## High Current Transformerless Power Supply Circuit

The simple configuration of a transformerless power supply circuit presented below is able to provide high current at any assigned fixed voltage level. The idea seems to have solved the problem of deriving high current from capacitive power supplies which earlier seemed a difficult proposition. I assume I am the first person to have invented this.

I have discussed a few transformerles power supply circuit in this blog which are good only with low power applications, and tend to become less effective or useless with high current loads.

The above concept utilizes high voltage PP capacitors for dropping the mains voltage to the required level,  however it is unable to raise current levels as per any desired particular application.

Although, since the current is directly proportional to the reactance of the capacitors, means the current can be lifted just by incorporating more capacitors in parallel. But this puts a risk of high initial surge currents which might destroy the involved electronic circuit instantly.

Therefore adding capacitors might help to increase the current specs of such power supplies but the surge factor must be first taken care of for making the circuit feasible for practical usage.

The circuit of a high current transformerless power supply explained here hopefully, effectively handles the surge developing from power transients such that the output becomes free from the dangers, and provides the required current supply at the rated voltage levels.

Everything in the circuit is kept just as its old counterpart, barring the inclusion of the triac and zener network which  actually is a crowbar network, used for grounding anything that goes above the rated voltage.

In this circuit the output would hopefully provide a stable voltage of around 12+ volts at around 500 mA of current without the dangers of any accidental voltage or current influx.

CAUTION: THE CIRCUIT IS NOT ISOLATED FROM MAINS AND THEREFORE INVOLVES HIGH RISK OF ELECTROCUTION, APPROPRIATE PRECAUTION NEEDS TO BE EXERCISED.

UPDATE:  A better and a more advanced design can be learned in this zero crossing controlled surge free transformerless power supply circuit

Parts List

R1 = 1M, 1/4W
R2,R3 = 1K, 1/4 WATT
C1----C5 = 2uF/400V PPC, EACH
C6 = 100uF/25V
All DIODES = 1N4007
Z1 = 15V, 1 watt
TRIAC = BT136

A neatly drawn PCB for the above high current transformerless power supply may be seen below, it was designed by Mr. Patrick Bruyn, one of the avid followers of this blog.

Update (August 22, 2014)

A deeper analysis of the circuit showed that the triac was dumping a significant amount of current while restricting the surge and controlling the current.

The approach taken in the above circuit for controlling voltage and the surge is negative in terms of efficiency.

In order to obtain the intended results as proposed in the above design and without dumping precious amps, a circuit with exactly opposite response needs to be implemented, as shown below:

Interestingly, here the triac is not configured to dump power rather it's wired in a such a way that it switches OFF power as soon as the output reaches the specified safe voltage limit, which is detected by the BJT stage.

#### Swagatam Majumdar

Hi Friends, Welcome to my site, a place where you will discover a massive collection of electronic circuit ideas, mostly requested by the dedicated readers and exclusively designed by me for their customized application needs. I have posted more than 1100 circuit designs in this site, if you have a personalized circuit requirement you may feel free to request it through the comment box, if it seems feasible to me then surely you may find it published here with your credentials attached in the post, thanks and please keep reading>

1. dear sir i have 230\12 15amp transformer. i need this transformer to make battery charger are you have a any circuit of battery charger circuit please send it.
thankyou

1. Hi Mubarak,

I have posted many battery charger circuits in this blog, kindly search them using the search box.

Regards.

2. Very good content

1. My pleasure...thanks.

3. dear sir,
what is the value of cap C1 shown in o/p?. C1 is repeated in drawing. is both same value? filter cap also same value as voltage drop cap? pls reply

thanks n regards

1. Hi Vineesh,
I have updated the parts list correctly, you can check it.

Regards.

4. Dear Swagatam,

Can this circuit illuminate 24Nos 5mm LED (series-parallel connection).

1. This circuit is capable of powering more than 500 LEDs, 24 is too small a number, you can use other simpler circuits for it, Just type transformerless power supply in the search box and you will find the right ones for your application.

Regards.

5. Great work Swagatam, u've trully outdone yourself with this one.

1. Thanks very much Bourgeoisie!!

6. Dear Swagatam,

After correction you have posted two circuit diagram. One of them consisting of opto-coupler another is consisting of only zener diode.

Which is right. Please provide the specification of the opto-coupler.

RAY

1. Dear Ray,

For simplicity sake you can try the one which is without a bridge rectifier.

The opto can be any phototriac.

7. hi Sir,
I have to know about power circuit for three phase power supply I have three phase power supply and in case of any of them disconnect then also my main circuit can do there work in any phase faultier. my main circuit is for earth leakage circuit breaker as you put in your blog give me the source of power supply when any phase faultier.

1. I'll try to design it soon and inform you when its done....

8. Dear Swagatam,

You have mentioned that one can use more than 500 LEDs with the above circuit. How do i connect the LEDs like parallel or series and should i add any resistor and where. Suppose if i connect 500 LEDs, what would be the brightness like when compared to Tube light or CFL and what would be the power consumption.

Regards
Nagaraj

1. Dear Nagaraj,

The above circuit has not been verified practically by me so I cannot say much until I test this circuit in my lab.

Regards.

9. which one is ok circuit please removed the wrong one

1. the middle circuit is good, just invert the zener, though....

