Voltage Droop Enough to be Concerned With?

[freeagent]

That unit could use a recap and refurb. That 3.3 is right in the basement with .288v drop. 5v is there too barely hanging in there.. 12v is dropping nearly .045v too.. still has a bit to go but looks pretty weak in the knees.

10 years is a good life.. I would just replace it.

But before that, I would try giving the main lead to the board a jiggle, rock it around a bit in its socket and see if you can get a better bite. That thing is pretty old and and may have seen a lot of use.

 

[lexluthermiester]

If the power architecture is done in a way resembling this one here:
View attachment 215825

then even if the input voltage fluctuates (within certain range) the output voltage should not track the input voltage, or even changing at all. This example is for FPGA, but I guess the general scheme is similar apart for the voltages and which rails the voltages are derived from.

Anyway I never actually measure the behavior of consumer motherboard VRMs. That would be a nice little project to do. It just seems so unintuitive and contrary to common electronic design practices to have VRM output that tracks input voltage outside of certain specific designs.

That graphic answers the question. See all the points where wattage loss is indicated?

 

[freeagent]

That graphic answers the question. See all the points where wattage loss is indicated?

No I just looked at the hwinfo screen.. it took me like 15 minutes to write that post as I try to get my kids to go to sleep.. now I have to go lay with them.

 

[lexluthermiester]

That 3.3 is right in the basement with .288v drop. 5v is there too barely hanging in there.. 12v is dropping nearly .045v too..

Those are all within norms.

 

[LordFarquaad]

You're using software to read hardware values, you will never get a true reading

 

[erek]

That unit could use a recap and refurb. That 3.3 is right in the basement with .288v drop. 5v is there too barely hanging in there.. 12v is dropping nearly .045v too.. still has a bit to go but looks pretty weak in the knees.

10 years is a good life.. I would just replace it.

But before that, I would try giving the main lead to the board a jiggle, rock it around a bit in its socket and see if you can get a better bite. That thing is pretty old and and may have seen a lot of use.

Those are all within norms.

 

[lexluthermiester]

View attachment 215841

Remember...

You're using software to read hardware values, you will never get a true reading

...this. The absolute values are not as important as the variances, which are not a concern. Your vdroops are not significant. Your PSU is fine. Your problem is elsewhere.

 

[ArdWar]

That graphic answers the question. See all the points where wattage loss is indicated?

I still don't get your point. The point of regulator (or VRM) is that even if the input voltage changes, the output doesn't. Of course conservation of energy still hold and input current also changes too to compensate.

In this case, if the VDRAM is regulated, of course it wouldn't change even if the supplied voltage from PSU varies. It can't be used as a sign that the PSU is okay or not.
For example looking at my diagram, the VCCINT being steady at 1.2V doesn't automatically mean the 5V supply being steady too. It might as well 12v, with significant ripple, and it'll still work fine. (with worse efficiency, but that's entirely different discussion)

 

[erek]

Remember...

...this. The absolute values are not as important as the variances, which are not a concern. Your vdroops are not significant. Your PSU is fine. Your problem is elsewhere.

rolled back to Game Ready Driver 471.68 (because 471.96 is when i began noticing the CS:GO random reboots).

also I have a webcam from my previous build on the Intel X99 Platform that was constantly reporting "USB Device Drawing Too Much Power" and causing random system restarts, I removed this from the equation today also.

(i didn't encounter the random restarts with the webcam on my AMD X570 Platform as far as i can tell, unless now)

---

 

[lexluthermiester]

I still don't get your point. The point of regulator (or VRM) is that even if the input voltage changes, the output doesn't. Of course conservation of energy still hold and input current also changes too to compensate.

In this case, if the VDRAM is regulated, of course it wouldn't change even if the supplied voltage from PSU varies. It can't be used as a sign that the PSU is okay or not.

Does that same circuit design apply to the CPU power lines?(yes) And what happens when you put a load on the CPU?(vdroop) Can RAM put a similar heavy load on it's VRM?(no) So can the RAM experience a load-level vdroop?(no) Can RAM experience vdroop at all?(yes) How?

If the PSU is weak and unable to provide the power the system requires under load, then the RAM will experience vdroop along with the rest of the system regardless of the VRM regulation employed. This is basic physics.

