D5 Pump flow rate?

Mr Mcgoo · 20 May 2020 at 19:24

So I have an Aqua Computer D5 Next 1500lph and love it. However at 100% I get 170lph. So my loop order is:
Res>Pump>240mm Rad>GPU Block>CPU Block>360mm Rad> Alphacool Quick Disconnect>Res.

I have 6 90 degree fittings one on each end of the rads, one on pump outlet and one on the res intake. I have a total of 105cm of 10mm inner diameter tubing (around 15cm tubing between each component). does this sound right?

What is everyone else getting?

Jokester · 20 May 2020 at 19:35

It's a bit low, the 90 degree fittings and quick disconnects are probably the biggest restrictions.

I'm getting 300lph with mine.

Mr Mcgoo · 20 May 2020 at 19:42

Yeah I guessed it was low. My cpu block is very restrictive. In fact, the most restrictive block I have seen on the market thus far. Temps are good. 9600K 4.8ghz 1.2v 60c, rtx 2070 2055mhz 981mv 52c, 1500rpm ml pro fans on the rads, 31.5c water temp and 80% 4350rpm pump. What you got in your loop for 300lph?

Jokester · 20 May 2020 at 19:48

I'm passive cooled, but with my 3700x + 2080ti I'm currently sat at 40degC water temps whilst gaming with 28degC ambient and 150-200W heat disapation from the radiators.

Mr Mcgoo · 20 May 2020 at 20:09

no fans wow. what temps you getting no the cpu and gpu while gaming?

Radox-0 · 20 May 2020 at 20:14

Your flow sounds about right.

I am at around 160 l/h with pumps at 60% RPM in main rig but use dual D5 serial pump for double head pressure. Loop is on the larger side though so about right for it.

Going to be hard to compare really as everyone's loop can very so significantly.

Jokester · 20 May 2020 at 20:16

Mr Mcgoo said:
no fans wow. what temps you getting no the cpu and gpu while gaming?

50 on GPU and 60 on CPU.

Too Tall · 20 May 2020 at 20:18

Sorry to hijack but what are you using to measure flow?

Mr Mcgoo · 20 May 2020 at 20:23

good temps for passive cooled

the pump has it's own flow metre built in

Jokester · 20 May 2020 at 20:49

I'm using an Aquacomputer MPS400.

Mr Mcgoo · 20 May 2020 at 21:57

according to aquacomputer a similar but smaller device is what is within the d5 next. supposedly it's almost as accurate.

Radox-0 · 21 May 2020 at 10:33

Yep using aquacomputer MPS400 hooked up to a Aquero 6 XT. Fantastic unit when made central hub of WC system.

ttaskmaster · 21 May 2020 at 10:36

Mr Mcgoo said:
according to aquacomputer a similar but smaller device is what is within the d5 next. supposedly it's almost as accurate.

It's just a virtual flow sensor, which calculates the flow rate based on a number of parameters (that you input), and you then have to set the initial calibration to get any semblance of accuracy. It only works for a simple-pump loop, too.

"Flow rate measurement using a mechanical flow sensor is significantly more accurate than calculating the flow rate from pump parameter readings".
Aquacomputer D5 Next manual.

Also, I think the highest it can read is 300lph anyway. You can set it to auto-adjust the pump speed to maintain 50-300lph, but whether that's the maximum it can achieve I cannot say.

Jokester · 21 May 2020 at 10:50

As a rough rule you want your flow rate to be 1 gpm(US) minimum, though I think it was about 0.8 gpm before temperatures really started to significantly start to increase, so that's 230l/hr for 1gpm, or 180l/hr for the 0.8gpm figure.

Mr Mcgoo · 21 May 2020 at 18:58

i'm not even close to 1 gpm. around 0.75 gpm based on 170lph

George Hudson · 22 May 2020 at 07:17

That flow rate is dependant on the diameter of the piping you use. Using a Barrow flow meter my D5 pump (with 10mm diameter hose) produces (litres/min) at the following power outputs 100% = 6.8, 90% = 6.5, 80% = 5.5, 70% = 4.5, 60% = 3.5, 50% = 2.5, 40% = 1.3 and 30% = 0.2. I have a 2 x 360mm radiators ( 1@ 38mm thick, the other @28mm thick) cooling the CPU (3900X) and GPU (RTX 2080TI) in my loop. With extensive testing of varying fan rates and pump flow rates to achieve as quiet a system as possible while maintaining a deltaT of <= 10 degC and CPU and GPU at ~50 degC while gaming/benchmarking, my pump works between 40% and 60%. So my maximum flow rate is 3.5 litres/min. Why not do some testing/benchmarking and data logging yourself and show your PC how much you LOVE it.

