Sparks when 24VDC power is supplied

[helper1234]

Hi All,
I have designed a circuit that can takes in 24AC/DC power supply and lights up an LED.

Below is the layout of the circuit

AC in-
FULL WAVE BRIDGE RECT -----Positive------ 220Micro farad elc cap----- 10V Volt reeg------ 22uF elec cap ----- LED--- resistor --- negative of rectifier
AC in-

The caps an reg are grounded to negative of rectifier.
The input of voltage regulator ranges upto 45 V in.

When I plug in 24AC wires to the AC inputs terminal there is no spark. Works fine. LED lights up.

When I plug in 24DC positive wire to one AC in and then plug in the other negative wire to other AC in. there is a very minute spark for a very very brief moment. Then the circuit works fine.
But I am worried about this spark because in long term this might cause the components on the PCB to go bad.
I tried decreasing the voltage and the spark goes away when my voltage is less than around 15 volts.

Can someone explain why this spark is happening and how do i get rid of it.
At one point the voltage regulator stopped working and I had to replace it.
Did it get damaged due to the continuous spark?

Any help or insight will be highly appreciated

Thanks

 

[kam1787]

Hi All,

When I plug in 24DC positive wire to one AC in and then plug in the other negative wire to other AC in. there is a very minute spark for a very very brief moment. Then the circuit works fine.

I am not sure what you are doing.........are you bypassing the rectifier?

 

[helper1234]

I am not sure what you are doing.........are you bypassing the rectifier?

The two ac lines goes into the rectifier and then. Smoothing capacitor of 220uf smooths out the positive ac with very little ripple. It then passes through the volt reg that gives out 10 volt DC. Then led is and resistor takes in 40ma current.

When DC is plugged it also goes through rectifier. Same process but its a DC so smoothing cap has no use. The DC volt goes thru regulator then to led and res

Ac works fine but for DC when I connect the wires there is an initial brief spark... Why is this happening?

 

[kam1787]

The two ac lines goes into the rectifier and then. Smoothing capacitor of 220uf smooths out the positive ac with very little ripple. It then passes through the volt reg that gives out 10 volt DC. Then led is and resistor takes in 40ma current.

yes I understand that. is the LED rated for 40mA? It can take 40mA?

- - - Updated - - -

When DC is plugged it also goes through rectifier. Same process but its a DC so smoothing cap has no use. The DC volt goes thru regulator then to led and res

Ac works fine but for DC when I connect the wires there is an initial brief spark... Why is this happening?

Ohhhhhhhhhhhhhh ok. Sorry. It is too early in the morning!

Most IC regulators need a capacitor in front - check the datasheet.

It is little odd you are getting a spark, but not so unusual. What regulator is it?

 

[Darkcobra]

Sounds like inrush. The 220uF capacitor starts off with no charge, and so briefly resembles a short as it quickly charges from 0V to 24V, drawing a spark.

The same spark will be visible if you charge a 220uF cap up to 24V, then short its leads together.

With the DC input supply, it almost certainly includes its own capacitor, which can supply enough current to support a visible spark as the charges between the two capacitors equalize.

With the AC input, it won't have that, and so current is limited by the transformer. Plus the voltage at moment of connection may be less than 24V anyway.

In any case, it shouldn't be a problem in your circuit. The bridge rectifier is the most stressed component, but can typically handle such pulses. If you want, you can include a small resistor between the bridge and the 220uF cap to limit the peak current.

 

[kam1787]

Sounds like inrush. The 220uF capacitor starts off with no charge, and so briefly resembles a short as it quickly charges from 0V to 24V, drawing a spark.

I think he omits the 220uF cap in the DC 'version'.

 

[helper1234]

I think he omits the 220uF cap in the DC 'version'.

Yes The Regulator recommends a small filter capacitor in front and a 22uf elec cap at the output.

Output has the 22uF.
At first since there is already 220uF cap i didnt put any filter cap...

then i added one more 0.47uF cap in parallel to the 220 cap to see if that filter cap does something but nthg happened... still there is spark when i use 24 DC.

basically anything above 15DC is giving me spark.

Yes i also think its an initial peak current inrush that cause the spark... But i dont know why nothing is happening at less than 15DC... or at 24AC....
???

 

[kam1787]

then i added one more 0.47uF cap in parallel to the 220 cap to see if that filter cap does something but nthg happened... still there is spark when i use 24 DC.

