Ever heard of PSU electrocution?

RegEdit

New Member
Reaction score
3
Location
Pacific Palisades, CA
I know you're not supposed to open up a PSU because of those charged capacitors that remain charged even after unplugged. But has anyone ever heard of any perfect storm stories when someone got electrocuted from just working inside a computer? Perhaps a piece of wire or metal got lodged inside a PSU and touched a capacitor and the casing of the PSU?
 
Its possible. When I was going through the electrical apprenticeship we came across stories and articles about people being killed by power supplies of all sorts... From microwave ovens to TVs. If I remember correctly, somewhere around 700mA is what kills. At that amperage it interferes with your heart causing a skipped beat or flutter...
The other thing is that the electricity needs to pass across the chest, as from one hand to the other. I always discharge caps with a screw driver.
 
With the old AT PSUs you could easily electrocute yourself if you weren't careful with the power switch and the connectors.
I remember always double & triple checking that the PSU wasn't plugged in before replacing one or replacing the switch itself.
 
The reason you're not supposed to open a PSU is because it's the only bit of a computer at the full mains voltage (when plugged in), not because the caps will kill you.

It's pretty unlikely that the caps in a PSU contain enough energy to electrocute you.
 
More people die each year from low-voltage electrocution than from high-voltage electrocution...mostly because they underestimate the killing potential of low-voltage.

Depending on resistance (wet skin v. dry skin) a current of 300mA-500mA DC can cause ventricular fibrillation of the heart...resulting in death.

Under the right conditions, your cell phone battery can kill you, and so can the capacitors in a computer PSU that has not been properly discharged.
 
Last edited:
Not seen any warnings of electrocution on low laptop/phone batteries.

I did once get electrocuted by an AT case the insulation to the switch must have been loose and I poked a screw driver behind it trying to find a screw forgetting the mains lead was in. I got quite a nasty shock but I was about 14.

Also never discharge capacitors with a screw driver, I did that once when trying to repair a camera flash and it exploded in my face. Thankfully the screw driver was double insulated otherwise I may have got a nasty shock too.
 
voltage doesn't kill, it's the amps man! DC current is the worst too.

I have been shocked by quite a few modems back in the day though - through that little bit of power running over the phone line when I would forget to unplug it before removing a modem from a case (in diagnosing a no post, or re-seating the modem/swapping slots with dial-up issues)

This also reminded me of once over a dozen years ago I had an (unbeknownst to me) improperly grounded ceiling fan with a very long dangling cord so I could pull it from my chair without getting up. I liked to prop my feet up on my full tower super micro case (well, case off always)...

Point being I pulled the cord one day with my feel still propped up on the case, and electrocuted myself pretty good. I was thrashing all around for about 30 of the longest seconds of my life because I couldn't let go of the cord to the fan when finally my foot jerked off the case. I could feel the AC surging from my hand down to my foot, what a jolt!
 
I am aware it is the ampage which kills and not the voltage ;) I do have a basic qualification in electronics. While other teenage lads were busy getting STDs, I was busy making LEDs flash with 555 timers etc :p
 
It actually takes only .1 amp to kill.


Your cellphone battery can't kill you though it can do much more than that. The reason is that much current will never flow through you at the voltage of your cellphone battery.


You will have about 150 K Ohms to 1 M Ohm of resistance from hand to hand.

I = E/R where I required is .1 to kill you


.1 = E/150,000


It is going to take 15,000 volts hand to hand to cause your body to draw .1 amps through the heart at 150 K Ohms. Obviously if you have just been swimming in the ocean and have wet hands it is much less... it also depends on the route.

If you get shocked from your stomach/abdomen to your feet your heart isn't in the circuit and you probably won't die.
 
I worked a lot in the service industry through the 70s and 80s. My understanding was 50ma can kill especially across the heart so you always prodded a live machine with your left hand behind your back in other words not near the case so the shock(if you got one) didtravel across your chest arm to arm. UK main board trips are around 30ma.
 
This is mostly a mute post, nearly all modern PSUs have discharge resisters built in so that theres no potential stored in the capacitors.

Even the old AT's had these. Its really the ancient ones that dont. I've always as a matter of course unplugged from mains and pressed start button to take anything left over out, more from habit than anything else.

I still cannot though understand why anyone wants to go poking inside a PSU given the cost of replacing one
 
Wait, whaaaat? Why in the world would you do that instead of using a load to discharge them?

Because it makes a nice little spark that I like, the screw driver has virtually NO resistance, and I don't have to wonder if my "load" is working correctly before I go touching caps with my body. So long as your not discharging a fly-back transformer or something, a screwdriver is fine.

Apple G4 (and perhaps G5) Systems do NOT discharge their loads, or not very fast anyway. I fix a bunch of these because the price of replacement is ridiculous. Price for caps is cheap.

Some PSU's are specialized enough to be rare, like the crappy ASUS Pundit systems.

Yes, modern PSU's GENERALLY discharge their loads with resistors, however, some cheap power supplies DO NOT DO THIS! It is not a manufacturing requirement. Also, are the resistors that drain the current WORKING CORRECTLY? I would assume you would be working on a BROKEN PSU. Never assume anything when working with stored energy of ANY KIND, this is basic safety 101. People die everyday from incorrect assumptions.
 
Last edited:
Back to my point, as long as you unplug the power cord and as long as you don't open up the PSU or poke into it you're not gonna get shocked working on a computer?

Yes you are correct with a modern ATX 20 pin or newer you will be ok. You can discharge the PSU after unplugging it from the wall by pressing the power button and that last bit of stored energy will crank a fan for half a second.
 
... Perhaps a piece of wire or metal got lodged inside a PSU and touched a capacitor and the casing of the PSU?
Since the PSU casing is grounded, if this situation occurred, the most likely thing to happen would be a breaker would pop or perhaps the PSU would fail but it shouldn't be possible for someone working on the computer to get shocked.

It really is a misnomer to say that current kills, not voltage. While it's true that current flow across the heart is what kills you, how much current flows is dependent upon the voltage and skin resistance. Skin resistance varies due to a number of factors such as whether wet or dry but the major contributing factor in electrocution is the voltage, the higher the voltage the more likely skin resistance will be overcome to allow sufficient current flow thru the heart.

The statement that a cellphone battery could kill you under the right conditions is true, but misleading, as the right conditions would have to be that the battery was connected to probes that pierced the skin on either side of the heart. That's about the only way the cellphone battery's low voltage could get enough current to flow to interrupt the heart.

Back in the 70s, I worked as a tech on huge power supplies that were capable of delivering 1000A @ 300V. I once got shocked by 480VDC thru my right arm from my elbow to my pinky finger. If that would have been from one hand to the other so that the current flow was across my chest, I wouldn't be here today. It hurt like he11 but I was able to continue working. Needless to say, I was a bit shaky the rest of the day.
 
Lets not forget the last most important factor. Time. The time you are exposed to an electrical current is also as important as the Voltage and current.
 
I've had a shock from a switch once. It wouldn't turn on, so I opened it up to take a look but left it plugged in. When I touched the psu, I had this weird tingling feeling up my arm, I was like wtf :o

Never done it since.
 
Last edited:
Back
Top