MeidokonWiki:

Contents

  1. Noise levels at different speeds
    1. Pump
    2. Pump speed vs flow
    3. Fans
  2. Heat loads
  3. Fan Control settings
  4. Results
    1. Low speeds
    2. Medium speeds
    3. High speeds
  5. GPU undervolting
  6. Links

Noise levels at different speeds

This is nicely supported by Fan Control now, so it's worth knowing what your RPM and PWM correlation is like.

Pump

This is an EK D5 pump in a FLT120 reservoir.

PWM

RPM

Noise

0%

800

Silent

10%

800

Silent

20%

800

Silent, 20% seems to be the floor for control

25%

1070

Silent

30%

1360

Silent

35%

1640

Some whine

40%

1950

Barely audible even in close proximity

45%

2200

Barely audible even in close proximity

50%

2500

Just noticeable if listening under desk

55%

2770

Somehow quieter than above, same as 45%

60%

3050

Quiet if listening under desk

65%

3330

Pump whine just audible from sitting position above desk

70%

3620

Pump whine higher pitched and more noticeable

75%

3950

Pump whine at maximum noise acceptable in general (quiet) conditions

78%

4075

Weird sweet spot, somehow better than 75% or 80%

85%

4450

Occasional liquid swishing noises

100%

4775

Frequent liquid swishing noises

Pump speed vs flow

Same EK D5 pump in a FLT120 reservoir.

I rebuilt the loop when I changed to the AP201 case, and that changes all the flow values because the loop is different. It's now much less impeded, meaning I get more actual flow for a given pump speed.

Pump PWM

Pump RPM

Flow meter volume

Flow meter tacho

Notes

100%

4775 rpm

3.6 L/min

83 rps

-

85%

4450 rpm

3.3 L/min

76 rps

-

78%

4075 rpm

2.9 L/min

67 rps

-

75%

3950 rpm

2.8 L/min

63 rps

-

65%

3330 rpm

2.3 L/min

52 rps

-

55%

2770 rpm

1.8 L/min

40 rps

-

45%

2200 rpm

1.3 L/min

30 rps

-

35%

1640 rpm

0.7 L/min

16 rps

-

30%

1360 rpm

0.4 L/min

9 rps

-

25%

1140 rpm

0.1 L/min

4 rps

Slowest speed that avoids low-flow alarm

Fans

A pair of Corsair ML140 fans on a Corsair 280mm radiator in the top of the case.

PWM

RPM

Noise

20%

400

Silent

35%

700

Silent

45%

915

Just noticeable if listening under desk

50%

1040

Just noticeable sitting normally above desk

60%

1240

Comfy noise level above desk

70%

1400

Modestly noisy, about as much as you'd want in general (quiet) conditions

80%

1590

Noisy

90%

1730

Very noisy

100%

1875

Too bloody noisy!

Heat loads

Components

Estimated load (Watts approx)

Idle

25

Full CPU

120

Full GPU

250

Fan Control settings

Having gotten a sense for lower and upper bounds that make me happy, I've gone with the following:

Toe of curve

Idle speed

Shoulder of curve

Load speed

Pump

30°C

1360 rpm

42°C

3600 rpm

Rad fans

30°C

700 rpm

42°C

1500 rpm

This is using the Auto curve, which takes an expected Idle temperature and maximum acceptable Load temperature, and interpolates between them. It'll maintain the Idle speed as long as the liquid temp is below the lower threshold, and ramp up as high as the Load speed as the liquid temp approaches the upper threshold. This ensures that the maximum noise will never be above my comfy levels, even though it means leaving potential cooling performance on the table.

Results

NB: these results are old, they're not for the current system as of 2024-03-22

This is mostly about finding a sweet spot for being quiet at idle, and sufficiently cool under load, without being too noisy.

This is the new loop layout in the Phanteks Evolv Shift Air, after Maru helped out with reworking everything. 280mm radiator with 2x 140mm Corsair maglev fans.

Low speeds

Load

CPU temp

GPU temp

Loop temp

Notes

CPU only

84 C

29 C

28 C

Cinebench R23

GPU only

63 C

54 C

33 C

Furmark

GPU and half-CPU

90 C

59 C

37 C

Furmark and Cinebench limited to half usual threads

GPU and full-CPU

n/a

n/a

n/a

Not tested because the CPU throttles at 90C

Loop speeds are simply too low to get heat out of the chip fast enough, the CPU will lower the boost ceiling to try and stay at 90 degrees.

With some GPU undervolting we get a very healthy improvement in temps:

Load

CPU temp

GPU temp

Loop temp

Notes

GPU only - stock

63 C

54 C

33 C

-

GPU only - undervolted

57 C

48 C

31 C

-

Medium speeds

Load

CPU temp

GPU temp

Loop temp

Notes

CPU only

82 C

27 C

27 C

Cinebench R23

GPU only

54 C

48 C

31 C

Furmark

GPU and half-CPU

82 C

51 C

33 C

Furmark and Cinebench limited to half usual threads

GPU and full-CPU

90 C

52 C

35 C

Furmark and Cinebench with unlimited threads

This is a fair bump in flow, which suggests that the CPU isn't being cooled effectively when running at fill-tilt, I think. I wish I'd taken note of the CPU speeds, because then the CPU-only and CPU+GPU test cases could be compared.

High speeds

Load

CPU temp

GPU temp

Loop temp

Notes

CPU only

83 C

-

29 C

Cinebench R23 - 140W

GPU only

55 C

48 C

33 C

Furmark - 280W total power (230W GPU + 50W CPU)

GPU and half-CPU

82 C

50 C

34 C

Furmark and Cinebench limited to half usual threads - 340W power (230W + 110W)

GPU and full-CPU

90 C

-

34 C

Furmark and Cinebench with unlimited threads

Not sure why I don't have GPU temps for all test cases. Very similar to medium flow case, suggesting no benefit from the higher speeds. CPU still hitting it's 90 degree ceiling.

GPU undervolting

Poking at this while running Furmark.

Stock voltage settings

Undervolt applied

Links

MeidokonWiki: servers/wa-chan/Watercooling (last edited 2024-03-22 13:51:23 by furinkan)