Cuplex-Evo Water block?

[Pug]

Originally posted by unregistered
but I suspect you do not understand just how small is the internal ID in the swivel section
have you actually measured it ? what is the ID ?
you made a statement, I am asking for the numbers on which your statement was based
(I have, its in the article eh ?)

The angled fittings, as supplied (by the manufacturer) with the blocks have an internal diameter in the swivel section of approximately 4.28mm by my measurements.
The straight ones are closer to 5mm
The angled fitting I have on my own block, by comparison, is 5.5mm.

I'm not sure what conclusion you think I'm trying to draw here apart from the fact that these connectors can serve a purpose if you choose to use them for efficient hose routing.

Oh and I subsequently received an apology from the forum member you cite.

[theetruscan]

Originally posted by unregistered
system resistance is the incremental sum of all the individual resistances
so increasing the bore anywhere will always help
more to your point, by how much ?
really depends on the rest of the system

in the case where the wb's internal flow path cross-section is the same as the fittings, not so great
-> but even so worthwhile for anyone with an interest in performance

...

hi ho (Vonnegut)

I am confused by something. Perhaps it is made clear in the article by this phrase: "connector chamfering mods." But as I have no idea what this means, I will ask what may well be a very stupid question.

I can't figure out if your above reference regarding increasing the barb size is speaking of improved flow because of using larger diameter tubing, or if 1/2 vs. 3/8 inch barbs alone make a significant difference when a block has 3/8 inch Cross sectional area internally.

I know this is neither really here nor there as this debate seems to be regarding swivel elbows, but I'm curious and confused.

[BillA]

to chamfer is simply to slightly bevel or radius the inlets and outlets to remove the abrupt shoulder

my previous post addresses the cross-section
bigger is better, generally

[Pug]

I was hoping the thread would return to the Evo, rather than focus on the swivel elbows.
I am still happy to discuss how we could get the best out of the block though (in the spirit of these forums).

I'm not _that_ concerned about redesigning the block (Aqua Computer have already done so with their Rev 1.1) but I do see a question raised frequently as to fitting 1/2" barbs to it.
I was genuinely interested in your opinion of this alternative mod (although I was of the opinion that it was kind of unnecessary).

The angled connectors I have are already chamfered where the tube narrows and the internal design of this block is so far removed from that of the 462-b, that I don't see that your particular mod is as relevant - but then I am always open to entertaining the alternate viewpoint.

You refer in the article to "those chromed push-in 90s" but I can't find a pic of them to compare.
Mine are Nickel-plated, not chromed. Maybe they aren't the same type. :shrug:
(I have already seen differences between connectors I have used from Festo, CompAir Maxam (as was), Legris (3000 series, 3600 series, etc), John Guest and others, so I wouldn't be surprised).

[theetruscan]

Originally posted by unregistered
to chamfer is simply to slightly bevel or radius the inlets and outlets to remove the abrupt shoulder

my previous post addresses the cross-section
bigger is better, generally

I'm sorry unregistered, one more question on the Barb size question.

My question was poorly phrased, sorry. Bigger is better is obvious. I just didn't understand whether you ran the 1/2 barbs with the same tubing and an adaptor, or with bigger tubing when generating those head loss charts, and if the change in tubing size played into results if there was one. Thanks. My apologies for the bad phrasing.

[BillA]

Pug
I can stipulate that there are many of these fittings, and that assuming they are all the same is not warranted
and additionally that I have MUCH more actual data than others, most of which is ‘unpublished’;
so what is quite clear to me is often not so to people without a background in the subject

jeez, the 462 mod is not relevant -> compare the effect of changing the connectors
did you not understand the previous post ? (or is this just about being 'right' now ?)

below are the 2 I have compared
the push-in 90 is that of Swiftech, marked "UK MADE 3/8"; min. ID ~5.2mm
the straight is from Innovatek (and also a Korean mfgr), no markings; min. ID ~7.1mm
while a ¼”NPT x ½” barb has an ID of ~8.7mm



now another of the graphs you love to hate,
this time comparing the connectors on the Innovatek



what are the conclusions that can be derived from all this ?
if you are content with the adequacy of the present installation -> do nothing, of course
if you wish to improve its performance, junk those 90s forthwith
and so we return, again, to my initial post

bye

[Pug]

So, bearing in mind that the thread is about the Cuplex Evo, do you happen to know the thread size of the Evo, or are you just assuming it's 1/4"?
Armed with the correct information, what is the largest ID available for a 1/8" thread? ([Edit] I meant 1/8", not 3/8" as origginally typed, sorry.)
Or are you suggesting people bore out and retap their blocks based on your findings with a totally different design of block?

The picture you showed appears to have a 1/4" BSP parallel thread compression fitting next to the push-fit angle. I wouldn't advocate using a taper thread to replace it but that's just me, (I've known people crack acrylic blocks this way).

It's not just about being right, it's about discussion until all doubt and confusion is removed from all sides (mine included).

I didn't say it wasn't relevant, I said it was unwarranted.
The original poster asked for people's experience with these particular blocks and was initially shown an example using one of the optional fittings choices (perhaps commonly known by all as the lowest performing configuration out of the options available - this point is moot).
If you look closely, you will see that the first version I posted a pic of doesn't use the angles at all.

I have run straight fittings in the block before. I can get similar temperature drops from relocating my case to the floor instead of on the desk.
Having designed my system around the possible future integration of my chiller into the loop, the angles aren't that massive of a concern for me.
I'm happy with it.

Oh and I have no aversion whatsoever to your graphs, I merely stated that _some people_ would rather know the conclusion to your findings illustrated with a link to the article (perhaps even with the graph, if it illustrates the point).
I read your posts and the article and still don't know what your temps were.
Not that it matters, of course. I had already got your point before you made it.

