Is this legal?

Unbelievable.

Graham

"Hammy"

** You are kidding ?

Can you supply a link to the actual note ?

** It break every basic safety rule in the damn book.

Incredibly lethal.

Plus, would trip any ELCB ( aka GFI ) immediately it was plugged in.

Strewth !!!!!!!!!!!

........ Phil

Yes I know.This is from the engineers from Microchip.Scary Eh?

As requested here is the full note.

"Hammy"

** For the life of me, I cannot see any ( even twisted ) logic that gives a sane purpose to that fuse.

Can you ?

...... Phil

Well they say it's supposed to provide protection for incorrect outlet wiring; line and neutral reversed.

Yup, the fuse will indeed blow if the house is wired improperly.

This is almost as goofy as the Pic architecture itself.

John

--
Not necessarily outlet wiring, since they say: 

"Note that the neutral is connected to ground through a fuse. This
would guard against improper AC wiring."

That could also mean the AC wiring on the plug side, and while it
does both jobs, more than likely it\'s intended to protect the
downstream stuff, including the user, from mains hot accidentally
connected to the common portions (grounds) of the circuit.

border1.nntp.dca.giganews.com!nntp.giganews.com!nx02.iad01.newshosting.com!newshosting.com!64.59.134.4.MISMATCH!pd7cy3no!pd7cy4no!shaw.ca!pd7urf1no.POSTED!53ab2750!not-for-mail

If you're lucky, the GFCI feeding the outlet might trip due to part of the Neutral current being diverted to the Safety Ground. The really dangerous result would occur if the AC plug or outlet feeding this device had the grounded and ungrounded conductors reversed. This could connect the device ground to the AC Line voltage - 120 VAC so that a user touching a metal part of the device sharing that ground and the AC Line Ground (Earth or Water Pipe or Conduit, etc.) might receive a shock. If the AC outlet contains a GFCI, the shock might be limited to the ~5 mA trip current, but the fuse would not likely fail. However, nothing in NEC rules stops a user from making an improper connection between the non-grounded (Neutral) conductor and the Safety Ground conductor. Good Luck if you do this.

Ken Fowler, KO6NO

"John Fields"

** You failed to explain HOW it protects anyone or anything.

Seems you suffer from the same delusion of the innate benefit of fuses the Microchip writer did.

...... Phil

"Ken Fowler"

** It is certain to trip.

** Nonsense - the device ground in on a separate pin. ** Nonsense - the fuse would immediately explode, as it is then connected from active to ground. ** More nonsense !!!!

Safety earth must not be linked to a current carrying conductor in an appliance.

Bloody ham.

...... Phil

That is an interesting set of circuits.

Would these be safe circuits for a beginner to experiment with, without the fuse of course?

Also, I think there's a typo in Figure 3. I'm pretty sure the top 25V cap is 330 microfarads, not 330 millifarads...

Michael

Protect equipment at all costs, up to the live of the operator?? Those engineers need a 20 million dollar lawsuit, and a better education.

border1.nntp.dca.giganews.com!nntp.giganews.com!nx02.iad01.newshosting.com!newshosting.com!64.59.134.4.MISMATCH!pd7cy3no!pd7cy4no!shaw.ca!pd7urf1no.POSTED!53ab2750!not-for-mail

I have no intention of doing this. I came across it while looking into how to generate a cheap low power supply on the primary side of an SMPS. I was curious as to whether are not it is even legal in any developed countries.

The fuse would blow if you plugged that into an improperly wired outlet; Line to ground dead short.

The way I think it's supposed to work is for load protection. The circuit that's getting power from this supply uses the neutral line as its ground, zero volt reference. So if you were to switch the lines around the ground would be at the RC attenuated line voltage (AC) assuming the load draws any current in this case. The zener wouldn't regulate (likely fail) the polarized electrolytic would blow overvoltage and reversed polarity take your pick. I did some simulations on it switching the ground around which is the same as the neutral conductor which is grounded at the utilities transformer C.T and that's what happened. With the fuse there depending if it's fast enough and where the AC line is at when you plug it in and the type of load you have; it may save the load. I could be wrong but that's my take on it.

But still even if the wiring is correct you're not supposed to be pumping 40mA in this case into the safety ground. When it is functioning correctly you have a neutral to safety ground short. If your A/V systems hums check if you have any devices designed by Microchip into the same branch as your stereo.

"Hammy"

** Absolutely NOT !!

**Correct.

** Nonsense.

** WRONG - it uses the safety ground.

YOU are just as confused as the dope from Microchip.

The fuse protects NOTHING !!

** Correct.

It is a 100% LETHAL arrangement.

...... Phil

Right my mistake no need to get worked up about it.

Not quite that bad. I would never do that.

I think the guy who designed this was on drugs while simultaneously watching 2001: A Space Odyssey

Dave.

"David L. Jones"

** Stan D'Souza ( the author ) was and still is Microchip's main applications engineer.

His bio says educated at the University of Cincinnati, between '81 and

83 - most likely in computer engineering.

So he is probably about 46 years old.

......... Phil

"Hammy" wrote in message news: snipped-for-privacy@4ax.com...

