Let's have a competition

[n00b 0f l337]

Havent done one in a while, I wouldn't mind putting up some capital for prizes either, competitions generally lead to new ideas.
Now how should we do this, I'm not feeling a budget competition, as that wont bring around new ideas, except how to ghetto up a system massively, and we seem to have autocascades under hand. I know this is outa section, but what havent we done?
A chiller competition? : Compete on temps, size, noise, price, power consumption, load capacity, looks
A cascade competition? (a little exclusive though and a bit more dangerous) : Compete on temps, size, noise, price, power consumption, load capacity, looks again
Or do we do another competition, possibly like an evap competition, or an HX competition, or something to that sorts?
I could put up possibly a evap or a lineset, maybe a plate HX for prizes.
I'm sorta liking the chiller idea, and I'm sure we can find some TXV's to play with ;)

[Jack]

A chiller competition would be nice.
But how will you loadtest it? Not everybody here has watercooling (pump,block etc.)

[n00b 0f l337]

Everyone has a load tester right? We could pass around a pump and such I guess? Any pump will work really to the same load tester.
Ive got an idea for a "Build a Product" competition, I'll write it up and post here again. Lets just say, make something, give it a price, and then be able to remake them and sell at that price.

[[XC] gomeler]

I'd rather not do a chiller competition as frankly chillers bore me :shrug: . Cascades would be a fun thing to have a competition for considering how many beautiful builds we've seen as of late. I just don't relish having to use CO2 as that seems to be the default 2nd stage gas for open competitions.

[Fhqwhgads6680]

gomeler;2362500']I'd rather not do a chiller competition as frankly chillers bore me :shrug: . Cascades would be a fun thing to have a competition for considering how many beautiful builds we've seen as of late. I just don't relish having to use CO2 as that seems to be the default 2nd stage gas for open competitions.

I second that.... chillers...are just... huge...more expensive...require wc gear... and I just find pointless... Just MHO but I never understood the use of chillers...

[Xeon th MG Pony]

A chiller is an add on to a water cooling loop, they are far superior flexibility.

They are not huge, they are made huge by poor design or ignorance in the proper requirements to make a compact unit (But then again huge is rather subjective, to me a 5TR system is small!) . Chillers are still one of the most critical of cooling systems in any industry!

[[XC] gomeler]

Size is rather subjective but I've created 400w-500w chillers within singlestage cases. Most of the chillers you see around here though are huge A/C conversion monstrosities :down: Chillers have a great purpose if you want to cool multiple sockets or sockets + GPUs/NB w/o having a multi-headed beast. For dual-socket setups chiller is the way to go.

edit: but I vote for a cascade competition or exposition but that's just because I'd enjoy it the most :)

[teyber]

dual head competition with custom evaps?

[Fhqwhgads6680]

I'm just saying there is more involved, more components to fail and more of a pain to get the same perofromance that you would out of a single stage... its not a bad idea, and I am not trying to thread crap. I just think this is the vapor phase forum so let keep this direct die oriented.

I do however like the Evap, or HX competition idea!

[[XC] gomeler]

Custom evaps would limit a lot of people as not all of us have lathes or CNC machines in our garages :( I'm stuck using either steppers or cryostars unless I purchase machine-time locally and that costs an arm and a leg. Dual-head could be fun, make a nice little SLI/CF setup but would have to assemble another load-tester.

[LardArse]

Toying with a hybrid LN2/Cascade cooler would be cool.
Evap with a portion outside of Cascade loop, piped out for LN2 to be poured in. So when you need to, you can pour in LN2 to drop temps. You can run it as it is, and you can save on LN2 use too. You just need an accumulator to catch all the Ethylene when evap is below boling point.

[Xeon th MG Pony]

I'm just saying there is more involved, more components to fail and more of a pain to get the same perofromance that you would out of a single stage... its not a bad idea, and I am not trying to thread crap. I just think this is the vapor phase forum so let keep this direct die oriented.

I do however like the Evap, or HX competition idea!

Some thing tells me you've never seen a proper chiller.

Allot of what you said is quite wrong

[phelan1777]

Guys, I know this is XS, but, for the sakes of a competition............K.I.S.S....

Beauty is in the simplicity.

[rogard]

How about a phase unit that increased its cooling as the processor increased its load?

[[XC] gomeler]

Don't we have valves to do that? TXV comes to mind.

[Fhqwhgads6680]

Some thing tells me you've never seen a proper chiller.

