Corsair RM850X (2018) Loud Click/Noise upon Shutdown

[boykale]

Hello. I just built my PC and everthing seems fine now. However I realized that a click sound is coming from PSU upon shutdown. Here is a sound of video which is exactly same with mine.

Is this normal? I experienced something like this first time.

https://youtu.be/_Pl6guNalHw?t=14

 

Also when I start my PC for the first time, it made similar sound and looked like starting but then powered off like I changed somethings in BIOS and then powered on again with a screen indicating You have changed CPU or memory please go BIOS. But then everything is working great. Is this also normal?

 

PC SPEC
GeForce RTX 2080 Duke 8G OC
Intel Core i7-9700K
MSI MPG Z390 Gaming Plus
Gskill 16 GB (2x8GB) SniperX DDR4 3200MHz CL16 1.35V

[LukeSavenije]

@jonnyGURU @Corsair Nick I think this one is going to be fun for you...

[Princess Luna]

As far as I am concerned that's just the relay switch and that's perfectly normal.

[Spotty]

Yep, completely normal, all RMx units will make that noise when it's switched off. I believe the click noise is a physical relay switch changing to 5VSB power when the system is turned off. You'll likely hear the same click when you push the power button and turn the unit back on and it switches back.

[bleu]

Thanks for making this thread. I was also a little worried about the loud popping coming from my Corsair rmx 750w during shutdown.

[OrionFOTL]

1 hour ago, Spotty said:

I believe the click noise is a physical relay switch changing to 5VSB power when the system is turned off.

It's disconnecting the NTC thermistor, but yeah.

[Guest]

1 hour ago, LukeSavenije said:

@jonnyGURU @Corsair Nick I think this one is going to be fun for you...

Why?  What's the problem here?  Seems normal to me.

 

[LukeSavenije]

12 minutes ago, jonnyGURU said:

Why?  What's the problem here?  Seems normal to me.

 

i wasn't sure, and since you kinda made that thing...

[Spotty]

1 minute ago, OrionFOTL said:

It's disconnecting the NTC thermistor, but yeah.

So the click is not for 5VSB instead it's a relay for the NTC Thermistor? *Googles what the heck a NTC Thermistor is*

 

Spoiler

 

Quote

 

Summary

Many applications today, including industrial machinery, power tools and other high current equipment, use limiting inrush current as a major design consideration to combat the problematic effects of inrush current. Inrush current occurs when a system powers on and experiences a spike in current. This current can be substantially higher than standard operating current. If not properly managed, it can reduce the effective operating life and impose damage to equipment. For example, inrush current could disable a cooling fan, eventually leading to total system failure.

 

Applications that are switched on and off quickly, such as welding equipment, present a particular concern for limiting inrush current. The limiting inrush current circuit must reset instantaneously during each power on to protect the system. This further complicates the management of inrush current.

 

Inrush Current Overview

During power on, a high inrush current can occur because the power supply’s link capacitor functions to dampen ripples in the output current. This capacitor acts like a short, causing an inrush of current. The inrush lasts until the capacitor is charged. Length of the inrush current depends upon the power supply and link capacitor.

The low internal resistance of the power supply aggravates this issue. Any resistance in the power supply introduces inefficiencies through heat. To minimize resistance, engineers typically use an inductive load. While this improves the overall operating efficiency of the power supply, the lack of resistance enables the inrush current to pass through to the main system when the power supply switches on.

Temporarily introducing a high resistance between the power supply and system at power on limits inrush current. The resistance switches out when the initial current surge at power on reaches completion.

 

NTC-based Limiting

For many systems, a negative temperature coefficient (NTC) thermistor can effectively limit inrush current. An NTC thermistor provides variable resistance based on its temperature. Placing an NTC thermistor between the power supply and system limits inrush current (see Figure 1). At first, the initial temperature of the NTC thermistor is low, providing high resistance. When the system is powered on, it energizes the NTC thermistor, causing the temperature to rise, and thus lowering resistance. As resistance drops to a low value, the current passes through without adversely affecting normal operation or power efficiency.

 

https://www.ametherm.com/inrush-current/ptc-thermistors-for-inrush-current-limiting

 

So principal of a NTC thermistor is to add resistance to a circuit and then as it heats up the resistance lowers. In PSUs it's used as a way to protect against inrush current when the unit is switched on... Is that correct? 
Also, why does it get disconnected when the unit is switched off? Is it always connected when the PSU is on? Any photos of one on a PSU so I know what I'm looking at?

(I'm certain people will ask about this again, so I just want to make sure I have the right answer for next time.)

