New Koolance *Dual D-5* Bay Reservoir! (RP-450X2)

[Diverge]

Unless I've mis-understood you, I don't think that won't give you the effect you want with a shared res feeding the pumps...

EDIT: It looks like I've mi-understood - are you suggesting tubing inside of the resevoir to give a direct feed from one external inlet to one of the pump internal inlets thus bypassing the shared res for one of the pumps? Are they threaded on the inside? is there clearance?

Yeah, a tube inside the res, if there is room, and if you can thread a barb into the holes. it might need modding. It might just work. Dunno really, since it's just an idea.

I only have one pump anyway

[Koolance]

Heh, I was thinking of the same splitter setup that Waterlogged illustrated.

Just a show of hands: how many people are currently (or would prefer to) run D-5's in series? We designed the RP-450X2 exclusively for multiple loops. If that's not a significantly more popular use of dual D-5's, I think we could consider another product altogether.

If series is desired for redundancy, how many people have actually had a D-5 die? Our own RMA list is extremely short.

Tim

[MagisD]

LoL that why I bought D5's in the first place ( and koolance Res's) but It seems on here Dual D5 is more about pressure/flow and wanting A dual top due to the bulkiness , Pita to mount for series flow.

If you could make one where you could redirect the flow with a simple insert so 1 top could do 1 pump 1 loop, 2 pump 1 loop , 2 pump 2 loop. And make it look nice like you did with this, you'd have a killer product on your hands. Make a RP-400 ver as well and you eat up some market share.

[Toolius]

Heh, I was thinking of the same splitter setup that Waterlogged illustrated.

Just a show of hands: how many people are currently (or would prefer to) run D-5's in series? We designed the RP-450X2 exclusively for multiple loops. If that's not a significantly more popular use of dual D-5's, I think we could consider another product altogether.

If series is desired for redundancy, how many people have actually had a D-5 die? Our own RMA list is extremely short.

Tim

Hello Tim
I would love a series version of this beauty
Count me in
Would swap both my Ek dual D-5 tops for this for sure. Thats precisely the reason i just got one of these right now to see if i could get it to work in series .
Might just run the one that i just ordered with a single pump in another machine and get 2 of the series versions if you decide on making them
Cheers again and kind regards !

[Diverge]

If you really want to get crazy with the design, you could make it configurable as:

1) 2x pumps in series
2) 2x pumps in parallel w/ shared res
3) 2x pumps in parallel w/ divider
4) 1x pump w/ whole res
5) 1x pump w/ 1/2 res (w/ divider in)

You could do this easily with a insert-able divider down the middle (if there is enough room between both pump inlets). And to make it in series, just add an internal tube like my ghetto ms paint drawing

Not that I would need any of this.. I'll just be happy to rid myself of the T3 I have.

[Sadasius]

Heh, I was thinking of the same splitter setup that Waterlogged illustrated.

Just a show of hands: how many people are currently (or would prefer to) run D-5's in series? We designed the RP-450X2 exclusively for multiple loops. If that's not a significantly more popular use of dual D-5's, I think we could consider another product altogether.

If series is desired for redundancy, how many people have actually had a D-5 die? Our own RMA list is extremely short.

Tim

Mine are D5's in series. You can see them in my avatar.

[vhaarr]

Just a show of hands: how many people are currently (or would prefer to) run D-5's in series?

I do; I have never seen the point in running more than one loop unless you have a crazy setup.
2 pumps in series allows me to run them on a lower setting and get redundancy in case of failure.

[Darth_Penguin]

Heh, I was thinking of the same splitter setup that Waterlogged illustrated.

Just a show of hands: how many people are currently (or would prefer to) run D-5's in series? We designed the RP-450X2 exclusively for multiple loops. If that's not a significantly more popular use of dual D-5's, I think we could consider another product altogether.

If series is desired for redundancy, how many people have actually had a D-5 die? Our own RMA list is extremely short.

Tim

I would also like to raise my hand as I am interested in running dual D5's in serial.

[halo112358]

I don't think that's going to fit in your :banana::banana::banana:ina! Can't you see it's black and is packing a double pumps?!? If it does fit then I am going to call you the local bicycle!

You might want to warm up your village bicycle jokes, I ordered a res & pump today - it should fit nicely.

[Sadasius]

Yup now that you mention it....I think it will fit too! That's all I got for now....I am drunk and I think the cat ate my sock. Funny think is I don't have a cat!

[ohms]

That looks DOPE!

Is it meant for 2 separate loops or chained pumps?

[halo112358]

That looks DOPE!