10. dear SIr,
I tested this. it is working. I m thankful you very much..
can i use this circuit for 30W Hight powe LED.]
can you post 10W to 100W hight power driver circuit...
thank you very much...!

1. As you have tested it, if it is working as per the proposed specs, then you can try it with a 30 watt led, it should work.

Thanks!

11. dear sir,
can i use BT139 triac instead of the BT136 ...?

if i use BT139 in this circuit will the LED burn..?

1. Which triac did you use in your tested circuit?

It should be BT136 only, not BT139.

12. sir I did not tested this triac circuit..I just tested only your 1st circuit for 1W led....now i m going to test this triac circuit..before I asked.thank you I will find the correct triac for this one...

also i wanna say ...the 1st ciruit's 1W Resister make too much hot...so i put there 22Ohoms..but it is also make hot...I think it is suitable...
thank you Sir...

1. Which circuit are you referring to....??

13. Hi Swagatam,

in your first modification, is it LED or diode?
if LED, is it normal LED like 5mm?

Regards,

1. It can be a 5mm LED, because it's only for indicating purpose....

14. Hi
If I email you a circuit could you tell me if your power supply could cope with it please
Regards
Larry - smithjollyroger@gmail.com

1. You may send it, I'll try my best to check it.

Regards.

15. Hi
Regards
Larry

1. Hi,

16. thanks have emailed today
Regards
Larry

17. Capacitors are used to increase voltage in 66kV substations so how capacitor can step down voltage in this ckt?

1. Different configurations produce different results, the above is designed for decreasing current.

18. hi sir you r doing great job
sir pls tell me the circuit to drive 1w 30ps poweR led. like china led driver.
my email vaibhao060878@gmail.com. thanks

1. Hi Vaibhav,

read the following article and make the circuit according to the explained principle:

19. please removed all wrong diagram and only put working one

20. Hii Swagatam,

What will Happen if I apply a load more than 500ma.. I mean if i connect 2 LEDs of 300ma in parallel.. Will the cap blow up or whether it will just provide a constant 500mA irrespective of the LOAD current required. Also the 2.2uF capacitor is got is huge..Its kind of used in FANS... So is there anyother type of capacitor that can be used..smaller one in size

Regards

Nikhil

1. Hi Nikhil,

The capacitor will not blow even if the output is short circuited...you can put any desired load, more leds will make the illuminations weaker.

you may use many 105/400v in parallel....

21. Hii Swagatam,
I have designed the above circuit with only one 2uF cap at the supply in. Also I have applied a load of 300mA LED at the output with a bit of modification. The modifications are as follow.
1. I placed the output cap between Vout and DC gnd rather than Vout and neutral.

When I checked at the output I get approx. 19. I applied a 300mA led in series with a resistor of 200ohms 1W.

When i powered it on things where working fine...I within a min or so i found resistor R2 burning... Can I know the reason. Also my LED's are glowing perfectly well... I am loving things circuit but that resistor is my concern. Also U have made us buy a 15V 1W zener but the resistor in series with them is just 1/4 watt... How does that Traic ckt works can you explain pl.

Regards and Thanks

Nikhil

1. Hi Nikhil,

There seems to be a mistake in the diagram, the output cap should be across the bridge outputs, I'll correct it soon.

The triac is basically working as an enhanced powerful zener diode, because an ordinary zener diode might be susceptible to surge currents just like an LED.

Actually the output without load should be around 15V (equal to the zener value), if it's not then you may try reducing R3 until this is reached.

Once the output load gets adjusted to around 15V, you may add three 300ma LEDs in series, this will stop the resistor from becoming hot.

For one 300 ma LED reduce the zener value to 4.7V

You may use a 1/4 watt zener diode if you want, it would still work, 1 watt is refereed for extra safety.

2. But Swagatam,

Why is R2 getting burned off..... and I am ok with the Vout as I can connect resistors in series for voltage drops. But why are these resistors getting heated up, when my LEDs consume all the current from the cap. As i have mentioned I am using a LED of 320mA. So when that is connected and the cap passing only 100ma, then practically almost nil current should flow through zener path, so what current is actually flowing through it.

Also the other end of R2 should terminated at neutral or across the gnd terminal of bridge.

Can U explain or elaborate the Traic part if you can

Thanks and Regards

Nikhil

3. I have given you the solution in the previous comment, reduce the voltage to 3 or 4V at the the output for a single LED, then the resistor won't burn.

check the voltage at the output without connecting any load, then multiply it with the output current, that would be the total wattage of the output, make the resistor higher than this value.

4. Ok.. will try to reduce the voltage.. But the resistor which is connected in series with LEDs doesnot burn off as I have already taken that precaution. I have made a series configuration of Resistors with 1/2W to equally distribute the wattage. To be frank I am planing to connect 5 LEDs in series. SO the voltage across load will be taken care off.

My Concern:

I found the resistor (R2) in series with the zener to be burned off. Is it that the amount of current flowing through the load will equal to the current flowing through the zener part. Because If that resistor burns off and the path of zener gets open, I will surely lose all my LEDs.

Hope I am not asking repeated question....

Regards

Nikhil

5. Try reducing R2 value, I think lower value will switch the triac hard ON and will restrict high voltage from generating across R2 and will stop heating it up.

Because making the triac conduct is important. As long as the triac is conducting the voltage would be maintained at the required correct levels.

