Transformerless Power Suppply

[Luciano]

Here, ground and neutral are not necessarily at the same potential.

The fuse will like that!

Best regards,

Luciano

[Melanie]

Actually Dave you quoted me as "both inaccurate and dangerous".

I challenge that.

I don't care what part of the world you're in, the Fuse in Figure 3 of TB008 is a complete waste of time, offering NO protection - let's repeat that in case anyone missed it first time around - offering NO protection to the User of that circuit. If that Fuse is in or out, the circuit will be LIVE. Microchip have dropped a bollock on this one - so let's direct the "inaccurate and dangerous" where it's needed - the author of TB008.

[Pic_User]

Hi All,

There seems to be a great deal of confusion among hobbyists and new professionals concerning ground and circuit ground.

I stopped calling the “common” connections in electrical / electronic circuits, “ground”, 30 years ago. Most of us, have worked out the dual meaning of ground. On one hand we actually mean earth ground, next we simply mean a common connection in our circuit.

This dual meaning is okay most of the time, but sometimes it causes misunderstandings and even dangerous situations. Transformerless power supplies are a common battle ground for the semantics of the term, “ground”. This is because, the misunderstandings lead to safety problems. Problems that cannot be ignored.

Melanie, as usual is the voice of reason and spot on. The TB008 application note, is right out of the twilight zone. The author, obviously has some confusion about “ground” and “common”.

I post this diagram to help illustrate what Melanie already pointed out.

[Luciano]

Hi,

What I have in my house.

(Click to enlarge).

Best regards,

Luciano

[Acetronics2]

Hi, Luciano

Seems not only for the Englishmen ...

3 Phases ... Your home really is a Castle !!!

Alain

[Luciano]

Hi Alain,

This is what we get here.
Only this option is available.

Our milking machines use three-phase motors.

Best regards,

Luciano

PS: En réalité j'habite dans un petit village au
sud des Alpes. Avec nos amis et voisins italien
on partage la langue et la culture. Ciao!

[JIM A]

The big difference between ground and "common" or "neutral" is the voltage drop that is developed on the neutral wire due to the fact that it has the same current flow as the line conductor (What goes out must return) (Load current x DC resistance of neutral conductor between receptacle and breaker box= Voltage difference between neutral and ground at the receptacle). Normally the voltage drop is neglible except in long circuit runs with higher current loads. The neutral and ground wires are normally connected together (bonded) at the breaker box. I totally agree that the fuse definitely needs to be in the "hot" or "line" or "mains" line to perform a safety function.

[mvs_sarma]

Hi all
All said and done, we have to accept that what we try to design and implement will be used by us only and that there would not a future occassion to extend the equipment using such PSU to other eqpt, which might create ground loops- a possible shock.

many times we will be pre occupied and would like forget the non isolated psu and take it for granted that we will be safe--

AS SUCH I SUUGEST THAT WE SOULD NOT PRACTICE THIS MODE AT ALL-- ok for academic interest.

[comwarrior]

Strangely, i have a design for a transformer-less PSU in the making...
My reason for not using a transformer is because to have a transformer that drops 110V and 240V down to a more usable voltage and supply the current i need would be massive and it costs >£50 each...

I'm not going to post details of the PSU because i intend on filing for international copyright.
The reason for this is because it will handle any single phase voltage, frequency and amount of noise (or noise only) you can throw at it.
It will even operate in a brown out with sustained unstable voltages as low as 24V (maybe lower).
It can also handle stepped AC, Square wave and saw tooth signals provided it is alternating...

Also, Mel is absolutely right, that AN is written for one purpose and one purpose only... TO KILL PEOPLE!

[rmteo]

Something like this PSAA60W-120

AC Input Voltage Rating: 100 to 240V AC
AC Input Frequency: 50 to 60Hz
Input Current: 1.6A at 100V AC, 0.7A at 240V AC
Output: 60W
Efficiency: 85% average

[comwarrior]

transformer-less... the above has a transformer, i can tell because of the input specification.

Here in the UK the current code for mains wiring dictates the use of an ELCB.
Earth Leakage Circuit Breaker. If more than 50mA flows up or down the earth wire the ELCB will trip isolating both live and neutral.
every time I've diagnosed wiring faults, their has always been a 50 to 60 V AC difference between earth and ether neutral or live...
If you use an isolation transformer you can connect the negative after rectification to ground. This is what most PC PSU's do.

edit:
That efficiency rating is crap!