Allot of what you said is quite wrong

I've seen many chillers, and I have messed with them myself when I started going sub ambient... I said its IMO but I just never like them for our application... I didn't say there aren't uses in industy...

And how would there not be more components to fail?.... the ARE more components... so...there ARE more to fail... and it IS generally more difficult to get the same performance out of them... you are never going to get 100% efficiency, so its only logical that a all things being equal take a direct die single stage's performance and take that same dd except convert it to chill liquid and your temps the the waterblock HAVE to be higher than the temps would have been at the evap of the DD....

anyway, I'm done aruging not trying to get OT I am just stating why I don't think we should do a chiller competition... this ISN'T the Chiller section...so I for one think we should do something else...

I was thinking maybe if we did an evap contest if we could find someone with access to a CNC and people could do the designing and provide the materials/dimensions required. Then have one the the judges or a volunteer with a prebuilt system with hand valves setup to change evaps do the testing with the same load tester in the same ambients with the same gases cap tube etc. I think we'd have some valuable information to gain from this contest aswell!! Just my ideas

edit: actually rogard and gom why not a TXV based SS competiton? I am sure we can get a TXV based SS to work decently and I for one have always been interested in how well a TXV based SS would hold temps at load. its just a thought I haven't messed with TXV's alot so I am not very knowledgable about them.

[[XC] gomeler]

Used a TXV before on a SS, 1/4 ton Danfoss with 00 orifice (think that was the number) and honestly I couldn't warrant the $40 increase in price and added complexity for a single-stage. Cascades and chillers due to the larger pulldown, perfect, single stages though TXV is overkill. I'm sure others would disagree but everyone has an opinion.

[Fhqwhgads6680]

I just always wondered if a tuned TXV system would have a lower Delta between load and idle temps then a cap tube or cpev based system.... cause in theory they should right? or in reality are the manufactured TXv's all too big for our application?

[teyber]

Gomeler, im 90% sure you have a drill press... no? I think the evap idea is great. Just get a pin-set of files, and my friend... I am in the process of making a mill out of my drill press.
http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=538
they have them in 5" too, hook it up to your drill press, get a mill bit at www.mcmaster.com (or is it www.mc-master.com ?) and a spiral evap would be really tricky, but just a simple design would be nice and not very hard..

Or, how about, somebody comes up with a design, and whoever can immatate it the best wins? something like that would be fun.

[tim-]

gomoler:
I agree it's useless for SS, but not for same reason as your's. I have try it and I don't recommend it 'cause it doesn't work good. it's to slow for work proper in a SS.

[Xeon th MG Pony]

to fast of a temp flux, a TXV would be plagued by hunting and other issues in a small system like that.

[[XC] 2long4u]

How about a competition for people who have never built a DD or chiller.

[teyber]

Im in! i got a 5250btu compressor i got for free, and for christmas im going to get the basic tools needed and some r22 :D im down for that!

[teyber]

or in age categories and skills.

[[XC] gomeler]

While it would be fun to see a spat of such builds, I'd be worried about the safety mplications. Setting a deadline for a first-time build and if any newbie was feeling pressured and took shorticuts :S

I have a drill press at home, little 1/2 HP with XYZ adjustable mount, perhaps I should arrange for it to be dropped off at my current townhouse. I used it for drilling out simple waterblocks and making precise dies and such but not attempting to mill out an evap. Wouldn't be able to be done anytime soon though :(

If we did it by age that'd be rather hilarious but I don't know if we have enough builders to fill out a bunch of "heats".

[teyber]

lol. Idk, i find it interesting to see peoples ages here.

Im thinking like 16 and under, 24 and under, 25 on up? something like that?

and yeah, if people like me took shortcuts... well, what walt would definatly say(for a good reason) it would definatly not be safe.

[teyber]

AH!
another idea!

BEST SS CASES!!! MADE BY YOURSELF!

[[XC] 2long4u]

gomeler;2362896']While it would be fun to see a spat of such builds, I'd be worried about the safety mplications. Setting a deadline for a first-time build and if any newbie was feeling pressured and took shorticuts :S

I have a drill press at home, little 1/2 HP with XYZ adjustable mount, perhaps I should arrange for it to be dropped off at my current townhouse. I used it for drilling out simple waterblocks and making precise dies and such but not attempting to mill out an evap. Wouldn't be able to be done anytime soon though :(

If we did it by age that'd be rather hilarious but I don't know if we have enough builders to fill out a bunch of "heats".