I am interested to know the voltage at the output without any load connected, if its around 15V would confirm the triac switching.

Regards.

6. I think it should be R3 which should heat up, and not R2.....

If it's R2 that's getting hot, simply remove it.

22. Hii Swagatam...

Had a blast today.... Actually when i tried to power on the supply, things were good. When I measured the voltage at the output,It was around 39V without load and suddenly boom.. The cap with 16V blew up...hahahaha.. A minor shock hit me but i am ok.. Will work on it again..I think all resistors and caps have burnt out. U need to check on the Traic ckt.. Calculations need to be reworked...

Regards

Nikhil

1. Hi Nikhil,

I'll check the circuit myself soon, because I think if the triac conducts it should restrict the voltage to the zener voltage value..

I'll report back soon.

23. Hi Swagatam,

Actually i wanted to power on my basic circuit with help of battery. So do i need any kind of protection for the same i.e. spikes or some other kind of noises. Can you guide me on the same.

Regards
Nikhil

1. Hi Nikhil,

If you can provide some more details about your circuit design I would be be able to suggest you better.

24. Basically I am trying to build a 12V/1A output using l4971 IC. I need to give the input through the battery 48V. SO i just need to check whether any surge or any noise can affect the IC l4971 during start up time due to which IC can get damaged.

Regards
Nikhil

1. If your regulator is correctly built you wouldn't require any surge protection. Also since the input voltage is not too high, and it's from a constant battery DC source the dangers are minimal.

25. mr swagatam, rely appreciate the awesome circuit diagram, the transformeless circuit can it be used to complemented the generator booster output circuit. and pls am looking for inductive ac booster circuit to boost 2.2kva to 5kva

1. thanks anonymous, the above circuit can be used for dropping mains voltage, not as a booster.
moreover no booster can boost VA or watts, it can boost only voltage, sacrificing equivalent current at the output.

26. can i use 5 nos of 1w led with 300ma to 320ma current ?

1. yes connect them in series with a calculated resistor

27. possible to 12v,1amp transformerless circuit ?
and
how to increase current in this circuit ?

1. use 12v zener, increase the number of caps at the input.

2. thank you sir

28. with 30v/1w zener diode
10 capacitor in parallel
100uf 63v capacitor

nine 1w led in series with 320ma current module in output

can other changing in this circuit ?
I am new in electronics

1. try the following circuit, but proceed gradually by increasing the number of caps one by one and by subsequently checking the results:

29. sir,

can i increase the voltage up to 36v with 500mA changing Z1 to 36V

Thanks

1. yes it can be done.

30. Sir ,

This Circuit is for AC 240 or 110?

1. it will work for both the supplies.

2. 1)I tried with polarity cap for C1 to C5 and they explode, is that they should be non-polarity?
2)beside gate pin, is there any polarity concern for the triac?

3. yes C1 must be a non polar cap.

yes all the triac pins are specific, with the triac held straight, the center pin will go to the R1, left to ground, and the right is the gate.

31. HI Swagatam,
Greetings !
I have tested the above circuit with single 20W Warm white LED and continuously ran for 72 hrs. Everything ran smooth. No heat-up. Tuned the zenner to 37.2 V (6 x 6.2/1W). Used 10uF paper type non-electrolytic. Used another NTC instead of series resistance. Wishing to make 30W driver if I get time.

Also following your other previous design (without Triac), I made 6x1W LED driver with 3uF polyester capacitor, 44 Ohm/2 W series resistance, 4x6.2V/1W zenner. It is also running quite smooth for last 24 hours. Still now no LED damage.Will think to put NTC if any LED breaks down.

Thanks for your effort. Infact, the concepts are having capability for patent. Be careful, somebody can take the your credentials.

1. Thanks friend!

It feels very satisfying when I see readers succeed with my projects!

Yes, I'll be careful about somebody stealing my credentials, I appreciate your suggestion very much!

Many thanks!

2. Let me share some info which may help others too. I was browsing a lot for high power LED in Ebay. Those, who implemented this kind of circuits can understand the essence and limitations and they already know a low current is more desired. But low current (<350mA) can not drive "big" LED (>10W) except a series of small LED (1W) are connected. A good solution is to use LED COB. There are many cheap LED COB (Circuit On Board, infact all are COB but still they are coming with that name only) with circular shape are available. Interestingly those are low cost than others ( 5W=1\$, 7W=1.7\$) and the voltage range increases controlling the current at 300-350mA and a 3uF-6uF will be enough to drive the 7W or 7+7W or more power.

And again, Its a tremendously nice job Swagatam and as I said, although protecting IPR is bit costly, but still you must keep it in mind, talk for commercialization and suspend/delete this whole page.

CM

3. Inteersting Info CM! I am sure the visitors would greatly benefit from this.

Actually I am already committed to sharing my exclusive ideas with the readers, if you think you have a better plans for me where you could use my innovations for generating some sort of income for both of us, then who knows I may slowly wind up here and start concentrating on the new job;)

Anyway, thanks a lot for showing your kind concerns! appreciate it.

4. Swagatam,
1. Can we place another current driver after voltage regulation for over precaution and maximum brightness of LED? May be we can proceed with LM317.
-CM

5. Yes it may be tried for further reinforcement, with LM317 in its current regulation mode as discussed here:

32. hello Swagatam,
can i use this circuit to run a 12V 0.2A dc cooling fan continuously ? if not why? what changes done with circuit, or what V/A rated fan i must use?

