Overclock.net banner

1 - 20 of 33 Posts

·
Registered
Joined
·
386 Posts
Discussion Starter #1
Hi friends,

After using various AIOs for a while, I'm starting to think of some experiments with them. Has any of you had any personal and direct experience with reusing an AIO pump and connecting it to a new rad? My concern is about the barb sizes on the pump. That should help me get the right soft tubing for those barb sizes and get the right compression fittings for the rad/tubing side too.

I'm expecting a thicker 240mm EKWB PE copper rad to arrive that I got for super cheap and I have a used Asetek 360 AIO (591LX), though the AIO is untested yet (LOL) :) The Asetek has a 4th gen pump, afaik.

So, does anyone know what are the sizes on the AIO pump barbs or on this particular pump?

My aim is something like that with my own rad, for starters:

 

·
9 Cans of Ravioli
Joined
·
19,150 Posts

Tube ID is 6mm. Standard 6-ID\8-OD tube works a treat. I've done the clear-tube mod and added a reservoir.
keep in mind the pump flow rate is abysmal. the vast majority of people doing this are swapping the tubing to clear tubing, not adding parts that are massively more restrictive than what the pump was designed to deal with.
 

·
Premium Member
Joined
·
6,719 Posts
I did this back in oh...2012-2013?.... I used an H50, H60 pump... It'll work and it'll do wonders for cooling. You'd be amazed how efficient those little pumps are... I think I was maxing out around 50C on my GPU I had the waterblock on...
 

·
Registered
Joined
·
386 Posts
Discussion Starter #4

keep in mind the pump flow rate is abysmal. the vast majority of people doing this are swapping the tubing to clear tubing, not adding parts that are massively more restrictive than what the pump was designed to deal with.
Good to hear from you again The Pook.

Super nice!!! very useful info! I couldn't find it during my search somehow. Thanks a lot!!! :)

I also couldn't find the exact 100% data on Gen4 Asetek pumps yet, but either Gen4 or Gen4.5 are somewhere over 3000rpm, as far as I understood. Of course a higher rpm than the current 2800rpm pumps is not a higher efficiency rating, but it gives me hope that it can move through a 240mm rad, even if the rad is thicker. On top of that this pump moved water through an aluminium 360 rad by design. Hopefully the internal construction of the EK PE 240 is not going to be too restrictive. Moving from a 360 aluminium rad to a 240 thicker copper rad is my hope to be sparing enough on the pump:)

Thanks again and I will let everyone know what I find out myself about the sizes (and my prelim measurements as well, before buying the tubing).
 

·
Registered
Joined
·
386 Posts
Discussion Starter #5
I did this back in oh...2012-2013?.... I used an H50, H60 pump... It'll work and it'll do wonders for cooling. You'd be amazed how efficient those little pumps are... I think I was maxing out around 50C on my GPU I had the waterblock on...
Thanks, I had that feeling, everyone confirms it, your input is appreciated!

Now I'll need to get the practical details with this tubing and possibly fittings in US, I'd prefer two compression fittings on the rad. So, I'll research and see :)
 

·
9 Cans of Ravioli
Joined
·
19,150 Posts

0.11GPM max flow rate is far too slow to do much expansion with.
considering an AIO on a GPU with a G12 gets 46c load with a 2080 Ti, saying you get 50c on a GPU with an AIO pump isn't really a glowing recommendation either, lol.


if you wanna do it just to do it then have at it, but I wouldn't do it if you're only doing it because you expect better temperatures.
 

·
Premium Member
Joined
·
6,719 Posts
DSCN7022.JPG
DSCN7039.JPG
IMG_20131128_095157.jpg


hehe...this is from the goodolde days...I forgot...I actually tried 2 radiators for 1 AIO pump on my GTX670 GPU.... I think you reach a level of diminishing return... I think I maxed at 40C...
 