If you can't build it safely you shouldn't be building one. The age idea would be good I guess except I wouldn't stand a chance since I got a late start. I'm 29.:wave:

[teyber]

o. I guess me and bretterbeck would be in the same league, and thast about it XD and that would be so funny seeing how bad i would fail... especially since im gunna be using r22.

What do you guys think about the cases?

[rogard]

What about a different mounting mechanism that didn't require dieeletric grease/neoprene etc?

[[XC] gomeler]

Goodluck with the mounting mechanism :) You basically need insulation against heat-leakage and then a vapor barrier. So you are pretty much always going to atleast need a material that'll form a tight seal.

[rogard]

gomeler;2363191']Goodluck with the mounting mechanism :) You basically need insulation against heat-leakage and then a vapor barrier. So you are pretty much always going to atleast need a material that'll form a tight seal.

hehe i didnt say it was easy :ROTF:

[Section8]

I would love to see what you guys could do with a dual head SLI/Crossfire setup :)

[teyber]

Drewmeister over at extremeoverclocking.com has an option for his prodcuction of units that has heaters in it, and requires absolutely zero condensation proofing...

[Stapler]

Chilly1's evap also has heaters, as well as vapos. Insulation is still a good idea

[[XC] gomeler]

I prefer not dumping 15 watts of heat into the system, I haven't had any problems with using a small shim around the socket and a backplate to protect against frost-up at idle. SLI/CF system would be interesting, if I get two votes than that gets my second.

[n00b 0f l337]

So looking thru....
Evap
Chiller
Cascade
Dual Evap

Cascades are to dangerous to sponsor a competition, we really shouldnt be stretching peoples abilities on them. Evaps we can definitly do though, or chillers. They're really not that hard to load test, all you need is a pump, some tube, and a block. Not to mention just about every builder offers chillers now and again and so should have those tools.
We can definitly do dual evap but I don't see that huge of a challenge, are we going for closest load temps? Its alot of money to drop off though into a unit for a competition..

[[XC] gomeler]

Same thing could be said of a chiller though, much more expensive than a SS.

[n00b 0f l337]

Eh not necesarily I guess, could make your own HX's, either way theres a cost though your right.
So what provides the greatest challenge then?

[teyber]

I think maybe a couple catergories of the competition. I think we should do
1) evap
2) chiller
3) single stage, and EVALUATE THE CASES IN ADDITION using a pre-determined cheap gas(or a few)

Nol, you want chillers cuz your good at building them XD

[[XC] gomeler]

Honestly greatest challenge so far has been autocascades but those never go well. Up next would be cascades :up:

[n00b 0f l337]

Nol, you want chillers cuz your good at building them XD
Actually I was thinking chillers because I havent done one in a long long time.
How about a small compressor cascade ;)

[[XC] gomeler]

I'm game :up: No rotaries for sure, but are you thinking sub 1/2HP compressors, sub 1/3HP? Always wanted to toss 2 NF11FXs into a cascade or are we talking smaller?

[n00b 0f l337]

How about as small as you can go and with any gas selection and such, then grade on temps at a load, and looks, and size, and power consumption?

[tim-]

if we are going for dual evap competition, count me in :D

[n00b 0f l337]

I think people might want to do a micro cascade

[[XC] gomeler]

I'm already planning a micro cascade even if it's just for :banana::banana::banana::banana:s and giggles. If we do end up having a contest for cascades then I'll submit it if it does actually make it past the planning stage.

[Xeon th MG Pony]

Same, a 100W -180c Cascade is what I'm going to be making, 100% hermetic and engineered :)

[[XC] gomeler]

Ooohhh, I want to see that in action ;)

[Xeon th MG Pony]

It's a long off project, been collecting compressors from water coolers for it :)

going to have to practice using cap tubes, I've only ever used TXVs as any thing I built was 3,000 BTU and up! It'll be water cooled too!

[[XC] 2long4u]

Did you figure it out yet? Whats the rules?
I just found a copeland scroll behind the dumpster at my work. It stands about 2 1/2 feet tall. I think it is from a 5 ton or so ac.:shrug:

[n00b 0f l337]

Just design? As in draw a schematic? I think sure that would be nice, but alot of people are just going to copy what we all do. We'll see a bunch of 30 stage cascades with repeating loops and such. I mean what is someone going to draw thats new?