Nothing that you can see happened, but the cap is needed for stability and proper functioning.

It is likely the current inrush - nothing to worry about. At <15VDC there is not enough voltage to allow enough current for a spark - so I would guess. It can still spark at 24VAC but perhaps it is not as visible. Again, nothing to worry about.

 

[helper1234]

at one point my voltage regulator went bad... could it be because of that constant touching and removal of wire? I was trying to observe peak current voltage and spark on the scope.... May be smthg got shorted out and it wasnt related to circuit... but stilll it iis worrying me...

The wires will be conneced to a switch. It can be a manual switch or a IC controlled switch... If its a IC controlled switch I am worried this might cause the IC to go bad.

 

[kam1787]

at one point my voltage regulator went bad... could it be because of that constant touching and removal of wire?

Yes. Why do you feel the need to keep 'playing' around!

- - - Updated - - -

The wires will be conneced to a switch. It can be a manual switch or a IC controlled switch... If its a IC controlled switch I am worried this might cause the IC to go bad.

Not if the IC was properly wired in - you need to give specifics on which IC and how it will be connected.

 

[helper1234]

Yes. Why do you feel the need to keep 'playing' around!

- - - Updated - - -



Not if the IC was properly wired in - you need to give specifics on which IC and how it will be connected.

It wil be an IC switch controlled by a high and low signal...

and no matter how much i play around the design should not blow up anything. If it s blowing up reguklator means there is a bproblem.
I need to find out because the led light is supposed to blink... continuosly for 1 minute which is turning off and on continuously

 

[Darkcobra]

at one point my voltage regulator went bad... could it be because of that constant touching and removal of wire?

Well maybe, if you have a lot of stray inductance, inrush current sufficient to cause a spark also caused a higher peak voltage. 24V is already close to the max voltage of most common linear regulators.

But that's just speculation. Run down other possibilities too. At 24V in, 10V out, and 40mA, your regulator is dissipating (24-10)*0.04=560mW of heat. Is the package/heatsinking sufficient for that?

 

[helper1234]

Well maybe, if you have a lot of stray inductance, inrush current sufficient to cause a spark also caused a higher peak voltage. 24V is already close to the max voltage of most common linear regulators.

But that's just speculation. Run down other possibilities too. At 24V in, 10V out, and 40mA, your regulator is dissipating (24-10)*0.04=560mW of heat. Is the package/heatsinking sufficient for that?

The voltage regulator is rated for 400mA of current. I am only using 40mA.
Also yes i even have a heat sink attached to the voltage regulator... there is no heat dissipation on the PCB circuit.

I am really confused on why does a spark happen. can some one explain?

 

[kam1787]

It wil be an IC switch controlled by a high and low signal...

YES, we assumed that.

- - - Updated - - -

I
and no matter how much i play around the design should not blow up anything.

Perhaps if I illustrated it using a different world.
You have a car.
The car keeps having problems.
The mechanic asks how you are driving the car.
You say, "I slam the gas pedal down and over-rev the engine, then I slam on the brakes, then I do that again and again and again and again and again and again"
The mechanic says "Do not do that"
You say, "But a car should go and should stop"

- - - Updated - - -

If it s blowing up reguklator means there is a bproblem.
I need to find out because the led light is supposed to blink... continuosly for 1 minute which is turning off and on continuously

You are mixing apples and bananas, or oil and water - 2 very different things and making them the same

Think about it. These are 2 very different conditions.

 

[helper1234]

I get the point that if something faulty is in the circuit then.. doing the faulty thing again and again may damage the components.... but thats the thing i need to figure out whats the faulty thing here. The circuit is supposed to light up with high signal and shut off with low. If too much on and off is going damage components then there is no use of the design because its main functionality has a fault.

 

[Darkcobra]

The voltage regulator is rated for 400mA of current. I am only using 40mA.
Also yes i even have a heat sink attached to the voltage regulator... there is no heat dissipation on the PCB circuit.

I can't tell if you're already aware of this from your wording, but it's important enough to state anyway. Regulators have separate limits for both current and power (heat) dissipation. You cannot exceed either. Linear regulators "burn off" excess voltage as heat, and the larger the voltage drop they must produce, the less current they can produce before they reach the dissipation limit. If you post the part and package you're using, I can verify it's being used within limits for you.