1. hello max,

yes i think you can give it a try. start with a 2uF/400v cap iniially for C1

33. Sir,
Are you test it.....

Because R2 is getting very hot.......

I am used in R2 1k 1 Watt......

PLZ solve my problem if you have any solution.,......

1. please don't become restless....the site is all about learning through discussions, experimentation, calculation etc.

Try using a higher value for R2, try a 10k, 1 watt resistor.

34. After running the 20W LED with 10uF for more than a month, it fuses suddenly. First LED fused, then C6 bursts (33uF/63V).

The reason I am suspecting:
I was driving another 12V/0.056A fan for cooling from same 36V O/P with resistance driver.
1. First resistance (1W) driver fails, which fails the fan to run.
2. LED get hot and burnt.
3. Peak voltage increased across C6 and it burnt.
Another possibility of damage: The power extraction may be too high and those Chinese LED will be certified someday for its voltage and current :)

Next modification I am thinking:
1. to use 30W LED to get 20W power. You will get 30% margin of protection.
2. Stable circuit for fan. 36->12V may be problematic with traditional LDO regulator. How about capacitor divider before rectifier?

BTW, my 3.3uF, 6W (6x1W) is running since last three months 24x7.

-CM

1. Was your 20 watt LED rated at 36V? In that case the fan wouldn't have done anything wrong, however if your LED was rated for a lower voltage then definitely the application was incorrect.

I think before connecting any LED it must be ensured first that the output voltage exactly matches with the LEd forward voltage.

35. why you have taken zener sensing voltage from dc and grounded it to AC line (through resistors)??

1. R2 is not crucial, it can be removed, the zener ensures that the triac conducts only after the desired zener voltage becomes available at the output, and thus keeps this voltage constant at the output.

36. sir please can i replace 105/400vpp in place of 2uf/400v can i also use this circuit to power my remote circuit which need 5v supply?

1. yes you can replace 1uF with 2uF, but this circuit is not recommended for delicate electronic circuits since it's not isolated from mains.

37. sir it 105 and 1uf the same if yes then it can be use in this circuit Simple Pure Sine Wave Inverter Circuit - 500 Watt Pure Sine sir can 105/400vpp use instead of 2uf or two 1uf? thank you.

1. yes that's correct, you can use two 105 caps, although it will become very bulky, it will work

38. what can i do for about 50v output..
and how can i increse the amps about 1amp..plz explain in proper manner so i can understand it,i am new in electronics..

1. use a 50v zener at the triac gate, and increase the no. of capacitors until you get the required 1amp from the circuit.

2. thanks, i will try it. :)

39. I need to change this circuit into 30v and 18 volt. please tell me how can me made?

1. use a 30V or a 18v zener for Z1, these will produce the specified outputs respectively.

40. I am using this circuit for a 12v 20 watt led, but it is not glowing completely as compared to 12v battery.the output of the circuit is 12v and 0.54Amps.10 watt led is working properly
what can i do for about 30 watt Led,can i use this circuit for 20 and 30 watt Led after increasing the amps upto 1amp?

1. I wouldn't recommend the above design for operating a 20 watt or higher LEds, it's better to go for an 12V/ 2amp SMPS readymade unit, which would provide a safe, reliable and permanent solution.......... and also will be free from lethal mains shocks.

41. i want to made this circuit voltage output of 30 voltge for glow 10 LED in the series. please tell me i need to change here capacitor or other component.

42. Hey, I tried and made the same circuit mentioned above, but the problem is that the output its giving is above 250v and I need it almost 50v. How can I get upto 50v??
Can I use 78xx series for the output?

1. which zener diode did you use in the circuit?

43. Sir,
You warned me not to use capacitors in parallel in the power supply discussed b4 since it causes high surge currents. Can i use this schematic ?
Or the one in the below link with one capacitor being replaced by parallel config

http://2.bp.blogspot.com/-z8OiKZpXIvg/UY3rG8dkgMI/AAAAAAAAEHg/9YMwKeEFwPw/s1600/variable+transformerless+power+supply+circuit.png

1. yes it will work in such circuits but it's always good to use smps designs for getting the correct and efficient results

44. In the intention of getting 12 V output, i have connected a 12V 1 W zener. But the problem is i am getting high voltages ( say 40 V ) even in single capacitor config. What 2 do ?
On checking your High current transformerless surge currentlimitted power supply ckt similar to thz, i didn't get any output

45. The fig. 4 which i am not getting output is

1. connect two 1N4007 diodes in series with the SCR anode and check whether it works or not.

46. Sir i have used BT136 instead of SCR. What 2 do now to make that ckt functional ?
And what abt this ckt, regarding the high voltage (40 V ) prolblem ???

1. try a 6V zener and check the results.

47. Sir, could u plz suggest me
4 Hrs ON 20 HRS Off timer ckt with caliberations which can be understood easily ( 4017 or 555 )

1. Arun, 555 and 4017 cannot be used for getting 20 hours delay, it will give inaccurate results

48. Sir i didn't get what u were suggested last time. Could u pls help me in modifying the current circuit by varying DC output using TRAIC BT136 with 24 V zener used
U have already posted based on SCR ( which is having varying 0-24 V DC ). But i need a one based on TRAIC.
If we can modify thz like that, will we get high current ouputs( say upto 500 mA ) like in the current ckt ?

