PSU install started smoking - any advice on my 2nd attempt with a new PSU?

[gamerboy]

A bit of a newbie here.

So I have a Dell Inspiron 620 with a Intel H61 MOBO. Has worked fine for last two years. Has 8g Ram and i5-2400 cpu.

I then get the itch to upgrade the GPU. So I buy a new PSU too (Corsair 600).

Steps I took on the install:

I remove the old PSU. Put in the new one and, to the best of my knowledge, connected the cables correctly.

I did not tie any of the cables together - so they were kinda a big mess in the case, touching who knows what.

When I put in the new GPU (GTX-650 ti boost) I did not close the little white plastic latch that is on the PCI slot onto the card because I couldn't reach it since the card is big compared to the space in the case. But the card felt snug. I connected PSU's 6 pin to the GPU.

I turned on the computer and nothing happened.

I turned it on again and then it started smoking.

I removed it. Later I reconnected the old PSU and when I boot up my computer says disk drive failure (I'm replacing the drive tomorrow).

So, a new PSU will show up tomorrow (550w 80 plus Gold Rosewill) and I'm just wondering if there are any tips or guides before I install it so that I don't fry the new one.

 

[Bubbleawsome]

I know the psu is another good one, so if it fails there is a problem. Just make sure cables aren't plugged into each other or touching metal.

 

[flimsy]

check inside the case to make sure nothing is shorting.

 

[Kckazdude]

Did you look for disfigured wires or burn spots to let you know where the magic smoke came from? Might give an indication of what went wrong before you have another problem.

 

[gamerboy]

I checked. No burn spots or burnt wires. Just smoke from the psu.

One thing that may or may not be worth noting is that I connect the computer to a power strip to the wall outlet via a 2 prong cheater plug because my apartment only has outlets with two holes - no grounding.

But I've used my computer for over a year with no problem here. The psu in my computer has been a 300w psu that dell supplies, which I'm sure is crap. So I would figure a good psu like the Corsair wouldn't be more prone to issues from a cheater plug. Keep in my I have lots of electronics including receivers, a powerful subwoofer, air conditioners, TVs, and a projector using cheater plugs with no issues.

 

[turn_pike]

My first instinct is to check whether you have the 110/220v switch in the correct position. That said it usually only causes catastrophic damage when you plug the psu in 110v mode to a 220v of.

 

[gamerboy]

The 110v/220v switch? Where is that? There is no such switch on the Corsair PSU case. Is it elsewhere in the computer?

 

[Torn Mind]

The 110v/220v switch is not on modern power supplies due to the implementation of the Active PFC(power factor correction) circuitry. One convenience this bestows is that it removes the need for the voltage doubler switch, even though active PFC itself has nothing to do with convenience and everything with not wasting power(you aren't affect directly by it, but the utility company is).

 

[westom]

But I've used my computer for over a year with no problem here. The psu in my computer has been a 300w psu that dell supplies, which I'm sure is crap.

Third prong does nothing for hardware safety or reliability. That safety ground is for human safety. Does nothing to avert hardware damage.

A Dell supply at 300 watts is electrically equivalent to some third party supplies rated at about 430 watts. Nobody lied. Each is using a different measurement to report watts.

That supply smoked because it was crap. Short together all outputs in that Dell supply and it will not smoke; will not be damaged. So why did your 'better' supply smoke? Ignore hearsay. A 300 watt Dell supply is probably more than sufficient for the load.

No previous failures proves nothing. Instead, learn what the third prong does before making any conclusion.

If the supply specifications state it is designed for voltages from 85 to 265 volts, then it does automatically what that switch would otherwise do.

Supply smoked if it was defective (by design or due to a manufacturing defect). Any properly designed supply will not smoke even when all outputs are shorted together. A standard for supplies even before the IBM PC existed. However some third party supplies smoke. Then many assume the short circuit - not a defective supply - caused that failure.

Is the Dell supply sufficient? Using a multimeter in combination with other knowledge can answer that question without any doubts.

 

[Bubbleawsome]

Third prong does nothing for hardware safety or reliability. That safety ground is for human safety. Does nothing to avert hardware damage.

