any of you guys U1 friendly, yet ???

#Rotor · 09-28-2002, 12:41 AM

I am... allmost....

the block looks like this, though all copper, the glassy top is just for showing the inerts....

it's a tough one, fitting massively extreme SMP cooling into a mere 43mm of overall vertical space. As if that ain't hard enough, no mounting holes.... I have a whapping 3mm between the top of the block body, and the top of the U1 case...

any one that can come up with a better idea for retention... I'm all ears.

Trick is, the blocks need to be tab-friendly, as well as hole friendly,

Puzzdre · 09-30-2002, 08:32 AM

Errrr, dunno about retention, but for the blocks I'm gonna say

WOW, WOW, WOW !!!

oh, yes, another WOW

are you going to be mad if we kids try to do something like that at home?

WOW!

#Rotor · 09-30-2002, 09:18 AM

as long as you are willing to share in the Millions you are going to make when Dell places an massive order for your creation, to use in there new OCer's friendly box...

I'm kidding.. about the Dell part... hehehe

sure have at it... it's very easy and you need no special tools for it....

one thing though.... you will be required to put up lotsa pictures of it...

Puzzdre · 09-30-2002, 10:21 AM

Yes Sensei !

jag_e_fattig · 09-30-2002, 11:00 AM

wow is right! my hat off to you!

#Rotor · 09-30-2002, 07:57 PM

a little update......

Some of you might know Machupo from over as OCN.
It is at his request, that I'm having all this fun here.....

He is building what is going to become the rig all others are going to be measured by....
things like SMP and Clustered and LINUX are key words that comes to mind...

truly an awesome expedition, and I'm honored to be a part of it.... once done, it will be the measure of what is going to follow, from there on....

dream caster · 10-06-2002, 11:49 AM

I do not understand your block and i don't understand what is your problem:

i don't see any path for water inside your block, just two flat blocks with dimples, touching, and no way for water between those half tubes.

you can mount your blocks using:
--an elastic piece of steel that goes over over the block and that could even fit in a shallow slot to make it even with block surface and to position block centered with procesor core and use tabs as usual (one, two or three as you see fit)
--holes, bolts and spring coils using only an extension of the lower half of block to save vertical space or using holes that keep all this inside the hole.
You can even braze a thin piece of copper below the lower part of your your block; the key would be using a strong enough brazing and have the surface of this added copper as low as posible.
--just holes and bolts, somewhat risky, but you need less space and can can use the same procedures from above

Puzzdre · 10-06-2002, 03:38 PM

Quote:

i don't see any path for water inside your block, just two flat blocks with dimples, touching, and no way for water between those half tubes.

I think those blocks are waiting for some dremel action...

#Rotor · 10-06-2002, 04:01 PM

indeed they are...

they might apear to be Dimples... but they are rather deep...
about 1/2" at their deepest point'

MadDogMe · 10-07-2002, 05:08 AM

Would you be able to make those hole 'already open' if you used a mill?. make them overlap if you know what I mean?. I know a drill would 'snatch' if tried, but what about a mill?.

PS, I love your work Rotor, but I have to say your 'quick links' are not quick to load on my 56k'er

.

PpS. do you put 'locator pins' in there to allign the threads?(barbs), when the two halves are assembled?. I could forsee some nasty thread stripping from someone who tries dis/re~assembling them thereselves

, iff they don't line them up perfect...

MadDogMe · 10-07-2002, 05:29 AM

I can think of a way of retaining it, I don't know if it's feasable though.

have two M4 holes drilled each end of the block over the three lugs at each end, have them 'like' countersunk to accomadate springs to push the M4 bolts upwards. have two peices of metal (call them retention straps) that fits under the three lugs at each end with two M4 threaded holes in each one for the bolts(either side of the middle lug).
position the block over the top and thread the bolts into the 'retention straps', the springs should act as a buffer to avoid overtightening. it would be fidley but doable.
you could probly make the retention straps into a one~peice square plate assembly that fits around the socket. to aid fitting/hole alignment and make it more secure(stop it slipping over the lugs!), make one end so it slides in under the lugs for fitting and slides back out for removal. use springs or tiny finger nuts for tightening/fixin in closed position.

:[X]: for fitting
:[ X]: for removal

X= socket, : = holes for mounting, [ = retention straps.

Doh!, thats what the top pic is ain't it?. I thought that 'bottom bit' was holes you were gonna put in the mobo!!, not a peice under the lugs [hides head in shame] oh well great minds (and fools) think alike

.

