Interpreting the data from Corsair Link, where is the standard?

Hi guys,

First of all, Merry Christmas to all!

We have a tower computer a core 10 with two video cards running Windows 10 64, nothing that fancy, just used for standard coding and internet services.

Anyway, we are experiencing for some time some freezes once in a while, we notice that after a cleaning the freezes stopped, now we are getting the freezes again, but these is nothing to clean really, all fans are good and all.

We have a Corsair Cooling system RM550X (H80i v2 Liquid Coler), and on the Corsair Link App (Firmware Ver. 2.8.0.0) we have several measurements, from all sort of temperatures and fan speeds.

Everything is red... so it must be the app standard color. What we want to know, is what is the standard temperatures, the SAFE temperatures that each particular area should be working under. It is no good if we have all this data but we can't make sense of it.

We appreciate very much any kind knowledgeable contributor that can help us understand what we are looking at, and perhaps help this way identify if there is something getting too hot.

Much appreciated for any input.

(P.S.: We have attached a print screen of the app as we see it here. Thx)

First things first ... your pump isn't being powered properly. The pump speed is two low and the fans are at zero. That's not good.

Please review section "B" of the Liquid Cooler FAQ linked in my signature.

First things first ... your pump isn't being powered properly. The pump speed is two low and the fans are at zero. That's not good.

 

Please review section "B" of the Liquid Cooler FAQ linked in my signature.

HI DevBiker,

Thank you very much for you reply!

I am going to check this information now, much appreciated!

I will let you know how it goes later, however the fans are shown as operational and we have speed for them...

Thank you

 

I will let you know how it goes later, however the fans are shown as operational and we have speed for them...

 

Thank you

That’s not the problem. You are cutting voltage to the cooler by using the BIOS fan control on the cooler. You need to control the cooler through the application. This is the equivalent of turning a dimmer switch to minimum on a ceiling fan, then wondering my the pull chain doesn’t change the speed. Set the motherboard header powering the cooler to 100%/Full Speed or disabled which also locks the fan control at 100%. You can then set the fans as low as you like with the Link/iCUE app and the pump has a chance at a longer life.

First things first ... your pump isn't being powered properly. The pump speed is two low and the fans are at zero. That's not good.

 

Please review section "B" of the Liquid Cooler FAQ linked in my signature.

That’s not the problem. You are cutting voltage to the cooler by using the BIOS fan control on the cooler. You need to control the cooler through the application. This is the equivalent of turning a dimmer switch to minimum on a ceiling fan, then wondering my the pull chain doesn’t change the speed. Set the motherboard header powering the cooler to 100%/Full Speed or disabled which also locks the fan control at 100%. You can then set the fans as low as you like with the Link/iCUE app and the pump has a chance at a longer life.

Hi Guys,

First of all, thank you for helping and happy holidays to all!

So we double checked the connections, and the connector seemed to be in the right place, there is a cable from the corsair control on the top of the processors going to the CPUFAN, and it's connected in the 3 right pins (CPU FAN IN/PWR/GNR).

We then went to the BIOS, Voltage control and found the CPU-FAN to be set for 'AUTO', we changed that to 100%, Manual? not sure. However as soon as we reset the PC we could hear a fan running. Now looking at the Corsair Link we can see a fan running.

Now, in the back of the PC, there is a mini radiator, with an fan inside and another outside, the one outside is running now, but the one inside is not, I am not sure if if comes one when it's hot...

Please check the pictures and we appreciate if you can guide us further.

Regards

Below are some pictures:

Hi Guys,

We can hear a fan now, not sure if it was working before...

Anyway, the machine had a temperature meter that now no longer shows temperature, it shows a code which means 'NOT USED'.

Also there is a fan inside the mini radiator that is not working, and another fan outside is... should both be working?

Thank you

I can't speak to the Code 00 - you'll have to ask Asus about that.

And the second fan should be running. Check to make sure that it's connected to the fan header on the pump correctly.

Also, on Asus motherboards, it's best to just set Q-FAN to disabled for the CPU Fan Header. Otherwise, it's possible that AISuite would override your BIOS settings.

I can't speak to the Code 00 - you'll have to ask Asus about that.

And the second fan should be running. Check to make sure that it's connected to the fan header on the pump correctly.

 

Also, on Asus motherboards, it's best to just set Q-FAN to disabled for the CPU Fan Header. Otherwise, it's possible that AISuite would override your BIOS settings.

Hi DevBiker,

Thank you for your support.

I will be looking into the points you mention.

Also, Can you tell me if the temperatures shown on the screenshots above overall are correct?

Thank you very much

I can't speak to the Code 00 - you'll have to ask Asus about that.

And the second fan should be running. Check to make sure that it's connected to the fan header on the pump correctly.

 

Also, on Asus motherboards, it's best to just set Q-FAN to disabled for the CPU Fan Header. Otherwise, it's possible that AISuite would override your BIOS settings.

