Oops. Got distracted by watching "Scary Movie 2" before completing the second post. :rolleyes:

With the Delrin, the XS is a viably machinable option. Actually I can't get over how clean it looks in comparison to the polycarb. Where the polycarb was visibly "stressed" from the machining process, with more than a few split tubes (not that the incorrect cutter depth helped much there), the Delrin looks near perfect. Some slight "fluffiness" at the edges of the tubes which can be expected on something this fine, but 90% of the tubes look perfect, with only 10% or so with fluff that needed to be cleaned away, and once done, also look perfect. Truly the piece is somewhat like looking at some sort of optical illusion, with the detail so small as to trick the human eye without some magnification assistance.

However, having said all that, it looks like I'll be scrapping the XS, in favor of the dark horse.

A while back at OCAU I hinted on a cheap to make design that would match a White Water, but would be far cheaper to make. The design started out life originally as one of my previously unshown GPU block prototypes. I scrapped further work on it due to limitations that I'm hesistant to explain here, as even saying that much may be giving too much away. Well I decided to re-prototype it, with a few added tricks and tweaks, but basically focused around a budget-block frame of mind. In simulations the figures hinted at something very good and I took the design a little further (I can't resist), not really trusting the figures so much in terms of absolute performance, but at least as a way of improving performance without really adding much cost.

Anyway, that block is what I've referred to as the Prototype #2, or P2 for short. The figures hinted at substantially better than Cascade performance within certain constraints, but overall I was basically expecting more like White Water performance.

I got the P2 and stuck it on, and on the first mount I was seeing (on the same test-bed as above) 1C lower CPU temperatures than with the Cascade SS, and ~5MHz shy of the Cascade SS's overclocking performance. I pulled the block off and noticed an uneven thermal paste imprint, which meant that there's basically more to offer. I remounted a couple more times and on the third try, the temperatures remained the same (1C better than the Cascade SS), but was now seeing 10-15MHz higher overclocks.

So basically the copper-based P2 is significantly out-cooling the Cascade SS, and allowing for stronger overclocks, however small the gains may be, it's still forwards progress.

Given the choice between the near-equalling performance but very expensive to make XS which is at the absolute end of the road for further improvements with it, and comparing that to the ever-so-slightly better P2 which is very cheap to make and has a lot of room for improvement (at a dollar cost), then basically this makes the XS a redundant block.

I can't believe it. Your actually getting close. Keep thinking. There is away and it has to do with the cups. Think about making this "secondary impingement effect" last longer within the cup. I would venture to say, the cups in your cascade are less than optimal but your limited in design choices being that small. This is something I'v been working on for the past year and a half trying to get it right with my limited machining , money(kids gotta eat). Your WWW is what got me thinking about a pure impingement block. But as you stated in your post, being restrictive has been the problem for me. The performance is shocking though paired with a strong pump. There is one more step beyond that can aid convection even more that was discussed between us last year. Making it practical is the challenge but it does work well.

thanks alot, that was alot more then i had hoped to get out of you.
not sure if you'v covered this or already refused to devulge, but have to ask anyways, can you give us a remote time frame to a first glimpse at the p2, i know it depends on alot of variables, many of which are not directly under your control, but can you let us know if you're looking at 1 month or 1 year until we can start to get a glimpse at this mystery block?

great work! sounds like it's going to be a winner, and also good to hear about the success witht the xs, even if it won't happen, always good to push the limits a bit.

Thanks Syscrusher. I have explored lot of options with the Cascade approach including what I described above, but have settled on this P2 thing as being the best way forwards from here.

As much as BillA will hate me saying it, am fairly sure I'm seeing h of over 100K here (on average), and am looking to boost that to the ~120K mark with the next set of enchancements.

Once I reach that level and am confident, then I will reveal all.

Honestly from the way you describe the P2 on OCAU as being so much simpler and also such a good performer I really hope to be able to see it someday. There is an elegence in something that achieves so much with very little. Market wise I'm sure it'll be the winner, a Lamborgini @ Volkswagens price hopefully. While the alure of the all out high dollar Uber block with the flashiest design and highest price tag will still be sought after by those more concerned with bling than performance, I think the mainstream enthusiast is more concerned with the almighty c/w rating and how much they have left over for some Tygon. Pesonally I spent pretty freely and just want the most bang, not the most bang for my buck, and from scoping posts for the past few years I am a lil' outside the mainstream. The P2 sounds as if it'll cover all three bases though. The most bang, the most bang for your buck, and the bling of owning a Cathar designed block.

Okay, it's really late here and I probably shouldn't give even this much of a hint away, but here it is with my mind being tired and all. Nature is such a wonderful teacher.

"What is one of the most violently turbulent and destructive forces in nature?"

WoooHooo TORNADO???!!! Run for your Aunty M DD

I would lean more towards a hurricane or typhoon. ;) Anyways, it sounds very interesting to say the least from the price/performance standpoint. Also the design....trying to think of what you are doing with the block design and not quite sure :)
Edit: maybe a waterspout? A tornado of water :eek: http://www.southdown-amateur-radio-s...aterspout.html

Ahhh but what natural occurence produces the highest velocities and LOCALIZED TURBULENCE ? Thats what impingement is all about.

Tsunami?
Taxmanian Devil?
Thermonuclear weapons?
Drunken frat boys in office?

A woman with PMS?

In that case the most effective form of cooling would be the "chill" that occurs before the storm. I did say that the design is nature based though, not chaos theory based.

On a side note, it appears that the Tasmanian Devils are all about to die out. There is some mysterious disease which is affecting the population. Their mouths start off with sores which progressively gets worse until they can't eat and die. This has just occurred in the last six months or so, and it appears that none of the Devils have any immunity to it. If you ever wanted to see one in real life, then in the next few years may be your last chance.

Their situation is also not helped by some idiot in the last year bringing over foxes into Tasmania (which was previously fox free) to hunt for sport, and the foxes have now started to multiply and kill off the local Tasmanian wildlife, which was one of the last most natural and unspoilt fauna reserves in the world.

Of the items on your list, the last two are basically human-based, and it seems sad that perhaps that may be the true answer to the riddle....

Well that was my depressing thought for the day.

Hurricane!!!

Hmmm, swirly bits...

Anyone ever gotten water to spin rapidly, and then watch for how long that characteristic continues even after turning corners?

Hmmm rifled jets?

Hey...does that mean you'll have to make 2 versions- one for south of the equator and one for above LOL

Nope. Olympics anyone? :dome:

And it has to swirl faster if you taper your chamber towards the outlet (conservation of angular momentum).

Ooh! I know! Mini spirals! (Slaps self, so others don't have to) jlrii: If you're forcing the spiral, it's irrelevant.

The sun?
Fusion?

...and...what happens after that...?

The water reaches relitivistic speeds, and reaches aboslute zero, right?

Stew, it will be interesting to see how you integrated a 'tornado' affect into the Cascade.

Actually the most destructive 'natural' force on earth is lightning (per area).

Still, can't wait to see the Cascade Tornado.

Well, if you get swirling really fast, centrifugal force opens up a partial vacuum in the center, which means more of your water molecules are closer to the metal, plus a small amount phase change fun. Of course you still have the overall goodness of higher pressure and higher velocity at your boundary layer.