arking, relay sound

[E2005]

Made a circuit board with dual opamp, and two voltage regulators. Has slots for microprocessor eval board to sit in. Before going further was testing the circuit( without putting the microprocessor board). Every time i connect the power there is arking at th contacts( for main power). I am not using relay or any other inductive component. Its not a high current cirucit dont understand the reason?

 

[geekEE]

Do you have large capacitors (or a lot of smaller ones) across the power supplies? Those can draw a lot of current until they charge up. Does the current draw go down to your expected value after a moment?

 

[E2005]

I have 22uF capacitor between supply and ground to prevent high frequency noise from interferening with supply and diodes to prevent buck emf( the system will control a servo with microprocesor). Are they the reason??

 

[geekEE]

Only one 22uF capacitor in a system that goes from mains down to two regulators? That doesn't sound right. Did you design the entire power supply circuit or are you using a purchased power supply? A power supply would certainly have some capacitors in it.

You still didn't say whether the current drops down to a normal steady state value.

What if you remove your microprocessor circuit? Or the regulators? Just pare it down and find out where the current is going.

 

[MacGyverS2000]

Leave it plugged in for a minute or two, unplug and immediately plug it back in. If there's no spark, you have some serious capacitance somewhere...

Dan - Owner

http://www.Hi-TecDesigns.com

 

[Skogsgurra]

Is this a Parallax system?

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[E2005]

Thanks all. Sorry geek2005, the current done's not go down. I found the problem and have fixed it but don't understand it completely.
I read somewhere that two capacitors 220n( i typed it wrong earlier as 22u) and 1u on the power supply removes the high frequency spike noise in the system, so I had put one.
I compared peak to peak mV on the ground plane using tecktronic oscilloscope. Interestingly putting the cap actually increased noise from 10 mV to 18 mV and also caused the spark on the input. Removing the cap has fixed the problem of sparking and reduced the noise in ground.

 

[MacGyverS2000]

You're saying removing a 220nF cap has stopped sparking on the mains line?

<sigh> Unplug what you have and look for someone who knows what they're doing. You're either not providing us with the information we've asked for or you're working with a very dangerous system (mains directly connected to circuit).


Dan - Owner

http://www.Hi-TecDesigns.com

 

[E2005]

Hi macgyvers2000, probably I did not provide sufficient information. The capacitor was 22uF, the sparking was very small but enough to make a very tiny noise. Every time i connected the power it got charged and dischared making that sound. The system is operated by battery( not main supply). When i said main power I ment where the battery connects to the circuit not mains.Sorry about nomenclature

 

[geekEE]

One big capacitor is NOT the same as multiple smaller capacitors. Location of the capacitors and the overall layout of your PCB is going to make a big difference too. Were you bypassing your supply close to the regulators or out at the devices that use those supplies? Also, hopefully you've laid out the PCB so that the return currents don't pass by sensitive analog circuits.

 

[E2005]

Hi GeekEE, i was bypassing it close to the switch where the battery connects to the circuit board. That is the first thing in the circuit then it goes to the regulators etc.Yes my PCB is ladi out to keep the sensitive circuit parts away

 

[geekEE]

And there are no capacitors on the input or output of your regulators? Check the data sheet for your regulators; I'm sure they'll show bypass capacitors. You'll probably need bypass caps at the chips that are fed by the supplies as well. The type and value of the caps will depend greatly on whether the device(s) are digital or analog. Higher inductance, due to long leads or traces, will make caps less effective for bypassing digital chips. Higher ESR (dependent on capacitor type and value) will also affect the effectiveness of the bypass caps, especially in switching regulators (although I'm assuming you're using linear regulators here even though you didn't specify).

So, when you say "I read somewhere that two capacitors 220n and 1u on the power supply removes the high frequency spike noise in the system", it's a rule of thumb that works in some places, but it's really dependent on your particular circuit. If you want better info, it would really help to give us a more complete description.