Hi There,

We reset the PC again, and double checked all connections, and the 00 code is gone... check the picture below, are those numbers good or bad?

We will now try to find why the corsair fan is not working.... (WORKING NOW!)

UPDATE:

We got the fan working! We found a connection behind a panel on the back of the tower that somehow was disconnected, probably when we pulled the fan out to look check for the other FAN connection.

ALL fans are working now!!!!

SO what we need to understand now, is what the measurements (values and safe levels) mean on the Corsair Link, and perhaps set some alarm warnings for a hot system, can you guys help out?

Much appreciated!

Thx

Where can I obtain the standard and instructions to use Corsair Link?

After some struggle I finally got the machine ready, now I need to setup Corsair Link correctly, where can I obtain guidance on this regard?

Website, documents, standard values... anything will help.

Thank you and happy new Year to all!:biggrin:

Hi there ... the temps above look absolutely fine.

One note - Link is End of Life; there will be no more updates. iCUE is the tool moving forward. You don't need it, at this point, as all of your hardware is supported in Link.

Hi there ... the temps above look absolutely fine.

 

One note - Link is End of Life; there will be no more updates. iCUE is the tool moving forward. You don't need it, at this point, as all of your hardware is supported in Link.

HI DevBiker,

Thank you for coming back to me.

Ok, understood, never the less it would be good to know what is the limit temperature for example for the processors and other things so I could set alarms in case something stopped working...

Happy new year!

Thermal shutdown on a 7900X should be at 95C, although you really don't want your 10 core in that neighborhood on a regular basis. It is somewhat subjective, but I would consider 80C on the package temp a general limitation. Your motherboard will do what I needs to do in order to protect the CPU for its temperature. With the H80i v2 or any other cooler, it really manages the coolant temperature and any notifications and alarms will be linked to that value. In Link, coolant temperature is often referred to as H80i v2 Temp and is displayed on the right side of the cooler device page. This value is somewhat unique to your environment and hardware, which also means it is going to change with room temperature if you have seasonal changes and certainly with case temp changes, although that may not be quite so much of an issue with this build.

Your baseline coolant temperature appears to be around 30C from the photos above. That is not really unexpected for a 10 core on a 120mm radiator. You will need to see where your normal maximum coolant temp lands, but I'll guess and say about 40-42C after some long workloads. I think the logical place to set coolant temp alarms is going to be 50C, about the same place I suggest everyone set them. When it's a hot day, you run extra hard, or some other minor variable change, the effect on coolant temperature is going to be small -- likely only a few degrees more or less. If you pump fails to start, malfunctions, or something of that sort, it is going to go 30-35-40-45-50-55 and up right in front of your eyes in about 20-30 seconds. If your coolant hits 50C, you need to see what's going on. I don't know that I would use a shutdown protocol for 50C. At 50C coolant temp you may still be under TJ Max for the CPU and any end CPU temps in that range will cause a shutdown anyway. The cooling system will not fail at 50C, so the goal is still to protect the CPU rather than the cooling system itself. Using the notification system in Link or iCUE is probably sufficient, but that is for you to decide.

Thermal shutdown on a 7900X should be at 95C, although you really don't want your 10 core in that neighborhood on a regular basis. It is somewhat subjective, but I would consider 80C on the package temp a general limitation. Your motherboard will do what I needs to do in order to protect the CPU for its temperature. With the H80i v2 or any other cooler, it really manages the coolant temperature and any notifications and alarms will be linked to that value. In Link, coolant temperature is often referred to as H80i v2 Temp and is displayed on the right side of the cooler device page. This value is somewhat unique to your environment and hardware, which also means it is going to change with room temperature if you have seasonal changes and certainly with case temp changes, although that may not be quite so much of an issue with this build.

 

Your baseline coolant temperature appears to be around 30C from the photos above. That is not really unexpected for a 10 core on a 120mm radiator. You will need to see where your normal maximum coolant temp lands, but I'll guess and say about 40-42C after some long workloads. I think the logical place to set coolant temp alarms is going to be 50C, about the same place I suggest everyone set them. When it's a hot day, you run extra hard, or some other minor variable change, the effect on coolant temperature is going to be small -- likely only a few degrees more or less. If you pump fails to start, malfunctions, or something of that sort, it is going to go 30-35-40-45-50-55 and up right in front of your eyes in about 20-30 seconds. If your coolant hits 50C, you need to see what's going on. I don't know that I would use a shutdown protocol for 50C. At 50C coolant temp you may still be under TJ Max for the CPU and any end CPU temps in that range will cause a shutdown anyway. The cooling system will not fail at 50C, so the goal is still to protect the CPU rather than the cooling system itself. Using the notification system in Link or iCUE is probably sufficient, but that is for you to decide.

Hi CAttack,

Thank you very much for the information provided! Now, It definitely gives us some understanding and parameters to start working.

Much Appreciated!

Happy new year!