So in absense of vdroop on the RAM when there is vdroop on the rest of the system, the PSU is clearly supplying the proper power to the system. The PSU is not the problem. The problem is elsewhere. Could be a motherboard instability, could be a RAM instability, or even the GPU. Whatever the problem, the PSU is unlikely the culprit.

rolled back to Game Ready Driver 471.68 (because 471.96 is when i began noticing the CS:GO random reboots).

also I have a webcam from my previous build on the Intel X99 Platform that was constantly reporting "USB Device Drawing Too Much Power" and causing random system restarts, I removed this from the equation today also.

(i didn't encounter the random restarts with the webcam on my AMD X570 Platform as far as i can tell, unless now)

---

Have you had any crashes since these changes?

 

[erek]

Does that same circuit design apply to the CPU power lines?(yes) And what happens when you put a load on the CPU?(vdroop) Can RAM put a similar heavy load on it's VRM?(no) So can the RAM experience a load-level vdroop?(no) Can RAM experience vdroop at all?(yes) How?

If the PSU is weak and unable to provide the power the system requires unload, then the RAM will experience vdroop along with the rest of the system regardless of the VRM regulation employed. This is basic physics.

So in absense of vdroop on the RAM when there is vdroop on the rest of the system, the PSU is clearly supplying the proper power to the system. The PSU is not the problem. The problem is elsewhere. Could be a motherboard instability, could be a RAM instability, or even the GPU. Whatever the problem, the PSU is unlikely the culprit.


Have you had any crashes since these changes?

not yet, been playing cs:go for a little bit this night... hoping

 

[lexluthermiester]

not yet, been playing cs:go for a little bit this night... hoping

View attachment 215842

The drivers might have been your problem. Fair common problem. Keep us posted..

 

[erek]

The drivers might have been your problem. Fair common problem. Keep us posted..

but can't find any similar reports about it

 

[ArdWar]

If the PSU is weak and unable to provide the power the system requires under load, then the RAM will experience vdroop along with the rest of the system regardless of the VRM regulation employed. This is basic physics.

Not if the PSU weakening is manifested as increase in output impedance, PSU rail's point of load voltage would lower slightly depends on the increase of internal impedance.

PSU like this would not revert to constant power mode where voltage will drop until load with highest voltage disconnects, relieving the load.

So in absense of vdroop on the RAM when there is vdroop on the rest of the system, the PSU is clearly supplying the proper power to the system. The PSU is not the problem. The problem is elsewhere. Could be a motherboard instability, could be a RAM instability, or even the GPU. Whatever the problem, the PSU is unlikely the culprit.

On the contrary with all the regulated voltages steady while almost all of the PSU voltages drooping I would argue that it's the PSU drooping while the regulators doing the job fine. Point of Load regulators will have sense lines that compensate for load-induced voltage droop.

 

[eidairaman1]

so majorly conflicting reports,

be worried or not? ^^^

If you're not freezing up or BSODing, no you should stop worrying about it

 

[erek]

If you're not freezing up or BSODing, no you should stop worrying about it

was getting random reboots w/ BSODs

"
The previous system shutdown at 11:23:08 PM on ‎9/‎5/‎2021 was unexpected.




[11:33 PM]
Dump file creation failed due to error during dump creation.




[11:33 PM]
The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.




[11:33 PM]
Audit events have been dropped by the transport. 0
"

 

[lexluthermiester]

but can't find any similar reports about it

That can happen too.

Not if the PSU weakening is manifested as increase in output impedance, PSU rail's point of load voltage would lower slightly depends on the increase of internal impedance.

PSU like this would not revert to constant power mode where voltage will drop until load with highest voltage disconnects, relieving the load.


On the contrary with all the regulated voltages steady while almost all of the PSU voltages drooping I would argue that it's the PSU drooping while the regulators doing the job fine. Point of Load regulators will have sense lines that compensate for load-induced voltage droop.

I'm not debating this any further. It's been debated to death elsewhere and if you don't understand, you need to do some research and learning.

 

[eidairaman1]

2011 psu was to this standard from 2007.
ATX12V v2.31

ATX - Wikipedia

en.m.wikipedia.org

+5% or -5% are absolute max. PSUs are measured in MTBF(Repair) or MTTF (Replace)

was getting random reboots w/ BSODs

View attachment 215843


"
The previous system shutdown at 11:23:08 PM on ‎9/‎5/‎2021 was unexpected.




[11:33 PM]
Dump file creation failed due to error during dump creation.




[11:33 PM]
The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.




[11:33 PM]
Audit events have been dropped by the transport. 0
"

What Codes were the BSODs?

