Transformerless power supply

I have a simple circuit on a breadboard that has been doing fine 24/7 for months which is --

120vac, hot into a 200k 1/2 watt that feeds a dropping cap (.47uf, 400v) that then feeds a bridge rectifier. The output of the bridge has a zener and a filter cap which supplies my DC.

I breadboarded it because I really never used a transformerless design in a project and just wanted to see how reliable it was. (so far -- very reliable).

My question is about the 100 ohm resistor (or around there) that is used in the neutral leg of the AC feeding the bridge that you often see in this circuit.

I'm not using one, just tying the neutral directly to the bridge.

So what is the purpose of this resistor? A safety precuation? Just to provide "some" isolation from ground (neutral 120vac?) Or maybe it acts as a cheap fuse?

Of course, you'd be stupid to not have the ground isolated from the chassis or anything that you could touch (like they did in the 50's and 60's).

I'll have to say that it's been very reliable on the breadboard anyway -- but I'm only drawing 25ma or so with my load.

Reply to
mkr5000
Loading thread data ...

months which is --

that then feeds a bridge rectifier. The output of the bridge has a zener an d a filter cap which supplies my DC.

a project and just wanted to see how reliable it was. (so far -- very reli able).

in the neutral leg of the AC feeding the bridge that you often see in this circuit.

vide "some" isolation from ground (neutral 120vac?) Or maybe it acts as a c heap fuse?

is or anything that you could touch (like they did in the 50's and 60's).

but I'm only drawing 25ma or so with my load.

I don't know about the 100 ohms. But is this a one of? You're using the AC 'neutral' as your circuit ground? (That's a bit scary if so.)

I'm also having a hard time adding up your numbers... the 200k ohm resistor is in series with the cap and then feeds the bridge? (25mA *

200k ohm = ~5kV?)

George H.

Reply to
George Herold

months which is --

then feeds a bridge rectifier. The output of the bridge has a zener and a filter cap which supplies my DC.

project and just wanted to see how reliable it was. (so far -- very reliable).

neutral leg of the AC feeding the bridge that you often see in this circuit.

"some" isolation from ground (neutral 120vac?) Or maybe it acts as a cheap fuse?

anything that you could touch (like they did in the 50's and 60's).

I'm only drawing 25ma or so with my load.

The "100 Ohm resistor" is usually there just to limit the amplitude of the surge current.

Post your circuit in ASCII or on A.B.S.E or by a URL link so we can properly help. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     | 
| Analog Innovations, Inc.                         |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

that's what I thought Jim -- won't stop a surge but at least help.

don't trust my numbers, actually can't remember the R value off hand -- even the

25ma is an exaggeration.

really just wanted to know the purpose of the resistor in the neutral side.

actually, it's been very reliable -- like I say, may end up using it in projects where i don't have room for a transformer.

using (4) 1000v rectifiers, and plenty of headroom with the resistor power ratings.

Reply to
mkr5000

months which is --

then feeds a bridge rectifier. The output of the bridge has a zener and a filter cap which supplies my DC.

--
What's the value of the filter cap, the Zener voltage, the load 
resistance, and the output voltage?
Reply to
John Fields

the 25ma is an exaggeration.

projects where i don't have room for a transformer.

ratings.

I've done a number of "capacitive dropper" supplies, mostly for appliance applications, where the user has no exposure to the "hot" circuits. They work just fine. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     | 
| Analog Innovations, Inc.                         |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

months which is --

then feeds a bridge rectifier. The output of the bridge has a zener and a filter cap which supplies my DC.

project and just wanted to see how reliable it was. (so far -- very reliable).

neutral leg of the AC feeding the bridge that you often see in this circuit.

"some" isolation from ground (neutral 120vac?) Or maybe it acts as a cheap fuse?

anything that you could touch (like they did in the 50's and 60's).

