[SOLVED] i think i shorted my power supply and i bought i new one.

[helpDiscoDrago]

system:
amd ryzen 7 3800x (3.9-4.5ghz)
ati rx 5700xt
16 gig ram 3400 mhz
dvd writer
7200 rpm western digital blue 64 mbyte cache
am4 gigabyte mobo b450m
power star 650w psu (i think i fried it’s fuse and i have a corsair 650w more expensive arriving tomorrow).

i initially forgot to plug in anything to the 8 plug cpu power on the mobo. all the fans and lights worked and the computer power button made that so.

then i realized i did not plug anything into cpu power. i plugged the 4 pin had from my cheap 650w psu into the 8 pin and it worked only to show lights and fans spinning.

i then thought the computer may not be showing anything because not enough power to my amazing awesome mystery ryzen 7 cpu. its a brand new build and i havent seen even bios yet.
i plugged in 6 pin to the 8 pin and...

the cpu fan lit up, monitor again showed nothing good, then the cpu fan light dimmed to about 40% of its luster. then it all went poof and will not turn on no matter what even.

i originally considered the hdmi cable (15 years old) to be broken. i have new one coming tomorrow.

i understand motherboard is not easy to fry at all especially after less than 20 seconds first turn on and less than 5 seconds for the 4 pin try and basically 3 seconds for the 6 pin failure.

i doubt the cpu is fried because i want to believe in good things and maybe also god.

please help i cannot wait until tomorrow to understand my life.
thanks.

 

[Karadjgne]

12v rail says everything you need. 12v runs everything in a pc nowadays, there's a few 5v like the usb and the power logic circuitry on the motherboard is 3.3v, but almost everything else is 12v. That includes a cpu that'll easily pull 100w (ish), a gpu that will pull 225w+, the fans, drives, rgb, and everything else that'll easily pull @ 100w for a grand total of @ 425w. On a psu of very unreliable quality and performance, stating it's 650w when reality has it at 30A x 12v = 360w, IF that's even an accurate measure. I'd suspect you'd be lucky to see anything above @ 25A on that rail, putting you much closer to 300w at best. Nowhere near the 650w lie. 30A at 12v and 36A at 5v was popular 20 years ago, when that platform was new.

You wasted $40 on a complete piece of (****) psu, try and get your money back or chalk it up to a learning experience.

The psu powers everything, it's literally the Heart of a pc. You literally asked a doctor to give you the heart of a 90 year old, obese chainsmoker because it is cheaper than the 20 year old vegan athlete.

You won't know what is or isn't affected until after you replace that psu, and hopefully the arriving Corsair is at least a 650w TXM, with a preference for a 650w RMx, and not the CV or VS models.

 

[helpDiscoDrago]

12v rail says everything you need. 12v runs everything in a pc nowadays, there's a few 5v like the usb and the power logic circuitry on the motherboard is 3.3v, but almost everything else is 12v. That includes a cpu that'll easily pull 100w (ish), a gpu that will pull 225w+, the fans, drives, rgb, and everything else that'll easily pull @ 100w for a grand total of @ 425w. On a psu of very unreliable quality and performance, stating it's 650w when reality has it at 30A x 12v = 360w, IF that's even an accurate measure. I'd suspect you'd be lucky to see anything above @ 25A on that rail, putting you much closer to 300w at best. Nowhere near the 650w lie. 30A at 12v and 36A at 5v was popular 20 years ago, when that platform was new.

You wasted $40 on a complete piece of (****) psu, try and get your money back or chalk it up to a learning experience.

The psu powers everything, it's literally the Heart of a pc. You literally asked a doctor to give you the heart of a 90 year old, obese chainsmoker because it is cheaper than the 20 year old vegan athlete.

You won't know what is or isn't affected until after you replace that psu, and hopefully the arriving Corsair is at least a 650w TXM, with a preference for a 650w RMx, and not the CV or VS models.

its golden bronze +80 according to the manuscript... i can only assume my brand new fair quality motherboard and my amazing processor is just fine then?

