Capacitive Reactance

[SpiderElectronics]

I'd love to see Dave do a video blog on Capacitive Reactance - you know, the kind of thing where you can build a mains ac to 5v DC pus using just a big capacitor and a few diodes.

[Psi]

It's pretty simple to do, you just put the info into the formula

Xc (Reactance) =  1 / (2 * pi * mains frequency * capacitor value)

So a 470nF 250v capacitor on a 50hz mains system is

Xc = 1 / (2 * 3.14 * 50 * 0.000000470)
Xc = 1 / 0.00014758
Xc = 6776 ohms

So 470nF in series with your zener/bridge will limit the current to a max of   230V / 6776ohms = 34mA


These sort of power supplies are very useful but not the sort of thing you want to encourage people to build.
Some designs have outputs that are floating at mains voltage so they're not for beginners.

[IanB]

So 470nF in series with your zener/bridge will limit the current to a max of   (230V * 1.41) / 6776ohms = 48mA

Why is the 1.41 there? Surely with an AC voltage applied to a reactance you will get an AC current with no extra adjustment factors needed: 230 V AC / 6776 ohms = 34 mA AC ?

[Zad]

They have a horrible habit of conducting high voltage high frequency spikes through...

[Mint.]

I reckon that's a good idea.

[amspire]

They have a horrible habit of conducting high voltage high frequency spikes through...

The supplies that rely on a capacitive divider are safest if you use an isolating transformer in front ...   and lower the transformer secondary voltage down to the voltage you need   ... and then get rid or the divider capacitor since you no longer need it.

Richard

[Psi]

So 470nF in series with your zener/bridge will limit the current to a max of   (230V * 1.41) / 6776ohms = 48mA

Why is the 1.41 there? Surely with an AC voltage applied to a reactance you will get an AC current with no extra adjustment factors needed: 230 V AC / 6776 ohms = 34 mA AC ?

Sorry.. yeah, now that i think of it, it should be a continuous current, so no peak.
I was thinking of it like a resistor.

[vk6zgo]

For a tiny current like that,there is very little advantage in using a capacitor instead of a small transformer.
Many transformer type "Wall warts" are quite small,& later switchmode types are even smaller.
Both types are a hell of a lot safer than a series capacitor,where you are playing Russian roulette as to which leg the cap is in, L or N

VK6ZGO

[ivan747]

You have to add some surge protection in a transformerless supply. Here's an application note from Microchip on capacitive power supplies. Be careful, this is dangerous stuff.

http://ww1.microchip.com/downloads/en/appnotes/00954A.pdf

FYI: I haven't built any of those circuits because I don't own an isolation transformer to test the supplies or a residual current circuit breaker installed.

Edit: Notice both of the supplies described are at mains potential (eg. if the line is at 180V at a particular point, the output will be 180V-5V, or whatever voltage you are using, when measured from the neutral wire).

[IanB]

I think testing those transformerless circuits is no more dangerous than testing a live mains circuit such as a lighting circuit or a domestic appliance. The suggestion of an isolation transformer is just guarding against unwise actions such as hooking an oscilloscope up to the mains.

That application note also contains some serious flaws, such as specifying a 250 V rated capacitor without also specifying the assumption of a 120 V mains supply. In fact the whole application note remains blissfully blind to the existence of 240 V mains voltages... 

[ivan747]

I think testing those transformerless circuits is no more dangerous than testing a live mains circuit such as a lighting circuit or a domestic appliance. The suggestion of an isolation transformer is just guarding against unwise actions such as hooking an oscilloscope up to the mains.

The most important application for an isolation transformer here is to allow you to touch the +5V line without being shocked, remember, all the resistance and capacitive reactance is on the lower side of the circuit.

[IanB]

The most important application for an isolation transformer here is to allow you to touch the +5V line without being shocked, remember, all the resistance and capacitive reactance is on the lower side of the circuit.

You're not supposed to touch any live conductors when you are testing a mains energized circuit. This applies whether you think it is safe or not. Good practice says "hands off"!

Plenty of people built, tested and repaired highly energized valve (vacuum tube) radios and other devices in the old days without electrocuting themselves. There are a multitude of safe techniques available including the use of clip-on probes attached before energizing the circuit, the keeping of one hand behind the back at all times, and above all, never touching any exposed metal bits.

