Custom Loop questions

[foolishlywise]

At present, I have a custom loop consisting of (EKWB) 240mm 40mm thick rad, Supremacy MX block and a PWM powered pump provided (400l/hr 6w).

 

I'm moving to a bay pump/res combo (the DBAY D5). However, I might want to run two pumps. 

 

Because of the way the rad is mounted (front of the case, with the inlet/outlet ports facing UP, the pressure from the rad to the CPU block IN is low it seems like. I've followed the EK recommended setup of
Block OUT -> Pump IN to Pump OUT -> Rad IN to Rad OUT -> Block IN. This ensures the block gets the coolest fluid possible straight from the rad. But, since the rad OUT port is upwards, gravity isn't helping. 

 

So, using the low powered pump between the Rad OUT and Block IN, does this have the potential to drop temps as water is reaching the block at a higher pressure or will the heat generated by the second pump make it a negligible difference? 

 

EDIT: forgot to add that its a single CPU loop only right now but looking at getting a better graphics card and adding that into the loop soon too. 

 

And, slightly random but reusing fluid - the one in there is around 2 weeks old and I have half a litre of it remaining. Best practise to completely drain and start again when adding in the bay pump/res or reuse the two-week stuff? 

 

 

[Leonard]

Not by much. If you want to lower the temperatures further thne you could add a 120mm radiator in the back for like maybe a 5-7C drop.

 

Another thing, the bay reservoirs are a hinderance to the removing of the front panel of a case.

 

The D5 will be more than enough for your present radiator and what you add, you won't need the old pump even with the reservoir inlets/outlets in the position they are at present. 

 

Granted you add another radiator, you could do a flow patter like this....pump/res combo ---> CPU---> back radiator ---> front radiator---> pump/res combo

[done12many2]

3 hours ago, FoolishlyWise said:

At present, I have a custom loop consisting of (EKWB) 240mm 40mm thick rad, Supremacy MX block and a PWM powered pump provided (400l/min 6w).

 

I'm moving to a bay pump/res combo (the DBAY D5). However, I might want to run two pumps. 

 

Because of the way the rad is mounted (front of the case, with the inlet/outlet ports facing UP, the pressure from the rad to the CPU block IN is low it seems like. I've followed the EK recommended setup of
Block OUT -> Pump IN to Pump OUT -> Rad IN to Rad OUT -> Block IN. This ensures the block gets the coolest fluid possible straight from the rad. But, since the rad OUT port is upwards, gravity isn't helping. 

 

So, using the low powered pump between the Rad OUT and Block IN, does this have the potential to drop temps as water is reaching the block at a higher pressure or will the heat generated by the second pump make it a negligible difference? 

 

EDIT: forgot to add that its a single CPU loop only right now but looking at getting a better graphics card and adding that into the loop soon too. 

 

And, slightly random but reusing fluid - the one in there is around 2 weeks old and I have half a litre of it remaining. Best practise to completely drain and start again when adding in the bay pump/res or reuse the two-week stuff? 

 

You only need to run one D5 PWM pump for your configuration.  Adding another will do nothing for you.  What CPU are you currently trying to cool?  I use to us a single EK PE 240 on an overclocked 4790k and it would still get a little toasty at high overclocks under full load, but other than that, the loop did fine.

 

If you are thinking of adding a GPU to the loop, you are definitely going to want to add more radiator surface area to your loop.  GPUs heat water a lot more than CPUs do. 

[foolishlywise]

18 minutes ago, Leonard said:

Not by much. If you want to lower the temperatures further thne you could add a 120mm radiator in the back for like maybe a 5-7C drop.

 

Another thing, the bay reservoirs are a hinderance to the removing of the front panel of a case.

 

The D5 will be more than enough for your present radiator and what you add, you won't need the old pump even with the reservoir inlets/outlets in the position they are at present. 