A Dell supply at 300 watts is electrically equivalent to some third party supplies rated at about 430 watts. Nobody lied. Each is using a different measurement to report watts.

That supply smoked because it was crap. Short together all outputs in that Dell supply and it will not smoke; will not be damaged. So why did your 'better' supply smoke? Ignore hearsay. A 300 watt Dell supply is probably more than sufficient for the load.

No previous failures proves nothing. Instead, learn what the third prong does before making any conclusion.

If the supply specifications state it is designed for voltages from 85 to 265 volts, then it does automatically what that switch would otherwise do.

Supply smoked if it was defective (by design or due to a manufacturing defect). Any properly designed supply will not smoke even when all outputs are shorted together. A standard for supplies even before the IBM PC existed. However some third party supplies smoke. Then many assume the short circuit - not a defective supply - caused that failure.

Is the Dell supply sufficient? Using a multimeter in combination with other knowledge can answer that question without any doubts.

Nooooo, the corsair is a good PSU, while the dell is subpar.
Just a defective unit probably. EVGA, Dell, Rosewill, and corsair PSUs can all blow up/smoke, it's just a thing that happens.

 

[westom]

EVGA, Dell, Rosewill, and corsair PSUs can all blow up/smoke, it's just a thing that happens.

It just 'happens' when one does not know power supplies; did not design them. Smoke is a condition associated with manufacturing defects with include design defects. Every smoke condition is explains by specific part failure.

Dell is a quite good supply. Unlike so many supplies marketed to computer assemblers, the Dell actually meets many ATX specs. I can tell horror stories of so many third party supplies that never meet ATX requirements - by listing the missing components, missing circuit, or in some cases, an intentionally deceptive design.

Corsair has promoted programs to separate them from so many other and inferior supplies often marketed to computer assemblers. Because most computer assembler have no idea how electricity works, what a power supply must do, or the always required specification numbers. No Dell, HP, or other engineer would select such inferior supplies. A problem that explains why Corsair is trying to not be painted by a broad brush that condemns so many supplies marketed to computer assemblers.

A Dell supply is more than sufficient. If is was not, then he lists specific defects in that supply by component and with numbers.

 

[ProphetMikey]

If the HDD was toasted there's always the chance that you jammed in a molex power connector the wrong around. I know it sounds difficult to do but I have seen it done.

Scratch that: showing my age there. Of course it would now be a SATA power connector.

 

[Rezist]

I believe over current circuitry should protect you from blowing your psu up if something shorted.

 

[bryanl]

Third prong does nothing for hardware safety or reliability. That safety ground is for human safety. Does nothing to avert hardware damage.

Not even if it lets the case to float at 60VAC due to the Y capacitors, when users plug in devices while the computer is running?

 

[bryanl]

Don't start a computer unless you know that no power connectors are touching metal and cannot touch metal by moving.

Always latch the video card since it can ride up in the PCIE slot when the video cable is plugged in or the case flexes, and the case often flexes when the cable is plugged in. A card not plugged in crooked can short adjacent contacts.

You need to verify whether the hard drive failure is due to the drive or the motherboard. Excessive voltage from the PSU can damage either, and motherboard damage is usually much more serious and much harder to repair. But damage to the drive may be a relatively simple matter of a shorted diode and opened fuse:

http://www.users.on.net/~fzabkar/HDD/TVS_diode_FAQ.html

 

[westom]

Not even if it lets the case to float at 60VAC due to the Y capacitors, when users plug in devices while the computer is running?

A case floating to 60 volts (a floating ground) means no change to DC voltages on electronics. Voltage is always a relationship between two points. 60 volts on a case is between the case and safety (or earth) ground. That relationship causes no current to flow through electronics. Voltage between individual DC voltages remains unchanged between each DC voltage an the digital (ie motherboard or PSU) ground. Voltage to digital ground is the only voltage relevant to electronics. And is not changed by a completely different voltage from case to equipment / safety / protection / earth ground.

Making it further irrelevant is a near zero (maybe 60 uamp) current that creates a floating 60 volts.