[psdayama]

Hi All,

There seems to be a great deal of confusion among hobbyists and new professionals concerning ground and circuit ground.


I post this diagram to help illustrate what Melanie already pointed out.

Nice to see the modified diagram but U have missed a vital component.
There should be small resistor like 22 Ohm( Ur LAW) to 47 ohm in serries
with 1uf/250V capacitor ( Melani wont approve of this ).Otherwise in few
days that capacitor will be reduced to open circuit (if of metallised film
type) or get short (if Film foil type) and whole ckt will go up in smoke.
The resistors should be atleast 1/2 watt type ( 1watt better). A high value
resistor like 220K is better placed across capacitor to protect user if he
accidently touches capacitor even when ckt is disconnected from mains.

[Pic_User]

Psdayama,

Thanks for the knowledgeable input. Makes the thread valuable to anyone reading it.
http://www.picbasic.co.uk/forum/atta...1&d=1163524317
-Adam-

[Melanie]

It's not cost saving, but size too...

When this thread started, I put forward an example of a Fan Controller. This fits inside a standard size wall recess which can be as shallow as 15mm. They just don't make transformer bobbins that small - and consider you have to wind around 8,000 turns of 0.01 or 0.008mm wire for a 0.5VA transformer primary - and even then it'll run warm because you really need about 12,000 turns at that rating and you quickly discover that there are minimum sizes for sensible transformer manufacture. Wire that thin is scary stuff to work with as it's so fragile.

Cost of course is a factor. A small Capacitor, a Diode and an Opto-isolator makes for a simple 'Fuse Blown' or 'Power Fail' detect circuit. Whilst a Transformer is nice, it just takes up too much PCB space - and you can't do it for 20 cents.

Finally, my choice of X2 Capacitors is deliberate. If you correctly chose your Capacitor, series Resistors are unnescessary. In all the years I've been designing these kinds of circuits, I've never had any failures of X2 Capacitors. Actually, I can't ever recall having a defective X2 Capacitor in any circuit. Chose your components wisely from reputable sources and you'll not have any troubles.

[shahidali55]

This App note has a lot of good information on transformerless PSU's

http://ww1.microchip.com/downloads/e...tes/00954A.pdf

[psdayama]

It's not cost saving, but size too...

When this thread started, I put forward an example of a Fan Controller. This fits inside a standard size wall recess which can be as shallow as 15mm. They just don't make transformer bobbins that small - and consider you have to wind around 8,000 turns of 0.01 or 0.008mm wire for a 0.5VA transformer primary - and even then it'll run warm because you really need about 12,000 turns at that rating and you quickly discover that there are minimum sizes for sensible transformer manufacture. Wire that thin is scary stuff to work with as it's so fragile.

Cost of course is a factor. A small Capacitor, a Diode and an Opto-isolator makes for a simple 'Fuse Blown' or 'Power Fail' detect circuit. Whilst a Transformer is nice, it just takes up too much PCB space - and you can't do it for 20 cents.

Finally, my choice of X2 Capacitors is deliberate. If you correctly chose your Capacitor, series Resistors are unnescessary. In all the years I've been designing these kinds of circuits, I've never had any failures of X2 Capacitors. Actually, I can't ever recall having a defective X2 Capacitor in any circuit. Chose your components wisely from reputable sources and you'll not have any troubles.

Your very first statement is exactly opposite to end one.
Check size of X2 capacitor and normal metallised film capacitor. You will
find X2 atleast 2times bigger than normal. Obivously no manufacturer can
provide X2 at same price so they are atleast 3times expensive.
Metallised film caps which are supplied by UL approved manufacturers will
NEVER short. They may become open with overload. So either way the
argument is pointless! A serries resistor costs less than a cent and it is
wise to limit inrush current.(In all situations). You may have not recieved any complaint with
X2 capacitor as they if not used properly become open circuit SLOWLY. If
anybody notices difference in EMI after 6months or years later, it will be
sheer accident!

I have designed and made a fridge controller (Blind type) with 555 and
similiar transformerless power supply some 3-4 years ago. I have used
normal 680n / 250V capacitor --( our system is 240V 50Hz) and those are
working nicely last 3-4 years and saving 100s of Rs. every month for the
users without any recall.

[sayzer]

...offering NO protection to the User of that circuit. If that Fuse is in or out, the circuit will be LIVE...

In this case, the circuit will protect itself but kill the user !


--------------------------