How did I fry my transistor?

I had a weird thing happen just now where I was experimenting with a transistor switch and after a little while it just stopped working entirely. Replacing the transistor fixed it, but I’m curious now what caused it.

I started with a very plain circuit like the first picture (attached to post.) I noticed that this was pretty susceptible to interference from say, touching the contact wires with my finger. Which I guess makes sense, as even a few hundred microamps ought to light up a 1ma LED, and I suppose the 22kΩ path to ground can only source so much current. I guess?

So then I thought maybe a small capacitor could help to prevent this interference, so I tried putting a 0.1uF capacitor in parellel with the 22kΩ resistor to ground, as shown in the second picture. This worked for a little while, but as stated above it suddenly stopped working and after testing each component in the chain the transistor was definitely what died.

So I guess I’m curious, what am I missing here? Did I somehow exceed the breakdown voltage momentarily, destroying the transistor? There was no smoke or sparks or heat that I could observe – just suddenly, nothing.

Help? What is wrong with the second circuit that might cause this?

Did you solder the components? (Too much heat?)

We do like schematics as a minimum, and yours is excellent, but in this case a (perfectly focused, no more than 1024 pixels wide) picture is called for.

The circuit is perfect; clumsiness of some sort is suspected.

Both are fine but the resistor to ground/cap's not required...

Circuit looks good. What transistor are you using?

Weedpharma

Thanks for the feedback. Transistor I am using is a PN2222A. Datasheet here: ON Semiconductor PN2222A Datasheet

Got a whole bunch of them, because at 5¢ per I figured it would be a decent practice transistor. I guess that was a good idea, because I clearly don't understand what I did to kill the first one. :astonished:

Paul__B:
Did you solder the components? (Too much heat?)

No, this is on a breadboard. This is not my first circuit by a long shot but my first time really poking at discrete transistors rather than using ICs. Some details that might be pertinent but that I left out thinking they were not relevant:

  • The supply is actually ~4.6V from a wall wart rated for 1.6A, should be overkill.
  • There is a 2200µF cap across the +/- supply. Again, overkill, but there for later experiments with neopixels
  • The 100nF cap is technically rated as 104Z, so I guess potentially anywhere from 80-180nF.

I will try to take a proper picture, the ones I'm taking now are blurry with my phone cam.

cjdelphi:
Both are fine but the resistor to ground/cap's not required...

Not required but I found the switch leaving the base of the transistor "Floating" left it very susceptible to random noise, flickering the LED on and off just by waving my finger nearby. I also used a 1MΩ at one point, but this basically seemed to act like an antenna. I figured this was a bit like a pull-down resistor, just something to help bring it back to ground when the switch opens? Is this not a good practice?

OK, managed to get a decent shot. Nothing that interesting I think. Rails are ~4.6VDC as I said before.

Maybe it was just a bad transistor…? I guess you get what you pay for sometimes…

Could be you accidentally shorted a point in the cct.
I have never had a new, bad semi conductor in the 45 years.

Except some diodes that were marked with the cathode on the wrong end!

I have never had a new, bad semi conductor in the 45 years.

Ditto for 30 years. Invariably, it turns out to be operator error (accidentally connecting a wire to a wrong pin etc..).

The led current limiting resistor only needs to be 220 ohms.
How did you test the transistor ? (exactly)

Static discharge? Highly unlikely but in the absence of any other explanation.

Static discharge?

2n2222's are not CMOS and therefore not static sensitive. In Southern California , in the summer , during Santa Anna wind conditions, the static is so extreme that you hear people yelling "ouch !" intermittedly all
day at work as they walk from one desk to another over a carpet and touch something at the other end.
It has been said that even TTL chips can be damaged under these conditions but I don't know if that is true.

raschemmel:
2n2222's are not CMOS and therefore not static sensitive.

It may take a strike of lightening to convince you but ALL electronic components are sensitive to Static discharge at some level.

raschemmel:
Ditto for 30 years. Invariably, it turns out to be operator error (accidentally connecting a wire to a wrong pin etc..).