·
Registered
Joined
·
386 Posts
Discussion Starter #8
In my case I'm going to use that on a CPU only. So, I'm only planning to go from my current 240 "regular" AIO to a ghetto-rigged 240mm AIO with a thicker 240mm cooper rad :) And, yes!!! I just want to do it for the sake of doing it :) I don't have a temp normalized environment anyway, so trying to compare this frankenstAIO to my previous home test on a Cougar Helor 240mm AIO (with 2150 PWM Gentle Typhoons and a 27 Alu rad) vs a Corsair H105 (1850rpm sluggish pump and 38mm Alu rad, with the same 2150 PWM GTs) may not be accurate anyway. Btw the h105 Corsair outperformed the Cougar then by 3c I think (unless I didn;t account for the environment), but then it's also a non-comparison, because the Cougar might have been used for my own OC more than the Corsair that I purchased used from someone else.

So, yeah, this is for my pure tinkering satisfaction, learning for maybe a future watercooling build, who knows, and hopefully for a few more years out of the AIO, which is also nothing to complain about ;)
 
  • Rep+
Reactions: The Pook

·
Xeon Overclocker
Joined
·
292 Posts
I did a similar mod earlier this year. In my case I was going from a 240 to a 120.
Going with a thicker rad gave an improvement to the amount of heat the cooler could soak, but wasn't life changing.
The extra ports did make re-filling and bleeding the cooler very easy.

I ended up going with 1/4in g1/4 barbs and zip ties. You can heat-shrink wrap over the barb and it'll look pretty much stock.
 

Attachments

·
Shooting down fallacies
Joined
·
8,074 Posts
I did a similar mod earlier this year. In my case I was going from a 240 to a 120.
Going with a thicker rad gave an improvement to the amount of heat the cooler could soak, but wasn't life changing.
The extra ports did make re-filling and bleeding the cooler very easy.

I ended up going with 1/4in g1/4 barbs and zip ties. You can heat-shrink wrap over the barb and it'll look pretty much stock.
The rad being thicker was not what made the improvement. If all things were equal, going from a slim 240mm to a thicker 120mm would result in a reduction of heat dissipated. Surface area of the facing is far, far more important than the rad's thickness. You saw an improvement due to using a better rad made from better materials.

There is a lot of misunderstanding of how the thickness of a rad affects performance and restriction going on in this thread, and it's only nine posts in.
 

·
Xeon Overclocker
Joined
·
292 Posts
I wasn't trying to imply any generic thicker smaller rad was overall better, only that it was able to go marginally longer before reaching thermal equilibrium in my particular use case.
I agree a 240 will beat a 120 all things equal. The low flow and less than stellar heatplate is more limiting in the Asetek AIO than the rads.

I don't see where anyone is stating misinformation on rad thickness, restriction, or performance. If anything, looks like others have set reasonable expectations: Don't expect miracles, don't add anything too restrictive, and while not recommended it is do-able.
 

·
Registered
Joined
·
386 Posts
Discussion Starter #12
I did a similar mod earlier this year. In my case I was going from a 240 to a 120.
Going with a thicker rad gave an improvement to the amount of heat the cooler could soak, but wasn't life changing.
The extra ports did make re-filling and bleeding the cooler very easy.

I ended up going with 1/4in g1/4 barbs and zip ties. You can heat-shrink wrap over the barb and it'll look pretty much stock.
Looks great. Thank you! A Youtube guy also suggested that it's a 1/4 barb. Someone else, I think on an older OCN post wrote that they had O-rings on the barb itself before pushing the tubing over. Orings acted as sealant. I found the tubing 8/6mm, even fittings for the rad, but was scratching my head about clamping it down, being skeptical about the zip-ties and about the final looks:) I kind of wanted a clear tubing look, but the shrink does make it look "factory" like. Thanks for the confirmation and input.
 

·
Registered
Joined
·
386 Posts
Discussion Starter #13
The rad being thicker was not what made the improvement. If all things were equal, going from a slim 240mm to a thicker 120mm would result in a reduction of heat dissipated. Surface area of the facing is far, far more important than the rad's thickness. You saw an improvement due to using a better rad made from better materials.

