104nF 50v decoupling capacitor that burns

Hi all, I am a hobbyst, i’ve made a project that uses an ATTINY85, it reads 3 digital inputs (0/1) and triggers 2 relays. It all works very well, but the decoupling capacitor i placed over the ATTINY85 VCC/GND simply burns out has soon has i power the PCB (4.3Vdc after the diode) .

Sure the boards work fine, without this capacitor, but i am puzzled by this burn out, does anyone have any clue or experience with this?

.

This is the bugger:

That is a 100nF capacitor allright. It burns out ? I have never had something like that. Perhaps it was a faulty batch... a very faulty batch, it is allmost too weird to be true. Do you have any other (ceramic) capacitor, perhaps from an old broken electronic device.

The 104 means: a '1', a '0', and 4 extra zeros in pF. That is 100000pF = 100nF

simply burns out

How do you know?

That makes no sense at all.

What are the symptoms of the "burn out"?

That 100nF capacitor is an online photo, it is not your 100nF capacitor.

The only thing that comes to mind is that it was defective. Besides being a really OLD part, it’s huge. You’d be ahead to just go out and buy 100 of the more modern Monolythic Ceramics. They are very inexpensive, 7c each.

http://www.ebay.com/itm/Kemet-C320C104M5U5CA7301-1uF-20-50V-radial-ceramic-capacitor-lot-100-/400676958218?pt=LH_DefaultDomain_0&hash=item5d4a35300a

The only time I have seen something like this happen is when something oscillated and sent too much AC current through the capacitor at a frequency where the ESR is high.

Unlikely in this case, though.

fungus: That makes no sense at all.

What are the symptoms of the "burn out"?

LarryD: How do you know?

It heats up red hot, smokes (ahhh... that unique transistor toast smell... :grin: ), and burns to a crisp, totally black! (i'll take a picture of my next assembly)

The C3 capacitor is exactly the same, a 100nF and it works like a charm on the reset circuit, no heat, no smoke.

I have a bag of 100 of these, so i'll make a sample test on a few, maybe there are some faulty ones in there and maybe i got really lucky on picking them for the decoupling circuit :D

KeithRB: The only time I have seen something like this happen is when something oscillated and sent too much AC current through the capacitor at a frequency where the ESR is high.

Unlikely in this case, though.

I suspect this!

The Attiny85 is clocking at 1Mhz internal. Maybe it's injecting signal on the capacitor, making it toast.

I'm gonna try out running the PCB without the tiny to see what happens on the capacitor... Stay tunned!

Check it with an ohmmeter to make sure it is really a capacitor.

Seconded , there are components that look similar that are actually overload or fuse devices. Forget what they are called.

Many multimeters have capacitance setting as well as resistance - if your has use
both to test your batch of devices - perhaps they are not capacitors, or are rejects.

Did you buy them suspiciously cheap on eBay?

KeithRB: The only time I have seen something like this happen is when something oscillated and sent too much AC current through the capacitor at a frequency where the ESR is high. Unlikely in this case, though.

More than unlikely - it just isn't going to happen because ...

Num3: The Attiny85 is clocking at 1Mhz internal. Maybe it's injecting signal on the capacitor, making it toast.

No. The microcontroller is not capable of generating sufficient power by any means.

The capacitor is faulty. Simple as that. Test it (an ohmmeter to start with), and the rest of the batch.

I was thinking more of a power supply oscillation - still unlikely, but more power available! 8^)

I was thinking more of a power supply oscillation

That's pretty close to the reason you put the 100nF capacitor in there...

I agree... it was a faulty part.