Is it meant for 2 separate loops or chained pumps?

C. All of the above.

[Church]

Koolance/Tim: People are going to dual(/multi) loops for 2 things - to workarround weak pump head pressure (which is solved in your product) and to thermally isolate loops (not done due to common res).

In light of Gabe's tests that single loop is preferable for most common usages (read gaming) i see rise of popularity of them.

I wonder if you can make "one product do all". As in some acetal "connector" to attach to restop to connect inlets instead of doing that with fittings & tubing, and some insertable wall in res to fully isolate halfs or reservoir (should be no problems, as there are already two fillports on top?). Probably both can be sold separately to not rise overall price of already not exactly cheapest restop.

[Tyranie909]

What I like about the single res option is that I can use one loop on parts that benefit from high flow such as my 2 rads ( 3x140 mm) and the second loop for the blocks. A parallel loop allows for the rads to be off the loop of the blocks and still receive the correct cooling of the liquid due to the speed at which the fluid in the res is exchanged out.

There was a thread here recently about the benefits of running split loops like this. I will see if I can dig it up.

[Church]

Tyranie909: rad flow restriction is MUCH smaller then that of waterblocks (i recall recent comparison that shown something like 17(or 27?) rads restriction was roughly comparable to that of single cpu waterblock) and by separating all of rads in one loop and all of waterblocks in another you'll get very unbalanced restriction(&resulting flow) wise in subloops. You'd get much better flow with more or less balanced/evened out components restriction-wise in those subloops. There is no gain in splitting the way you are proposing to.

[malkiewicz]

Nice looking unit but I think it would look better with the Koolance logo subdude / its a bit bulky IMO. Like the RP-450X2

[Tyranie909]

Interesting. I am still lookiing for the thread as well. My problem is the number of blocks I am running. 2xgpu (470 gtx), EK cpu, MIPS ram cooler, Full cover asus rampage extreme III EK, and the 2 3x140mm rads. From what I was told before, splitting these up into 2 loops using a common res (seperate pumps) would increase flow. I am aware that the rads are relative low restriction, but they also gain the most from higher flow. Blocks usually do well in a more restrictive lower flow loop together. In my case I was just going to put the mobo block, ram block, and rads together. Then let the cpu and gpu blocks stay in one loop together.

Oh and you don't have to spell out my name each time, just call me Tyr . Makes life easier that way.

[Church]

I'm guessing that minimal gains in rad heat dissipation because of bigger flow will be more then offset with worse heat transfer from waterblocks because of slow flow because of too much restrictive things in one subloop.

[MagisD]

If he was using 1 pump as in the hated t3. But a seperate pump for each loop less in each loop better flow.

Gabes post convinced me to go serial in considering it spends 80% of the time idle/vpn to work allmost no load i can auto ramp my pumps. But for him split loops with 2 pumps would increase flow in each loop. But how much of an actual effect that's gona have is debatable. Martins sounded of on that and high vs low flow in rads makes a very minor difference.

Sent from my X10a using Tapatalk

[Utnorris]

Yup now that you mention it....I think it will fit too! That's all I got for now....I am drunk and I think the cat ate my sock. Funny think is I don't have a cat!

Nice.

[Darth_Penguin]

This should give some redundancy.



...and for those worried about back flow, I suggest a pair of these.

http://www.mcmaster.com/#7775k52/=96awof


Tim, If I may suggest a product. Good low restriction, low crack pressure check valves with G1/4" swivel fittings.

WL, would one of these setups be better than the other in regards to reducing backflow? and is backflow guaranteed to happen when using the y adapters?

[century child]

Nice.

I don't think anyone will complain about quoting that pic. Something like that you just can't see enough of.

[halo112358]

What's the thinking with dual loops sharing the res - you run one high flow loop (rads) and another moderate restriction loop (cpu + gpu) ? I ended up ordering the bay res and one pump, I'm considering ordering a second pump if I'll get $75 worth of benefit out of it

[Gwiggles]

Possible to pay more for one without any writing on it? :/ Almost surely not, but may as well ask. I really like thiiisss

[MagisD]

What's the thinking with dual loops sharing the res - you run one high flow loop (rads) and another moderate restriction loop (cpu + gpu) ? I ended up ordering the bay res and one pump, I'm considering ordering a second pump if I'll get $75 worth of benefit out of it

This all depends on you blocks. But in most cases a rad only loop has minimal impact if any.
a couple people have tested this unless you have some insanly high restrictions somewhere high vers low flow in rads pretty much dump the same amount of heat in a closed system.

Sent from my X10a using Tapatalk