 

Granted you add another radiator, you could do a flow patter like this....pump/res combo ---> CPU---> back radiator ---> front radiator---> pump/res combo

I hear this. I haven't got my GPU yet but soon. If it would shave off 5-7degC from the overall temps, I'd be happy. Albeit, looking at my Define XL R2, I'm trying to guess if a 140mm rad will fit in the back. 

 

10 minutes ago, done12many2 said:

 

You only need to run one D5 PWM pump for your configuration.  Adding another will do nothing for you.  What CPU are you currently trying to cool?  I use to us a single EK PE 240 on an overclocked 4790k and it would still get a little toasty at high overclocks under full load, but other than that, the loop did fine.

 

If you are thinking of adding a GPU to the loop, you are definitely going to want to add more radiator surface area to your loop.  GPUs heat water a lot more than CPUs do. 

Exactly the same as you had - OC'ed 4790k. Toasty, indeed (I miss Sandybridge soldered IHS's). In your experience (sick rig looking at your sig), would adding a second rad bring down those CPU temps? (In the interim of now and getting a GTX 1080 maybe - three months). Ambients aren't high (UK's cold 99% of the time) so trying to think if that might be something worth doing. 

[done12many2]

3 hours ago, FoolishlyWise said:

Exactly the same as you had - OC'ed 4790k. Toasty, indeed (I miss Sandybridge soldered IHS's). In your experience (sick rig looking at your sig), would adding a second rad bring down those CPU temps? (In the interim of now and getting a GTX 1080 maybe - three months). Ambients aren't high (UK's cold 99% of the time) so trying to think if that might be something worth doing. 

 

Agreed on the soldered IHS!  Good think they still do this for the extreme chip such as my 5960x.  

 

As far as would an additional radiator help?  More than likely yes, but it will be largely based on how bad the TIM is on your particular 4790k.  Mine was decent and responded well when I tested it in a larger radiator configuration.  I was about to permanently add the additional radiator to my 4790k build, but instead I sold the whole PC.  

 

In your case I'd say it's definitely worth doing since you're going to be adding a GPU to the loop soon.

 

I wish we had UK ambient temps all year long, but no such luck.  It gets hot and humid here during the summer, which I have to offset with AC.

[Leonard]

21 minutes ago, FoolishlyWise said:

I hear this. I haven't got my GPU yet but soon. If it would shave off 5-7degC from the overall temps, I'd be happy. Albeit, looking at my Define XL R2, I'm trying to guess if a 140mm rad will fit in the back. 

 

---gone---

That 5-7C decrese is like the most that MIGHT happen, anyway if you want that then go for it but make sure to seal the tubes and fittings properly as the added pressure may cause a leak.

 

The 140mm radiator is sure to fit in the back as the Define XL R2 supports 120/140mm fans in the back.

 

IMO, a 240/280mm radiator on the top would be a better choice and you have the clearence so it should be fine.

[foolishlywise]

4 hours ago, done12many2 said:

 

Agreed on the soldered IHS!  Good think they still do this for the extreme chip such as my 5960x.  

 

As far as would an additional radiator help?  More than likely yes, but it will be largely based on how bad the TIM is on your particular 4790k.  Mine was decent and responded well when I tested it in a larger radiator configuration.  I was about to permanently add the additional radiator to my 4790k build, but instead I sold the whole PC.  

 

In your case I'd say it's definitely worth doing since you're going to be adding a GPU to the loop soon.

 

I wish we had UK ambient temps all year long, but no such luck.  It gets hot and humid here during the summer, which I have to offset with AC.

I was in silent prayer hoping Skylake would have solder. Nope 

 

Indeedy. I (personally) think the TIM is horrible on mine, it hits high 80's as far as I remember on Large FFT's so.. still need to fully play with OC'ing it with LLC settings etc. 

 

UK ambient temps, yes.

UK rainfall? No. Hell no. 

 

3 hours ago, Leonard said:

That 5-7C decrese is like the most that MIGHT happen, anyway if you want that then go for it but make sure to seal the tubes and fittings properly as the added pressure may cause a leak.