As for the spark:

1) If it is caused by inrush, there really should be no problem, and it should also be able to handle any number of power cycles. Linear regulators are made to handle sudden application of power with no special precautions. The previous failure may have simply been a fluke, or slip of a test lead. If there is also an unusually high inductance somewhere causing a voltage spike exceeding the regulator's max voltage, it can be damped by adding a 4.7 ohm 1/4W resistor before the input capacitor to limit the peak current.

2) If it is not caused by inrush, it could be a faulty part, wiring error, or almost anything - which unfortunately we have little chance of discovering.

 

[Venkadesh_M]

The spark is due to the sudden charging current of filtering capacitor in starting...You have to combine a small 10 to 50 ohm resistor in series with filter capacitor...

and Also it will protect the Diode from damaging due higher surge current.....

 

[helper1234]

I can't tell if you're already aware of this from your wording, but it's important enough to state anyway. Regulators have separate limits for both current and power (heat) dissipation. You cannot exceed either. Linear regulators "burn off" excess voltage as heat, and the larger the voltage drop they must produce, the less current they can produce before they reach the dissipation limit. If you post the part and package you're using, I can verify it's being used within limits for you.

As for the spark:

1) If it is caused by inrush, there really should be no problem, and it should also be able to handle any number of power cycles. Linear regulators are made to handle sudden application of power with no special precautions. The previous failure may have simply been a fluke, or slip of a test lead. If there is also an unusually high inductance somewhere causing a voltage spike exceeding the regulator's max voltage, it can be damped by adding a 4.7 ohm 1/4W resistor before the input capacitor to limit the peak current.

2) If it is not caused by inrush, it could be a faulty part, wiring error, or almost anything - which unfortunately we have little chance of discovering.

Thank you all guys so far for the replies. I highly appreciate your input nd efforts,...

The voltage regulator links is as below
https://www.digikey.com/product-detail/en/IFX25001TS V10/IFX25001TS V10-ND/2270483
Part number : IFX25001TS V10

I am sure its not faulty part or wiring error because i made 5 pCBs.. and each and every one is causing the spark to occur at 24VDC... basically anything higher than 15 DC...
Not all of the parts could be bad.

For,
Inductance, could you give me a hint on where in the circuit could the inductance occur? and how?
I will try out the resistor option tomo.

I have also heard about surge resistors that protects the circuit against such unwanted initial surge of high current? Does anyopne know where i can buy them???

- - - Updated - - -

The spark is due to the sudden charging current of filtering capacitor in starting...You have to combine a small 10 to 50 ohm resistor in series with filter capacitor...

and Also it will protect the Diode from damaging due higher surge current.....

Can you explain how will the 10-50 ohm help the diode. and get rid of the spark?
So you mean th resistor is place after the bridge rectifier and before the capacitor that is connected to the voltage regulator IN.
???

 

[helper1234]

As for the spark:

1) If it is caused by inrush, there really should be no problem, and it should also be able to handle any number of power cycles. Linear regulators are made to handle sudden application of power with no special precautions. The previous failure may have simply been a fluke, or slip of a test lead. If there is also an unusually high inductance somewhere causing a voltage spike exceeding the regulator's max voltage, it can be damped by adding a 4.7 ohm 1/4W resistor before the input capacitor to limit the peak current.

The spark is due to the sudden charging current of filtering capacitor in starting...You have to combine a small 10 to 50 ohm resistor in series with filter capacitor...

and Also it will protect the Diode from damaging due higher surge current.....

Thanks Guys, that was it. Without the resistance the capacitor was trying to get charged instantly so it was taking up huge inrush current. Once the resistor or was put in then the current was limited so it didnt have the huge inrush, thus its working now with no spark.

Thanks everyone who helped out. Seems like any resistance value will work here.
Is there any requirement or formula i need to use to calculate the resistance value needed? that changes with the value of the capcitor?
I am using the 220uF cap so what value will be good?

Thanks again

 

[adi0803199]

The value of resistance will determine the limit of current that can flow through this circuit and thus prevent from sudden inrush of current,on the other hand a high value of resistance will cause a higher delay to charge which should be less than half the frequency of the input signal.This means that the regulator may give an error in 10V dc supply as generally LM series regulators have an error of ±0.2 V.Basically a trade off between the range of frequency of input signal to safety.Hope this helps.