[OrionFOTL]

32 minutes ago, Spotty said:

In PSUs it's used as a way to protect against inrush current when the unit is switched on... Is that correct? 
Also, why does it get disconnected when the unit is switched off? Is it always connected when the PSU is on? Any photos of one on a PSU so I know what I'm looking at?

Yes, an NTC thermistor is used to limit inrush current when the PSU starts and main caps charge. After the startup phase it doesn't serve any purpose and needlessly consumes some energy, decreasing overall efficiency. So some PSUs also use an electromechanical relay, which disconnects the thermistor after the PSU starts, and reconnects it again when the PSU shuts down, so the thermistor can serve its role during the next time the unit starts.

 

I kinda wrote my post backwards, should've said it's reconnecting the thermistor on shutdown.

Here's one in Corsair RM850x 2018. Green is the thermistor, black cuboid is its relay.

 

[Spotty]

Just now, OrionFOTL said:

Yes, and NTC thermistor is used to limit inrush current when the PSU starts and main caps charge. After the startup phase it doesn't serve any purpose and needlessly consumes some energy, decreasing overall efficiency. So some PSUs also use an electromechanical relay, which disconnects the thermistor after the PSU starts, and reconnects it again when the PSU shuts down, so the thermistor can serve its role during the next time the unit starts.

 

I kinda wrote my post backwards, should've said it's reconnecting the thermistor on shutdown.

Here's one in Corsair RM850x 2018. Green is the thermistor, black cuboid is its relay.

I was confused trying to figure out why it would need a relay to switch it off when the PSU was switched off, but that makes a lot more sense if it's off when the unit is running and reconnects when it shuts down.

Exactly what I needed to know. Thanks, I appreciate it.

[Rexper]

9 hours ago, Spotty said:

it's a relay for the NTC Thermistor?

It's resistance decreases the higher the temperature. The more it resists, the more it builds temperature.

Basically, it creates a softer start, as opposed to a short, high burst of inrush current.

 

9 hours ago, OrionFOTL said:

needlessly consumes some energy, decreasing overall efficiency.

The relay is also used so the NTC thermistor works in a hot start.

Disconnecting the thermistor while the PSU is operating allows it to cool down.

[Frenky]

My Be Quiet Straight Power 11 makes the same noise 

[Stefan Payne]

So we're talking about the noise of an electro mechanical switch again??
Why doesn't Corsair just write on the Label "this device contains an electromechanical switch. Noise that sounds like a mechanical switch switching is normal and intended" @jonnyGURU

[OrionFOTL]

7 hours ago, Stefan Payne said:

Why doesn't Corsair just write on the Label "this device contains an electromechanical switch. Noise that sounds like a mechanical switch switching is normal and intended" @jonnyGURU

I think it's because clueless people will then buy non-Corsair units without such label, not knowing they can still have the switch inside.

[comicsansms]

53 minutes ago, OrionFOTL said:

I think it's because clueless people will then buy non-Corsair units without such label, not knowing they can still have the switch inside.

Have it written on a sheet of paper in the box of the unit. Doesn't need to be on the product listing.

[Stefan Payne]

7 hours ago, OrionFOTL said:

I think it's because clueless people will then buy non-Corsair units without such label, not knowing they can still have the switch inside.

Naa, if you're good at marketing you can sell it as a plus, like not spinning fans *scnr*

 

 

You know, like Enermax back in the day, when they sold standard feature as their own in Marketingspeech...

 

You could claim that its a switch to reduce the inrush current, something like that.

 

[Ankerson]

15 hours ago, Stefan Payne said:

So we're talking about the noise of an electro mechanical switch again??
Why doesn't Corsair just write on the Label "this device contains an electromechanical switch. Noise that sounds like a mechanical switch switching is normal and intended" @jonnyGURU

 

I agree, maybe they need to make it known so the end users can see it and know about it so they don't panic.

 

As most people know any NEW sounds in a PC is not normally a good thing.

 

So making it known would be good.

[Stefan Payne]

but on the other hand I know that nobody reads a red DIN A5 piece of paper IN RED with WHITE Letters that tells you to NOT use the drive Cables of the old PSU...

[Ankerson]

4 minutes ago, Stefan Payne said:

but on the other hand I know that nobody reads a red DIN A5 piece of paper IN RED with WHITE Letters that tells you to NOT use the drive Cables of the old PSU...

 

And there is always that. ROFL 

 

I laugh every time I see someone do it.

 

I just don't get people these days. 

 

I think they are getting dumber.

 

PCs today are ridiculously easy to build compared to the old days you know. It's almost impossible to screw it up, but they still do....