 

The 140mm radiator is sure to fit in the back as the Define XL R2 supports 120/140mm fans in the back.

 

IMO, a 240/280mm radiator on the top would be a better choice and you have the clearence so it should be fine.

Living in hope is better than not living it at all, eh? Good call about sealing tubes - EK ACF compression fittings should be up to the task. Ended up ordering a 140mm EK rad because the amount of textbooks, refills pads etc I leave on top of the case (and how the habit formed) suggested it isn't a good idea to mount it up top just yet. 

 

Scan are meant to deliver the rad tomorrow. But, I don't think that's going to happen - technically I did get the order in before the cutoff time. 30 seconds before to be precise...

 

Many thanks guys. I shall update when everything comes! 

[foolishlywise]

On 22/06/2016 at 4:32 PM, Leonard said:

-snip-

 

Granted you add another radiator, you could do a flow patter like this....pump/res combo ---> CPU---> back radiator ---> front radiator---> pump/res combo

At the risk of sounding stupid, the flow pattern (as far as I understand it) is like this:

 

Pump/Res OUT -> CPU IN -> CPU OUT -> Rear Rad IN -> Rear Rad OUT -> Front Rad IN -> Front Rad OUT -> Pump/Res IN ?

 

By this, the CPU block would be in receipt of the highest pressure of water, right? 

 

I may or may not have been up all night following The UK's European Union Referendum (spoiler: disappointing result) so a little sleep deprived. And I don't want to get it wrong! 

 

EDIT: Also forgot to ask about the current fluid in there. Its less than 3 weeks old and I still have spare fluid left from the same batch. Is it worth reusing the existing stuff or should I just get rid and use totally new fluid? 

[Leonard]

1 hour ago, FoolishlyWise said:

At the risk of sounding stupid, the flow pattern (as far as I understand it) is like this:

 

Pump/Res OUT -> CPU IN -> CPU OUT -> Rear Rad IN -> Rear Rad OUT -> Front Rad IN -> Front Rad OUT -> Pump/Res IN ?

 

By this, the CPU block would be in receipt of the highest pressure of water, right? 

 

I may or may not have been up all night following The UK's European Union Referendum (spoiler: disappointing result) so a little sleep deprived. And I don't want to get it wrong! 

 

EDIT: Also forgot to ask about the current fluid in there. Its less than 3 weeks old and I still have spare fluid left from the same batch. Is it worth reusing the existing stuff or should I just get rid and use totally new fluid? 

No risk of sounding stupid IMO. Yes that is the flow pattern. Now if you add the GPU the flow pattern would/could be pump/res out ---> CPU in--->CPU out--->back rad in---> back rad out---->GPU in---> GPU out----> front rad in---> front rad out--->pump/res in.

 

Not only would the CPU get the highest pressure but the fluid will have considerable time to cool down by going into the 2 radiators.

 

You can reuse the fluid but i am sure you will see small black/dark grey specs in the fluid, that is just flux from the radiator, so just don't put that back in. Meaning you can reuse most of the present fluid but don't put the last bit of the present fluid because of the flux in the water and you can top off with new fluid. You can use that fluid for about at least six months, i change mine every eight months but i use deionized water only. Do you use silver coils/biocide in the fluid? I use both but i have found that kill coils is the better way to go because biocides like the PT Nuke tends to make the alge build up on the reservoir as a hardish brittle coating inside the reservoir, it can be cleaned easily but just takes much more time.

 

One thing to note is that you should make sure the new radiator is of the same internal metal to avoid issues that come with mixing metals and water.

[W-L]

On 6/22/2016 at 11:19 AM, FoolishlyWise said:

-SNIP-

One thing to mention if your planning on using a DDC and D5 I don't recommend it as they are different pump types and designs your going to have different flow rates and pressure between the two, you would be better off with just a single D5 ideally or DDC. Personally I don't really recommend bay res units if your wanting dead silence as they tend to readily transfer vibration through into the case. 

