Production FormD T1 Classic (READ FIRST POST)

[xkruz]

Ok these are the result

Alphacool Eisbaer LT Solo @100% with TX240 Radiator



Syscooling ddc @50% with TX240 (I set it to 50% since I wouldn't run it above 50% in the case


Test setup:
Asus x570i
Ryzen 3950x
Alphacool eisbaer lt solo block
Even with ddc pump, the block was still eisbaer lt solo
Hspc tx240 radiator
2x slim noctua 12x15mm
1x artic p12 fan for test no3
Blender run 30min, thanks to @fabio who teach me how to do blender render
Ambient was 23°C
The fan set to 70% on noctua 12x15mm
And 80% on arctic p12

Based on average temp
Alphacool eisbaer solo lt pump block are 68.1°C
Same block but with ddc pump 64.4°C
With 2° difference on max temp

I think flow rate DO affect cooling performance, BUT after passing the minimum flow rate required in the loop, the difference will be smaller.
I don't know exact number of minimum flow rate since I don't have a flow meter, but as I observed in the reservoir the alphacool pump barely make a splash when compared to the ddc.

Bonus:
Syscooling ddc @50% with TX240 but with extra 1 arctic p12 25mm fan
Though its only 20min, the temp are stabilized IMO


Push pull config with 1 extra arctic p12 fan

Need to do more test on lower rpm etc..

 

[Luke NK]

I hope someone who has the T1 is interested in having EK's AIO and test temps with it, plenty of tests with different rads and pumps I'm wondering what temp differences might be with an AIO (if any!)

 

[fabio]

Ok these are the result

Alphacool Eisbaer LT Solo @100% with TX240 Radiator



Syscooling ddc @50% with TX240 (I set it to 50% since I wouldn't run it above 50% in the case


Test setup:
Asus x570i
Ryzen 3950x
Alphacool eisbaer lt solo block
Even with ddc pump, the block was still eisbaer lt solo
Hspc tx240 radiator
2x slim noctua 12x15mm
1x artic p12 fan for test no3
Blender run 30min, thanks to @fabio who teach me how to do blender render
Ambient was 23°C
The fan set to 70% on noctua 12x15mm
And 80% on arctic p12

Based on average temp
Alphacool eisbaer solo lt pump block are 68.1°C
Same block but with ddc pump 64.4°C
With 2° difference on max temp

I think flow rate DO affect cooling performance, BUT after passing the minimum flow rate required in the loop, the difference will be smaller.
I don't know exact number of minimum flow rate since I don't have a flow meter, but as I observed in the reservoir the alphacool pump barely make a splash when compared to the ddc.

Bonus:
Syscooling ddc @50% with TX240 but with extra 1 arctic p12 25mm fan
Though its only 20min, the temp are stabilized IMO


Push pull config with 1 extra arctic p12 fan

Need to do more test on lower rpm etc..

Nice!!! How is setup the CPU? I mean, Stock, Auto, PBO Auto or Enabled.

You can try to enable the PBO in auto, and change in Ryzen master the value you want to change.

You should try 1slim only and 1 fan 25mm. The PushPull I think will be the same.

BTW, when you say "same block, ddc pump" means you left the eisbaer lt solo like block and having it off?

 

[xkruz]

@fabio the cpu was on stock, pbo auto in the bios

Same block means I just turn off the pump on the eisbaer lt solo and turn on the ddc without changing the cpu block.

I will try later with 1x slim 1x 25mm, and try lower rpm with push pull config

 

[DrHudacris]

I did the @fabio benchmark!

8086k 4.8Ghz all core at 1.11V static, but gets 1.28V for some reason (about 130 watt power draw)
TX240
2x A12x15 (1500RPM) + 1x A12x25 (1200RPM)
GPX pro solo @2200 RPM


CPU max at 70C, coolant temp at 39C. My room temp is 20.5C. @fabio are your ambient sensors inside the case? Because the delta for me is 18.5C ?

 

[fabio]

I did the @fabio benchmark!