I had a Hallicrafters S38 SW receiver that had one side of the (unpolarized!) AC line soldered to the chassis. The chassis was isolated from the exterior metal case by a few rubber grommets that had worn through, so the case was hot. I checked it with a neon bulb (hold one lead, touch the case with the other) and reversed the plug if it lit.

I wonder how many people were killed by stuff like that, like a hot-chassis radio falling into a bathtub or something.

--
John Larkin         Highland Technology, Inc 

jlarkin at highlandtechnology dot com 
 Click to see the full signature
Reply to
John Larkin

Is it possible to reverse the polarity of the input? That is, to feed the hot directly into the bridge and have the cap and 200K resistor connected to the neutral?

If so, I'd guess that the 100 ohm resistor is a current limiter, should the polarity be reversed and some point in the circuit inadvertently becomes grounded. Otherwise, 100 ohms is effectively in series with 200K, so its effect on the normal operation of the circuit would not be noticeable.

--
Paul Hovnanian     mailto:Paul@Hovnanian.com 
------------------------------------------------------------------ 
 Click to see the full signature
Reply to
Paul Hovnanian P.E.

"mkr5000" <

** What is the 0.47 cap for when the 200k does all the dropping ??

I think you have stuffed up the schem.

The 200k is meant to go across the 0.47 uF ( so it discharges when disconnected from the AC supply ) - the cap alone will pass about 20mA AC to the bridge and zener.

The 100 ohm is simply a surge limiter and can be in series with the A or N conductors.

... Phil

Reply to
Phil Allison

On a sunny day (Fri, 25 Jan 2013 18:31:37 -0800) it happened "Paul Hovnanian P.E." wrote in :

There was a discussion about that many years ago, and IIRC it involved switching the device on when the mains period is at maximum. Steep du/dt, makes the cap a virtual short. So you need peak current limiting, and that is why the series resistor. I have used such a circuit myself a couple of times. OK if you are 100% double insulated in EU.

Reply to
Jan Panteltje

maximum.

--
With 200k in series with 120V mains, at the peak voltage (170V) the 
maximum current into a short would be 848.4µA. 
 Click to see the full signature
Reply to
John Fields

My Mother has a baby photo of me and she is giving me a bath in the kitchen sink. (late 50's)

Right above me (and above the water), is a kitchen AM radio and guess what? It's plugged in !

The power cord was hanging on the wall right behind me.

Reply to
mkr5000

Like I say, forget about my numbers, I was talking out of my &*& -- in a haste just referenced a schematic I saw that had a 200k in it.

My circuit has (2) 20K 1/4w in parallel right before a .47uf cap and I'm just driving a couple of LED's with a 5.1v zener.

Just a test circuit.

You guys answered my question about the 100 ohm -- I'm not using it but sounds like good practice.

Another thing about this circuit --

I know it's used a lot for driving LED's. Sure seems a lot more practical sucking the 120vac for a few mils rather than a high frequency driver.

?
Reply to
mkr5000

maximum.

In which countries is this products sold ?

In countries with polarized plugs, it is usually forbidden to put a fuse in the neutral wire.

In an area with non-polarized plugs and thus possibly the full mains voltage to ground on each terminal, in case of a isolation fault, in the bad case, about 1 A will flow through the 100 ohm resistor, quickly evaporating the low power resistor.

As long as you do not call that device a "fuse" it can be used in countries with both polarized as well as non-polarized plugs :-).

Reply to
upsidedown

On a sunny day (Sat, 26 Jan 2013 08:08:44 -0600) it happened John Fields wrote in :

maximum.

True, but as he specifies 1/2 Watt I assumed he ment 200 Ohm.

200k would not even work after a bridge with any load, and you would not need the cap anymore either, and it is likely below the zener normal current.
Reply to
Jan Panteltje

months which is --

then feeds a bridge rectifier. The output of the bridge has a zener and a filter cap which supplies my DC.

project and just wanted to see how reliable it was. (so far -- very reliable).

the neutral leg of the AC feeding the bridge that you often see in this circuit.