There is a lot of misunderstanding of how the thickness of a rad affects performance and restriction going on in this thread, and it's only nine posts in.
Hi Ciarlatano, I appreciate your comments on OCN. It's all good this time, I think there is no misunderstanding. And @Servos is also right, his information did not imply lesser understanding of surface/dissipation area, one did not contradict the other ;) I am also aware of the pros and cons, materials (tubing life span/permeation and clouding / material depositing, liquid, good copper to copper, fittings material), of the surface area and mechanics. I agree with Servos on general limitation of the heat transfer on the subpar CPU block and pump. I am aware of pump's low flow and possible challenges it may encounter in a thicker EKWB CoolStream PE 240 dual (parallel flow). I agree and am aware that the surface area on this rad is bigger, one of the reasons I went with it. But your input will always be appreciated.

Anyway, this is a ghetto-rigging project, so... ;) I am doing it for both my watercooling learning curve, as well as to see if I can pull some more years and performance out of an outgoing AIO parts;) And the experience of other users, their advice, shows that reviving an AIO one way or another is feasible, which is also interesting and useful if I want to reuse that in my kids PCs etc.

Oh, boy, I still didn't test the 360mm AIO itself and I'm talking about rigging it!!! LOL
 

·
Registered
Joined
·
324 Posts
trying to compare this frankenstAIO to my previous home test on a Cougar Helor 240mm AIO
Does the Helor block would contaminate the fluid with aluminum? I heard people say the base (not the coldplate) is made of aluminum, as well as the barbed fittins on the block. I'm looking for a block to do the same thing to an Alphacool 360 rad. If you could tell, I would appreciate.

I think all these blocks, Asetek included, can be used with 5mm (internal size) tubing. I have done a previous mod with a 360 Barrow rad and I used a 5X10X2.5 tubing. The block was from DeepCool. I read somewhere that these barbed fittings are universal in most AIOs. I also used a barbed fitting to connect to the G1/4 of the Barrow. It looked neat and I kept the radiator outside the case (which is small).

The Helor block looks good, and perhaps it has a little more pressure? The things is huge. I also considered the LIQTECH II from Enermax, which claims to have a unique pump with up to 450 L/h. But reviewers are reporting gunk in the fluid out of the box and damaged o-rings.

Anyways, I think the 5mm tubing will fit. On my mod, it wasn't necessary any clamping. Once the tubing is in place, you can only take it out with a blade, just like the DeepCool one, which does not come glued, just pressure.

And it would be great to find a good block for this. Too much hassle to get a pump, a res, a block and all the fittings and movable parts. And the black tubing looks neat.
 

·
Registered
Joined
·
386 Posts
Discussion Starter #15
Does the Helor block would contaminate the fluid with aluminum? I heard people say the base (not the coldplate) is made of aluminum, as well as the barbed fittins on the block. I'm looking for a block to do the same thing to an Alphacool 360 rad. If you could tell, I would appreciate.

I think all these blocks, Asetek included, can be used with 5mm (internal size) tubing. I have done a previous mod with a 360 Barrow rad and I used a 5X10X2.5 tubing. The block was from DeepCool. I read somewhere that these barbed fittings are universal in most AIOs. I also used a barbed fitting to connect to the G1/4 of the Barrow. It looked neat and I kept the radiator outside the case (which is small).

The Helor block looks good, and perhaps it has a little more pressure? The things is huge. I also considered the LIQTECH II from Enermax, which claims to have a unique pump with up to 450 L/h. But reviewers are reporting gunk in the fluid out of the box and damaged o-rings.

Anyways, I think the 5mm tubing will fit. On my mod, it wasn't necessary any clamping. Once the tubing is in place, you can only take it out with a blade, just like the DeepCool one, which does not come glued, just pressure.