Perhaps the reason I can't let go of this is that anyone glancing at this thread may think that the block only comes with those fittings options and from your statement they might infer that the block itself is not a performer.

I must admit, all the way through this thread, I have been assuming you have no experience with this particular block.
If I am wrong, let me know.
If I'm right, however, let's see what we can do to change that.
YGM.

[pHaestus]

One can look at the pictures of the block and know that they are not 3/8" NPT or BSP. If they were, it would be possible to use a 5/8" ID barb on them. They are not, however.

How big is the temperature difference? I am not comfortable enough with the block design to throw a number out. I AM, however, quite surprised at the vehemence of the watercooling community to "free" performance. It has been shown several times by Bill (the original 462 data, the innovatek blocks, and again with the -U vs. UH swiftech blocks) that substantial performance can be gained by changing fittings with an eye to waterblock design parameters. I also wrote an article on changing fittings and researvoir placement to make better use of existing pumps. FREE performance! And the community just argues with the Darcy equation rather than assimilating it into their own design plans. Splendid.

Glad to see you guys have managed to keep the thread civil. I don't think moderators are really even needed around here

[Pug]

D'oh! Thanks pHaestus, I meant 1/8", not 3/8" :blush:

Didn't mean to argue this much, just didn't want to be misunderstood or misquoted.

[BillA]

lol

well, I'm hoping you meant 1/4"

[pHaestus]

As was I

[Pug]

Nope, sorry. G 1/8 BSP parallel it is, hence my point.

[Whomp]

So, no that it's calmed down a bit, I need this block to run in a 1/2 inch system. Someone told me I could just get NPT connecters and crank 'em in.
I want the the innovatech type connectors shown in the pic above, but it has to be able to accept 1/2" ID - 5/8" OD Tygon.
Think this is something I could pick up at the "local" hardware store (80 miles from here), or is there somewhere special I'll hae to go to get them?

Thanks,
WhOMP!!!

[BillA]

Originally posted by Pug
Nope, sorry. G 1/8 BSP parallel it is, hence my point.

yes Pug
you have clearly demonstrated that you are more clever than anyone here
my congratulations

you showed your cleverness by gaming us:
putting out the bait
jigging it up and down
getting us to take the hook
-> and then pulling the switch (providing the essential piece of information to make your point)
saying with your smiley 'gotcha'

is this what you learned on that other forum ?
well, here you WILL get the redass for it
congrats Pug, you made my ignore list

Whomp
pretty sure no such fitting exists
particularly G 1/8 BSP parallel to 1/2" barb

[Pug]

Hmmph.
Not a lot of facts in that statement then, were there?
I suppose a simple apology is beneath you. Shame.

Facts:
Topic = Cuplex Evo
1st line of post = "Has Anyone used one of these?"
1st line of reply = "They're reported to be quite good, nearly as good as Innovatek blocks. Their only drawback is they're in 10mm (3/8") max, you'll have to mod them to put larger barbs if your system is in 12mm.
(Emphasis mine)

There was no switch, the blocks have had 1/8" fittings since the Cuplex (before the Evo).
The big smile was because I thought I might have finally got you to understand the point I was trying to make.
It's not my fault if you're too busy insulting me to notice (or check your facts).
I hope your threat was a hollow one and you do happen to read this but if not, I can't force you to.

My cleverness is immaterial, I may have shown my calmness (and that even surprised me, under the circumstances) and perhaps a little tact and diplomacy where necessary.
I learnt on the other forum not to be arrogant because those who do, often end up with egg on their face.
To quote somebody, somewhere - Stick to the facts and ask questions if you're not sure (although I may have made that last bit up myself).


Whomp - the only thing I have found that may be of use is a G1/8 male to G1/4 female BSPP increaser.
Here's a pic of a couple fitted to a reservoir

Trouble is, I don't know how much they will increase flow restriction and if you use a large pump it may even cause cavitation.
(Happy to hear input on this btw.)
Min ID is 5mm (chamfered inlet) for a depth of approx 6.5mm.
I do know of someone using them in a reservoir with a 10mm ID pushfit system on a White Water block who hasn't reported any problems yet but I don't remember what pump he was using off the top of my head.

[Edit] Actually, Whomp's question is one I recieve a lot.
If you look closely at the first pic I posted of the dual Evos, the more eagle-eyed of you may have spotted another potential solution, although it may not be as effective in a serial or single cpu configuration.
Essentially, they are 10mm barbs, sleeved with a short section of 3/8" hose, with 1/2" pipe clamped over it.

And btw, if I'm not welcome here, LMK and I'll bow out gracefully.
Don't wish to make waves.
I have full respect for many posters and all staff on this board.

[oc192sonet]

Greetings everyone, I am back on the boards, nice to see you all again. Here is some eye candy for you all:

http://www.cray.com/company/video/

I know it is not subject related, but ideas come from strange places. Enjoy!

[oc192sonet]

Hey check this out:

http://www.vita.com/vso/vso200209/18...-Spraycool.pdf

Also there is a conference going to take place May 13, in Scottsdale, Arizona:

http://www.vita.com/cool/alcws.html

Is this the appropriate forum for this?

[redleader]

Hit "New Topic" next time instead of reply

[funktional]

Hi there,
I'm german so I can read all the test with the cuplex EVO
The main Problem is that most test are very bad because the testers usually mess around with aircoolers and have no idea how to test an Waterblock and how to compare different waterblocks. But one test had been done on a professional test station, and although the test station didn't seem to work correct all the time it looks as if the evo is not that good. Its Better than the older Generation of German Waterblocks but the actual Generation of many companys is better than the EVO or the Innovatek Waterblock. Revision 1.1 comes closer to these but still can't hit them.