[n00b 0f l337]

Ooo now that does sound good, however at least I know I doubt I could be able to do anything like that, I design empirically out of pocket. Mind teaching us? I know I'm not the only one who works that way. I've always found that "theoretically" is not a word I like, "theoretically" the Hindenberg could have been the most amazing aviation invention ever.

[n00b 0f l337]

//The Hindenburg was designed and engineered to use helium, politics caused it to use hydrogen...
Okay not the best example, though I do believe it was hydrogen designed as all the other blimpies at the time were hydrogen. Post Hindenburg was a major switch to helium. However that didn't stop major evidence showing the Hindenberg's paint also contributed to thermite reactions during its burning.

All I'm saying is, theoretically is great on paper, but it also has a % error like everything else on paper. I'd love to learn how to do it all on paper though. Sounds like a fun brain scrambler.

[n00b 0f l337]

Right, in some sense theoretical design is empirical at least in part.

[HoriWanderFuLL]

How about a quadcore Killer machine single stage comp

best temps after 2 hours with duel Orthos at 1.7vcore wins :-)

its easy get to work start building ..I'll have a look on the weekend see sum new pics :D

Hori sits n waits to get Flamed :p:

[n00b 0f l337]

O thats easy, a quad core killer with best temps after 2 hours? Bah, I'll send you a intel stock heatsink, talk about a quad core killer ;)

[HoriWanderFuLL]

O thats easy, a quad core killer with best temps after 2 hours? Bah, I'll send you a intel stock heatsink, talk about a quad core killer ;)

so you have obviously owned 1 and had it to 1.8vcore and try to benchmark
1024 wprime or 32 Mb super Pi with it >>At 4300 Mhz?.

i doubt it or you wouldnt post such crap..because my compresor last about 2 minutes with these soughts of vcores and Fsb range

Cheers Hori

[n00b 0f l337]

I was joking, instead of a quad unit, I was saying a good way to KILL a quad core.
I've been building units holding quads past 4300 with 4 instances of orthos.

[HoriWanderFuLL]

I was joking, instead of a quad unit, I was saying a good way to KILL a quad core.
I've been building units holding quads past 4300 with 4 instances of orthos.

sorry if i got cranky just thought was good idea..you unit sounds good holding 4 orthos at those speeds ..you got a link to it for me to look ??

Cheers Hori

[phelan1777]

I was joking, instead of a quad unit, I was saying a good way to KILL a quad core.
I've been building units holding quads past 4300 with 4 instances of orthos.

Yeah, I am curious when I can get in line for one :-P.

[n00b 0f l337]

Sdumpers build did that, he's got a great quad, then again that was with a 3/4-7/8hp rotary compressor.

[[XC] gomeler]

Holding a quad at stupid-fast speeds with stupid-high vcore for 24/7 operation isn't a problem at all, just requires larger compressors and larger condensers.. Any idiot can build an enormous rotary powered single-stage that dumps 400w+ of heat and holds a quad with those requirements. Hell even our "lowly" NF11FX can hold a quad at such clockspeeds albeit at much warmer evap temps, it's all in the sizing of components. I'm still partial towards a cascade contest of sorts, I like the combination of space used (perhaps measure the volume) along with performance and aesthetics.

[n00b 0f l337]

But instead of the stupidly huge compressors I think a small compressor competition would be good.

[gosmeyer]

Smaller the better

[HoriWanderFuLL]

But instead of the stupidly huge compressors I think a small compressor competition would be good.


Hori sits n waits for these build pics ?? ll the ones u built that hold these Quads easilly....:rofl:

[Shalamay]

I think the quad core is a good idea, since that seems where the cpu market is heading. Small, smiple, and quite, thats what I want as a buyer.

[HoriWanderFuLL]

I think the quad core is a good idea, since that seems where the cpu market is heading. Small, smiple, and quite, thats what I want as a buyer.

I agree mate ..But didnt you know that that is to easy and they have built heaps of em....coming from a couple of guys who have posted couple thousand post each yet all i get is answers like yer mate to easy with a rotary...well shows me the builds n pics...

single stage phase unit thats gunna hold a quadcore at xtreme overclocks for long periods...
me posting like thismakes me look like a d ickhead and i knowit...but i just wanna say what i think is good..show me this Nfl compresor with quadcore and coretemps..because all i wanna know at the end of the day is whats req to hold 1 of these cpu's so i can blow u all away withsome benchmarks :D

This is a pic of a vapo at its limits http://img148.imageshack.us/img148/2677/q66004000mhzwprime1024rj9.jpg (http://imageshack.us)
Shot at 2007-08-04

http://img174.imageshack.us/img174/6530/4400mhzst2.jpg (http://imageshackus)
around 1.54vcore ...this same unit at 1.8vcore u cant even run a benchmark....

