[Review] PrimoChill Typhoon III

[Erasmus354]

Y fitting will only add restriction when converting two waterpaths into 1, never when converting 1 waterpath into 2 such as on the outlet of a pump top because the restriction of the components in the loop will always be greater than that of the Y.

Think of it this way you have 1 serial loop total flow through components = 1.5gpm flow through pump = 1.5gpm.

Now you throw a Y on the pump outlet and divide the serial loop in two parallel loops. You now have two loops capable of flowing at least 1.5gpm each 1.5gpm loop a + 1.5 gpm loopb = 3gpm flow through pump and Y.

Read what I wrote again, then bang your head against the wall for not paying attention.

It isn't the Y fitting that is reducing restriction, it is the act of the parallel loops. If you were to have a parallel loop with something like the T3 where there are two outlets on the pump it would have less restriction than if you made that parallel loop with a Y fitting. It is because the Y fitting typically will have a smaller water path than the tubing.

Now you're not talking about much mind you, but there is some, probably equivalent to the restriction of a barbed fitting.

 

[Erasmus354]

Yea, problem is alot of people ran out and bought this thing due to the whole greatest thing since sliced bread propoganda that was put forth and now you see people giving thier reviews stating "My GPU temps dropped 2 degrees!" The reason is because the majority of these folks had no clue what they were getting into and ran out and bought these things for simple cpu/gpu loops and in that configuration the reason the gpu temps dropped is because there is so much less restriction than the cpu loop that you now have more flow through your gpu but LESS flow through the cpu block so the most important temp which we all care about (CPU) is actually going to go up. Now add to it not enough restriction for the pump and as you have said premature pump failure. Pump failure can be disastrous if you are not around when the pump stops running it can destroy your computer in no time at all!

So now I am left wondering how long we have before all the "T3 blew up my computer" threads start coming from all these unsuspecting folks that ran out and bought this thing.

As far as the folks with complex loops restrictive enough to benefit from this, for the same price of the T3 they can pickup another DDC and run 2 loops with seperate pumps and see far greater benefit than using this.

The only advantage I see this product offering is being able to save some space by having a pump/res all in one type deal but you sacrifice performance in order to achieve that not gain it because on a simple loop you should be running the thing in serial mode in which the EK and Detroit tops are significantly better or else you have a complex loop able to take advantage of the parallel loop configuration but you sacrifice performance over just going with 2 seperate loops with DDC's.

I'm not aware of any modern CPU's/Mobo's that people would be watercooling that don't have a temperature shut down. You're not going to fry anything because your pump failed anymore.

 

[bluehaze013]

Read what I wrote again, then bang your head against the wall for not paying attention.

It isn't the Y fitting that is reducing restriction, it is the act of the parallel loops. If you were to have a parallel loop with something like the T3 where there are two outlets on the pump it would have less restriction than if you made that parallel loop with a Y fitting. It is because the Y fitting typically will have a smaller water path than the tubing.

Now you're not talking about much mind you, but there is some, probably equivalent to the restriction of a barbed fitting.

The Y is not a restriction, the flow capability of the Y is greater than that of the two loops combined so it's not going to restrict flow simple as that. The total flow through the pump with the T3 is 3.47gpm the Y is capable of flowing much more than that. What your saying is equivalent to saying plugging a 3/8" tube into a 1/2" tube the 1/2" tube poses a restriction, it's just doesn't work like that.

I think you are confusing the Y with the actual pump top the diameter of the outlet on the serial pump tops being smaller than the two that are on the T3 could have an impact we would need to see some actual testing to determine that which is one of the main issues not addressed in the review and the basis of many of the complaints.

 

[maeveth]

I am going to have to chime in here and support the fact that the Typhoon 3 may indeed by snake oil. I bought one recently and just received it today. Upon opening the box I immediatly commented on the somewhat shoddy feeling of the unit.

Problems:
-There are also small fractures on the top of the unit(picture below).
-The fitting holes are not really very cleanly drilled.
-There is no channel to allow insertion into 2 5.25" bays without breaking off the rails on most cases
-The seal between the 2 halves is much to hard and feels as if with not too much age would start to crack
-There are no included directions on how to connect the D5 to it.