"some" isolation from ground (neutral 120vac?) Or maybe it acts as a cheap fuse?

anything that you could touch (like they did in the 50's and 60's).

I'm only drawing 25ma or so with my load.

formatting link

Reply to
bloggs.fredbloggs.fred

just referenced a schematic I saw that had a 200k in it.

driving a couple of LED's with a 5.1v zener.

like good practice.

sucking the 120vac for a few mils rather than a high frequency driver.

A common LED night-light circuit is a cap current limiter in series with a small spike/surge-limiting resistor, then an LED with an anti-parellel diode. Variants are two leds back-to-back, or a bridge rectified driving one or more led's. It's fairly efficient because the line current is mostly unreal reactive power. It's pobably more reliable than an electronic converter.

The extra resistor in the neutral leg isn't necessary. Some additional surge protection, like your zener, would be good for higher current situations. Your

10K series is pretty high, lowers efficiency, but that's a lot of surge/spike limiting. The cap's reactance is 5K, so your current limiting is mostly resistive/lossy.

I wonder if anything interesting can be done with a capacitive voltage divider. If the coffee were done dripping, I might look into that.

And an anti-phase-control thing might be interesting, a series device that only conducts at low voltage and turns off at high voltage.

--
John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    
 Click to see the full signature
Reply to
John Larkin

Your 10K series is pretty high, lowers efficiency, but that's a lot of surg e/spike limiting. The cap's reactance is 5K, so your current limiting is mo stly resistive/lossy.

so, that tradeoff may be a good one? protection vs efficiency?

but of course, for higher current circuits, resistance plays a bigger role because of the heat generated in the resistor?

what value of resistance would offer surge protection worth anything on an AC line? or -- forget about it? because a typical surge is going to wipe y ou out regardless? isn't that the nature of a transient on an AC line? (no t talking over voltage conditions but say a lightning induced spike)

John, that brings me to another question, I'm sure you know the answer to -

-

can there really be smaller amplitude transients on an AC line depending on your circumstances? (location etc from the problem) or is a spike generall y "pretty nasty" close to the source of the problem and drop off quickly wi th distance?

I hope I'm making sense. In other words, can the amplitude from a bad spike be anywhere from a 1 to a 10 or is generally a 9 or 10 and then die off ra pidly?

Reply to
mkr5000

now that I read my post, kind of a stupid question really --

1 through 10 no doubt but I guess it's "when and where" along the length of the power line that I'm thinking about.
Reply to
mkr5000

surge/spike limiting. The cap's reactance is 5K, so your current limiting is mostly resistive/lossy.

because of the heat generated in the resistor?

line? or -- forget about it? because a typical surge is going to wipe you out regardless? isn't that the nature of a transient on an AC line? (not talking over voltage conditions but say a lightning induced spike)

your circumstances? (location etc from the problem) or is a spike generally "pretty nasty" close to the source of the problem and drop off quickly with distance?

anywhere from a 1 to a 10 or is generally a 9 or 10 and then die off rapidly?

One surge is when you connect it to the line. You might see the full peak, 150 volts or so. If the series cap was left charged from the last disconnect, double that. That's one reason to add a 10M or such discharge resistor.

And an AC line will have nasty spikes from other loads being switched and, sometimes, lightning.

You've got to do the math. Take a case: 0.47 uF and 1K. Apply 150 volts and you get 150 mA with a decay time constant of 470 microseconds. A 500 volt spike would make 500 mA for however long it lasts, maybe 100 us. The led or whatever has to survive that.

What I've seen for night lights is a cap and a resistor, into back-back LEDs or an LED and an antiparellel diode. For bigger stuff, it's an RC on the line side, a full bridge rectifier, and electrolytic filter cap, a series 1-watt sort of resistor, and the LED string. That elecrolytic cap + series resistor reduces flicker and absorbs transients.

If you search the web, there are a lot of really silly night-light circuits.

--
John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    
 Click to see the full signature
Reply to
John Larkin

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.