[[XC] gomeler]

Maybe if you opened your eyes a bit you'd see all the quad-core threads and all the builds associated with them.. NF9FX and NF11FX and other ~1/3hp hermetics do a decent job of holding quads in the lower 4GHz range with associated vcore. There are literally dozens of build threads that show quad-core capable machines, use the search function.

edit: and it's no wonder your VapoLS won't hold a chip at 4.4Ghz, that is way outside it's envelope for its tune.

[n00b 0f l337]

I dont think he realizes or load tested on a load tester, so he has no idea the wattage hes working wth, or the change in wattage from going from 1.54 to 1.8

[[XC] gomeler]

Just noticed, in those images the speedfan temps are the same. Was so confused as to how a vapo was holding a 4.4Ghz q6600 with 1.79vcore at -48C

Anyways, back on the subject of this thread. Small compressor cascades? I'd rather not build something so emaciated that it's useless for my own personal purposes, so can we make ~1/3HP the maximum so I can use NF11FXs? :p:

[n00b 0f l337]

NF11FX are 3/8, but I think I'm going to need to finish writing up a possible rules and regulations chart. We all have killawatts and such to measure wattage right? Why not measure compressor input power ;)

[DetroitAC]

How about allowing either cascades or autocascades, and limit the total compressor displacement?

The problem is 50Hz countries and 60Hz countries need to be on equal footing.

It'll take some calculations to get a figure, since some guys are using 50Hz, some 60Hz. I think something like:
For 60Hz: (spec sheet displacement (cc/rev))*(3450 (rev/min))= 77000 cc/min
For 50Hz: (spec sheet displacement (cc/rev))*(2875 (rev/min))= 77000 cc/min

60Hz entries can use about 22.32cc total compressors displacement
50Hz entries can use about 26.78 cc total compressors displacement

That's off the top of my head, we need a little research to make sure those rev/min are representative.

[n00b 0f l337]

Could do that, or just power input. I like your idea though at capping that way.

[tim-]

if we have 2 compressors with same btu rate it doesn't matter if its' a 50 or 60hz motor? or I'm a thinking wrong?

regards
Tim

[DetroitAC]

if we have 2 compressors with same btu rate it doesn't matter if its' a 50 or 60hz motor? or I'm a thinking wrong?

regards
Tim
You're right Tim, but that's if you have performance data at the same conditions (refrigerant, evap temp, cond temp, subcool, superheat, air over shell, etc.). Not all pump makers use the same rating conditions. Copeland uses conditions that I think are a bit more sensible than Danfoss for LBP application. Capacity comparisons will be difficult and at times a bit subjective unless one person does all "capacity correction calculations", displacement is clear, no interpretation needed.

[[XC] gomeler]

Displacement caps would be a fun way to do things, input power would also be interesting. With displacement being the limiting factor though then you don't have to worry about toeing the line in regards to compressors if your particular tune really pushes the compressors to the limit and pushes you past your input power limitation.

[n00b 0f l337]

Thats true, so whats a small displacement we can work with? Want to keep it quite small remember, less then normal cascades.
How about 10-15cc USA? So like 12-18cc EU?

[HoriWanderFuLL]

gomeler;2371756']Maybe if you opened your eyes a bit you'd see all the quad-core threads and all the builds associated with them.. NF9FX and NF11FX and other ~1/3hp hermetics do a decent job of holding quads in the lower 4GHz range with associated vcore. There are literally dozens of build threads that show quad-core capable machines, use the search function.

edit: and it's no wonder your VapoLS won't hold a chip at 4.4Ghz, that is way outside it's envelope for its tune.

POINTLESS GOING ANY FURTHER.

like i said u cant show me any 1/3rd Hp capable of holding a quad at high Vcore you say there are dozens in the thread i say there are none in that thread unless you call 1.55vcore high ? i call it moderate and lucky.

I respect your position Gomeler and dont want to push your buttons but a quad on low 4ghz range wasnt what i asked for if you Opend your eyes you would know this.

