By Pass Capacitor Issue. Request Help and Solution

[raymondq]

Hi,

I have make a custom board with cPCI interface which have 3.3, 5 +/-12v power supply input

I have used a pair of bypass capacitor(in paralled) for all these power input.

The capacitor value are 10uf, 50v and 0.1uf, 50v.

However, i realised that the -12V supply fluculate (current fluculate) when i power on the board.
I did not use -12V in my design. the -12V are onli connected to the bypass capacitor.

when i remove the bypass capacitor for the -12V supply. the current drawn is stable.

Is this problem caused by insuffienct cap vaule?

Can anyone advise on the vaule of bypass capacitors for all these input supply for cpci design?

 

[rohitkhanna]

It's not clear whether you HAVE an output load on the -12v, or DO NOT HAVE a load.
What current are you measuring if there is no load ??? What type of fluctuations are you seeing ?

Next -
what method/ circuit are you using to generate these voltages ? Are you using standard 3-pin regulators ? Please post complete circuit with test setup so that we can understand & try and point out where the problem could be.

 

[raymondq]

Hi I have attach the block here.
basically my card is power up by a cPCI chassis.
The chassis have a voltage monitoring circuit which will show the status of all the low voltage 3.3 5 -/+12v
The chassis is power up by an external 28v power supply

when i insert my card in the chassis and power up. The -12v led lit up on the chassis. The current level display on the 28v power supply start to fluctuate.

my custom board did have any circuit that used the -12v. only the decouping circuits.

Is the vaule i used for the bypass cap/ decoupling circuit wrong that could have casue this issue?

when remove both the cap for -12v, the issue is sovled.

 

[BradtheRad]

If the problem goes away when you remove a capacitor, then my first thought would be the capacitor has developed low resistance.

This can happen although it doesn't occur often.

I would remove the capacitor and test it. I would connect at least 12 VDC for a long time, and watch whether it allows current through it.

If it lets through more than a few uA, then it is becoming leaky and should be replaced.

 

[rohitkhanna]

If the problem goes away when you remove a capacitor, then my first thought would be the capacitor has developed low resistance.

This can happen although it doesn't occur often.

I would remove the capacitor and test it. I would connect at least 12 VDC for a long time, and watch whether it allows current through it.

If it lets through more than a few uA, then it is becoming leaky and should be replaced.

Exactly brad !!

And to add to that, do remember that electrolytics (your 10uF) have a polarity. With a -12v, you would connect +ve terminal to GROUND

cheers!

 

[godfreyl]

Another possibility is that the regulator or converter becomes unstable and oscillates when there is no load. If it were a leaky cap, I would not expect the current to fluctuate as described.

At this stage we can only guess since we have no idea what the circuit looks like.

 

[raymondq]

Exactly brad !!

And to add to that, do remember that electrolytics (your 10uF) have a polarity. With a -12v, you would connect +ve terminal to GROUND

cheers!

I think you have hit the jackpot .....
mine cap + is connected to the -12v input ....
I tink that is the problem.

Thank you

 

[rohitkhanna]

......cap + is connected to the -12v input ....

Wow.... you're lucky it hasn't blown up yet!