1. Arun, triacs don't respond well to DC variable inputs so it's better to use a transistor instead as shown in the following article:

Just modify a few things in this design:

Provide the input voltage through capacitors
replace D2 with a bridge rectifier.
Make R1 = 10k, 2 watt
and R2 = 10k pot
The zener diode may be completely eliminated.

49. Sir,
Could u plz tell me what are the lower cut off voltagr and higher cut off voltage of a portable emergency battery showing 4.17 V

1. High = 4.66V
Low = 3.66V

50. SIR I HAVE SENT U A CKT. 2 hitman2008@live.in
Could u plz check it out ? I have some doubts

51. I have replied to ur mail. Could u plz check it out as earlier as possible ????

1. I have reverted....

52. Since the power supply is not able to draw above 80mA from the 1.5 uf capacitor and higher value capacitors causing high surge current problems, i have
designed the circuit in such a manner that No two relays activates simultaneously during charging time to avail sufficient charging current to the battery, after charging no current is required to battery so the availablr current is used to activate the two relays effectively. Similarly during discharging time, both relays are idle due to the absent of power supply therby saving the current from battery.
THEREBY I HAVE AVOIDED THE LOW BATTERY CUT OFF BY OPAMP STAGE.

53. After checking my circuut i found the following current consumption

For parallel config. Of batteries ( 4.5 V )
1). Current taken from the source to battery for charging 39 mA
2). Current drawn from the battery during discharging is 80 mA (little brightness)

For series config. ( 8.8 V )
1). 39 mA
2). 160 mA ( more brightness )

Since in both cases charging current is same, i chose the series config. because i need more brightness.
Is my selection fruitfull Sir ?

Also i could understand that a single power supply will not deliver sufficent current during charging phase ( 2* 30 mA of a relay + 39 mA for charging the battery + 7 mA for operating the circuit components = 106 mA whereas the 1.5 uF alone can only draw a maximum of 85 mA ).

So could you please tell me whether it is possible to use any combination of two 12 V 1 W zener (parallel or series) could be used in association with two parallel 1.5 uF capacitors in a single power supply. Surge current to be avoided by a MOV and a Surge limiting resistance ???????????????

1. Arun,
With a low voltage DC source such as a battery, series or parallel connection both will give same consumption and brightness for the LEDs if configured correctly, I dont understand how it is differing in your case...strange.

I could not understand this question:

Also i could understand that a single power supply will not deliver sufficent current during charging phase ( 2* 30 mA of a relay + 39 mA for charging the battery + 7 mA for operating the circuit components = 106 mA whereas the 1.5 uF alone can only draw a maximum of 85 mA ).

If you use zener diodes which regulate voltage by sinking current with a capacitive power supply, it will cause it too become highly inefficient and will drop its current greatly

I think you should use a cell phone charger as the input instead of a capacitive power supply.

54. Sir would you mind checking your hitman2008 inbox again ?

THEN AFTER THAT
:-
If i am getting that much current (160 mA ) from that config.,
will it be possible to add several capacitors in parallel with each additional zener diode taken place in the circuit along with a MOV and safety resistance ( for surge protection ) deliver higher value currents ??????

1. Arun with 8V you eed to connect the parallel LEDs in strings of two and possibly add a 50 ohm resistor i series with each string.

if you add a zener diode ,the current may not increase as per the requirements and calculations.

with low voltage loads a capacitive powers supply will not work correctly ad efficiently.

55. Good morning Sir.
Is it possible to take current flowing to the battery as reference to make charging cut off at fully charged condition ???
Means, considering the current flowing to to the battery from source during charging, and select a threshold voltage level at which the current bcomes zero ( no more current flows to the battery ) as the cut off point.

But when i am noticing the charging of my two series connected 4.66 V batteries, i have found that, initially the charging current rises to a high level ( say 130 mA ) and slowly it is getting reduced . At last it bounces back and forth between some current level ( say 47.2 to 47.6 mA ) it is not becoming zero.

1. Hi Arun,

the voltage that is been fixed is referred as the reference voltage and the sensing voltage needs to be taken from the battery.

add a feedback resistor 100k from the output of the opamp to pin3 for restricting the fluctuations at the thresholds.
I have covered all these comprehensively in all the 741 battery charger circuits in my blog.

56. Sir.
You missunderstood my intention.
I mean ;
Eventhough the sensing voltage is to be taken for the opamp stage , it will get affected by the reference 12 V supply when the battery is under charging and a resulting voltage appears as the voltage for the opamp. Am i right sir?
If yes why couldn't it be possible to feed this resulting voltage to the opamp for sensing ???
In my design, i have chosen the maximum power saving mode for the battery.
If i am feeding the battery voltage as the sensing one, the opamp stage will function always even during discharging period ( which is not recommended ) drawing a current of 9 mA from the battery.
But in my design, i have selected the 12 V rail as the sensing voltage itself since a resulting voltage between this and the battery voltage appears during charging, and this resulting voltage is getting compared till certain cut off. Thereby by my design, no unnecessary current drawings are there from the battery during power failure.

You didn't reply to ma last question.
CAN I MAKE THE CUT OFF POINT BY CONSIDERING CURRENT FLOW TO THE BATTERY DURING CHARGING ( selecting a cut off voltage at the point when the current flown to the battery has been completely stopped ) ??????

