PSU 5V rail issues (reported 4.74V in Gigabyte SIV and

[Chiyawa]

Well, it happens again. The 5V rail reported by my motherboard BIOS is 4.74V, which is very low. The system has issue running external HDD. I began to suspect my PSU is having issue. Refer to this thread:

 

Looks like I need a new PSU then, but which should I choose? I have narrow down some of the PSU that is available from my local shop.

- Corsair RM 750W - MYR549 (a bit pricy)

- Cooler Master MWE Gold 750W - MYR429

- ThermalTake Toughpower Grand (non-rgb) 750W - MYR489

- EVGA Supernova GT 750W - MYR529

- Gigabyte Aorus P 750W - MYR529 (I'm not sure if this is the unit with explosive feature)

- Silverstone DA 750W - MYR429

- ThermalTake ToughPower GF 750W - MYR499

 

*Above listed price are from the shop, which includes tax

 

Regards,

Chiyawa

[An0maly_76]

My old man was an electrician for 40 years and told me that generally, 10% variance is acceptable, which would be 0.5 volt, or 4.5 volts output before there should be an issue. So I'm not sure you have an issue here.

[BiotechBen]

+/- 5% is within spec for it. 

[Chiyawa]

1 minute ago, An0maly_76 said:

My old man was an electrician for 40 years and told me that generally, 10% variance is acceptable, which would be 0.5 volt, or 4.5 volts output before there should be an issue. So I'm not sure you have an issue here.

The issue is my external HDD is having issues and can't seems to power on. I try the external HDD in my laptop and it works, so I suspect that the problem is that it couldn't supply adequate power to power up the external HDD (4TB WD My Passport).

[Chiyawa]

Just now, BiotechBen said:

+/- 5% is within spec for it. 

The issue is my external HDD is having issues and can't seems to power on. I try the external HDD in my laptop and it works, so I suspect that the problem is that it couldn't supply adequate power to power up the external HDD (4TB WD My Passport).

[BiotechBen]

Just now, Chiyawa said:

The issue is my external HDD is having issues and can't seems to power on. I try the external HDD in my laptop and it works, so I suspect that the problem is that it couldn't supply adequate power to power up the external HDD (4TB WD My Passport).

Usually for that it's a matter of current. Your mobo might only be outputting a few hundred milliamps while the laptop outputs 1.5A.

my b450 will only do 500ma while my laptop will do 1.5A

[An0maly_76]

Was going to say, this could be an amperage issue, USB varies widely, some up to 3.1A. I remember some older WD Passport drives that literally required a splitter to draw their power from TWO USB ports, as one simply didn't have enough to keep it running.

[Chiyawa]

17 minutes ago, BiotechBen said:

Usually for that it's a matter of current. Your mobo might only be outputting a few hundred milliamps while the laptop outputs 1.5A.

my b450 will only do 500ma while my laptop will do 1.5A

5 minutes ago, An0maly_76 said:

Was going to say, this could be an amperage issue, USB varies widely, some up to 3.1A. I remember some older WD Passport drives that literally required a splitter to draw their power from TWO USB ports, as one simply didn't have enough to keep it running.

I see. So my PSU is okay? Still, with 4.74V, it's very marginal, don't you think? I have a USB-C hub that also couldn't be power up unless I plug in a USB-C power into it. Anyway, thanks for your suggestion. Now I might need to find a USB hub with external power if that's the case.

[An0maly_76]

Just now, Chiyawa said:

I see. So my PSU is okay? Still, with 4.74V, it's very marginal, don't you think? I have a USB-C hub that also couldn't be power up unless I plug in a USB-C power into it. Anyway, thanks for your suggestion. Now I might need to find a USB hub with external power if that's the case.

In light of BTB's revelation of -5%+, that would put the 'out of spec' right at 4.75V, so it does seem like it might be a bit suspect. But would the manufacturer RMA it, is the question.

 

It does seem a powered USB hub would solve your immediate problem while you sort out whether the manufacturer will RMA the PSU.

[mariushm]

I would first look into using a different cable, with thicker wires.

 

Don't rely on what the bios says. Get a multimeter, set it on DC reading (20v or more if it's manual range) and measure the voltage at a mole connector  (old hdd connectors, the pinout it  5v - ground - ground - 12v ... so put the black probe in one of the middle two pins, then use the red probe in the pins on the sides of the connector. One will read 5v and the other 12v or very close to it.

 

Yeah, gave the same advice in that other thread as well ... A 5$ multimeter is cheaper than $100+ power supply and it's a one time purchase that would be useful for decades.

[Guest]

17 hours ago, Chiyawa said:

I see. So my PSU is okay? Still, with 4.74V, it's very marginal, don't you think? I have a USB-C hub that also couldn't be power up unless I plug in a USB-C power into it. Anyway, thanks for your suggestion. Now I might need to find a USB hub with external power if that's the case.

First off, the +4.74V you're seeing is at the board.  Not the PSU.  If you check the PSU's voltage at an actual PSU output connector with a DMM, you'll probably find the voltage a lot closer to +5V than what you're seeing in Gigglebyte's software.

 

Secondly, the USB's output voltage is regulated by the motherboard.  It's not a straight shot from the PSU, through the motherboard, through the USB controller IC and out the back.  There's a lot more to it than that.

 

[Chiyawa]

22 hours ago, An0maly_76 said:

But would the manufacturer RMA it, is the question.

I don't think so. I remember my PSU only have 3 years warranty (but I need to check). It's Gigabyte G750H.