Cya Hori

[[XC] gomeler]

Do this, tell me how much a Q6600 at 4.4GHz with 1.8vcore dumps and I'll set my loadtester to it on my next build and show you what happens. I personally won't be throwing 1.8vcore at any chips that aren't under a cascade or better so a load-tester will have to do.

With such small displacement cascades, are we going to go for best load temp with a subdued load (125w?) or some other metric to compete with? Smallest delta between idle and load with a set dummyload, extra points for colder temps?

[n00b 0f l337]

I vote on quad core load, 250 watts minimum, but its a cascade, so lets do 300. K?

[HoriWanderFuLL]

Gomeler i would not know how to test the wattage of a cpu mate...Out of my capabilities.

i presume though at 1.8vcore it would be around 350 watt ??
but this is only a guess..

i know my ls wasnt and isnt capable of holding my Quad at 1.8 vcore but i would like to see a system single stage that could...even at 1.6 vcore my Ls was in shutdown mode after 3 kminutes of benchmarking and the Ls is a 1/3rd Hp isnt it ?? or 1/4 ?
so im thinking alot of these units are capable of only to around 1.55vcore.

Cheers Hori

[Stapler]

Hori I gotta ask, why are benching a quad at 1.8v on an SS. I gotta believe you'd get better results with lower VCore. 1.8v quad is no problem for a good pot with dry ice or LN2, but it's to much for phase.

[n00b 0f l337]

Actually even dry ice, I've never seen people use 1.8 with dry ice, at least not on quads. Quite silly to use more then 1.65 on phase, and 1.7 or so for dry ice. Then the fearless can use up to 2 or 2.1v on ln2. I'd only use 1.8 on cascade possibly. And even then, thats pushing it.

[Stapler]

I don't think it would be to difficult to run 1.8v, it's really a matter of the pot. For dual core with a rev. 2 mousepot 1.8v is no problem. Quad I think would be possible as well, but I don't have first hand experience. The pot makes a huge difference.

Kinc and Elmor are certainly crazy running 2-2.1 volts (not quad)

[n00b 0f l337]

Sadly we have tiny little evaps tho ;)

[Starkiller42]

Hori: units of this size just need to be bigger. A chiller I just started construction on should be able to hold -40*C or colder for 400 watts of heat; pretty much the same thing if you just replace the HX with the evap. You just need to properly account for the larger load in the design ;).

But for the quad, the basic rule of thumb I learned for heat loads was stock_watts * (voltage^2/stock_voltage) * (speed/stock_speed). You can at least guess the wattage range. Say 125 watts stock (IIRC), that's 125*(1.8^2/1.325)*(4.3/2.4)=547W. Ok, so the volts are a little high; check it at 1.65V: 125*(1.65^2/1.325)*(4.3/2.4)=460W. Seems a little high to me, but my assumption of 125 could be off ;).

[n00b 0f l337]

Bit different Star, a chiller can have a much much much more efficienct evaporator. Not to mention you can provide alot more refrigerant into a larger evap for a higher load. And you could use a TXV or the like

[[XC] gomeler]

300w with such a small displacement limitation on a cascade? I imagine a single rotary then would do much better than a cascade with such a "high" load :shrug:
I'll entertain the concept though, I guess it'll make things interesting. So..
1) Displacement limitation, Need to settle on a number/range.
2) 300w load on evap
3) Measure idle and load evap temps
4) Measure input watts
5) Make as compact as possible
Missing anything?

[n00b 0f l337]

Why have a cascade if your not pushing the proc and such, maybe 250 watts then, but I'm thinking 300.
But I think you got everything.

[HoriWanderFuLL]

Hori I gotta ask, why are benching a quad at 1.8v on an SS. I gotta believe you'd get better results with lower VCore. 1.8v quad is no problem for a good pot with dry ice or LN2, but it's to much for phase.


Just for some Pi suicide runs mate..also quick 32Mb wpime runs....
yer i need to get a dice pot...Kayl has just made sum nice ones up .
sought of like my Hardware in a box though


you see im not just benching the quads at high vcore im also benching my phase units and capabilities at these speeds as well...gives me a good idea on what unit can hold vcore and Oc..So i have a pretty good idea what im talking about...as i have had more units than some of these guys have had Wanks..so thats alot eheh

[[XC] gomeler]

Well I'm not disagreeing on the load, I'd actually like to see it pushed up a bit higher to 350w+ as benching a quad under cascade generally entails even higher loads. I'm just worried of the performance we'd be able to achieve with such limited displacement. I imagine it'll be idle:-90C to -80C, 300w: -35C or something outrageous. Trying to find some P/T charts so I can start designing a potential build with a 15cc limitation and having some trouble but it's looking ugly. Guess once you get on AIM I'll have to give you a holler.