8086k 4.8Ghz all core at 1.11V static, but gets 1.28V for some reason (about 130 watt power draw)
TX240
2x A12x15 (1500RPM) + 1x A12x25 (1200RPM)
GPX pro solo @2200 RPM


CPU max at 70C, coolant temp at 39C. My room temp is 20.5C. @fabio are your ambient sensors inside the case? Because the delta for me is 18.5C ?

No, I have two, one on the back/outside and one outside near the monitor base.

My delta is based on the last one because the back one get a bit hotter under load and I didn't want to have the delta based on this.

So Coolant Temp-Ambient Exterior

The delta is ok, maybe 12/15 could be better. I am wondering the reason for it.

But the general temp are really good!

 

[DrHudacris]

No, I have two, one on the back/outside and one outside near the monitor base.

My delta is based on the last one because the back one get a bit hotter under load and I didn't want to have the delta based on this.

So Coolant Temp-Ambient Exterior

The delta is ok, maybe 12/15 could be better. I am wondering the reason for it.

But the general temp are really good!

I might have a defective thermistor or something, because at idle I see coolant temps at 32C

 

[fabio]

I might have a defective thermistor or something, because at idle I see coolant temps at 32C

I've doubled check mine. My internal MPS sensor say 32, but I've place a thermometer inside the reservoir, and under load the temp is 38C, so 13/14 of delta.


I've added a constant of 6.5C to the coolant temp inside the aquasuite, so now it's showed properly!


I think to have another confirmation I think I will try another sensor type, an inline of alphacool maybe. The AquaComputer one gave me the same lower temp.

At Idle I have as well 27/30. I think is just mine!

Good to know it!
But curious how 16core+Titan RTX put the same amount of heat in the water like your GPU and 6 core CPU!

 

[SpaceCowboy]

Hey everyone, been lurking following this project for a while not wanting to clog up the discussion, but need to post to fill out the survey.

Excited to get my hands on this case!

 

[DrHudacris]

I've doubled check mine. My internal MPS sensor say 32, but I've place a thermometer inside the reservoir, and under load the temp is 38C, so 13/14 of delta.


I've added a constant of 6.5C to the coolant temp inside the aquasuite, so now it's showed properly!


I think to have another confirmation I think I will try another sensor type, an inline of alphacool maybe. The AquaComputer one gave me the same lower temp.

At Idle I have as well 27/30. I think is just mine!

Good to know it!
But curious how 16core+Titan RTX put the same amount of heat in the water like your GPU and 6 core CPU!

Oh interesting! I'm using an inline phobya fitting.

Just to clarify, I only stressed CPU here, no GPU load on my test. Are you stressing both at the same time in your earlier post? Or just CPU test with blender and later GPU test with heaven?

Even my little 6 core can pull MASSIVE wattage. Before I decided to undervolt, I had all core 5.1GHz easily 190 watt with AVX workload.

 

[fabio]

Oh interesting! I'm using an inline phobya fitting.

Just to clarify, I only stressed CPU here, no GPU load on my test. Are you stressing both at the same time in your earlier post? Or just CPU test with blender and later GPU test with heaven?

Even my little 6 core can pull MASSIVE wattage. Before I decided to undervolt, I had all core 5.1GHz easily 190 watt with AVX workload.

I do first the CPU with blende (that is also an AVX/AVX2 workload) and after Heaven, that use a 1/2 threads and the GPU. These are real world scenario test, that's why I do them!

I should do a test during gaming though

BTW, in idle the temp is correct like in the sensor. I will add an inline one next week.

But definitively this is a great build!

 

[Llamakob]

As many other been a long time lurker for this project. Just now realized that I have to post to enter the survey and I can’t remember if I posted something on reddit. Anyways great project wahaha and good luck with it.

 

[Cruella Devill]

EK-CoolStream Classic SE 240

EK-CoolStream Classic SE 240 slim copper radiators are designed for optimal airflow throughout the radiator surface with medium density fin structure. Built for EK-Vardar 120mm high static pressure fans, these radiators are optimized for maximum heat dissipation across entire fan operational...

www.ekwb.com

EK-CoolStream SE 240 (Slim Dual)

EK-CoolStream SE is a slim, high-performance computer liquid cooling radiator which combines EK's unique CSQ design with latest radiator core engine. Best cooling capacity in the 28mm thick radiator class!

www.ekwb.com

Please tell me if there is a difference in real tests?
Thanks.