And it would be great to find a good block for this. Too much hassle to get a pump, a res, a block and all the fittings and movable parts. And the black tubing looks neat.
Hi Contiusa,

All AIOs use a copper block with an aluminium radiator, so to avoid galvanic corrosion (contamination), they use a specialized coolant to stop such corrosion and algae. It's basically, in most of the AIOs, a mix of ethylene glycol (sometimes propylene glycol - such as in Cougar Helor or Fractal Celsius) with bactericide and distilled water. Cougar Helor pump has a fill port and the AIO comes with a small bottle of coolant to top it off. For that reason, I actually didn't intend to use the Helor block for this experiment, as I can still refill it, but an older Asetek pump - however it seems I was wrong about the Asetek pump speeds, so I will have to make a new decision. I didn't research the Helor's block for reasons mentioned above, but I would anyway use a premixed 1L EKWB Cryofuel (which is actually one of the few liquid premixes that does not use ethylene or propylene and) - it would inhibit corrosion and bacteria. So, as long as we use a proper coolant (distilled water with antifreeze that contains bactericide - or a premix from EKWB or Corsair X5 etc) the mix of metals becomes less worrisome, just like in an AIO.

Indeed, I'd rather avoid Enermax Liqtech pumps. GamersNexus did a video where people repeatedly reported them as clogging, possibly due to a poor mix of materials - particularly the rubber/plastic gasket over the copper block seems to contaminate the fluid. And, as you mentioned, there are multiple reports elsewhere. I actually sold one some time ago, for that reason :)

Thanks for the info on barbs.

To everyone - I'll update those interested on the situation in the next days, there is more info on a few older AIOs and their pumps that might help someone avoid my mistakes.
 

·
Premium Member
Joined
·
6,719 Posts
Hi Contiusa,

All AIOs use a copper block with an aluminium radiator, so to avoid galvanic corrosion (contamination), they use a specialized coolant to stop such corrosion and algae. It's basically, in most of the AIOs, a mix of ethylene glycol (sometimes propylene glycol - such as in Cougar Helor or Fractal Celsius) with bactericide and distilled water. Cougar Helor pump has a fill port and the AIO comes with a small bottle of coolant to top it off. For that reason, I actually didn't intend to use the Helor block for this experiment, as I can still refill it, but an older Asetek pump - however it seems I was wrong about the Asetek pump speeds, so I will have to make a new decision. I didn't research the Helor's block for reasons mentioned above, but I would anyway use a premixed 1L EKWB Cryofuel (which is actually one of the few liquid premixes that does not use ethylene or propylene and) - it would inhibit corrosion and bacteria. So, as long as we use a proper coolant (distilled water with antifreeze that contains bactericide - or a premix from EKWB or Corsair X5 etc) the mix of metals becomes less worrisome, just like in an AIO.

Indeed, I'd rather avoid Enermax Liqtech pumps. GamersNexus did a video where people repeatedly reported them as clogging, possibly due to a poor mix of materials - particularly the rubber/plastic gasket over the copper block seems to contaminate the fluid. And, as you mentioned, there are multiple reports elsewhere. I actually sold one some time ago, for that reason :)

Thanks for the info on barbs.

To everyone - I'll update those interested on the situation in the next days, there is more info on a few older AIOs and their pumps that might help someone avoid my mistakes.
I forgot about this part... yea so what I did was fill the radiator and run the pump to circulate the water...I left one barb off with the hose dangling on the outlet of the pump and essentially stuck the radiator in a water 25% antifreeze mix solution that gets rid of corrosion/algae etc... I connected the barb to the hose after shaking the rad and running the pump for say 10-15 minutes all the while in the bucket of solution till I knew there were no air bubbles in the loop... oh and I think I used a crockpot for the tub of solution...whops...
 

·
Registered
Joined
·
324 Posts
Hi Contiusa,

All AIOs use a copper block with an aluminium radiator, so to avoid galvanic corrosion (contamination), they use a specialized coolant to stop such corrosion and algae. It's basically, in most of the AIOs, a mix of ethylene glycol (sometimes propylene glycol - such as in Cougar Helor or Fractal Celsius) with bactericide and distilled water. Cougar Helor pump has a fill port and the AIO comes with a small bottle of coolant to top it off. For that reason, I actually didn't intend to use the Helor block for this experiment, as I can still refill it, but an older Asetek pump - however it seems I was wrong about the Asetek pump speeds, so I will have to make a new decision. I didn't research the Helor's block for reasons mentioned above, but I would anyway use a premixed 1L EKWB Cryofuel (which is actually one of the few liquid premixes that does not use ethylene or propylene and) - it would inhibit corrosion and bacteria. So, as long as we use a proper coolant (distilled water with antifreeze that contains bactericide - or a premix from EKWB or Corsair X5 etc) the mix of metals becomes less worrisome, just like in an AIO.