[HoriWanderFuLL]

300-320 watt -30c would be nice wouldnt it ?? core tmp of cpu would be nearly close to 0c ??

[n00b 0f l337]

core tmp of cpu would be nearly close to 0c ??
Good luck, I think you need to check your numbers. Kayl has -80C evap temps with -8C cpu temps. Do the math on the delta at our heat density and you'll know its not really about cpu temps, its about keeping a subzero cpu and removing the heat. Thats the main reason ln2 and such is so much greater on the oc's, they actually have -100C cpu temps due to the greater size of there heat exchangers.

[HoriWanderFuLL]

yer thats why i asked because i dont know...
Thanks Hori

Edit so the Ln2 actually gets the core of the cpu that low ??

[n00b 0f l337]

Yes as they actulaly have to fight the coldbug around -127C. But its not a huge difference really in performance, sure a good bit, few hundred mhz, but compared to water to phase or such its not as huge.
Its a bit odd how they scale.

[HoriWanderFuLL]

Amazing :) anyways im finished here ther threads spreading in wrong direction,,

Cheers Hori

[DetroitAC]

How about 10-15cc USA? So like 12-18cc EU?
300Watts sounds good to me, my models show quite reasonable temps for a well built system, but I would still rather build an autocascade, it just seems a more practical unit, perhaps two categories?

[[XC] gomeler]

The idea of a total displacement limitation sounds really good to me unless there is a particular reason why we should separate cascade/autocascade. Maybe b/c the autoc has higher suction pressures give it a slightly higher advantage in any sort of scoring?

[n00b 0f l337]

We could do displacement limitations and another autocascade competition...
Or just, best temps at 250 watts with a displacement cap, get there however you like.

[[XC] gomeler]

OK just had my airconditioning fixed so temperatures in my townhouse are creeping down from the 98F it peaked at so I can finally think straight. Since this will be a competition and not an exposition, there should be some sort of metric by which we are judged. I haven't built enough autocascades and cascades of similar complete displacements to be able to tell if the cascades out perform due to lower suction pressures. Just right off the bat though I'd say have each build rated from maximum number of points split amongst various categories. This is just an idea so don't fret :D

Performance, does it meet the dummyload requirement and a certain minimum temperature. Guess we'd all have to agree on a certain point, 250w @-30C on evap face or something like that. Bonus points for getting colder, don't adjust dummyload though, keep that static.

Size, combined cubic inches/cm with the smallest unit getting full points and staggering it down from there. Volume I'd say should be tallest component not including suction line, widest points in the x and y direction to form a rectangle.

Input power, how much power does the system draw compared to the amount it is shifting. Figure out some sort of formula that'll reward highly efficient units.

Idle/Load Delta? Maybe? Reward units that don't flucuate much? Just worried as this would encourage builders to make something that idles at like -35C @ 250w and loads at -30C, not really pushing the boundaries in regards to performance. :shrug:

Implementation, This is iffy as it'd be very opinionated. Basically you'd be judging on beauty which is hard as you can't just take a ruler and measure elegance. Panel of a range of judges perhaps?

Just throwing out ideas, if they stink let me know.

[Freddie123]

I like those ideas. I do agree with you on the last point though, beauty is in the eye of the beholder after all.

[[XC] 2long4u]

Come on already! 105 posts and no rules yet?!

[PhilippF]

The most fair and interesting competition would indeed be to limit input power and specify heat load. So everybody is absolutely free to do what he wants between the ac-plug and the dummy load.
E.g. Max. electrical power (cos-phi corrected) 1000 Watts and 300 Watts heat load. The one with the lower loaded temps wins :)

A nice rule would also be to limit space, e.g. it has to fit in 15x15x25 inch casing...

[DetroitAC]

I don't like the displacement idea because of the different characteristics of the different types of compressors. E.g. a typical 15cc A/C rotary will annihilate a 15cc recip unless the recip is running a very low compression ratio, which it probably won't be.