 

As for your loop order it doesn't matter as long as the pump is below and right after the res.

 

[foolishlywise]

On 24/06/2016 at 2:05 PM, Leonard said:

-SNIP-

 

Do you use silver coils/biocide in the fluid? I use both but i have found that kill coils is the better way to go because biocides like the PT Nuke tends to make the alge build up on the reservoir as a hardish brittle coating inside the reservoir, it can be cleaned easily but just takes much more time.

 

One thing to note is that you should make sure the new radiator is of the same internal metal to avoid issues that come with mixing metals and water.

I've drained the fluid, nothing black in there but I had to leave it out of the loop for over 36 hours (and counting). So I'm leaning toward replacing it. 

 

I use the EK stuff (forgot what it is called) but not a silver kill coil. Apparently, they test for Silver oxide as part of their RMA procedure (according to someone on Reddit) so I don't, just in case they do.  No idea if yourself (or anyone else) has thought this way but what if I were to put a UV lamp shining into the res for a few hours once every two weeks. Would that kill anything that forms before it causes issues or would it be stopped by the acetate (I think) of the res?

 

Luckily, I ordered another EK rad with the same metals. 

 

On 25/06/2016 at 3:49 PM, W-L said:

One thing to mention if your planning on using a DDC and D5 I don't recommend it as they are different pump types and designs your going to have different flow rates and pressure between the two, you would be better off with just a single D5 ideally or DDC. Personally I don't really recommend bay res units if your wanting dead silence as they tend to readily transfer vibration through into the case. 

 

As for your loop order it doesn't matter as long as the pump is below and right after the res.

 

Ah, I see! Going with the D5 in the end. The EK bay res seems to have (looking at it and reading the blurb) rubber grommets that effectively isolate it from the case so it shouldn't be too bad, I hope! 

 

Unfortunately, the rad ordered (EK CoolStream CE 140) doesn't actually fit the rear of the case. Something I discovered once I'd drained the coolant out of the loop. Ordered a XE 120 which arrived today (Sunday delivery is amazing) BUT the retailer sent me a used one. Used as in the box was genuinely soaked as the rad had leaked and on closer inspection, there was algae inside it. Nice. Albeit with this rad, I managed to see that it will fit in my case! 

 

..new one coming 12pm on Tuesday. 

[Leonard]

8 hours ago, FoolishlyWise said:

I've drained the fluid, nothing black in there but I had to leave it out of the loop for over 36 hours (and counting). So I'm leaning toward replacing it. 

 

I use the EK stuff (forgot what it is called) but not a silver kill coil. Apparently, they test for Silver oxide as part of their RMA procedure (according to someone on Reddit) so I don't, just in case they do.  No idea if yourself (or anyone else) has thought this way but what if I were to put a UV lamp shining into the res for a few hours once every two weeks. Would that kill anything that forms before it causes issues or would it be stopped by the acetate (I think) of the res?

 

Luckily, I ordered another EK rad with the same metals. 

---gone--

Glad to hear the fluid was clean, if you kept the fluid in a clean glass bottle and covered it then it would be fine to use again.   FYI, you should keep all your fluid in glass bottles rather than plastic because all plastic leaches, some faster than others, but all leach. How ever there are some plastic and glass bottles with a internal coating that is designed to make all the liquid repel from he bottles, can't remember what it is called but there are no tests about how that reacts to metals so i would stay away from those too. I use a 2 gallon version of this to store and collect my deionized water.

 

The silver oxide they are testing for would be to know if you have used like silver coils in your loop. They also test for other stuff as well. So i would follow what ever EK says is best to do or  find out if PT Nuke and silver coils are covered under the warranty before you use it The UV lamp may help but i don't know what type of lamp you will use, if it has plastic on the ends or if you will use just say the bulb, the plastic is sure to cause an issue.