[vk6zgo]

This kind of circuit was very rarely used in the past,because its inherent disadvantages outweighed any savings.
It also is an AC only circuit,so didn't even have the excuse that transformerless supplies had in the past--that they could be used on DC mains supplies.

I have always been bemused by the liking for AC/DC transformerless designs in old UK & USA made radios.
The standard in Australia,until the advent of switchmode supplies was always a transformer type.
The radios were not all that much heavier,or expensive.
OK,perhaps there were more DC mains systems still surviving in those countries,but special DC models  could have been made,as they were in Australia.

I have used series caps to run LEDS off an isolated  65 volt control supply circuit,but the idea of using them as part of a mains supply makes my hair stand on end.
I have worked on many pieces of equipment with transformer & switchmode type supplies.
In most cases,the likelihood of injury is minimal,if you keep on the isolated side of the supply.

VK6ZGO

PS: I just had a look at the app note---These things are OK inside the guts of a double insulated appliance,but NOT on the bench of a person with little experience.
The app note gives all the information you would ever need about them.I would strongly advise against building one!

[IanB]

I have always been bemused by the liking for AC/DC transformerless designs in old UK & USA made radios.

I never saw any in the UK. Valve radios had a honking big transformer with windings for the low voltage heater supply and the high voltage active part of the circuit. When transistors came along there was a simple transformer based low voltage supply. I took loads of them apart to find out what was inside them.

The transformerless supplies are useful where some low power device like a microcontroller is integrated with high voltage parts like triacs. There is no safety hazard because the whole circuit is considered live and dangerous, just as if there were no low voltage parts at all. I would think most wall mounted dimmer switches fall into this category.

[vk6zgo]

Very old UK radios had transformerless supplies--I remember seeing them being brought in for repair at my first job.
"Poms" would bring the horrible things with them when they came to Oz.
They were commonly known as "Death radios".
A few American radios of this type turned up,but more Brits.

UK TVs were transformerless at around the same period,& a bit later.

VK6ZGO

[IanB]

I think some of the radios with big transformers I had my hands on were from the thirties or forties. I have an idea maybe that designs from the 50's or 60's might have dropped the big transformer as a cost saving measure. I think a 50's valve radio belonging to my parents may not have had a power supply transformer in it. I also recall that valve TV sets I took apart never had transformers in them except for the impedance matching transformer for the speaker. It makes me wonder how the valve heaters were powered? I didn't have enough knowledge to figure out the circuits at the time. I do recall there were some big wire wound resistors on ceramic formers that were clearly intended to get hot. (When you turned on an old TV there was that wonderful smell of burning dust rising from the back and a creaking noise as things warmed up, plus the gentle glow of the valves lighting up visible through the vent holes in the back.)

[vk6zgo]

They used series strings of valve heaters--the valves had higher than normal filament voltages.
They did the same in the USA.
VK6ZGO

[Psi]

The technique is quite often used in modular units for wall plates.

for example, digital timers that have to fit inside a tiny box and run directly from 230v


It works well for this application because it only interacts with the outside world through a relay so there's no problems keeping it isolated.

[ivan747]

The technique is quite often used in modular units for wall plates.

for example, digital timers that have to fit inside a tiny box and run directly from 230v


It works well for this application because it only interacts with the outside world through a relay so there's no problems keeping it isolated.

I have though of a few designs like this for some time, specially remote controlled outlets, X10 systems and automatiation systems like that. The only thing stopping me from doing stuff like that is that I don't trust some of my equipment (specially my Velleman scope for mains work), I don't have an isolation transformer to test stuff on and, overall, I don't have enough self-confidence to work on mains because of details like that. Maybe I have to start dealing with the mains more frequently. 

By the way, here are some more Microchip application notes on transformerless supplies and their applications:

"AN236: X-10® Home Automation Using the PIC16F877A"
   http://ww1.microchip.com/downloads/en/appnotes/00236a.pdf

"AN958: Low-Cost Electric Range Control Using a Triac"
   http://ww1.microchip.com/downloads/en/AppNotes/00958A.pdf

[Psi]

The technique is quite often used in modular units for wall plates.

for example, digital timers that have to fit inside a tiny box and run directly from 230v


It works well for this application because it only interacts with the outside world through a relay so there's no problems keeping it isolated.

I have though of a few designs like this for some time, specially remote controlled outlets, X10 systems and automatiation systems like that.

Me too, i'm working on a UPB module for remote controlling outlets.