 

[helpDiscoDrago]

its golden bronze +80 according to the manuscript... i can only assume my brand new fair quality motherboard and my amazing processor is just fine then?

its cv model... but it has the 8 plug cpu powering 4+4 am i making a mistake?

 

[helpDiscoDrago]

i do not have the monies 🙁
i do what any boy do with nice cpu parts, i put together with playdoh and hope for best results!

 

[helpDiscoDrago]

i want to believe in 7200 rpm hdd because the discs are solid metal. i hope cache works well with fast ram and i cannot disregard game code infrastructure that makes all ends meet. ssd storage not enough at slightly higher price so i cherish hard disc with hope for good future. the motherboard is not fancy 125$ plus range but i feel complete lack of pcie 4.0 on everything but 270$ motherboard is a bit outrageous. this is my buying potential for now and later i can even reuse key ingredients. i am not handy with cash so I hope power supply works best case. i game pro level. good day.

 

[Phaaze88]

i do not have the monies 🙁
i do what any boy do with nice cpu parts, i put together with playdoh and hope for best results!

Ok, but as a counter argument, you bought a 40USD power supply, but you could afford to purchase:
3800X: 330USD - almost 8X the psu's cost
2x 8GB 3400mhz ram: 85USD - about 2X
RX 5700XT: 400USD - 10X
... but couldn't spend at least the cost of the ram on the power supply?

It would've been simple enough to knock it down to a 3700X, or even a 3600X, and offload that difference towards a high quality power supply, and you would've been golden!

It's a common mistake new builders make: too much excitement/focus on the cpu + gpu, that they neglect the very heart needed to supply lifeblood to the system, because it's 'dull', or not as exciting to invest into.
... until things go south.

These scenarios are very common. Learn from the potentially expensive mistake and move on - never cheap out on power supplies.

 

[Karadjgne]

A Sata3 6Gb SSD is @ 5x faster to move data than a 7200rpm hdd. A NVMe is at least 1.5x and upto @ 3x faster than the Sata HDD. That makes everything from web pages to downloads to gaming map load times anywhere upto @ 10x faster. So that makes a huge difference, for most its worth the price difference. It's the difference between CSGO map loading in 30 seconds, or 3-4 seconds.

You way overpaid on the cpu. The 3800x is nothing more than a higher energy consuming, better binned version of the cheaper 3700x. Their performance at stock is almost identical. It's only at full boost you might get an extra 100MHz from the 3800x, depending on temps.

Should have opted for 3600MHz ram, the 3400MHz on that 3800x is worse performance than a 3700x with 3600MHz ram.

Hope the CV works out for you, but as a pro gamer, you should understand that by opting for the lowest value psu its like somebody requiring you to perform using a stock HP keyboard and 2-button ball mouse, when you really should be on a mechanical and 7 button macro gaming optical mouse with 10k+ dpi.

 

[jonnyguru]

i then thought the computer may not be showing anything because not enough power to my amazing awesome mystery ryzen 7 cpu. its a brand new build and i havent seen even bios yet.
i plugged in 6 pin to the 8 pin and...

Yeah.

The 6-pin PCIe and 8-pin have reversed polarity pins.

You probably fried a bunch of stuff by doing that.

 

[jonnyguru]

its cv model... but it has the 8 plug cpu powering 4+4 am i making a mistake?

4+4-pin and 8-pin are the same thing.

 

[Karadjgne]

Some motherboards have 4pin cpu, some have 8pin cpu, some even have 8+4pin cpu, so the EPS is split 4+4 so that it can be used by all motherboards. It's the same theory as the PCIE for the gpu is 6+2pin, can be used as 6pin or 8pin depending on the gpu.

 

[jonnyguru]

Some motherboards have 4pin cpu, some have 8pin cpu, some even have 8+4pin cpu, so the EPS is split 4+4 so that it can be used by all motherboards. It's the same theory as the PCIE for the gpu is 6+2pin, can be used as 6pin or 8pin depending on the gpu.

You talking to me or OP?

Because the way I read OP's post.. He took a 6-pin PCIe and plugged it into an 8-pin CPU connection on his motherboard.