I ordered this unit directly from PrimoChill and selected to have it shipped with a Laning D5 pump. While I know a Swiftech MCP655 is actually a Laning D5 they said "Laning D5" and not Swiftech MCP655 which I some degree object to since I did not actually get what I ordered.

In addition my order was split into two seperate shipments without any notification except a hand written note in one box "Shipped separately".

I am thus far unsure I am going to actually install the unit in my case due to the above reasons. I will be contacting Primochill tomorrow to try and get this sorted out.

 

[bluehaze013]

I'm not aware of any modern CPU's/Mobo's that people would be watercooling that don't have a temperature shut down. You're not going to fry anything because your pump failed anymore.

Evga Classified and Gigabyte UD5 are 2 of the most recent i've used with no thermal shutdown. Personally experienced a ddc with a bad plug stop running on Classified thankfully I was using the computer and managed to shut it off but it still hit 120c before I managed to shut it down and I noticed right away!

Problems:
-There are also small fractures on the top of the unit(picture below).
-The fitting holes are not really very cleanly drilled.
-There is no channel to allow insertion into 2 5.25" bays without breaking off the rails on most cases
-The seal between the 2 halves is much to hard and feels as if with not too much age would start to crack
-There are no included directions on how to connect the D5 to it.

Second one i've seen with those stress cracks yours doesn't look as bad as this one though!

 

[SonDa5]

Will this work with a radiator that 1/4" female connections?

 

[Archmage]

I was about to make an excruciatingly long post explaining a lot of things (I even typed it out), but I decided... screw it. I'm tired of this, and it will not end until further testing is done anyway.

I have some different comments:

- Manufacturing defects and lack of instructions = not acceptable. I'm very sorry Maeveth. From the first several pictures of samples on XS it seemed to be manufactured well enough (at least compared to what I've seen from XSPC's earlier batches). My friend's EK rev2 DDC 3.2 top was manufactured well.

- I don't approve of posting the T3 review across many forums. I thought it was just here and XS. I see the "campaign," but the data is still useful.

- Pump life is a major concern. My friend and I ran a D5 with very low pressure drop for 2 years before it failed. Hopefully the worst case scenario (Loop #1) is enough restriction.

- I'm an idiot for prematurely suggesting this product. Not because it doesn't perform as I expected, but because of the aforementioned concerns. Pump life isn't a concern to me because I'd have enough restriction in my loops, but that doesn't excuse the possible issue to others who I did NOT warn. I didn't even consider the manufacturing because the originals looked fine.
I'm sorry. I really am sorry. Someone with my level of experience with watercooling products (and products in general) should have known better than to suggest a premature product.

- At no time did I misinterpret the performance of the T3.

This next one is not directed at anyone here (thank you all for keeping this discussion relatively clean):

- The behavior of some others is laughable. Either they haven't read the review, they're just pissed off about the marketing campaign and lack of objectivity of language in the review, and/or they're too dense to understand basic concepts. I saw a lot of immature hate-mongering dated before it was deserved.

All of that said... I do like the format of the RRR forum in particular. I won't dare read through all of that thread though because it's long as hell and from the first few pages I imagine it'd be obscene.

Later.

Edit: If anyone else has discussed the D5 WRT low-restriction and pump-life, then let me hear your thoughts. I know it may be an issue from experience and conventional wisdom, but I found it odd that people were quick to agree with me? WTH? heh... I guess it's an obvious concern then. D'oh.

 

[maeveth]

- Manufacturing defects and lack of instructions = not acceptable. I'm very sorry Maeveth. From the first several pictures of samples on XS it seemed to be manufactured well enough (at least compared to what I've seen from XSPC's earlier batches). My friend's EK rev2 DDC 3.2 top was manufactured well.

No worries, live and learn.

 

[Erasmus354]

Evga Classified and Gigabyte UD5 are 2 of the most recent i've used with no thermal shutdown. Personally experienced a ddc with a bad plug stop running on Classified thankfully I was using the computer and managed to shut it off but it still hit 120c before I managed to shut it down and I noticed right away!

That's odd because EVGA says that all their boards have auto temp shutdown...

Our boards have built in safety features that will shut the board off if the CPU temperature reaches an unsafe threshold. There might be software to do it as well but I do not know of any offhand.

