Good Flow Rates?

What's your setup? What kind of flow meter is it? Is it supposed to be accurate, or just one of those for general flow confirmation purposes?

I don't personally use a flow meter (but I have one on order for my Aquaero which will be installed when all the parts arrive) and I have read temperature performance increases until about 1 GPM, at which point it levels off, so this should ideally be everyone's target. Convert 1 GPM into L/H and you wind up with ~227 L/H, so yours seems a little low. It converts to ~0.54 - 0.60 GPH.

I haven't done much testing myself though, so I don't know how much would be gained by upping it.

Do you have a very restrictive loop? Many radiators/blocks? Many tight bends? What pump are you using? What setting is it running at, full speed? Are there any kinks in your tubing?
 
My loop is about 145l/h according to my AquaComputer "high flow" meter with my D5 at 3000rpm.

It's close to 200l/h with my pump at 4500rpm, but I can see no discernable cooling performance chance between the two settings.
 
I was getting 227 just with the pump and radiators on the loop themselves. It's an external setup.

I probably have more fittings than most people. No Kinks.
img_4525_zpsczqyevum-jpg.jpg
 
I think you're probably in good shape. I'm not sure where the 1g/m figure came from, though I've seen it elsewhere, but my limited testing doesn't really agree with it.

IIRC, the context of the figure was that beyond 1 gallon per minute you don't really see any additional cooling capacity. I can believe that, but it seems a strange benchmark to me. My loop is simple, typical, one CPU, one GPU, two radiators and a D5 - and I can't even hit one gallon per minute.
 
My loop is about 145l/h according to my AquaComputer "high flow" meter with my D5 at 3000rpm.

It's close to 200l/h with my pump at 4500rpm, but I can see no discernable cooling performance chance between the two settings.

Interesting. I have heard that 1 GPH number as being the magic number for some time, looks like it may be a little off.

As I have mentined in another thread, I have 4x Calitetemp inline temperature sensors and one of those "high flow" meters on order. Once I get them I plan to measure the temperature before and after my GPU and CPU. I'm going to do a benchmark with the pump at full blast and each component at full load and see if there is a measurable temperature difference across the blocks, then I am going to set pump speed automatically based on the max of the delta T across each block. This way at idle the pump will probably run at its minimum 20% speed, and then at load it may come close to maxing out.
 
I think 1 GPM came mostly because it's an easy number to remember. I think Martin's Liquid Lab did some testing on it back in the day as well.
 
Back
Top