LiquidRulez · 10-07-2002, 10:17 PM

wheres the o ring groove?
or are you soldering the 2 halves together??

Oh and the pics in the first post arent showing up....id like to see them too.
you got a link??

#Rotor · 10-08-2002, 07:34 PM

I do not solder or use o-ring... there can only be GOOP

well none of my customers are morons

so I'm not worried about them screwing up something like that...

I am however taking care of that possibility that the two halves might miss- align, by being very precise in my drilling... the holes align perfectly, and since the bolts are counter-sunk, they re-center the top block as you tighten them... Works really good actually..

a milling machine would indeed be very preferable to actually be ably to have each hole overlap, with the next, and in doing so, nullify the need for all the Dremelling... why you wana donate me a mill? Maybe?? Maybe?

I dono.... the links on the first post seem to be in perfect order...
can not see why they would not work....

MDM your idea sound prety close to what I had going in that rendered picture of the block...

MadDogMe · 10-09-2002, 04:57 AM

Yep I could have saved me two index fingie tips, well my right one mostly

.

Do you think there'd be room to coutersink some springs in there?, if they know what they're doing they could just hardmount them, but if you ever go public they could do with a safety margin. you'd have to find the 'oh so just right ones' though

...

Quote:

I am however taking care of that possibility that the two halves might miss- align, by being very precise in my drilling... the holes align perfectly, and since the bolts are counter-sunk, they re-center the top block as you tighten them... Works really good actually..

I'd still have to put 2 stainless pins in if it were me for peace of mind. what size are the bolts that hold it together?, is there any 'flex' with them?.

************************************************** *************

This sounds like it's gona be some system, is there gona be a article of some kind put up about it?. what kinda pump is gona ppower that little lot?

.
I adore rackmounts!, the more industrial the better, mattblack paint and grab handles, oh!, and loads of that 'non~slip' ally plating

/// \\\ /// \\\ kinda stuff

.

PS, links show up fine for me, have you tried refreshing?, or skipping back a page then forward?...

#Rotor · 10-11-2002, 09:22 AM

the bolts are (SS 8-32nc CS) which means they go in tight through a 11/64" hole . which is what I'm using in the top plate.
obviously the bottom plate has the thread cut into it...

the tolerance for radial movement of the two plate is in order of less than 1sec and at the radius on which the threads are, from the furthers pivot (bolt) in the block, that constitutes to less than 1 "thow" (0.001") I think that's good enough...

what I'm saying in Shade tree terms... them two plates don't move when the bolts are in and threaded

for re assembly, it is recommended too, that before tightening the block down finally, to screw in the two fittings, then to tighten the block with them in place.... It will insure a perfect seat of all components involved... access sealant will fill up any microscopic imperfections as it gets tightened.

I am after about 3 years of doing it this way, still 100% leak free..... and I am proud of that....

but I should watch my mouth, and not say that so loud ... there is always a first time....

MadDogMe · 10-12-2002, 04:56 AM

I did'nt think about putting the barbs in loosly before final tightening

. That was my only worry

...

FRAGN'STIEN · 10-16-2002, 10:57 PM

Well #Rotor I have to say you have inspired me the build my blocks using the crosshatch pattern. I ordered some copper from online metals monday and it arrived here today. WOW that was fast, but seeing that I only live about 6 hours away from Seattle... I'm going to use a 3 barb setup, one 1/2" center inlet and 2 1/4" outlets. The stock I bought is 2"x3.1875x1/4". I bought 6 pieces so I have 3 shots at it. I hope I get it right the first time. Also what type of GOOP do you use? I was thinking of soldering mine, but I have a set of taps so I may just go that route.

MadDogMe · 10-17-2002, 03:57 AM

It's got to be best to be able to dismantle and clean a block has'nt it?. especialy one with lots of surface area...

#Rotor · 10-18-2002, 10:40 PM

I use Marine Goop....

soldering does work for when you have the fittings in from the top.... but from the side, like that... the taper of the fitting, tends to sheer the solder, as it is very soft and pliable...

now I suggest you take your time, with marking out the grid... the more accurate you are with that first step, the easier the rest will be, choose the drill-bit you want to use for the grid, then design the spacing between the holes so they just just do not cut into each other, believe me, you will have a hard time to drill straight with ordinary bit's , if half of it is going through air, and the other half through copper....

oh and for tapping the assembly holes, don't go with too thin, Copper is just as hard on taps, as on anything else... I use 8nc32 CS-SS bolts for it... works gr8.

