A Simple AC powered LED Lamp

[RCinFLA]

The large ripple voltage on 100 uF causes it to overheat. Electrolytics do not like large ripple voltage. The ballast cap approach works but there is a potential problem with 'bouncing' the A.C. either by plug-in or switch. If the AC is momentarily interrupted by contact bounce the ballast cap charge may add to the newly applied AC peak creating a current surge to LED's. Besides filtering some of the 100 Hz ripple to LED, the primary reason to have the cap across the LED is to absorb the potential bounce spike. The 100 to 200 ohm resistor should be between rectifier and electrolytic to help limit the current spike on contact bounce. Using a higher voltage electrolytic will help a bit on the electrolytic ripple issue.

 

[RODALCO]

Hi Zmint.

I did read your thread but was busy with other works as I' am spending too much time on the web and don't get around anything else at home anymore.
Just had to repair the Buzz buttons broken wiring on the kids PS-2 game, besides of other house hold choirs etc.

The design flaw in that schema is the oversized Cap of 470 nF which can yield in a current up to 40 mA!!
Then the inrush current will be too high, a 1 or 2 k.ohm resistor of at least 1 to 2 Watts needs to be in series with the cap. to limit the led start current.
Then the elco rated at 35V is underrated, should be at least 63V and 105°. designed for high ripple loading.

I aim usually for 10-15 mA for LED circuits with the resistive design as discussed in my YT video, for long term reliability.

Kind regards

Raymond

 

[zmint]

Better Idea...

Thanks to all of you for helping me. From all the suggestions I got, I think this is close enough to a better circuit I came across :


This circuit employs capacitive reactance for limiting the current flow through the LEDs on application of mains voltage to the circuit. If we use only a series resistor for limiting the current with mains operation, the limiting resistor itself will dissipate around 2 to 3 watts of power, whereas no power is dissipated in a capacitor. The value of capacitor is calculated by using the following relationships:
XC = 1/(2πfC) ohms —————(a)
XC = VRMS
/I ohms ———— (b)
where XC is capacitive reactance in ohms,C is capacitance in farads, I is the current through the LED in amperes, f is the mains frequency in Hz, and Vrms is the input mains voltage. The 100-ohm, 2W series resistor avoids heavy ‘inrush’ current during transients. MOV at the input prevents surges or spikes,
protecting the circuit. The 390-kilo-ohm, ½-watt resistor acts as a bleeder to pro-vide discharge path for capacitor Cx when mains supply is disconnected. The zener diode at the output section prevents ex-
cess reverse voltage levels appearing across the LEDs during negative half cycles. During positive half cycle, the voltage across LEDs is limited to zener voltage. Use AC capacitors for Cx. Filter capaci-
tor C1 across the output provides flickerfree light.

Is it all right now ? or are there still many errors to be corrected.
What can be substituted for MOV 270/20, if I don’t get hold of it. ?

Thanks
Zmint. (Trying to gain, a positive charge from electronics)

 

[3lectrokid]

may i ask what is so good having an ac lamp instead of a dc lamp. does it increase performace. i thought that using a rectifying bridge it would stop flicker

 

[audioguru]

A rectifier bridge doubles the frequency of the flicker. It still flickers on and off. The capacitor following the rectifier filters the flickering into DC that does not flicker.

 

[zmint]

The value of capacitor is calculated by using the following relationships:
XC = 1/(2πfC) ohms —————(a)
XC = VRMS
/I ohms ———— (b)

1) Can someone please elaborate, how the vlaue .22uF for Cx* was reached, for using 16 LED
2) Secondly is there any programmable shunt regulator which can be programmed for 48v, 1 w or 69V 1 W.
3) The "mov 270/20" what does it signify .. 270 ? 20 ? I was not able to get it. There are some kind available but they have different labeling system.

 

[RODALCO]

1) 0.22 uF yields in 16 mA at 230 Volts
2) No
3)MOV clamping voltage 270 Volts 20 kA rating, you can fit another one as long it has the 270 Volts rating or just leave it out. ( a power spike could take the lamp out)

 

[zmint]

MOV clamping voltage 270 Volts 20 kA rating, you can fit another one as long it has the 270 Volts rating or just leave it out. ( a power spike could take the lamp out)

The one I have got is 47D471K will it work ?

