Low Flow and A-C Question

[madmat]

A-C gear is cool if you can afford it. The whole low flow thing boggles my mind though because you're starting with a low head pump and adding fairly restrictive blocks which adds head loss and you add them in a fairly high quantity if you're going to cool every heat source.

Then to further complicate matters you use small diameter tubing which further hinders flow by increasing head loss. It just seems to me that after a certain point that unless you go multiple pumps/loops you'll end up with close to zero flow through the loop.

That's not a knock against anyone, it just doesn't make good sense to me. I could see compensating for the restriction by running parallel loops to cut resistence down (since water reacts a lot like electricity) or using a larger diameter tube.

No, I'm not trying to incite a riot nor induce flaming against myself nor anyone else but would someone care to elaborate?

Happy Mashie?

 

[BellaCroix]

The difference is that "low-flow" waterblocks are designed to work with those considerably lower flow rates. According to their website, Innovatek blocks need only .3 LPM of flow to perform well. Most loops with CPU and GPU pull between .7 and 1.5 LPM which is just fine for an XX-Flow based waterblock to work.

Now if you try to pull .7 LPM through a Storm or DD you're going to get temp numbers that will make you wonder why you didn't just stick with air cooling. The designs of those blocks benefit greatly from the added flow rates and many of those loops are at or above 3 LPM.

Basically, as long as you're using gear designed to work with whatever your approach is, it'll work just about the same in most real-world applications. High-flow has the potential to deliver a few degrees better temperatures when properly configured, but in a moderate overclocked machine, a few degrees one way or the other doesn't really make too much of a difference except in bragging rights.

My second forray into WC was with a very slight overclock using DD gear and 1/2" hose, it was rock solid and I was happy with it's performance. I decided to give low flow a try and switched to Innova/Aqua, all I changed was the watercooling. My overclocked remained the same, all equipment remained the same, stability remained the same, but (IMO) it looked a whole lot better.

 

[theseeker]

The difference is that "low-flow" waterblocks are designed to work with those considerably lower flow rates. According to their website, Innovatek blocks need only .3 LPM of flow to perform well. Most loops with CPU and GPU pull between .7 and 1.5 LPM which is just fine for an XX-Flow based waterblock to work.

Now if you try to pull .7 LPM through a Storm or DD you're going to get temp numbers that will make you wonder why you didn't just stick with air cooling. The designs of those blocks benefit greatly from the added flow rates and many of those loops are at or above 3 LPM.

Basically, as long as you're using gear designed to work with whatever your approach is, it'll work just about the same in most real-world applications. High-flow has the potential to deliver a few degrees better temperatures when properly configured, but in a moderate overclocked machine, a few degrees one way or the other doesn't really make too much of a difference except in bragging rights.

My second forray into WC was with a very slight overclock using DD gear and 1/2" hose, it was rock solid and I was happy with it's performance. I decided to give low flow a try and switched to Innova/Aqua, all I changed was the watercooling. My overclocked remained the same, all equipment remained the same, stability remained the same, but (IMO) it looked a whole lot better.

BC,

A very well written explanation and non-biased as well. When I saw this thread, I expected a flame war to break out. It is nice to see someone who has first hand experience with both approaches relate their experiences.

 

[madmat]

As I stated in my question, I only seek enlightenment. The A-C blocks I've seen have seemed to be fairly restrictive so I was always kinda when I thought about using them with a 2m head pump, especially one with sub 200GPM flow rates.

Thanks for taking my question as a non troll type of question (since I'm honestly not trolling) and taking the time to answer me.

 

[topcat989]

.........Basically, as long as you're using gear designed to work with whatever your approach is, it'll work just about the same in most real-world applications. High-flow has the potential to deliver a few degrees better temperatures when properly configured, but in a moderate overclocked machine, a few degrees one way or the other doesn't really make too much of a difference except in bragging rights.

My second forray into WC was with a very slight overclock using DD gear and 1/2" hose, it was rock solid and I was happy with it's performance. I decided to give low flow a try and switched to Innova/Aqua, all I changed was the watercooling. My overclocked remained the same, all equipment remained the same, stability remained the same, but (IMO) it looked a whole lot better.

This is basically what I've been saying all along

 

[R1ckCa1n]

My second forray into WC was with a very slight overclock using DD gear and 1/2" hose, it was rock solid and I was happy with it's performance. I decided to give low flow a try and switched to Innova/Aqua, all I changed was the watercooling. My overclocked remained the same, all equipment remained the same, stability remained the same, but (IMO) it looked a whole lot better.

Yes, I hope more people come to this realization. Hell I didn't get any worse overclocks with my Cuplex EVO as I did with my Cuplex XT, Storm, G4, Cuplex Pro.

 

[Erasmus354]

Yes, I hope more people come to this realization. Hell I didn't get any worse overclocks with my Cuplex EVO as I did with my Cuplex XT, Storm, G4, Cuplex Pro.

I dont think many people argue that in every case you will get extra overclock out of that last few degrees. However, you could, you never know. Your argument that the overclock is the same so why not get AC is circular. If the overclock is the same, why spend more money? The fact of the matter is, get AC stuff if you think the looks or the hobnobs they sell with it is worth the price premium.

If you want performance and the best chance (overclocking is never a guarantee) at getting the most out of your system go for a performance setup. If you just want quiet cooling and dont care about looks, there is no reason to spend the premium on AC. You can get quiet cooling with much cheaper gear. If you want the looks of AC, then by all means spend the extra to get AC.

Once Systemcooling finishes the review we may be able to recommend performance systems which utilize the Cuplex XT, provided you pair it with a good pump to utilize the merits of its design.


About the question of how the restrictive blocks can perform at low flow, they create a lot of turbulence in the water. This helps to decrease the laminar flow, which helps prevent stagnant water. The Storm performs very well as low as 1gpm, and in tests it still performs pretty well down to .5gpm. Most restrictive designs do a good job of giving a lot of turbulence to the flow, which helps even at low flow rates.

 

[Top Nurse]

As I stated in my question, I only seek enlightenment. The A-C blocks I've seen have seemed to be fairly restrictive so I was always kinda when I thought about using them with a 2m head pump, especially one with sub 200GPM flow rates.

Don't know about using the AC blocks with ultra flow pumps, but I do know that the Cuplex XT runs really well on a DDC. Much more so than the Cuplex Pro probably due to it being less restrictive. However, as others have pointed out the Cuplex XT seems to run just fine on the Aquastream pump as well. I suppose it will work "better" with a DDC, but most people prefer to give up the couple of degrees (if that) in higher temps for a pump that mates with other AC electronics, is relatively small, and performs adequately to the task at hand.

Now as to your original question I really can't say as to why it works as well as it does, but it does. Perhaps there is something wrong with the commonly held hypothesis's regarding water cooling?

 

[Top Nurse]

About the question of how the restrictive blocks can perform at low flow, they create a lot of turbulence in the water. This helps to decrease the laminar flow, which helps prevent stagnant water. Most restrictive designs do a good job of giving a lot of turbulence to the flow, which helps even at low flow rates.

I think this is the reason that AC thinks their TurboPlex USB device works so well. The pumping action of the water directly on the HS creates a tremendous amount of turbulence and effectively destroys the boundary layer creating a better transfer of heat into the water.