Jump to content

Arctic Liquid Freezer III

Ralfi
By Ralfi
in Cooling
 Share
Posted

Edit: Updated the chart legend with additional details

 

The chart legend does not include the three 120mm AIO fans as those are there by default (intake pull mode). This was just a way to see if the 5.25" bays with the covers removed improve the temperatures. After that I added two 140mm front fans pushing air through the radiator with the three 120mm fans pulling air through the radiator for improved airflow and ran the same two tests again, this time the temperatures dropped ~2 degrees or so.

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
4 hours ago, Jax-Ur said:

I was unable to find correct screws in any of the hardware stores I visited today so I had to McGyver the 140mm fans in front with zip ties. I ran the test without the front dust cover as it would not fit due to the zip ties.

 

After running Cinebench twice 60 mins with 5.25 bays closed and open here are the results and with two 140mm front pushing air through the radiator (and the three 120mm pulling air through the radiators here are the result. I have split the graphs into two for easier reading.

 

Room temperature ~23.4°C

3x120mm 5.25 COVERED = 5.25" bays covered, three default AIO fans pulling cold air through the radiation

3x120mm 5.25 OPEN = 5.25" bays open, three default AIO fans pulling cold air through the radiation

 

3x120mm + 2x140mm 5.25 COVERED = two 140mm fans in front pushing air through the radiators, 5.25" bays covered, three default AIO fans pulling cold air through the radiation

3x120mm + 2x140mm 5.25 OPEN= two 140mm fans in front pushing air through the radiators, 5.25" bays open, three default AIO fans pulling cold air through the radiation

 

image.thumb.png.b4c70a0199b5b337ce2c8024cdff2602.png

 

 

image.thumb.png.4bfd772486f4154f58e14144fd782d05.png

 

 

Edit: for whatever reason running with the 140mm fans pushing air and the two 5.25" bays closed looks to yield a slightly better result.

Edit2: new charts with hopefully better colors

 

@RevGAM@NorKris@Ralfi

 

Thanks very much for sharing this info! As I thought, there was a minor, 2-3oC improvement using push-pull. I wonder if the fans were all the same, what the difference would be. Certainly, for someone in an area where electricity is very expensive, it's probably not worth that small decrease.

I've been using computers since around 1978, started learning programming in 1980 on Apple IIs, started learning about hardware in 1990, ran a BBS from 1990-95, built my first Windows PC around 2000, taught myself malware removal starting in 2005 (also learned on Bleeping Computer), learned web dev starting in 2017, and I think I can fill a thimble with all that knowledge. 😉 I'm not an expert, which is why I keep investigating the answers that others give to try and improve my knowledge, so feel free to double-check the advice I give.

My phone's auto-correct is named Otto Rong.🤪😂

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
  • Author

Ok so to confirm, you’ve combined a push/pull setup & haven’t done an actual push v pull with the default fans?

^ Thanks for the added testing, though, this was the comparison I meant earlier. Sorry if I wasn’t clear.

https://au.pcpartpicker.com/b/Tw9G3C

 

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
18 hours ago, Ralfi said:

Ok so to confirm, you’ve combined a push/pull setup & haven’t done an actual push v pull with the default fans?

^ Thanks for the added testing, though, this was the comparison I meant earlier. Sorry if I wasn’t clear.

No unfortunately I did not record the original setup values (fans pushing the air from inside case through the radiator), but the temperatures were very, very close. That was of course without stressing the GPU which would have created hot air inside the case -> worse cooling.

 

The radiator is currently "sandwiched" if you will in a pull/push mode, with the 3x120mm fans pulling air and the 2x140mm fans pushing air. Like this one but mounted front in an R5 case and the 360 version.

 

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
  • Author
14 hours ago, Jax-Ur said:

No unfortunately I did not record the original setup values (fans pushing the air from inside case through the radiator), but the temperatures were very, very close. That was of course without stressing the GPU which would have created hot air inside the case -> worse cooling.

 

The radiator is currently "sandwiched" if you will in a pull/push mode, with the 3x120mm fans pulling air and the 2x140mm fans pushing air. Like this one but mounted front in an R5 case and the 360 version.

 

Ok, now I’m curious ( again 😂 ) to know if 3 same sized fans on either side work better than 3 x 120 & 2 x 140 fans.

 

Ideally, I’d want all fans to be the same make/model so the airflow’s more ‘in sync’ & not potentially causing air to get stuck in the sandwiched area.

 

I did a push/pull test with two mismatched 120mm fans on an air cooler & the performance was worse with the two fans compared to a single push fan. Not sure how AIO’s are impacted by fan mismatches though…

https://au.pcpartpicker.com/b/Tw9G3C

 

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
1 hour ago, Ralfi said:

Ok, now I’m curious ( again 😂 ) to know if 3 same sized fans on either side work better than 3 x 120 & 2 x 140 fans.

 

Ideally, I’d want all fans to be the same make/model so the airflow’s more ‘in sync’ & not potentially causing air to get stuck in the sandwiched area.