There is software that can auto shutdown on temps as well, but obviously doesn't work as well as a hardware solution.

 

[bluehaze013]

That's odd because EVGA says that all their boards have auto temp shutdown...

Originally Posted by EVGATech_Kyle
Our boards have built in safety features that will shut the board off if the CPU temperature reaches an unsafe threshold. There might be software to do it as well but I do not know of any offhand.

There is software that can auto shutdown on temps as well, but obviously doesn't work as well as a hardware solution.

Do you have a link to that post? It's news to me, it's not in the bios anywhere and I can assure you from personal experience it doesn't work if it's supposed to be there

I used software to shutdown if temps exceeded 100c after having the initial scare but yea it does not really work because when CPU overheats the computer bluescreens so the software can't actually shutdown the computer.

I have since installed a bigng with sensor hub and flowmeter. It employs a hardware shutdown through the atx switch, basically presses the power button for you if flow falls below a certain threshold or temps exceed certain thresholds etc... many ways to configure it. It's a really great piece of kit.

Edit: Found the thread http://www.evga.com/forums/tm.asp?m=100800141 if you read through the thread you will see several users stating they have also gone over 100c without any shutdown on the Classified and it seems the "EVGAtech_Kyle" is now recommending software to shutdown the computer if temps get to high... Glad you pointed out this thread though, should be interesting if EVGA responds.

 

[Archmage]

I went to search the EVGA issue (my friend has an evga board and is watercooled) and saw your posts in that Evga thread, Bluehaze.

That's a bummer. I think you could also use the software set to a lower temperature (like 80C) before it BSOD/hangs. That way it'll successfully shut down. This is what I've told my friend to do, but that was before I assumed the hardware feature worked. Hell I hope my solution works...

Edit: You found that thread as I was posting... hilarious

 

[bluehaze013]

I went to search the EVGA issue (my friend has an evga board and is watercooled) and saw your posts in that Evga thread, Bluehaze.

That's a bummer. I think you could also use the software set to a lower temperature (like 80C) before it BSOD/hangs. That way it'll successfully shut down. This is what I've told my friend to do, but that was before I assumed the hardware feature worked. Hell I hope my solution works...

Edit: You found that thread as I was posting... hilarious

Yea I did that for awhile, Kept Core Temp open and running with thermal shutdown, I think I had it set to 90c. I run i7 920@ 4.6 temps in low 80's so have to set it higher and it does work but just seeing what could happen I didn't trust it for a final solution. Really glad I picked up the bigng the thing works great set profiles for my fans and pump on it and it automatically ramps fan and pump speed up and down with water temps so PC is dead silent when i'm just surfing the net or emailing whatever and then they ramp up when gaming or benching and it all works independent of OS so no programs to run and take up resources. Really happy with it.

 

[Erasmus354]

There really is no excuse for any of these boards to not support thermal shutdown on the i7. Intel makes it pretty simple, the processor trips a signal when it wants to get shutdown. All the motherboard has to do is watch for that one signal.

6.2.3 THERMTRIP# Signal
Regardless of whether or not Adaptive Thermal Monitor is enabled, in the event of a
catastrophic cooling failure, the processor will automatically shut down when the silicon
has reached an elevated temperature (refer to the THERMTRIP# definition in
Table 5-1). THERMTRIP# activation is independent of processor activity. The
temperature at which THERMTRIP# asserts is not user configurable and is not software
visible.

That said, the processor does shut itself off, it disables the internal clock. However the motherboard still has to shut off power to the processor.

 

[skinnee]

The System flow Rate comparisons I ran will be added to my Pump Comparison pitching both D5's and DDC's in one set of scatter plots, we'll finally get to see PQ's together. Its going to be an eye chart, but it will show the full sprectrum of PQ. I've already been IM'd numerous times by someone who wants RD-30 testing included and really wants this whole Y/parallel loop discussion to be over. But there is still more testing and information that needs to be done before this parallel discussion can be boxed up and set on the shelf...

For each loop the configuration and components can be found in the OP. Y loop were split up the same for the Y tests as the were for the T3.

On to the system flow rate comparisons...