Indeed, I'd rather avoid Enermax Liqtech pumps. GamersNexus did a video where people repeatedly reported them as clogging, possibly due to a poor mix of materials - particularly the rubber/plastic gasket over the copper block seems to contaminate the fluid. And, as you mentioned, there are multiple reports elsewhere. I actually sold one some time ago, for that reason :)

Thanks for the info on barbs.

To everyone - I'll update those interested on the situation in the next days, there is more info on a few older AIOs and their pumps that might help someone avoid my mistakes.
Yes, I know AIOs have a copper block and aluminum rad, but as I said, I do have a copper radiator (Alphacool) and I don't want to use coolant, just distilled water and fungicide, so I wanted to know if the Helor block has any aluminum in it. Reviewers mentioned it has an aluminum base (not the coldplate). Another reviewer said that it has aluminum fittings on the block.

Do you still have the Helor? Could you check if the block would contaminate the fluid? If the fittings are indeed made of aluminum?

This is the mod I did with a DeepCool block and a Barrow 360 rad. I run it for a couple years only with distilled water and fungicide. You can see that there is no clamping on the tubing.



And these are the fittings I used to plug onto the radiator (G1/4). They are nickel plated. If I'm not wrong, I used the 7.2mm ones (I ordered a few sizes do check). As I said, the AIOs generally have a 5mm internal diameter, so a 7.2mm fitting (I think it has 5mm of internal diameter as well) is a good size for a 5X10X2.5 tubing. Once you connect the tubing, you have to cut them off. I have some clamping and they were not necessary. I think you can find these at e-Bay as well.

 

·
Registered
Joined
·
324 Posts
And I'm really considering the LIQTECH II from Enermax. There is no other option out there other than the Asetek / DeepCool ones, which have very little pressure. I might bet on it. Open the block and clean all the gunk if any with distilled water and vinegar. Back then I was even looking for o-rings in case it came with a defective one.

I'm really trying to avoid any extra parts in my loop. Just a block and a rad.
 

·
Shooting down fallacies
Joined
·
8,074 Posts
And I'm really considering the LIQTECH II from Enermax. There is no other option out there other than the Asetek / DeepCool ones, which have very little pressure. I might bet on it. Open the block and clean all the gunk if any with distilled water and vinegar. Back then I was even looking for o-rings in case it came with a defective one.

I'm really trying to avoid any extra parts in my loop. Just a block and a rad.
I'd be wary of the Enermax. Why not an Arctic LF II? In the GN teardown the pump appeared to be much higher quality than an Asetek or DeepCool.
 

·
Registered
Joined
·
324 Posts
I'd be wary of the Enermax. Why not an Arctic LF II? In the GN teardown the pump appeared to be much higher quality than an Asetek or DeepCool.
From the pics and blue prints, it sounds like an inpeller simular to Asetek or the other brands. The pump has only 2000rpm as well.

There is the Alphacool Eisbaer LT (Solo), but it says up to 100 L/h, a bit higher than Asetek, which people says hovers around 50/60 L/h, if I'm not mistaken, but if they are saying "up to", it is more likely to be around 60/70 L/h. Then it is just more of the same. The Enermax pump is rated at 450 L/h. So even if it runs at 350/400 l/h, it is way above and would be more on par with a 360 XT45 copper rad.

I'm not sure if the higher flow would make a difference, but the Asetek / DeepCool ones are so low that we might see a few degrees difference. Perhaps it won't make a difference on my i7-9700K, but I have been tweaking hardware for a decade, making tests,and it is good to hold in hands a better piece of hardware. I want to test the Alphacool with a good pump as well.

I'll see. And from the looks of the Heler, it does not sound like it has a low flow, hence why I'm trying to solve this aluminum riddle.
 
1 - 20 of 33 Posts
Top