 

I did a push/pull test with two mismatched 120mm fans on an air cooler & the performance was worse with the two fans compared to a single push fan. Not sure how AIO’s are impacted by fan mismatches though…

From my testing, I can confirm that mismatched fans in push- pull can do worse than using just the stronger model on the push (usually, although there are fans that seemed to do better on pull) side. I put a T30 in tandem with an A12x25, but the T30 on push alone was better. 

I've been using computers since around 1978, started learning programming in 1980 on Apple IIs, started learning about hardware in 1990, ran a BBS from 1990-95, built my first Windows PC around 2000, taught myself malware removal starting in 2005 (also learned on Bleeping Computer), learned web dev starting in 2017, and I think I can fill a thimble with all that knowledge. 😉 I'm not an expert, which is why I keep investigating the answers that others give to try and improve my knowledge, so feel free to double-check the advice I give.

My phone's auto-correct is named Otto Rong.🤪😂

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
On 3/12/2024 at 11:44 AM, Ralfi said:

Ok, now I’m curious ( again 😂 ) to know if 3 same sized fans on either side work better than 3 x 120 & 2 x 140 fans.

 

Ideally, I’d want all fans to be the same make/model so the airflow’s more ‘in sync’ & not potentially causing air to get stuck in the sandwiched area.

 

I did a push/pull test with two mismatched 120mm fans on an air cooler & the performance was worse with the two fans compared to a single push fan. Not sure how AIO’s are impacted by fan mismatches though…

I guess I could (but won't because this works for me now) test that but it would not be a comparable to the other tests as the radiator would have to be outside the case for everything to fit.

 

Looking at the 420 in the video the results should be pretty good though and more importantly you could possibly run the fans at lower RPM to achieve better thermals.

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
On 3/12/2024 at 5:44 AM, Ralfi said:

Ok, now I’m curious ( again 😂 ) to know if 3 same sized fans on either side work better than 3 x 120 & 2 x 140 fans.

 

Ideally, I’d want all fans to be the same make/model so the airflow’s more ‘in sync’ & not potentially causing air to get stuck in the sandwiched area.

 

I did a push/pull test with two mismatched 120mm fans on an air cooler & the performance was worse with the two fans compared to a single push fan. Not sure how AIO’s are impacted by fan mismatches though…

Actually, in addition to the tests I did, STS pointed out the weakness in the Arctic LF3 (one of them, anyways): the fans. He then proceeded (I believe he had the 420) to add 3 more identical fans to the other side and tested it and, lo, the results were better.

I've been using computers since around 1978, started learning programming in 1980 on Apple IIs, started learning about hardware in 1990, ran a BBS from 1990-95, built my first Windows PC around 2000, taught myself malware removal starting in 2005 (also learned on Bleeping Computer), learned web dev starting in 2017, and I think I can fill a thimble with all that knowledge. 😉 I'm not an expert, which is why I keep investigating the answers that others give to try and improve my knowledge, so feel free to double-check the advice I give.

My phone's auto-correct is named Otto Rong.🤪😂

Link to comment
Share on other sites
Link to post
Share on other sites
  • 2 weeks later...
Posted

I was looking at this AIO - as the max. price to pay and only considering this one.   However, there's also a Thermalright Frozen Edge 360 for approx. *half* the price - $73 (for the FE) vs $180 CAD for the Arctic LF III.

 

I imagine the Arctic is the better product - but, I am not 100% sure it's 2x the price better.   Is it?   Is it noisier and the fans aren't as good as the Arctic?

My cpu is a 12700K but I'm just mentioning that - to make sure the Thermalright AIO would be sufficient for it - but, I want something pretty quiet - and if there's a big difference - I might consider the Artic although for double the price, that's a lot to pay?  

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
53 minutes ago, Paul17 said:

I was looking at this AIO - as the max. price to pay and only considering this one.   However, there's also a Thermalright Frozen Edge 360 for approx. *half* the price - $73 (for the FE) vs $180 CAD for the Arctic LF III.

 

I imagine the Arctic is the better product - but, I am not 100% sure it's 2x the price better.   Is it?   Is it noisier and the fans aren't as good as the Arctic?

My cpu is a 12700K but I'm just mentioning that - to make sure the Thermalright AIO would be sufficient for it - but, I want something pretty quiet - and if there's a big difference - I might consider the Artic although for double the price, that's a lot to pay?  

I have the same CPU and I've been running the TR Frozen Notte 360 since shortly after Black Friday. It has a stronger pump than the Edge, I believe. Even with the pump at 100%,I don't hear much although lately there's been a little bit more noise, and sometimes the fans hit the right RPM to produce a harmonic resonance. Doesn't bother me but if you're sensitive...

I've been using computers since around 1978, started learning programming in 1980 on Apple IIs, started learning about hardware in 1990, ran a BBS from 1990-95, built my first Windows PC around 2000, taught myself malware removal starting in 2005 (also learned on Bleeping Computer), learned web dev starting in 2017, and I think I can fill a thimble with all that knowledge. 😉 I'm not an expert, which is why I keep investigating the answers that others give to try and improve my knowledge, so feel free to double-check the advice I give.