The A/C rotary will be drawing a sh*t load more power and running way out of spec but in performance it will win because it's volumetric efficiency will stay high, around ~70% over the possible compression ratio (1:5 to 1:15) range whereas the recip might vary between 70 - 30% over the same range.

I think that if you want to limit compressor size then limit the input power.

Do you agree Erik?

Tom
Good point Tom, I honestly have never dealt with a rotary, from what you described it behaves a lot like a scroll. I'll take your word for it, and I suppose the rules need to deal with that problem in some way, otherwise it's a rotary winner and I'd prefer to see nicer, quieter units (easier to sell when the competition is over), and not just a competition to see who can find a rotary that is just under the limit. Maybe a separate category, maybe a penalty factor, maybe a different displacement limit?

@PhilippF, I assume you're talking about the apparent power vs. actual power problem? We call it power factor here, (the problem is there is a phase lag between current and voltage) I'm not an electrical guy, but do you know a good way to correct, or are compressor power factors generally close to each other? Basically, please explain (cos-phi corrected)

-Erik

[n00b 0f l337]

Very good point Tom, thast why the power requirements as well. But yes we might run this with no rotaries allowed. So how about judged on size, power consumption, parts selection/difficulty, and of course temperature. With bonus temps for looks.

[rogard]

Heres another idea off the top of my head - how about a Quick-connect system for phasers?

Like the quick connector air cons do?

So you could swap out evaps and condensors etc without having to rebraze?

[tim-]

rogard:
you still need to vacuumate the line so I don't see any reason for quick connection if it's not ment as an experiment unit.

[PhilippF]

@DetroitAC:

If you have a load which is only a resistor, voltage and current are always proportional (the two sine waves are equal). But if you have an inductive or capacitive load, the current and voltage sine waves get out of phase and are different by a certain angle. A cos phi = 1 means you have only resistive load, a cos phi smaller than 1 means your load consumes power which is later in the cycle again given away into the ac line again.

http://en.wikipedia.org/wiki/Power_factor

There are a lot of cheap power meters around which can determine cos phi and show the real load (taken power - returned power).

e.g.: http://www.smarthome.com/9034.html

If there are any clarities left, please drop me a not :)

[[XC] gomeler]

I'd rather see this limited to your standard hermetic compressors as has already been stated. I think we've dropped enough ideas that a set of premature rules can be formed and we can work them into a final set to which we all agree are fair. Dibs not on that though, crunch time in preparation for school in the fall so I'm rather busy.

[DetroitAC]

PhilippF, thank you it's clear. I didn't know the kill-a-watt meters calculated actual power (thought they were just VA).

I was picturing everyone measuring VA, and the results messed up.

P.S.
I like your sig, but the tape is duct tape, as in tape used to close seams in air ducts. It's confusing since there was a brand here in the US called "Duck Tape" with a picture of a duck on the packaging. It's one of those jokes that just gets lost with time.

[n00b 0f l337]

How might I calculate the difference or measure any returned current? That is indeed a major tuning concern!

[Xeon th MG Pony]

It's called power factor, thats why most inductive motors do have a run cap and start cap, one to start it and one to realign the current to voltage SIN waves. Coser you are to a 1 the better, computer PSUs where realy bad for that, they would look like they ate allot more then they do becuase they had poor power factor correction.

[PhilippF]

And the big power supply companies hated the increase in bad PC power supplies, because they have to run large capacitor farms to correct the power factor again :)

In Germany small hoseholds have only to pay for the real power they consume, big industrial plants have also to pay for their blind power (makes them to correct their power factor).

Btw: http://groups.google.com.au/group/rec.music.gdead/msg/cb7b5453068f86cc?hl=en&lr=&ie=UTF-8&oe=UTF-8

I think both "duct tape" and "duct tape" is correct^^

[Xeon th MG Pony]

Ya, The PC PSU companies got smart and started adding factor correction into the suplies and marketed them as a pluse, now days I think it is compulsory, if not, any one should make sure to get one with active PFC.

By adding a small run cap to any inductive load you can seriously save money and clean up the noise on the AC line in one shot! Erf AC is all ways a head ache so much weird stuff about it compared to DC!!

What most doen't realize is there is no power shortage or issues with the companies, the issue is the wanting waste and carelesness of the users! If every one took a little effort to reduce power wastage and comercial companies utilized PFC and such there'd be zero energy issues!

[[XC] 2long4u]

How about the smallest best performing chiller. It would be easy take the demention/proformance ratio.