​

As I stated in the original T3 review, the Single Loop (TITO used in the charts) performance on the T3 isn't earth shattering and does not match the DDC w/ XSPC V3 top or PQ boasting D5 tops. Parallel in general shows how seperating restriction from a serial loop to two legs can boost flow rate for each individual leg, but there is quite a bit of testing that needs to be done to either support or rule out the level of balancing. From what I am seeing on a purely flow rate perspective it doesn't appear that the balancing or planning is as crucial as we once thought...but we'll find out for sure soon enough as PQ testing on the Y configs has to be done now.

 

[magoo]

Just as most suspected...........apples to apples..........meh.

 

[Archmage]

Just as most suspected...........apples to apples..........meh.

Are you referring to the serial flow rates (expected) or the fact that the parallel "Y" flow rates of the other pumps are competitive with, but in general did not outperform those of the T3 in skinnee's specific configurations?

I think we need to know if this is a "dual-Y" setup or single-Y. I'd think we'd see a slight improvement with single-Y, however it'd require a multi-port res.

In any case... from what I've see we can now get:

- top fill-port
- more aesthetically pleasing bonding strip
- fewer manufacturing defects? I've read some cases of RMA's of some of the original batches. Hopefully Primochill is taking care of customers. Too early to tell...

We still need to wait for further test results I suppose.

 

[maeveth]

- fewer manufacturing defects? I've read some cases of RMA's of some of the original batches. Hopefully Primochill is taking care of customers. Too early to tell...

We still need to wait for further test results I suppose.

After contacting Primochill they took back the unit. I am waiting for the refund to come through. So far it was a good experience. However I am still waiting for the actual credit to come through.

I hope Primochill is able to significantly raise the production quality of the unit.

 

[Archangel7]

Been a while since I posted about a water cooling product, but I thought that my 2 cetns may prevent someone from my experience with this unit. Caveat, I did use Feser cooling liquid in unit, and even though I do not believe this was the cause in fact of the problem I had, I want to be perfectly candid regarding the subject. Also, given that I did use same, I am not entitled to be angry or otherwise upset that something went wrong.

That being said, I reviewed skinnee's review of the Typhoon prior to buying, and ran distilled water through the unit for approximately 2 months, or a little less. I switched to feser when I added another vid to the second loop. My setup is as follows:

1st Loop: 1/2 inch/T3 >rad>Y>cpu & NB>Y >T3; 2nd Loop: 3/8 inch/T3>vid sli>T3.

All was 24hr tested, and the other day I arrived home, booted up and started puting about on the net. Few minutes in, I hear snap/crackle/pop, not good! Shut down and looked inside to find a pool of feser, black no less< at bottom of case. Now for a litle more embaressment on my part. Have a 5.25 bay powering vids, and yes I took a chance and placed typhoon above, due to it being the best config. So again, my error.

What I found was that 1/2 barb out had broken off at the base of the ghost fitting where thread begins. when I disassembled, the in 3/8 bust off at the base of the barb. Now I do not know if this was because of the tube tension, or whether or not the feser weakened the plastic used, but it was not something I expected to occur after reading skinnee's review. Not blaming skinnee, but I did rely somewhat on that portion of the review that stated the fittings could take punishment.

Anyway, I paid dearly for my errors. Power supply is dead, which really does not bother me that much, but unfortunately I have beige carpet in my residence. Side note, do not use black feser if you do not want to spend 5+ hours trying to clean it up if your system leaks. Still did not get it all out, and that stuff spreads when water is applied. So now I have a nice big blue spot, one less power supply, and a lot of mental anguish to go along with it (mostly anger at my own stupidity).

I also have several severe cracks appearing in and around the embedded threads, and was an early purchaser, so the front face defect is present, too. I will say that it worked well for my setup, and my temps were good in all respects. I suspect that the material used by PC is not the best, and do not understand why PC would make a clear bay res without making sure that the material was resistant or impervious to chemical degradation from the popular cooling liquids out there. Again, I do not know the cause, and it may only be a coincidence that the barbs cracked at this time.

 

[vjcsmoke]

Looks like this product has been discontinued as of May 19th 2010. That's too bad, it sounded like a great idea. Will there be a Typhoon IV??