As I stated above, at the time I was poking my finger about (gently) between the base of the transistor and the resistors/cap to see if I could cause interference which I was seeing before. In that general area, as you can see from the picture, there's really not a lot of risk of shorting, but I suppose that's the best explanation there is.

The led current limiting resistor only needs to be 220 ohms.

Aware of this, thanks, but I prefer my LEDs dimmer than most. I find fully lit LEDs a bit hard to look at (and I mean really, if we're being nitpicky, 220 ohms is a tiny bit short since it provides 20.9mA @ 4.6VDC and the LED is rated for 20mA.. ±5% for a carbon film resistor and... ok ok I'm done.. :slight_smile: )

How did you test the transistor ? (exactly)

Two methods, the first being to replace the transistor in the circuit. The circuit started working again. That was a big hint.

Testing now with a multimeter, a working transistor shows:

BE Vf 0.664
BC Vf 0.659

The broken transistor shows:

BE Vf 1 (Can't get a reading)
BC Vf 0.654

So yeah, I guess I will have to settle for "briefly shorted it somehow" as the explanation. No idea how really, but I will just exercise more caution in the future.

Thanks everyone for your help!

KenF:
It may take a strike of lightening to convince you but ALL electronic components are sensitive to Static discharge at some level.

I've got a few boxes of valves, and I am fairly certain it would take an actual lightning strike to fry one.

ChilliTronix:
I've got a few boxes of valves, and I am fairly certain it would take an actual lightning strike to fry one.

Some years ago I threw out about half a dozen tea chests full of valves believing I'd never use them again. Literally just a few weeks later I realised I'd thrown away an absolute gold mine. They were all in full working condition and over half of them were still in their original packing. :frowning:

jotapeh:
The broken transistor shows:
BE Vf 1 (Can't get a reading)
BC Vf 0.654

Open base-emitter junction.

KenF:
Some years ago I threw out about half a dozen tea chests full of valves believing I'd never use them again. Literally just a few weeks later I realised I'd thrown away an absolute gold mine. They were all in full working condition and over half of them were still in their original packing. :frowning:

I'd give you a slap for that if you lived close by.. I don't so I hope you have suitably chastised yourself! The least you could have done is stick them on ebay!

It may take a strike of lightening to convince you

Oh come on. When someone talks about "static discharge" they are not talking about lightning . (which is a plasma discharge in the atmosphere)

If his transistor got hit by lightning I rather doubt he would be around to post on the forum.

I threw out about half a dozen tea chests full of valves

Those don't sound like electronic components. How do they apply to the current example ?

raschemmel:
Oh come on. When someone talks about "static discharge" they are not talking about lightning . (which is a plasma discharge in the atmosphere)

NO it's a static discharge in the atmosphere that creates plasma.

In my original statement I said it was unlikely to be the original posters problem. Why do you feel the need to pick a fight over this?

Those don't sound like electronic components.

So are you now criticising me for my slapdash use of the term valve instead of "Thermionic Valves."

How do they apply to the current example ?

They don't. I was answering a prior post in this thread.

Sounds like you've forgotten to take your meds again :slight_smile:

@KenF,

So are you now criticising me for my slapdash use of the term valve instead of "Thermionic Valves."

Did it occur to you that I might be in the USA where we call them vacuum tubes (and always have) ?
I don't want to continue this discussion but I would like to say that you shouldn't accuse people of trying to
pick a fight because they comment that your association of static discharge with respect to a transistor and lightning is not relevant.

Sounds like you've forgotten to take your meds again

Really ? Let's keep it civil shall we ?

Open base tie. Clearly the emitter tie is functional, as quite likely is the die itself.

It is not unusual for the encapsulation to be less than sturdy, particularly after soldering (LEDs) and an internal weld fails.

KenF:
In my original statement I said it was unlikely to be the original posters problem. Why do you feel the need to pick a fight over this?

Internet convention? :zipper_mouth_face: