Can anyone rate the primochill Typhoon III?

[surfhick]

Nope, I'm saying regulating flow has an effect on temps period and i'm also saying that temps raise when you lower flow because in a closed loop system when you lower flow through the radiator you also lower flow through the blocks. That is all i'm saying because that is all I know for sure. I haven't actually tested flow through radiator seperate from the flow through the blocks and that is what I have been trying to figure out all this time.

Anyways I give up, dont feel like arguing anymore...

Cheers

Couldn't you just put a valve in the radiator loop and partially close it? That would, I think, slow the flow through the radiator and speed flow through the blocks.

[millertime359]

Couldn't you just put a valve in the radiator loop and partially close it? That would, I think, slow the flow through the radiator and speed flow through the blocks.

Still going to slow flow through rest of loop.

[Vapor]

Flow through the pump will slow down overall, but if you do it *just* right you can increase flow through a specific subloop. Intentionally limiting your flow through one part is something that's kind of hard to recommend

[bluehaze]

Couldn't you just put a valve in the radiator loop and partially close it? That would, I think, slow the flow through the radiator and speed flow through the blocks.

No it doesn't work because it's a closed loop system and your dealing with a fixed volume of water any change to flow through the rad is insignificant. If you halve the flow through the rad your also halving the volume of water your dissipating the heat from so in the end any change you make basically cancels itself out.

Then to make matters worse you have now halved your heat dissipation while increasing your thermal transfer through the block so temps will be worse not better.

It's kind of difficult to understand because the whole process goes against intuition, I still don't have a full grasp of it but I think I am starting to get it

There is some more info in this thread:
http://www.xtremesystems.org/forums/...d.php?t=228901

[Utnorris]

Hey Boxmods,
Any chance of making a dual D5 or dual MCP355 res with separate chambers for two individual loops?
Thanks.

[Altspacen]

I am going to chime in here only because I'm curious and after pushing my 920 to 4.2GHz I couldn't keep it stable w/o temps over 90C in a room thats 76F.

Here is a little rundown of my setup - Bay Res - 355 w/ xspc top - HK3.0 - MCR120 - 2 Alphacool GPU blocks for my 4850x2 - MCR320

I believe that splitting up possibly enlarging my 120 to a 220 if room permits inside my case and increasing to a D5 pump would help me keep temps like this in check. I have since de-clocked to 3.8 (200x19) HT Off and hitting 64C max load. I would like to get this box up to 4.0 with HT on with reasonable temps, given the information provided does this seem to be a feasible possibility with running this new T3? I don't see how it would hinder my setup and increasing my flow would totally help, if I could fit more radiator in my box that would also help a good bit but one can only stuff so much in a box and still have it look pretty.

[rambler358]

I am going to chime in here only because I'm curious and after pushing my 920 to 4.2GHz I couldn't keep it stable w/o temps over 90C in a room thats 76F.

Same here, my CPU core temps and ambient temps are almost the same but my cores were just hitting 90. I have my OC at 4.0GHz 24/7 currently. When folding at 77*F ambient temps, my cores go into the low 80s.

Here is a little rundown of my setup - Bay Res - 355 w/ xspc top - HK3.0 - MCR120 - 2 Alphacool GPU blocks for my 4850x2 - MCR320

I'd suggest putting that mcr320 before the CPU. The heated water from the pump output is going directly to the CPU the way it is currently, and not helping your CPU temps.

[Utnorris]

I would separate the loops into individual loops with a second pump and res personally. I can see where this res will help in some situations, but not a single loop with multiple GPU blocks and an I7. The heat from the GPU's will just bleed over to the CPU sub-loop and while the GPU's have a high tolerance for heat the I7 does not.

[rambler358]

I can see where this res will help in some situations, but not a single loop with multiple GPU blocks and an I7. The heat from the GPU's will just bleed over to the CPU sub-loop and while the GPU's have a high tolerance for heat the I7 does not.

True, but I believe the heat from the i7's sub-loop would be hotter than from the GPUs sub-loop, so there should be no issues with the GPU's sub-loop increasing the CPU loop temps. And I think perhaps just the opposite.

[skinnee]

Just a quick update...

Testing is complete and I am working on the data analysis now. Need to snap some better photos (I don't like the ones I took earlier), and get cracking on the review write-up. I'm going to push this in front of all the other reviews to get the data published ASAP.

[quattro_]

good news skinnee thanks

[Utnorris]

Thanks Skinnee.

[Utnorris]

True, but I believe the heat from the i7's sub-loop would be hotter than from the GPUs sub-loop, so there should be no issues with the GPU's sub-loop increasing the CPU loop temps. And I think perhaps just the opposite.

Either way, you are going to be mixing the fluid, which means the temps from one will bleed to the other. If you have several hot components then the will just add to each other's heat output. I think this is a great for most loops with a single GPU, CPU and NB/Mosfet setup, but I don't see how it could lower the temps of a multi-GPU and CPU loop just by splitting them into two sub loops. Would it be better than a single loop, sure, but not better than two discreet loops. Would love to be proven wrong on this since I would rather have one pump and one res instead of two each.

[rambler358]

Either way, you are going to be mixing the fluid, which means the temps from one will bleed to the other. If you have several hot components then the will just add to each other's heat output.

As you stated, the GPUs have a higher tolerance for heat than the CPU. If the GPUs' cooler water can help cool the hotter CPU loop by mixing into the single res, then all the better I think.

Also looking forward to seeing Skinnee's results!

[BoxGods]

Hey Boxmods,
Any chance of making a dual D5 or dual MCP355 res with separate chambers for two individual loops?
Thanks.

I do have a design for a dual D5 and a Dual DDC version. The issue for the dual D5, for me at least, is the loss of efficiency from an overall design approach. The dual D5 design is aimed more at redundancy for mission critical implementations and still uses the dual loop design with PEC.

The DDC version is similar in layout to the other DDC dual pump with a few important tweaks. Vapor is mostly right when he mentioned that the DDC does not scale to the dual loop setup on the D5 exactly as it is, but I did find a way to get it VERY close. That design is a machined POM (delrin) part so it can enter production pretty fast once the design is sold and green lighted. I am not pushing it too much right now because I have a D5 pump top (single loop type) coming out shortly as well as a CPU block that is a totally new direction for blocks. Then there is Rain Maker, a complete line of radiators that I honestly believe will blow you guys away. If you think T3 is a different approach...Rain Maker is 10 times more innovative.

The point is The dual pump versions are down on the list a bit.

I am going to chime in here only because I'm curious and after pushing my 920 to 4.2GHz I couldn't keep it stable w/o temps over 90C in a room thats 76F.

Here is a little rundown of my setup - Bay Res - 355 w/ xspc top - HK3.0 - MCR120 - 2 Alphacool GPU blocks for my 4850x2 - MCR320

I believe that splitting up possibly enlarging my 120 to a 220 if room permits inside my case and increasing to a D5 pump would help me keep temps like this in check. I have since de-clocked to 3.8 (200x19) HT Off and hitting 64C max load. I would like to get this box up to 4.0 with HT on with reasonable temps, given the information provided does this seem to be a feasible possibility with running this new T3? I don't see how it would hinder my setup and increasing my flow would totally help, if I could fit more radiator in my box that would also help a good bit but one can only stuff so much in a box and still have it look pretty.

I think it would yes. Vapor has access to better numbers on all your components then I do so he could probably suggest better loops out of the gate...my napkin math suggests the GPU's on one loop. I also agree with you about keeping it pretty. If you have a rad mounted externally...your just not trying hard enough

Other posts are saying that the GPU's will dump heat into the CPU loop. Though that is not incorrect a more correct way to think about it is the loops reaching equilibrium. You can't just view it as the loops hitting their balance and stop there though as you have to factor in the much more efficient flow rate you also get. Of course more rad (up to a point) helps. Valor is a few orders of magnitude smarter then me so I would 100% wait for his take and for sure go with his loop recommendations.

[Kibbler]

Valor is a few orders of magnitude smarter then me so I would 100% wait for his take and for sure go with his loop recommendations.

Nah that Valor guy is kind of a douche. Vapor on the other hand is a WC guru.

It sounds great that there are so many new designs in the pipeline, I do hope most if not all of them go into full commercial production. Looking forward to it!

[BoxGods]

As you stated, the GPUs have a higher tolerance for heat than the CPU. If the GPUs' cooler water can help cool the hotter CPU loop by mixing into the single res, then all the better I think.

Also looking forward to seeing Skinnee's results!

I bet nobody is looking forward to them as much as me =)

Yes they do equalize.

[BoxGods]

Nah that Valor guy is kind of a douche. Vapor on the other hand is a WC guru.

It sounds great that there are so many new designs in the pipeline, I do hope most if not all of them go into full commercial production. Looking forward to it!

Sigh. You wanna know the WORST part? Honest to God when I proofed that I saw I had put Valor (I watched 4 episodes of Angel back to back so Valor was on my mind I guess) and I went back and changed it...to Valor....again.

Sorry VAPOR. Everyone please feel free to call me BoxMods

That's why when you date multiple girls...try to find ones with the same name =)

Yeah, that pipeline is long and slow as hell sometimes. Fortunately I am old and patient...sorta.

[Altspacen]

The real reason I have placed my GPU's under water was that the 4850x2 sounds like a rocketship taking off when the fans were on. Even with cpu core temps holding at 82C the gpu temps would rise about 1 or 2 degrees from where they were sitting idle, so the GPU temps I don't bother with.

As far as my radiator going into my res, I am a large tonnage chiller mechanic by trade so a little bit of thought went into that. I would much rather have a res full of the coldest water I can produce than to continually cool down a large heated pool. Next time you drive past some of those city buildings and see the "smoke" coming from the tops of them in the winter, that is more often than not their cooling tower. Though its an open loop the same principal applies. Pump -> Condenser of chiller -> Cooling tower -> Cooling tower sump. The added heat of the pump is negligible, the tower produces the coolest water possible and it sits in the sump ready to get pumped through.

[Utnorris]

One more question Box Gods, if I was to drill and tap a hole in the top of the res for a fill port would this effect the efficiency of the res in anyway? I wouldn't think so, but I thought I would check before doing so.
Thanks.

Also, thanks for all the hard work in bringing some really interesting and innovative products to water cooling. I definitely am interested to see what you do with rads and blocks. A completely new design, in my opinion, would be a welcome change.
Thank you again.

[BoxGods]

One more question Box Gods, if I was to drill and tap a hole in the top of the res for a fill port would this effect the efficiency of the res in anyway? I wouldn't think so, but I thought I would check before doing so.
Thanks.

Also, thanks for all the hard work in bringing some really interesting and innovative products to water cooling. I definitely am interested to see what you do with rads and blocks. A completely new design, in my opinion, would be a welcome change.
Thank you again.

It can be done. A few things to keep in mind though.

Try to place the hole as close to the front face as you can--because of the draft needed for molded parts, the wall thickness is greater towards the front. There is also no room to make the plug flush so you will most likely loose the bay directly above the T3, and you will have to pull that bay cover when you are going to slide the unit out to fill it.

Have you tried using the front ports? In almost all installations when you tip your case onto its back the T3 becomes the highest point in the loop. To my mind, even the largest heaviest cases (think DB's back breakers) are pretty easy to rotate onto their backs. Especially compared to moving the case so you have access to both side panels, removing both side panels, removing the rez mounting screws, sliding the rez out (meaning extra tube length needed) and then not being able to bleed ALL of the air out, awkward pouring with a chance of any spillage migrating into your machine, then reversing the whole process to button everything back up.

To drain the system I usually tip the machine onto its back, replace the plugs with fittings and a few feet of scrap tube, tilt the case onto its face and drain right into a bucket. Because there are 2 ports you can puff some air into one side to really speed things up.

Because of the front ports and the three over flow basins I am totally comfortable with running the machine for a few hours on a new install with the plugs out so I can replace any air that accumulates in the T3's domed inner face.

I sent Skinnee a "mock case" so he could more easily use the front ports as they were intended and see if they really help. I am thinking a guy that has to fill and drain loop after loop for testing will find out pretty fast if it is really a useful feature or not lol.

[Utnorris]

Yeah, I have a MM UF2O case and a top port would actually be easier for me to fill then to tip it on it's back. It's quite heavy with all the parts I have installed (check my sig). With a top port I would just remove my top panel unscrew the new port and fill. As far as draining goes, I use QD's and a t-fitting at the bottom port on each rad making it the lowest point and it drians very easy and in a matter of seconds, best damn thing I have ever installed.

Ok, so tapping the top with a port would be doable. Thank you for the answer and recommendation Boxgods.

[skinnee]

I sent Skinnee a "mock case" so he could more easily use the front ports as they were intended and see if they really help. I am thinking a guy that has to fill and drain loop after loop for testing will find out pretty fast if it is really a useful feature or not lol.

The mock case or rack for mounting the TIII into worked really well. What was even better was I managed to get through testing a pump or radiator without getting wet socks for once (BlueAqua, Dangals, turtletrax and Charles_H know how often that happens).

The spill trays on the front of the TIII work great, all I needed was a paper towel handy to dab up my spills that were caught in the trays. Filling and draining were very easy when I tipped it back.

UTnorris, if you're going to be tapping the top, go towards one of the front corners, this will help get as much of the air out and TIII filled with as much water as possible. You will still have to do a bit of tilting, but not near as much.

[BoxGods]

The mock case or rack for mounting the TIII into worked really well. What was even better was I managed to get through testing a pump or radiator without getting wet socks for once (BlueAqua, Dangals, turtletrax and Charles_H know how often that happens).

The spill trays on the front of the TIII work great, all I needed was a paper towel handy to dab up my spills that were caught in the trays. Filling and draining were very easy when I tipped it back.

UTnorris, if you're going to be tapping the top, go towards one of the front corners, this will help get as much of the air out and TIII filled with as much water as possible. You will still have to do a bit of tilting, but not near as much.

Doh! I should have thought of the corner thing, sorry.

[Utnorris]

Thanks guys for the suggestions.