Am I wrong?

 

[Karadjgne]

Op, and you? wrong? Please lol. No, I read that too, but there's no way to tell what, if anything, got fried because of that until he replaces the psu. With as poor a quality as that is, it's almost it's own protection, hopefully it blew before anything else could.

i plugged in 6 pin to the 8 pin and...

the cpu fan lit up, monitor again showed nothing good, then the cpu fan light dimmed to about 40% of its luster. then it all went poof and will not turn on no matter what even.

Chances don't look good though. Actually somewhat curious as to how that happened, keyed connectors usually prevent such, but I guess some muscle power was involved.

 

[helpDiscoDrago]

because power for the cpu is about 105 w and max output for a 4 pin i read was half of the 150 w the 8 plug can produce, it seems likely the motherboard just refused the power supply from the surge protection feature of its inanimate fuse. this i would think is simply the only thing that happened (similiar to the internal clock going battery free). i also read that the cpu has no fuse. seems the most sensitive item is the psu which is an outdated and outmoded caveat. i was foolish to believe 650w units would be compatible with such common modern 8 pin expectation (however uncommon the processor is as of now). i beoieve the new psu will do fine and i also think the hdmi cable was out.

can the motherboard boot without the cpu being powered? when i first tried to boot i didnt have any plug in and all system fans were ago. now the psu seems shotty and i cant get
nothing but my mouse led to light up. thabks be to MSI for a pretty nice gaming mouse.

 

[Karadjgne]

What you are reading isn't right or you are misinterpreting stuff and it's leading you astray.

The EPS cpu 8pin and the 6+2pin pcie are not the same at all. The EPS is @ 336w, the pcie is 75/150w. On top of that is the wattage the cpu pulls from the motherboard mains supply, so it's not pulling the entire 105w from the EPS, which is closer to 163w for the 4pin.

You caused the issue by inserting the 6pin pcie. The EPS has yellow wires on one side black on the other. On the pcie, the wires are opposite. You put 12v to ground, and ground to 12v. A dead short.

Because that power supply is so cheaply built, it doesn't have much in the way of protective circuitry, if any at all. No fuse, no breaker, nothing. By putting 12v to the motherboard ground, you send power backwards to everything that's connected. Stuff like drives that are only supposed to be 5v, you gave it 12v. The circuitry on the motherboard that's supposed to be 3.3v, you gave it 12v. The cpu is supposed to be less than 1.5v and you gave it 12v.

A good 550w psu can power your pc. That piece of junk that's really only 300w didn't stand a chance of powering your pc.

 

[helpDiscoDrago]

i think the corsair cv 650w will do the trick. at 80 bronze it seems basic but reliable enough. i do declare you’ve help me understand that i have toasted the power supply. this is acceptable to me but:

Motherboards use special fuses called polymeric positive temperature coefficient thermistors; in English, this translates into "a fuse that will reset itself." These fuses protect the motherboard from frying every time a larger amount of electricity surges through the computer power supply.

—— from Techwalla. this also makes sense because the motherboard expects current to flow at very high rates involving heating / cooling and potential surges. it encourages me to believe that the silicon holds everything very nicely and the fiberglass wouldnt allow too much spiking. from a materials science perspective it seems sound, that i did not possibly actually fry the motherboard. my understanding is Volts = Current x Resistance so the actual 15 volt value you spoke of could be either sharp current (unlikely because its senseless engineering for a psu to produce dangerous amounts of current when all that is needed is a balance with decent resistors) versus the likely path taken -> the silicon and wiring has natural resistance as well as resistor units to cause an increase valuation of R while the I ampere value is slightly higher and possibly reversed. then the voltages would read a string negative off of moment / as well as a near triplication of reverse flow of current.

it seems wild that i even thought id make any sense trying to figure that out...

according to Quora - cpu processors do not have fuses. fuses are considered servicable parts. the hard wiring of a cpu, however delicate, is designed to withstand extreme conditions.

if current is able flow for yearhours with a circuit designed to balance voltages long term, the actual current I in amperes is actually factored by significant amounts of exposure. with heat and ion electricity considered, the motherboard must be unbelievably resilient and the cpu must be like a tank. resistors resist current which gives apparent voltage multipliers, but resistors will stop a current dead in its track as fuses truly example - zero snap electricity. therefore, the resistor would have to appear as a more significant resistor unit (air) and the current would stop and the values would fail as mathematics. i think this is true because i just recently found out the two details about cpu and motherboard - one has no fuse at all and is itself a potential air fuse, the other has fuses specifically designed to reseat itself to avoid repeated harm.

