools, fwiw

NT

Reply to
meow2222

e:

te:

55?
h

full isolation. Touch the load and you can still get a shock. Cs on both L &N give a load floating at half mains V, ground the load and i doubles.

rent depends on the capacitance. A cap that passed 1A to a load would also pass about 1A to a human, frying them. Its only safe when, among other thin gs, the cap is small enough to limit current to a touch-safe value.

the mains directly.

supplies for sub-mA [touchable] loads, and could be allowed quite safely.

t control for safety from either C or R in the psu, you get something needi ng multiple failures for any risk to occur, which is perfectly satisfactory safety-wise. At that point there is no need for any insulation from the us er.

s, and perhaps audio devices using a piezo type speaker.

And what I wrote was clearly about touchable psus... that's life. Not sure where you get the straw man bit from. Yeah, the shift in the thread seems c lear enough, a few paragraphs up.

NT

Reply to
meow2222

I like the last one. Use the incoming power to enable a PC to auto-trade in the stock market buying and selling making enough money to buy a battery factory which produces a product to power the isolated side of the interface after delivery by UPS... er, that's United Parcel Service, not Uninterruptable Power Supply. I don't think there is anything uninterruptable about UPS the carrier.

--

Rick
Reply to
rickman

functioning

manufactured

all

course,

have

switching supplies at 120V via the Y capacitors

into a TV

In a lot of places where the device is electric supply is from two hot wire without any neutral (including such instances here in the US) they use an X capacitor instead of a Y capacitor.

?-)

Reply to
josephkk

full isolation. Touch the load and you can still get a shock. Cs on both L&N give a load floating at half mains V, ground the load and i doubles.

supplies for sub-mA [touchable] loads, and could be allowed quite safely.

Reply to
josephkk

full isolation. Touch the load and you can still get a shock. Cs on both L&N give a load floating at half mains V, ground the load and i doubles.

I disagree. Due to the possibly substantial currents that may flow through the cap dropper it is not isolated. Nor do i think that is within the meaning intended by the NEC and regulators.

?-)

supplies for sub-mA [touchable] loads, and could be allowed quite safely.

Reply to
josephkk

On a sunny day (Sun, 30 Nov 2014 15:00:46 -0800 (PST)) it happened snipped-for-privacy@gmail.com wrote in :

Symmetry. Ever worked with 1kW 80 MHz?

Reply to
Jan Panteltje 

--
I believe that, strictly speaking, "isolated" means that there's no 
ohmic path from the mains to the load, which is what the capacitors' 
dielectric assures, regardless of the current allowed through the 
load by the capacitors' reactances.  

Note also that even if direct contact is made - by a human, say - 
with mains hot, as long as there's no path from the human back to 
the other side of the mains, there'll be no shock hazard, no matter 
the meaning "intended" by the NEC and regulators. 

Not that it would be a good thing to leave exposed mains connections 
flying in the breeze, I'm just trying to make a point.
Reply to
John Fields

Care to elaborate how dis-symmetry causes RF burns?

No. I have worked with 8MW at 4.5kHz. Does that have anything to do with this??

Regards

Klaus

Reply to
Klaus Kragelund

If there are no transformers, then each side of the bridge rectifier will be driven with a very stiff source of 80 MHz, ground referred, via a capacitor whose reactance at 80 MHz is small. The output of the bridge will be connected alternately to the two phases of the RF, but if you touch the output, even in the most favourable case, you'll have many many volts of 160 MHz ripple current going through your body. Grounding one side of the bridge output will produce some interesting fireworks.

At 60 Hz, with a transformer, the secondary winding's average voltage bounces up and down by half the p-p voltage, but since the inter-winding capacitance is small, this doesn't cause a lot of current to flow.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net
Reply to
Phil Hobbs

Usually there's a path from human to earth/ground, which serves very effectively for shocking purposes.

NT

Reply to
meow2222

Compared to you, Rube Goldberg was a piker!

Mark L. Fergerson

Reply to
Alien8752

What? I'm just describing a carrier based power delivery system like several others here... but with a different type of carrier. ;)

--

Rick
Reply to
rickman

On a sunny day (Mon, 1 Dec 2014 09:26:14 -0800 (PST)) it happened Klaus Kragelund wrote in :

How about you posting a cirucit diagram first of you 80 MHz 'isolation'?

Yes, its not high enough. RF goes where no .. has gone before.

You are talking audio.

I know, I did get RF burns on a regular basis, from less power.

Reply to
Jan Panteltje

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.