 

As i said before i use the PT Nuke(the blue one) and 4 silver coils in my reservoir and i have no issues other than oxidation of the CPU block's channels which happens after a year of use regardless what fluid you use and oh yeah, you should open and clean all blocks after a year of use or at least just reverse the fluid flow when cleaning the blocks to remove these particles. Note that opening blocks and bay reservoirs can be a pain in the ass to reassemble.

[foolishlywise]

On 27/06/2016 at 5:30 AM, Leonard said:

Glad to hear the fluid was clean, if you kept the fluid in a clean glass bottle and covered it then it would be fine to use again.   FYI, you should keep all your fluid in glass bottles rather than plastic because all plastic leaches, some faster than others, but all leach. How ever there are some plastic and glass bottles with a internal coating that is designed to make all the liquid repel from he bottles, can't remember what it is called but there are no tests about how that reacts to metals so i would stay away from those too. I use a 2 gallon version of this to store and collect my deionized water.

 

The silver oxide they are testing for would be to know if you have used like silver coils in your loop. They also test for other stuff as well. So i would follow what ever EK says is best to do or  find out if PT Nuke and silver coils are covered under the warranty before you use it The UV lamp may help but i don't know what type of lamp you will use, if it has plastic on the ends or if you will use just say the bulb, the plastic is sure to cause an issue.

 

As i said before i use the PT Nuke(the blue one) and 4 silver coils in my reservoir and i have no issues other than oxidation of the CPU block's channels which happens after a year of use regardless what fluid you use and oh yeah, you should open and clean all blocks after a year of use or at least just reverse the fluid flow when cleaning the blocks to remove these particles. Note that opening blocks and bay reservoirs can be a pain in the ass to reassemble.

 

I need a jug. Oh man getting the liquid into the res this morning was like peeing in the dark.Good shout about the containers - glass obviously doesn't leech so defo something to look into. 

 

About block and bay res disassemblies, I like a challenge 

 

Albeit I think I have a problem. The pump is making a clicking noise akin to a HDD - 

It's been decoupled from the chassis using its own rubber isolation mounting bits as well as foam. Taking it out of the case and running results in the same clicks. Is this normal for a bay res? 

 

 

[W-L]

2 minutes ago, FoolishlyWise said:

-SNIP-

Sounds like those small air bubbles to me it takes times for them to work out of the system sometimes a few days to a week, try getting a large syringe those work quite well for filling the loop.  

[Leonard]

I would agree with @W-L. The D5 has a slight issue removing air trapped in the pump itself when installed horizontally like it is in the bay reservoir and this sounds just like trapped air. I recommend you make sure all the ports are securely fastened and then turn your case so that the front of the case is facing upwards when the pump is running and then just give the case a slight shake to send the air out and or shutting the pump off and then on again a few times.

 

I assume you didn't use the drop of dish washing liquid trick to "lubricate" the pump on first use. This most times, prevents the air from being trapped in the pump.

 

Also, be sure to full the reservoir to the top to avoid air being reintroduced to the fluid buy the force of the D5 pump.

[foolishlywise]

I managed to reduce down the noise of the air bubbles thanks to the advice guys, much appreciated! However, as suggested by W-L earlier, the noise from the bay res/pump is loud. Even with the rubber grommets that EK have and my sponge mod around it, the vibration was pretty bad (ie: it could be heard in the living room - directly under mine). As such, I decided to not go with the bay res/pump and go with a standard XRES D5 Revo pump/res convo. Since with this setup, I could later on split the pump, get a top and use a bay res. 

 

I'm trying to now mount the Revo inside the case. As big as the XL R2 is, the mounting is semi dreadful. If I were to mount the pump/res on the rad in the front with a bracket, is there any obvious problems with this? 

 

EDIT: 

System all working! Managed to get the drive cage to the centre of the PC rather than its normal position and used the EK UNI bracket to mount the pump/res. Safe to say it all fitted! 

 

Thank you for your help guys, much much appreciated!