 

[RODALCO]

Can't find any data on it, sorry.

 

[Hero999]

Yes it's highly likely that C1 has blown and should be replaced with a 250VAC X2 rated capacitor.

I would also add a 470R resistor in series with the whole circuit to limit the surge current and reduce the harmonic currents to a safe level.

I suggest you throw them out. That circuit is EXTREMELY dangerous. I'm surprised they'd let anything like that on the market. It's a shock hazard, and a fire hazard. You can be grateful it only blew a capacitor without starting a fire.

The lack of a step down isolation xfmr is totally irresponsible.

It's not dangerous at all if all the components are insulated from the user. It's no more hazardous than an SMPS or any other mains operated device.

 

[zmint]

Yes it's highly likely that C1 has blown and should be replaced with a 250VAC X2 rated capacitor.

Sir, I guess this for the first circuit I posted ?
Anyhow, what exactly is x2 rated capacitor ?

I would also add a 470R resistor in series with the whole circuit to limit the surge current and reduce the harmonic currents to a safe level.

Have a look at the second circuit I posted, Is it the way it should be done ?

Thanks

 

[zmint]

?

I have also made a lamp as described in one for your post https://www.electro-tech-online.com/threads/240-volts-100-hertz-29-led-lamp.96363/#post791590
I used 4 x 3.3 K 2 W resistor for 15 LED. Is it Ok ?
Secondly, is it must that the resistors number (quantity) should be in even number and they have to be connected on each side. Can they all be connected in a single series also ? When we connect resistor in series or in parallel (same value) do we add up their wattage to get their total working wattage in both the cases ?

Your original thread was closed, so I asked here.
I have also attached a varistor datasheet which my dull electronic head could not understand.
View attachment Metal Oxide Varistors datasheet(2).pdf

Thanks

 

[Hero999]

Sir, I guess this for the first circuit I posted ?
Anyhow, what exactly is x2 rated capacitor ?

Yes.

X2 is an over voltage rating. Normally this kind of rating is given to capacitors used for RF suppression.
**broken link removed**

Have a look at the second circuit I posted, Is it the way it should be done ?

Thanks

There is nothing wrong with doing it that way but it's better to use a full wave rectifier because it reduces the ripple, especially if a capacitor is connected in parallel with the LEDs.

 

[tcmtech]

This is the basic circuit I have been using for many years without problems.

 

[Hero999]

I think you should put a 1M bleeder across C1 to discharge it when the power is removed and a 470R resistor in series to reduce the on surge and harmonics.

 

[tcmtech]

I could but given that the original LED light I made with it has been on non stop for about 5 1/2 years now I am content with how its been working as is.

 

[zmint]

This is the basic circuit I have been using for many years without problems.
1

What are the values you use for C1, C2 and R1.
Basically, how would I calculate them, for a 230 V 50Hz AC. power supply.

 

[tcmtech]

The method I used was to subtract the total voltage drop of all of the LED's in series from the RMS voltage of the lines. Then use that value to calculate the needed impedance of the capacitor at the normal line frequency and voltage I have in order to limit the LED current like a resistor normally would but without the power losses associated with it.

For the R1 and C2 values I just use any spare capacitor I have laying around thats between 20 - 250 uf and has a higher voltage rating than the voltage drop the LEDS produce.
Its mainly just to take the flicker out and give it a short standby power source during momentary power blinks. It also absorbs line spikes that everyone seems to get their underwear so knotted up over in regards to line powered LED circuits as well.

R1 is for additional flicker and spike smoothing of sorts and allows for any short voltage spikes being dumped into the C2 capacitor to be further trimmed down. I use around 200 - 500 ohms depending on how much current I am driving the LED's at. But its not critical so anything from 100 to 1K will work just fine.
If you think you may have higher levels of line noise a larger value of C2 and R1 will further filter it out.

 

[Hero999]

I could but given that the original LED light I made with it has been on non stop for about 5 1/2 years now I am content with how its been working as is.

Hang on I've realised that you have a resistor in series with the LED and the capacitor will help to remove spikes and surges.

 

[tcmtech]

I never was any good at doing things just like everyone else.

The theory I used (for whatever reason) was to absorb the spikes and then use their energy instead of block them.