1. Arun, Your circuit is too lengthy, and due to lack of time I couldn't go through it all at once....if possible send it in parts and put the relevant questions there.... I'll try to check them out.

for current sense triggering you may refer to this post

57. Sir, after a long time i came back again. This time seeking your valuable help in one of my previous project..... I have sent it to hitman2008....
Plz check it and reply me soon sir

58. That memory element shown in the diagram is intented to inform us the current status of the overhead tank.The red LED associated with that portion will be turned on once that tank is filled and retains that ON condition for a long time. It willn't be affected by any power failures. So we can get know whether the tank had been filled or not at any time..
It was working perfectly at first weak. But after that, it is not getting turned on automatically when water overfowing detected by the Auto off trigger. But when a power failure occurs after the overflowing detected and current returns, it is getting turned on. I think sufficient voltage is not reaching the sensing node of the memory cell which may be causing this problem... So by the by what z ur opinion abt thz sir ??

1. Arun, to save time and confusions, send the diagram in parts, i'll try to solve them stepwise

59. Sir, i have to ask you one doubt. This is not related to the current problem. My question is : Any other means exist to step up a 120 V AC voltage to 230 V AC instead of a step up transformer ?? ( like a popup capacitor in a transformerless power supply )

1. Arun, yes you can do it using charge pump circuits, but the current will not be sufficient.

http://en.wikipedia.org/wiki/Voltage_doubler

60. Sir Arun again.
All the problems of my circuit have been solved myself.
Now i am in the middle of constructing a wireless audio system.
My intention is stated below:
A blower ( instead of fan ) is placed nearer to TV both sits in the hall of my house. So when i am turning on TV fot entertainment, it would be necessary to increase the volume of TV to overcome the sound from the blower. So i have decided to make a wireless audio control system which is intented to place at a comfortable distance from me and help me in hearing the exact sound as heared from the TV.
It is supposed to take the audo from the audio output jack ( To VCR ) of the TV and let it be processed and transmitted to air by a homemade FM transmitter. Then a FM receiver kept at the other place captures the aired signals and reproduces the original sound signals. I want to include a jack for headset also so that i could enjoy TV at night without disturbing others......
So could you plz help me in starting this project.... I am exopecting your help throughout this project

1. Congrats on that Arun!

As for the wireless transmitter, I already have one such design in my blog, I have tested it perfectly, it'll give a distortion-free reception in your cell phone FM radio, you may refer to the following diagram:

61. The cost of PP capacitor is high here in Nepal so using 5 of them is way costly, can you propose another circuit with at least 500mA current?

1. A capacitive power supply will require many PP capacitors without it you can make a 500mA unit, the best thing would be to buy a cheap reliable SMPS circuit from the market costing just Rs.100/-

62. I have successfully made some LED project from you blog Thanks very much for that. but, in this project though I have not made any mistake arranged according to the diagram checked several times but getting 95 volts DC at C6 terminal output, for a single 225k PF and 15v Zenor.

I used 1pcs 225K capacitor 2 run 1 watt LED 4 pcs serially but flickering also producing audible flickering noise replaced 100 mfd by 475mfd 63v Polarized capacitor visual flickering diminished but faint audible noise still remain. Is BT136 producing sound or capacitor no heat at no components Triac, R2 R3 and 15v 1/2 watt Zenor diode after 8 hours running.

But consuming 156-160 miliamp at 230v AC Adding more 225k 250v PF producing more flickering audible noise as well as current and intensity of led and current consummation.

I have tested this project in Breadboard as well as Vero board found same result. can you explain How can be fixed?

1. Thanks Prosenjit,

Change C6 to 10uF/400V and check, it's the capacitor that could be causing the flickering noise.

I'll update a new circuit soon in the above article, you can try that if you are having problems with the above.

63. I have tried using 475uF 40v Cap in C6 Flickering lowered but a audible sound still there.

Also Please check why Consuming 157miliAmp at 230v AC that is around 36 watt in a Good quality AC Ameter aw well as in mulitmeter selecting AC amp.

I am trying to build a led Lamp for my Planted aquarium with 1 watt or 3 watt LED in RED, BLUE, WHITE Color because I cannot afford imported High lumen High power LED amp.

1. please try the last circuit shown in the following article, you can expect a much better response from it than the above design according to me, you can try it out.

the high consumption in the above design could be because of the triac conduction which shorts the entire capacitor voltage to ground while regulating the output

64. What will happen if I lower its resistance R3 and or R2

I have lowered R3 to 470 ohm still working....

1. If you lower R2 too much the triac will stop working.

If R3 lowered too much, the zener and the triac will burn.

Actually I want to control brightness of LED by using a Potentiometer

Please tell me where to incorporate a POT safely... in this circuit.

1. you can add the left side circuit given in the following article:

connect the LEDs between the transistor emitter and the negative of the supply.

66. sir how to increase current at output upto 5 amp? Is this ckt is very safe & easy as use transformer for drive LED? can we use "Universal High Watt LED Current Limiter Circuit - Constant Current Circuit for LEDs" circuit as in your blog on right side of this ckt for constant current generator for LED?

1. Ashok, capacitive power supplies are never safe or efficient for low voltage DC applications.

You must use an SMPS for your requirement as explained by me in your other comment.