 

[DrHudacris]

I do first the CPU with blende (that is also an AVX/AVX2 workload) and after Heaven, that use a 1/2 threads and the GPU. These are real world scenario test, that's why I do them!

I should do a test during gaming though

BTW, in idle the temp is correct like in the sensor. I will add an inline one next week.

But definitively this is a great build!

Ahh, without a cool down period between? I will try this! Still, your temps are great and mine are not! Very strange

 

[aexeq]

EK-CoolStream Classic SE 240

EK-CoolStream Classic SE 240 slim copper radiators are designed for optimal airflow throughout the radiator surface with medium density fin structure. Built for EK-Vardar 120mm high static pressure fans, these radiators are optimized for maximum heat dissipation across entire fan operational...

www.ekwb.com

EK-CoolStream SE 240 (Slim Dual)

EK-CoolStream SE is a slim, high-performance computer liquid cooling radiator which combines EK's unique CSQ design with latest radiator core engine. Best cooling capacity in the 28mm thick radiator class!

www.ekwb.com

Please tell me if there is a difference in real tests?
Thanks.

You should just use the classic, as the temperature will be within margain of error and I am not sure if the non classic will even fit the T1.

@DrHudacris is your 8086k delidded? This actually makes a huge difference with this CPU.

@fabio as far as I know, the EK Magnitude on Ryzen 3000 is one of the, if not the best water block, because it was specially made to account for Ryzen 3000 dye offset from the middle. If you want to compare the Eisbear LT block on Ryzen 3000, you should test it against an EK Velocity or other water blocks, that are not optimized for Ryzen 3000.

 

[fabio]

Ahh, without a cool down period between? I will try this! Still, your temps are great and mine are not! Very strange

I just wait a few minutes and launch it.
But our temp are almost similar. Of course different CPU, GPU, pump, fan setup, tubing, fittings ecc, so, it’s ok!

 

[fabio]

You should just use the classic, as the temperature will be within margain of error and I am not sure if the non classic will even fit the T1.

Everything above 27mm will be a pain to fit.
Stay with the Classic

 

[fabio]

You should just use the classic, as the temperature will be within margain of error and I am not sure if the non classic will even fit the T1.

@DrHudacris is your 8086k delidded? This actually makes a huge difference with this CPU.

@fabio as far as I know, the EK Magnitude on Ryzen 3000 is one of the, if not the best water block, because it was specially made to account for Ryzen 3000 dye offset from the middle. If you want to compare the Eisbear LT block on Ryzen 3000, you should test it against an EK Velocity or other water blocks, that are not optimized for Ryzen 3000.

I think the Velocity or the Heatkillers IV (IMO the best block in general for flow, price, performance) will have just 1/2 degrees if difference from the magnitude. But there is literally NO review with a ryzen 3000 chip. Such a shame!

 

[aexeq]

I think the Velocity or the Heatkillers IV (IMO the best block in general for flow, price, performance) will have just 1/2 degrees if difference from the magnitude. But there is literally NO review with a ryzen 3000 chip. Such a shame!

Magnitude vs Velocity vs Heatkiller would be very nice to see. Imo the EK blocks look better, but in terms of performance to price you can't beat a second hand Heatkiller

Edit: On Ryzen 3000 of course!

 

[DrHudacris]

@DrHudacris is your 8086k delidded? This actually makes a huge difference with this CPU.

I bought it used, and first owner claims it was delidded, yes! When I rebuild my T1 (when I finally have some time!) I think I will repaste and double check mounting pressure of the monoblock. Might even go direct die... But right now, not worried about the component temps, just surprised with coolant deltas. But I will test CPU/GPU like @fabio did in his tests to emulate. Won't be apples to apples, because my GPU pulls 250W and CPU 130W, and monoblock vs CPU block, and different pumps and, and and etc