 

10 hours ago, mariushm said:

Don't rely on what the bios says. Get a multimeter, set it on DC reading (20v or more if it's manual range) and measure the voltage at a mole connector  (old hdd connectors, the pinout it  5v - ground - ground - 12v

I see. I got a cheap multimeter but not sure about the accuracy because it doesn't have any certification. Should I get Fluke multimeter though?

 

5 hours ago, jonnyGURU said:

First off, the +4.74V you're seeing is at the board.  Not the PSU.  If you check the PSU's voltage at an actual PSU output connector with a DMM, you'll probably find the voltage a lot closer to +5V than what you're seeing in Gigglebyte's software.

Gigglebyte's software. Ha ha, you made my day.

 

Yeah, I'll be checking with my cheap multimeter. Not sure how accurate it is, though, because there doesn't seems to have any certification on it (yeah, it's very cheap). Should I get Fluke instead for better accuracy?

[An0maly_76]

6 minutes ago, Chiyawa said:

I don't think so. I remember my PSU only have 3 years warranty (but I need to check). It's Gigabyte G750H.

Especially if @jonnyGURU is right, and I believe they are.

6 minutes ago, Chiyawa said:

Gigglebyte's software. Ha ha, you made my day.

Gigglebyte's...

 

Huh-huh... Huh-huh... YOU said, 'bites'... Huh-huh... Huh-huh...

[Guest]

2 hours ago, Chiyawa said:

Gigglebyte's software. Ha ha, you made my day.

 

Yeah, I'll be checking with my cheap multimeter. Not sure how accurate it is, though, because there doesn't seems to have any certification on it (yeah, it's very cheap). Should I get Fluke instead for better accuracy?

Your cheap DMM is going to be 3000x better than Gigglebyte's software.  Even if it's not a Fluke.

 

And as @BiotechBen pointed out:  It's more about current, not voltage. So even if your PSU and USB port are dead on 5V, it doesn't mean you're getting enough power to the drive.

 

 

[Chiyawa]

@jonnyGURU, @BiotechBen, @mariushm, @An0maly_76

 

Huh, this is the second time it does this and it's very weird.

 

This morning, I switch on my computer and look into the voltage rail from the BIOS. the PSU 5V voltage is reported around 4.83V (and the system struggles to detect my HDD when I plug in). But when I try the method suggested by @mariushmand my multimeter read 5.00V, the BIOS immediately reported the 5V rail as 4.98V! It's strange. Does my motherboard had a secret sensor that detects me checking the voltage and recalibrate its report? Or does the test somehow ground the residue of the sensor and now it's working fine? Or could it be that the PSU really provide 4.8V until I test the voltage and somehow discharge some residue current and they function well again? That's strange. This happens the second time (the first time is from the previous post).

 

Now my external HDD works just fine. The system can detect it normally, the disk is not clicking (restarting because not enough power), and it works even through a USB 3.0 extender.

 

Hmm... What a surprise. I'm not sure how long this voltage will stay 5V, but this is, in my 15 years of troubleshooting of computers, the first time I see PC behave this way.

 

Anyway, any of you face this issue before?

 

As a precaution, I think I might get a new PSU just to be safe. Gigabyte PSU is sketchy at best, and I'm sure I don't want the feature they added into their PSU that can display fireworks out of the blue.

 

On a side note, should I get AVR? The electricity in my area is good, but you might never know if power surge will strike. I mean, I usually unplug my PC during thunder storm or when I'm not using it.

[An0maly_76]

A battery backup will generally have some sort of voltage regulator built-in, and it is possible for line noise and other conditioning issues to cause issue like you describe.

[Guest]

16 hours ago, An0maly_76 said:

A battery backup will generally have some sort of voltage regulator built-in, and it is possible for line noise and other conditioning issues to cause issue like you describe.

That's not going to help the OP.

17 hours ago, Chiyawa said:

On a side note, should I get AVR? The electricity in my area is good, but you might never know if power surge will strike. I mean, I usually unplug my PC during thunder storm or when I'm not using it.

AVR isn't going to help you there.  All an AVR does is bucks or boost your voltage if it drops or rises from a nominal value.  If you're in fear of surges, you get a surge protector.  These have MOVs in them that absorb surges.

18 hours ago, Chiyawa said:

This morning, I switch on my computer and look into the voltage rail from the BIOS. the PSU 5V voltage is reported around 4.83V (and the system struggles to detect my HDD when I plug in). But when I try the method suggested by @mariushmand my multimeter read 5.00V, the BIOS immediately reported the 5V rail as 4.98V! It's strange. Does my motherboard had a secret sensor that detects me checking the voltage and recalibrate its report? Or does the test somehow ground the residue of the sensor and now it's working fine? Or could it be that the PSU really provide 4.8V until I test the voltage and somehow discharge some residue current and they function well again? That's strange. This happens the second time (the first time is from the previous post).

I think you're on to something here!  Yes.  Poor ground could be the problem and your DMM could be improving that loop.

 

Set the DMM to ohms.  Keep that negative probe in that connector and put the positive probe on any unpainted metal surface of the PC case or PSU.  Is the reading under 5 Ohms?

 

[Chiyawa]

On 8/14/2022 at 3:38 AM, jonnyGURU said:

Set the DMM to ohms.  Keep that negative probe in that connector and put the positive probe on any unpainted metal surface of the PC case or PSU.  Is the reading under 5 Ohms?

I see. Well, I'll be trying that when I have free time. Thanks for your suggestion.

 

But anyway, should I leave the system on or off to test this?

[Guest]

30 minutes ago, Chiyawa said:

But anyway, should I leave the system on or off to test this?

Doesn't matter.