67. Why did Triac use for? pls explain...

1. for controlling voltage and surge current

68. Can i use 22V, 1W Zener & TRIAC BT134 ?

69. Or could you just design a power supply than can produce (3) +24v leads and (3)+24vgnd with a transformer.... something that can push about 10 amps i already have a center top 120-12/12 at 5amps transformer

1. sorry, it doesn't look feasible to me...

70. so your saying there is no way to build a power supply like that, or should I just assume that they just connected all +24v and 24vgnd's together for the sake of the type of connection used?

1. yes you are right, just terminate the outputs separately through different wires, that's all....at the most you could use separate diodes for the positives in order to create some sort of isolation.

71. Dear Sir how can i convert 62volt .30mA ( this is telephone line dc supply ) to 12 volt & .750 mA current please give circuit s about this ,thanks

1. Dear Bijendra, Volts can be reduced using a transistorized circuit but amps cannot be increased...

72. Sir,
According to Second schematic LED blown out First switch on 2 times tested.
tested using single 225k(2.2uf) Cap and 36v and 12v zenor alternating result same used 50k pot instead 10k.
LED not lighted completely just a flash and gone but not all 150mA LEDs used.

Let me know how to overcome this problem?

1st Circuit is tested okay really protecting from surge created by Capacitor by delay effect used 470uf Cap.

1. Bubai, you shouldn't have connected the LEd directly at the output, you should have first confirmed the voltage by adjusting the pot, and once it was adjusted to the LED level, only then the LEd included.

2. ...in the second circuit please connect a 0.1/400V cap parallel to the triac MT1, MT2 otherwise the triac will fail to sustain its conduction.

73. Dear Respected Sir,
Can I use BC546 to run 25pcs 150mA LED and 2 pcs 39v Zenor Diode serially in the Second like circuit? What is the use of 1uf ceramic Cap at output? what is BJT stage?

Could you tell me which Transformerless simple circuit is suitable to run 25 pcs of 150mA LED serially/paralleled initial switch on surge protected. I am unable to find suitable circuit in your oceanic blog. I have tried NTC 10D ( 10hom) before bridge rectifier but it is not feeling much warmth NTC after immediate disconnection, post Bridge rectifier serially seems to have not effect of NTC.I used a 225k(2.2uf) Cap and 10uf 250v cap as filter.

1. Dear Bubai,

Don't use any of the above circuit, instead you can a simple design as shown here

....use the second circuit, connect all the 25 LEDs in series at the output of the circuit...and and use a coil in place of the NTC

make this coil by winding about 300 turns of 30SWG enameled copper wire on any ferrite core, or you can use the coil from any old relay for the same but the relay coil resistance should not be more than 100 ohms for good brightness.

74. how to increase current in this circuit?

1. remove triac stage and use a fan dimmer circuit in series...then increase the capacitor value as per the required current

2. i want to run 8 mm led total 50 nos. in series with transformer less smps.
led vf = 2.2 min & 2.5 max. @ 25 ma per led.
But i need stable output with stable in ac voltage spikes & fluctuation.
i used MPP type capacitor but not satisfied due to break di-electric after some time .
no problem with use of triac or any part in with ac line. But ckt must be stable .

3. a god quality capacitor will never break....use a god quality capacitor as shown below:

http://www.xenonflashtubes.com/ebay/400v105.jpg

for voltage regulation you could try a triac dimmer in series with the capacitor input supply as shown here:

4. Respected sir
i read your artical dimmer in series with capacitor.But it is not suite for me.
pl inform can i run 8mm*60 led ( vf= 2.2 v Toal string current is 30 ma) with update august 22,2014 ckt ? may i use single 1uf/400v instead of 5 ?

5. Nitesh, the above circuit may not be suitable either for your application, you can try the following one:

75. Kudos to your work Swagatam ! I'm looking for deriving 12v dc to operate a relay in a wireless switch rated around 150 ma. Also, want to further step down 12v dc to 3.3v dc (precise) for the sensitive wireless circuit to work. Current required is around 500ma here. Is it safe to use a regulator IC like LM1117 to buck 12v to 3.3v in this non-isolated power supply circuit, without blowing the IC off ? Point to be noted here is that, the load is not always present as its a switch, in contrast to the LED load discussions above which is always present. Total load current will vary from 50ma to 650ma peak, which should be managable as per discussions above. Your expert guidance awaited .....

1. Thanks Pushkar, I don't think the above circuit wold be suitable for anything other than LEDs, because a mains shock isolation may not be necessary for LED applications and also LEDs do not require highly sophisticated DC for operating.

Therefore I would rather recommend an SMPS version for your unit and not the above design.

2. ...you may try the following one instead

76. That was a pretty quick response, i wonder how you must be coping up with tons of queries you might be getting in various threads you've started :)
Now, I agree with your response, and was of the same opinion to start with. Well my aim was to try and see if size and weight can be reduced, hence started exploring this alternative. SMPS uses a transformer as well, so i was thinking of going for regular mains transformer 12v 1Amp, with a LM1117 ahead to get the regulated 3.3v as desired. I do understand that SMPS is better, but seems to me to be a overkill for the application. Let me know if you think otherwise. Thanks a ton again !

1. thanks pushkar,
SMPS concepts utilize a small ferrite transformer and the whole unit could be five times lighter than an identically rated iron transformer. Moreover these are cheaper and much long lasting than their transformer counterparts.