My phone's auto-correct is named Otto Rong.🤪😂

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
1 hour ago, Paul17 said:

I was looking at this AIO - as the max. price to pay and only considering this one.   However, there's also a Thermalright Frozen Edge 360 for approx. *half* the price - $73 (for the FE) vs $180 CAD for the Arctic LF III.

 

I imagine the Arctic is the better product - but, I am not 100% sure it's 2x the price better.   Is it?   Is it noisier and the fans aren't as good as the Arctic?

My cpu is a 12700K but I'm just mentioning that - to make sure the Thermalright AIO would be sufficient for it - but, I want something pretty quiet - and if there's a big difference - I might consider the Artic although for double the price, that's a lot to pay?  

I have the Aqua Elite 360 V3, same unit, different fans. It is pretty decent. It does react well to stronger fans, though they are not needed. It can deal with 260w though my 7nm 5900X. You should be ok. It is stronger than my Thermalright air coolers.. I have a few.

AMD R9 5900X | Thermalright Phantom Spirit 120 EVO, T30,TL-C12 Pro
Asus Crosshair VIII Dark Hero | 4x8GB G.Skill Trident Z @ 3733C14 1.5v
Zotac 4070 Ti Trinity OC @ 3045/1496 | WD SN850, SN850X, SN770
Seasonic Vertex GX-1000 | Fractal Torrent Compact RGB, Many CFM's

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
9 minutes ago, freeagent said:

I have the Aqua Elite 360 V3, same unit, different fans. It is pretty decent. It does react well to stronger fans, though they are not needed. It can deal with 260w though my 7nm 5900X. You should be ok. It is stronger than my Thermalright air coolers.. I have a few.

BTW, I forgot to mention what my bro did... the Notte doesn't need stronger fans according to my preliminary testing. 

I've been using computers since around 1978, started learning programming in 1980 on Apple IIs, started learning about hardware in 1990, ran a BBS from 1990-95, built my first Windows PC around 2000, taught myself malware removal starting in 2005 (also learned on Bleeping Computer), learned web dev starting in 2017, and I think I can fill a thimble with all that knowledge. 😉 I'm not an expert, which is why I keep investigating the answers that others give to try and improve my knowledge, so feel free to double-check the advice I give.

My phone's auto-correct is named Otto Rong.🤪😂

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
20 minutes ago, RevGAM said:

the Notte doesn't need stronger fans according to my preliminary testing. 

You are just running at stock clocks right? Should be pretty easy to cool.

AMD R9 5900X | Thermalright Phantom Spirit 120 EVO, T30,TL-C12 Pro
Asus Crosshair VIII Dark Hero | 4x8GB G.Skill Trident Z @ 3733C14 1.5v
Zotac 4070 Ti Trinity OC @ 3045/1496 | WD SN850, SN850X, SN770
Seasonic Vertex GX-1000 | Fractal Torrent Compact RGB, Many CFM's

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
Just now, freeagent said:

You are just running at stock clocks right? Should be pretty easy to cool.

It is. When OC'd, it got hot and performed worse, iirc.

I've been using computers since around 1978, started learning programming in 1980 on Apple IIs, started learning about hardware in 1990, ran a BBS from 1990-95, built my first Windows PC around 2000, taught myself malware removal starting in 2005 (also learned on Bleeping Computer), learned web dev starting in 2017, and I think I can fill a thimble with all that knowledge. 😉 I'm not an expert, which is why I keep investigating the answers that others give to try and improve my knowledge, so feel free to double-check the advice I give.

My phone's auto-correct is named Otto Rong.🤪😂

Link to comment
Share on other sites
Link to post
Share on other sites
Posted

Ok, sorry!

 

If you're having trouble with your VRMs overheating, the VRM fan on this and other AIOs can potentially be crucial.

I've been using computers since around 1978, started learning programming in 1980 on Apple IIs, started learning about hardware in 1990, ran a BBS from 1990-95, built my first Windows PC around 2000, taught myself malware removal starting in 2005 (also learned on Bleeping Computer), learned web dev starting in 2017, and I think I can fill a thimble with all that knowledge. 😉 I'm not an expert, which is why I keep investigating the answers that others give to try and improve my knowledge, so feel free to double-check the advice I give.

My phone's auto-correct is named Otto Rong.🤪😂

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
7 hours ago, Paul17 said:

What type of change?  

Couldn't tell you.. I would imagine a complete redesign somewhere.

AMD R9 5900X | Thermalright Phantom Spirit 120 EVO, T30,TL-C12 Pro
Asus Crosshair VIII Dark Hero | 4x8GB G.Skill Trident Z @ 3733C14 1.5v
Zotac 4070 Ti Trinity OC @ 3045/1496 | WD SN850, SN850X, SN770
Seasonic Vertex GX-1000 | Fractal Torrent Compact RGB, Many CFM's

Link to comment
Share on other sites
Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share
Go to topic listing Cooling

×