 

[erek]

2011 psu was to this standard from 2007.
ATX12V v2.31

ATX - Wikipedia

en.m.wikipedia.org

+5% or -5% are absolute max. PSUs are measured in MTBF(Repair) or MTTF (Replace)


What Codes were the BSODs?

happens too quick to tell what the codes are, but that's the same exact repeated pattern in the event log

 

[eidairaman1]

Do you know how to turn off auto restarts of Windows after a crash?

happens too quick to tell what the codes are, but that's the same exact repeated pattern in the event log

 

[erek]

Do you know how to turn off auto restarts of Windows after a crash?

WHEA_UNCORRECTABLE_ERROR

This is while playing Watch_Dogs Legion

@lexluthermiester

Here's what HWiNFO64 Logs look like leading up to the Crash:

 

[lexluthermiester]

WHEA_UNCORRECTABLE_ERROR

Oh now that's a thing! I have seen this when dialing in an OC. Are you OCing anything? Gotta ask.. I've also seen that error when drivers are throwing fits..

This is while playing Watch_Dogs Legion

Does this happen in any other games?

@lexluthermiester

Here's what HWiNFO64 Logs look like leading up to the Crash:

Those charts look ok to me. Seems like you've got a ton of load going on, but you have a 3090 AND a Ryaen 9 5950X, so that is expected.

 

[erek]

Oh now that's a thing! I have seen this when dialing in an OC. Are you OCing anything? Gotta ask.. I've also seen that error when drivers are throwing fits..

Does this happen in any other games?

Those charts look ok to me. Seems like you've got a ton of load going on, but you have a 3090 AND a Ryaen 9 5950X, so that is expected.

all stock, no OC.

no other games,


---

already on the hunt for a new PSU. Struggling because of all the problems with RTX 30 series:

Top Tier PSU:

• Seasonic | Focus [OneSeasonic refreshes] GX / GM / PX / SGX - Prime [8] GX (<=750W / =>1000W) / (Ultra) Gold =>1000W / PX =>1000W / (Ultra) Platinum =>1000W / TX =>1000W / (Ultra) Titanium =>1000W / PRIME Snow Silent Gold [550W] - Connect

[8] Seasonic PRIME based units experience shutdowns with RTX3080/3090 (and possibly RX6900 XT) GPUs, especially ones with unlocked power limit like FE and ASUS Strix. It is recommended, if going with such units, to overprovision wattage, 1kW for RTX3080 and 1.2kW for RTX3090, or power-limit the GPU, or performing a 'dirty' fix of disconnecting a pin of 12V V-sense wire from PSU-side connector of 24-pin motherboard cable (courtesy of Jonny Guru). Units based on post-2018 revisions of Seasonic Focus platform and majority of units by other OEMs are not affected.

 

[lexluthermiester]

all stock, no OC.

no other games,


---

already on the hunt for a new PSU. Struggling because of all the problems with RTX 30 series:

Top Tier PSU:

• Seasonic | Focus [OneSeasonic refreshes] GX / GM / PX / SGX - Prime [8] GX (<=750W / =>1000W) / (Ultra) Gold =>1000W / PX =>1000W / (Ultra) Platinum =>1000W / TX =>1000W / (Ultra) Titanium =>1000W / PRIME Snow Silent Gold [550W] - Connect

[8] Seasonic PRIME based units experience shutdowns with RTX3080/3090 (and possibly RX6900 XT) GPUs, especially ones with unlocked power limit like FE and ASUS Strix. It is recommended, if going with such units, to overprovision wattage, 1kW for RTX3080 and 1.2kW for RTX3090, or power-limit the GPU, or performing a 'dirty' fix of disconnecting a pin of 12V V-sense wire from PSU-side connector of 24-pin motherboard cable (courtesy of Jonny Guru). Units based on post-2018 revisions of Seasonic Focus platform and majority of units by other OEMs are not affected.

My arguments have been to troubleshoot the problem and make a choice based on evidence & merit. I'd hate to see you spend a chunk of money on a new PSU only for the problem to persist after it arrived.

That said, what's your budget? Let's see if we can find you a new PSU that will last you another 10 years!

 

[erek]

My arguments have been to troubleshoot the problem and make a choice based on evidence & merit. I'd hate to see you spend a chunk of money on a new PSU only for the problem to persist after it arrived.

That said, what's your budget? Let's see if we can find you a new PSU that will last you another 10 years!

due to the concerns over the Seasonic units, it's hard to trust much

Tempted to just go with this:

Corsair AX1600i 1600W Digital ATX Power Supply | eBay

Find many great new & used options and get the best deals for Corsair AX1600i 1600W Digital ATX Power Supply at the best online prices at eBay! Free shipping for many products!

www.ebay.com