. I would not worry about the "actual" flow rate. It is the combination of the "relative" flow rate and the fan speeds on the radiators required to achieve the desired goals.

Mr Mcgoo · 22 May 2020 at 14:17

That was a deep learning curve right there. will do some benchmarks later today and post my results

Mr Mcgoo · 22 May 2020 at 17:23

Results are in. I used 3dmark port royal for 6 x 20 min tests and cpu-z running 4 cores in the background giving me 95%+ GPU usage at all times and 92%+ CPU usage at all times. At each 20 minute test I increase pump power 10% starting at 50% and monitored cpu gpu water and ambient temps, followed by pump rpm and flow rate. I left the pc to idle for 10 minutes after each 20 minute test to cool back down to ambient before starting the next test. Results are as follows:

50% pump 3450rpm
cpu temp 60c
gpu temp 47c
water temp 29.8c
flow rate 82lph
ambient temp 22.5c

60% pump 3750rpm
cpu temp 60c
gpu temp 46c
water temp 29.9c
flow rate 114lph
ambient temp 22.6c

70% pump 4000rpm
cpu temp 60c
gpu temp 46c
water temp 30.0c
flow rate 126lph
ambient temp 22.8c

80% pump 4300rpm
cpu temp 60c
gpu temp 46c
water temp 30.1c
flow rate 141lph
ambient temp 22.9c

90% pump 4600rpm
cpu temp 60c
gpu temp 46c
water temp 30.1c
flow rate 154lph
ambient temp 23.0c

100% pump 4900rpm
cpu temp 60c
gpu temp 45c
water temp 30.3c
flow rate 168lph
ambient temp 23.1c

So, it would seem doubling the flow rate did nothing for temps as within margin of error. At 100% pump power the coolant moves a small but noticable amount from the small amount of air in my res and at 50% it barely moves at all. So maybe possible to have a bad flow rate reading for the pump itself? If anyone else has this pump could they list their loop order and fittings so as to compare flow rate?

Radox-0 · 22 May 2020 at 17:34

That 1 gpm is a rule of thumb. If your not seeing a benefit, then no big deal, for me IMO critical thing to look at would be how the water delta over ambient behaves when ramping up / down. In your case to be honest your water temps under load are fine. 6-8 degrees water delta temp over ambient is not anything really to worry about IMO.

George Hudson · 24 May 2020 at 16:22

Pleased to see you doing some data logging. That Delta-T of 7 degrees C you are seeing should put a smile on your face. The next data logging test is your fan speeds. Remove any ramping on your water pump, setting it at 50% speed permanently as you were seeing no benefit of it running any faster. Like wise, remove ramping and lock ALL the fan speeds at say 50% and perform the test again recording temperatures. If temps start getting too high, abort the test, allow system to cool, increase ALL fans by 10% and repeat. What you are wanting to do is find the fan speed that keeps your CPU and GPU at a temperature you are happy with when they are stressed. When datalogging, record the temps of the sensors that the fans are monitoring. You will find this in your BIOS where you setup fan curves. You will want to try different fans speeds for each set of fans. Think of a set of fans as all those on one radiator. The important thing is to ensure that fans remain locked at a set percentage while benchmarking/testing, including any additional fans which are not on radiators. You want as few variables as possible. Try adjusting one set of fans while leaving the others, so as to understand the effects each set has within your system. You may find an extra 10% or 20% on the set of fans moving cool ambient air across your radiator is quieter than the extra 30% speed required on the exhaust fan radiator in order to achieve the same Delta-T, CPU and GPU temps you are happy with. Eventually, what you want is a set of fan and pump power percentages that keeps your CPU and GPU at temps you are happy with when gaming/stressing them so you can then setup your ramping curves in the BIOS.