 

[digitalgriffin]

i think the corsair cv 650w will do the trick. at 80 bronze it seems basic but reliable enough. i do declare you’ve help me understand that i have toasted the power supply. this is acceptable to me but:

Motherboards use special fuses called polymeric positive temperature coefficient thermistors; in English, this translates into "a fuse that will reset itself." These fuses protect the motherboard from frying every time a larger amount of electricity surges through the computer power supply.

—— from Techwalla. this also makes sense because the motherboard expects current to flow at very high rates involving heating / cooling and potential surges. it encourages me to believe that the silicon holds everything very nicely and the fiberglass wouldnt allow too much spiking. from a materials science perspective it seems sound, that i did not possibly actually fry the motherboard. my understanding is Volts = Current x Resistance so the actual 15 volt value you spoke of could be either sharp current (unlikely because its senseless engineering for a psu to produce dangerous amounts of current when all that is needed is a balance with decent resistors) versus the likely path taken -> the silicon and wiring has natural resistance as well as resistor units to cause an increase valuation of R while the I ampere value is slightly higher and possibly reversed. then the voltages would read a string negative off of moment / as well as a near triplication of reverse flow of current.

it seems wild that i even thought id make any sense trying to figure that out...

according to Quora - cpu processors do not have fuses. fuses are considered servicable parts. the hard wiring of a cpu, however delicate, is designed to withstand extreme conditions.

if current is able flow for yearhours with a circuit designed to balance voltages long term, the actual current I in amperes is actually factored by significant amounts of exposure. with heat and ion electricity considered, the motherboard must be unbelievably resilient and the cpu must be like a tank. resistors resist current which gives apparent voltage multipliers, but resistors will stop a current dead in its track as fuses truly example - zero snap electricity. therefore, the resistor would have to appear as a more significant resistor unit (air) and the current would stop and the values would fail as mathematics. i think this is true because i just recently found out the two details about cpu and motherboard - one has no fuse at all and is itself a potential air fuse, the other has fuses specifically designed to reseat itself to avoid repeated harm.

I agree with the other people here. You 95 % likely blew out your cpu and motherboard. Maybe the gpu is blown out too depending on how the backplane is setup.

Six pin Pcie power cables were never meant to be used on the cpu power connector.

You might get lucky, but as someone who knows electrical circuits, i seriously doubt it. A polyfuse that you referred to will NOTsave you from bad voltages. It will only help protect you from over currents.

Sorry to sound harsh, but talking to us isnt going to solve your problem one way or another. You are just going to have to wait it out and try and pray there was some protection diodes on the ground plane. As I tell my son, "You messed up. I make mistakes too. Learn from it. Tomorrow is another day."

There is no if ands or buts on this. I suggest this thread be closed.

 

[helpDiscoDrago]

but then what is pcie 4.0? i dont know how voltages could be forced unto a processor designed for such strict management of electricity. i read on reddit that its current that destroys a cpu processor and high stress testing yielded maximums at 200watts or better for a cpu (intel in that case). does that even allow for a 175 watt power connector to destroy something that is protected from over currents as you mentioned? i would be horrified if i destroyed my cpu and motherboard and even possibly my gpu. i was just pluging the 6 pin into the 8 pin with hopes it would work and maybe it wouldnt. i think the wiring layout of the 6 pin being known as “different” just means the power supply is capable of reading reverse voltage and ground for what is ultimately a simple wire (power supplies i dont think are smart tech). i think the resulting ground/live wires are just indicative of the motherboard lay itself and not the forced inputs over a basic wire. how could a psu even perform that? it just limits and breaks voltage up over wires that are not always used.

i hope i didnt bust my cpu or mobo or gpu i really do. the power star psu has ovp and uvp protection so over voltage and under voltage values are factored out upon circuit formation. am i wrong in this?

