Console hardware hacks (such as improving cooling) *spawn

[orangpelupa]

I replaced the stock thermal paste on my Pro with liquid metal and cut an intake hole in the top cover above the fan. It made a noticable difference but it's still loud as hell. I don't play it much anymore or I would have probably added a 120-140mm fan to the other side. The stock heatsink is a joke. I'm not paying retail for a PS5 unless the cooling solution is solid. I hope the XBX shape rubs off on them.

what fan do you get? Pop the top cover off, if it have 12 segments (the one circling the center of the fan, before the blades), its DELTA and its the loud one. Replace it with a Nidec (8 segments). If you already have nidec... i have no idea how can it still be loud. Did you also replaced the thermal pads?
do we have a thread talking about making PS4 (and pro) quiet?

 

[Shifty Geezer]

do we have a thread talking about making PS4 (and pro) quiet?

We do now.

 

[Deleted member 11852]

do we have a thread talking about making PS4 (and pro) quiet?

Disassembling and putting some decent thermal paste in there will make massive difference, as will clearing out the dust that inevitably builds up over time. It's never going to Xbox One-level quiet but you can bring to noise level down to barely-imperceptible in an average room.

 

[orangpelupa]

Disassembling and putting some decent thermal paste in there will make massive difference, as will clearing out the dust that inevitably builds up over time. It's never going to Xbox One-level quiet but you can bring to noise level down to barely-imperceptible in an average room.

Yeah it's crazy how horrible the stock TIM are. Btw I used 3mm thermal pad instead of 2mm that are usually recommended by YouTube videos and people on social media.

 

[Deleted member 11852]

Yeah it's crazy how horrible the stock TIM are. Btw I used 3mm thermal pad instead of 2mm that are usually recommended by YouTube videos and people on social media.

The thicker you go, the less thermal conductivity your have. I.e, the harder it is to dissipate heat from the core through the pad.

 

[orangpelupa]

The thicker you go, the less thermal conductivity your have. I.e, the harder it is to dissipate heat from the core through the pad.

That's the opposite of the test result

"There is also a cheaper trick to fix the temperatures – increase the thickness of the pad. Temperatures with Arctic 1.5 mm are similar to those with TG 1 mm. This is obviously due to a combination of several factors. After compression, there is more material (with higher concentration/absorption) for the heat transfer, the MOSFETs are better “submerged“ in the pad, which leads to heat transfer from the edges too, and also more pressure plays its role."

https://www.hwcooling.net/en/test-of-chewing-gums-3-arctic-and-thermal-grizzly-minus-pads/3/

 

[shiznit]

what fan do you get? Pop the top cover off, if it have 12 segments (the one circling the center of the fan, before the blades), its DELTA and its the loud one. Replace it with a Nidec (8 segments). If you already have nidec... i have no idea how can it still be loud. Did you also replaced the thermal pads?
do we have a thread talking about making PS4 (and pro) quiet?

I have one of the original batch Pros so it's definitely the loud one. I checked Ebay for fan swap parts but they weren't cheap. I removed the original TIM, applied clear nail polish to the components around the SoC, and used one tiny drop of Condoctonaut on the SoC. It also helps to add what's left on the cue tip to the heatsink contact area. I also ran it with the top over off but I found a cheap spare on Ebay and cut a hole around the fan area.

Honestly, the $20 or so I spent on one tube of Condoctonaut has been a great investment. My NUC is quiter, my 1080 Ti overclocks better, my Macbook is quieter and doesn't throttle as much, and I tamed the Pro somewhat.

 

[Shifty Geezer]

applied clear nail polish to the components around the SoC,

Why?

 

[MrFox]

Liquid metal short circuits everything and disintegrate aluminum on contact because of it's gallium content. So it needs some conformal coating everywhere around it.

 

[orangpelupa]

I have one of the original batch Pros so it's definitely the loud one. I checked Ebay for fan swap parts but they weren't cheap. I removed the original TIM, applied clear nail polish to the components around the SoC, and used one tiny drop of Condoctonaut on the SoC. It also helps to add what's left on the cue tip to the heatsink contact area. I also ran it with the top over off but I found a cheap spare on Ebay and cut a hole around the fan area.

Honestly, the $20 or so I spent on one tube of Condoctonaut has been a great investment. My NUC is quiter, my 1080 Ti overclocks better, my Macbook is quieter and doesn't throttle as much, and I tamed the Pro somewhat.

My pro is also the original batch. I bought it at launch.

But Sony use 2 fan manufacturers, nidec and delta. I was lucky, I got nidec, it's the quiet one. Despite that it sounds like a jet engine until I replaced the TIM (paste and pads)

That's why I'm curious with your fan. Maybe you got a delta.

Btw did you also replaced the thermal pads?

 

[shiznit]

I did this two years ago and I don't currently have the console, but I'm pretty sure I counted the fins and I have a Delta. I don't remember what I did with the GDDR5 thermal pads.

Liquid metal is probably overkill vs a proper amount of regular thermal paste in most applications. But for devices like consoles and laptops where changing the TIM it's often the only thing you can do, the incremental gain seemed worth it. I don't want to dismantle these devices more than once.

Here's a Macbook mod video for those interested, you probably don't need liquid metal on the L4 and southbridge but I'm not sure how it would interact with adjacent thermal paste.

 

[MrFox]

Digital Foundry reported that the second revision (late 2017, early 2018) was quiet, including the ones they personally own. There was no significant hardware change so the first batch problem isn't about the heatsink or fan size. It could be a silicon revision, or mechanical tolerances on the surface mating.

 

[orangpelupa]

Yeah the delta fan was simply loud, according to various testimonials on reddit.

Digital Foundry reported that the second revision (late 2017, early 2018) was quiet, including the ones they personally own. There was no significant hardware change so the first batch problem isn't about the heatsink or fan size. It could be a silicon revision, or mechanical tolerances on the surface mating.

Aren't they also use tiny bit less watt, and less aggressive dan curve? They got higher running temperature

 

[MrFox]

Aren't they also use tiny bit less watt, and less aggressive dan curve? They got higher running temperature

Doesn't seem to be much less, I think I remember DF tested it? I can't test this as I only have the launch one, it's difficult to compare apples to apples on the internet without having physically access to the two brand new specimens.

The fan curve must take into account many many sensors inside the chip and around the enclosure, they can go conservative or more agressive based on the amount of data they have. Launching in 2016 was quite an agressive move. It's possible there was a hotspot or something else causing one of the internal sensors to make the fan spin, it takes only one anywhere. Or it's possible the silicon was tested more later and they had the data needed to removed some margins on the revision. Hard to tell.

The launch batch of PS4 had horrible heatsink compound but it didn't seem to be the case on ps4 pro (not on mine anyway, it was perfectly applied). It's certainly can't hurt to change it and many said it helped.

 

[eastmen]

I put an active heatsink on my intellivision's Cp1610.. don't hate guys its the sickest mod.

On the more serious side about 3 years ago I repasted my dreamcast heatsinks and put in a new fan. Its virtually silent now