You can buy a 12V 1amp adapter from the market if you find it difficult to build yourself.

LM1117 may not be necessary, just use a 7805 IC to drop the 12V to 5V and then followed by three of 1N4007 diodes to further drop it to 3.3V approximately

77. Hello Sir can i use this circuit for microcontroller based project , i am in search for Mc based fan regulator transformerless supply , will u plz tell me how can i take output for zero crossing detector through it , plz reply soon, Thanks

1. Hello Tariq, the above circuit does not include a zero crossing detector so I am afraid it might not be suitable for an MCU,

by the way I have a PWM transformerless fan regulator circuit in this site but without using a MCU, you can see it below

78. Good day Swagatam,
I enjoy electronic circuits. This is my first visit to this site and already burning candles reading articles. There are numerous circuits and versioning for transformerless power supply from you and others in this website. Can you recommend me a stable circuit you or others designed and the picture of components (because when I buy the components, they come in funny shapes), non surge around 12V (or selectable) and update here or can email me at june1629@yahoo.com.

Thank you.

1. Thanks so much june1629,

if you are looking for a good surge free transformerless power supply then probabaly you would want to try the following design:

It's much sophisticated and reliable than the previous ones, but kindly be informed it has not yet been tested by me practically. though I strongly believe that it's technically sound and without any errors.

79. Hello Swagatam...
first i want to thank you for sharing your knowledge with us....
my friend can you propose me a transformer-less power supply that can operate with 230v. i hope to use 1W x 3 or 1W x 4 LEDs. i think the LED need 350mA for higher brightness. please propose me a power supply that can use very long time without problems...

best regards...

1. Thank you madusanka, for any high watt LED it's always recommended to use an SMPS adapter, however you can also try the following circuit and see how well it performs:

use 3 nos of 105/40v capacitors in parallel for the input capacitor in order to get at least 200mA current at the output

80. Hello Swagatam,

Thank you for publishing your design.
I made a PCB design of it with almost identical parts you suggested, but I cannot get it working. Could you please look at my design for what mistake I obviously made?

1. Thanks very much Patrick, I'll update the info in the above article soon...

2. In PCB by mistake Triac gate is connected with BC547( Emitter ) instead of Collector that is why circuit is not working, if I am not wrong. rotate NPN transistor 180 degree in PCB design software hope it will work.

3. OK thanks, I'll check it and correct it soon.

4. Hi,

I change the transistor and the overall layout. Could you guys please take a look if this looks oke?

Thank again!

5. Again I build up this PCB but it just doesn't work. Can anybody take a look at my design and maybe tell me what I am doing wrong?

81. Capacitor increase so current increase i am right?????? Explain triac and zener function in 1 st circuit

1. yes that's correct

2. the triac grounds the excess voltage that surpasses the selected zener voltage level and thus maintains a constant voltage at the o/p

82. Do you think i can use this circuit for the 100W LED's? Because transformer 24V 3 amp they dont have here :(

1. No this circuit is not recommended for a 100 watt LED module.

2. ok, thanks, i am going to try to lower the winding's from the coil.

83. Dear Mr Majumdar,
Your electronic skill is superior. Thanks for showing everything with diagram.etc. My question is:
Do you have a schematic diagram of a transformerless power supply 220v AC in. and 3.3v 400mA DC out?

1. Thanks Christer,

May I know for what application you are planning to use it?

2. Thank you for responding so fast. It is a project for school. Burglar alarm, and I'm using Arduino.
I need to supply with 3.3 v.

Thanks again

3. OK, in that case the following SMPS design could be recommended:

the zener at the output can be adjusted to achieve the required 3.3V instead of the shown 5V

a capacitive transformerless power supply could be dangerous, and therefore may not be recommended for your application

4. ...By the way you can simply use you mobile charger for the purpose by adding a couple of 1N4007 diodes in series with the positive to get the same...

84. dear Swagatam,
it's good AC-DC converter. Can you explain more details transistor, triac stage?
when circuit work normally, current through zener, 100k resistor and 100k variable resistor about 10mA, voltage at base of NPN transistor about 1V cause open transistor. Therefore, voltage at G gate of triac about 0V, triac off => it's not logic, am i wrong??
thank you.

1. dear unknown, actually there should be a 1K 2 watt resistor across the triac MT1/MT2....so when power is switched ON, the circuit will switch ON with a low current via the 1K, and switch ON the triac for a full conduction with the rated current of the capacitor....however as long as 30V is not reached the transistor will not conduct, but when this happens the transistor will conduct and cut off the triac...this sequence will go ON very rapidly at microsecond rate, making sure that the output voltage never exceeds the 30V mark or above the zener voltage mark.

2. dear Swagatam,
Thank you very much, you're so enthusiastically.
I understand now.

3. you are welcome dear Ngan!

85. Can i use the above circuit for building a cell phone charger which will give 5V and 2.2A of current??

1. technically it may be possible, but strictly not recommended because the above design is not isolated from mains and is too crude and dangerous to be used as a cell phone charger

86. Hi Mr. Majumdar,
Can I use this circuit for three phase input supply.
Input voltage will be ~440V, which will be phase to phase.

1. Hi Vikram, it could be possible if all the vital components are rated at 1kv or at least 700V