 

[DSzymborski]

but then what is pcie 4.0? i dont know how voltages could be forced unto a processor designed for such strict management of electricity. i read on reddit that its current that destroys a cpu processor and high stress testing yielded maximums at 200watts or better for a cpu (intel in that case). does that even allow for a 175 watt power connector to destroy something that is protected from over currents as you mentioned?

That's only for a CPU that's connected properly getting the right power, which is not at all your situation. +12V in the wrong place can easily be instant death.

 

[digitalgriffin]

As I said, "There is NO IF's AND's or BUT's" here.

There is a standard that both motherboard manufacturers and PSU makes agree too called the ATX power supply spec. This defines what connector does what and what pin has to carry what voltage and how much. Everybody conforms to this. A PCIe power cord is NOT wired up the same as a ATX CPU AUX 4+4 connector. Circuits can burn out with as little as 0.3V wired in the wrong direction. (Even less on Chip (IC) circuits) You used 12 Volts on a ground. That's 12 volts backwards. Way over .3V

How do you think we are going to solve your problem?

 

[helpDiscoDrago]

That's only for a CPU that's connected properly getting the right power, which is not at all your situation. +12V in the wrong place can easily be instant death.

thats a good point i will try and return the processor if it doesnt work. i havent seen it work once as of yet so i dont know what harm did or did not occur. my motherboard bios didnt boot up when i had nothing plugged into the 8 pin eps at first (i had forgotten to plug anything in). doesnt the motherboard boot itself initially without the processor (which i assume requires some sort of drive installation?)? only all the fans were working well.

 

[DSzymborski]

thats a good point i will try and return the processor if it doesnt work. i havent seen it work once as of yet so i dont know what harm did or did not occur. my motherboard bios didnt boot up when i had nothing plugged into the 8 pin eps at first (i had forgotten to plug anything in). doesnt the motherboard boot itself initially without the processor (which i assume requires some sort of drive installation?)? only all the fans were working well.

Do you normally return merchandise that you personally broke? I guarantee that was not my point.

 

[helpDiscoDrago]

i didnt receive any instructions for installation i had no clue i was going to break or not break something... the plugs seemed well enough and the 4 pin was explained as “sufficient unless the cpu requires a good deal”. i literally thought the 6 pin was just a more fancy 4 pin and it caused this grief. its scary because the power seemed to have dulled and then it shut off, as the fans were working and then suddenly not after less bright lights. i cant believe there arent a huge amount of protective systems to insure safe and stable processing. i am really upset thinking i broke the cpu but limited warranty should cover the fact that it was attached to the motherboard properly. i do accept if i busted the motherboard but what else can i do right now im not all that well off and i just wanted to game..

 

[digitalgriffin]

i didnt receive any instructions for installation i had no clue i was going to break or not break something... the plugs seemed well enough and the 4 pin was explained as “sufficient unless the cpu requires a good deal”. i literally thought the 6 pin was just a more fancy 4 pin and it caused this grief. its scary because the power seemed to have dulled and then it shut off, as the fans were working and then suddenly not after less bright lights. i cant believe there arent a huge amount of protective systems to insure safe and stable processing. i am really upset thinking i broke the cpu but limited warranty should cover the fact that it was attached to the motherboard properly. i do accept if i busted the motherboard but what else can i do right now im not all that well off and i just wanted to game..

In the end, this is your responsibility. If you didn't know how to hook it up, it was your responsibility to get someone whom did. Returning it because you broke it is unethical.

Like I said, "You messed up. I sometimes mess up too. Learn from it. Tomorrow is another day."

 

[Corwin65]

Once the test any components you may try to return, chances are they will deny a refund.

It's not an offending statement so please do not take it that way but AMD/Intel/NVidia do not care about your ignorance.

Nor should they to be honest.