I thought this might be a forgery, but here is the same app note on the official site:

Sometimes it is OK to use a transformerless power supply for a PIC. I have done this for a voltage relay, which determines if line voltage is within a certain permissible range, and it provides an output to the control circuitry via an optoisolator. It is enclosed in a plastic octal relay enclosure, and is externally fused at about 1 amp. It is important that the capacitor is rated for across the line duty, but the small fuse limits any major disaster, and the opto is rated at 2500 volts, so it is safe. There is still some chance of failure, as always, but as long as the external circuitry is grounded, or you use a GFCI, it is OK.

But, I have had problems with using a series capacitor (2 uF for 120 VAC, two in series for 240 VAC), wherein an 18 ohm 1 watt series resistor overheats, and the zener diode in series shorts out. At 60 Hz, the current should be about 90 mA and less than 1/4 W. But a square wave of 240 VRMS produces 800 mA and 12.8 watts in the resistor. I think the customer may have used the test set on a generator or possibly an inverter supply, which may have had considerable distortion and noise, so that is another consideration for a circuit like this.

It's strange, but I recently did a simulation for this, and I did not get much increase in current with a square wave. I will have to go back to the circuit I used for that and see where the discrepancy may be. It is greatly dependent on the rise time of the square wave.

I did an LTSpice simulation for the exact circuit proposed by Microchip, and I get about 232 mA with a 120 VRMS square wave, which puts 2.5 watts into the 47 ohm resistor.

But this is really an extreme condition. A square wave into a standard switching power supply would also cause an overload. And a square wave with

800 uSec rise and fall puts out only 89 mA, or 0.37 watts in the 47 ohm resistor.

Paul

ASCII file follows:

Version 4 SHEET 1 880 680 WIRE -192 128 -208 128 WIRE -80 128 -112 128 WIRE -32 128 -80 128 WIRE -208 160 -208 128 WIRE -32 224 -32 192 WIRE 32 224 -32 224 WIRE 160 224 96 224 WIRE 304 224 160 224 WIRE 448 224 304 224 WIRE -80 256 -80 208 WIRE -32 256 -32 224 WIRE -32 256 -80 256 WIRE 160 272 160 224 WIRE 304 272 304 224 WIRE 448 272 448 224 WIRE -208 288 -208 240 WIRE -32 288 -32 256 WIRE -208 416 -208 368 WIRE -32 416 -32 352 WIRE -32 416 -208 416 WIRE 160 416 160 336 WIRE 160 416 -32 416 WIRE 304 416 304 336 WIRE 304 416 160 416 WIRE 448 416 448 352 WIRE 448 416 304 416 FLAG 304 224 Vout FLAG 448 416 0 SYMBOL cap -48 128 R0 SYMATTR InstName C1 SYMATTR Value 1µ SYMATTR SpiceLine V=600 SYMBOL voltage -208 144 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 WINDOW 3 -473 111 Left 0 WINDOW 0 -43 10 Left 0 SYMATTR Value PULSE(0 240 1m 40u 40u 8.3m 16.6m 1000) SYMATTR InstName V1 SYMBOL zener 320 336 R180 WINDOW 0 -17 64 Left 0 WINDOW 3 -56 -8 Left 0 SYMATTR InstName D3 SYMATTR Value 1N750 SYMATTR Description Diode SYMATTR Type diode SYMBOL polcap 144 272 R0 WINDOW 3 24 64 Left 0 SYMATTR Value 330µ SYMATTR InstName C2 SYMATTR Description Capacitor SYMATTR Type cap SYMATTR SpiceLine V=25 Irms=760m Rser=0.09 MTBF=1000 Lser=0 mfg="Panasonic" pn="ECA1EFQ331" type="Al electrolytic" ppPkg=1 SYMBOL res 432 256 R0 SYMATTR InstName R3 SYMATTR Value 1k SYMBOL res -96 112 R90 WINDOW 0 0 56 VBottom 0 WINDOW 3 32 56 VTop 0 SYMATTR InstName R2 SYMATTR Value 47 SYMBOL voltage -208 272 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V2 SYMATTR Value -120 SYMBOL diode 32 240 R270 WINDOW 0 32 32 VTop 0 WINDOW 3 0 32 VBottom 0 SYMATTR InstName D1 SYMATTR Value UPSC600 SYMBOL res -96 112 R0 WINDOW 0 -32 75 Left 0 WINDOW 3 -49 107 Left 0 SYMATTR InstName R1 SYMATTR Value 1Meg SYMBOL diode -16 352 R180 WINDOW 0 24 72 Left 0 WINDOW 3 24 0 Left 0 SYMATTR InstName D2 SYMATTR Value UPSC600 TEXT -304 456 Left 0 !.tran 10s startup

The point there is to blow the fuse if the outlet is wired up backwards. (replace N with L and you get a short to ground which will blow the fuse).

I have no idea why this matters except that the first diode to ground would be shorted if the outlet is wired up wrong(hence that is the reason for the fuse). But the circuit definitly could be improved by placing a diode before the first diode so that if its wired up wrong it won't short the diode... that or place another fuse there.

In any case its not really a good circuit and in some cases N is not g and substantial current could flow. (I'm going by the note on why the fuse exists and its easy to see why they added it to that circuit but makes no sense why they made such a sorry circuit in the first place.