FRAGN'STIEN · 10-19-2002, 12:52 PM

Great. Thanks for the reply #Rotor. I'm going to use a 7/32 bit with a 143° tip. It's only black oxide, but should be OK as long as I keep it lubed durring cutting. Oh and I'll be starting it here in just a couple minutes so I'll keep you guys posted

09-28-2002, 12:41 AM	#1
#Rotor Cooling Savant Join Date: Feb 2002 Location: Dione, sector 4s1256 Posts: 852	any of you guys U1 friendly, yet ??? I am... allmost.... the block looks like this, though all copper, the glassy top is just for showing the inerts.... it's a tough one, fitting massively extreme SMP cooling into a mere 43mm of overall vertical space. As if that ain't hard enough, no mounting holes.... I have a whapping 3mm between the top of the block body, and the top of the U1 case... any one that can come up with a better idea for retention... I'm all ears. Trick is, the blocks need to be tab-friendly, as well as hole friendly, __________________ There is no Spoon....

09-30-2002, 08:32 AM	#2
Puzzdre Cooling Savant Join Date: Aug 2002 Location: Croatia Posts: 969	Errrr, dunno about retention, but for the blocks I'm gonna say WOW, WOW, WOW !!! oh, yes, another WOW are you going to be mad if we kids try to do something like that at home? WOW! __________________ 'Out of cheese error... ...please reboot the universe (press the GBL to continue)'

09-30-2002, 09:18 AM	#3
#Rotor Cooling Savant Join Date: Feb 2002 Location: Dione, sector 4s1256 Posts: 852	as long as you are willing to share in the Millions you are going to make when Dell places an massive order for your creation, to use in there new OCer's friendly box... I'm kidding.. about the Dell part... hehehe sure have at it... it's very easy and you need no special tools for it.... one thing though.... you will be required to put up lotsa pictures of it... __________________ There is no Spoon....

09-30-2002, 10:21 AM	#4
Puzzdre Cooling Savant Join Date: Aug 2002 Location: Croatia Posts: 969	Yes Sensei ! __________________ 'Out of cheese error... ...please reboot the universe (press the GBL to continue)'

09-30-2002, 11:00 AM	#5
jag_e_fattig Cooling Savant Join Date: Mar 2002 Location: SoCal, formerly Sweden. Posts: 110	wow is right! my hat off to you! __________________ Tearing it up AER - 05/29/87-10/26/03, I'll never forget.

09-30-2002, 07:57 PM	#6
#Rotor Cooling Savant Join Date: Feb 2002 Location: Dione, sector 4s1256 Posts: 852	a little update...... Some of you might know Machupo from over as OCN. It is at his request, that I'm having all this fun here..... He is building what is going to become the rig all others are going to be measured by.... things like SMP and Clustered and LINUX are key words that comes to mind... truly an awesome expedition, and I'm honored to be a part of it.... once done, it will be the measure of what is going to follow, from there on.... __________________ There is no Spoon....

10-06-2002, 11:49 AM	#7
dream caster Cooling Neophyte Join Date: Aug 2002 Location: Atacama desert, Chile Posts: 43	I do not understand your block and i don't understand what is your problem: i don't see any path for water inside your block, just two flat blocks with dimples, touching, and no way for water between those half tubes. you can mount your blocks using: --an elastic piece of steel that goes over over the block and that could even fit in a shallow slot to make it even with block surface and to position block centered with procesor core and use tabs as usual (one, two or three as you see fit) --holes, bolts and spring coils using only an extension of the lower half of block to save vertical space or using holes that keep all this inside the hole. You can even braze a thin piece of copper below the lower part of your your block; the key would be using a strong enough brazing and have the surface of this added copper as low as posible. --just holes and bolts, somewhat risky, but you need less space and can can use the same procedures from above

10-06-2002, 04:01 PM	#9
#Rotor Cooling Savant Join Date: Feb 2002 Location: Dione, sector 4s1256 Posts: 852	indeed they are... they might apear to be Dimples... but they are rather deep... about 1/2" at their deepest point' __________________ There is no Spoon....

10-07-2002, 05:08 AM	#10
MadDogMe Thermophile Join Date: Aug 2002 Location: Just shut up ;) ... Posts: 1,068	Would you be able to make those hole 'already open' if you used a mill?. make them overlap if you know what I mean?. I know a drill would 'snatch' if tried, but what about a mill?. PS, I love your work Rotor, but I have to say your 'quick links' are not quick to load on my 56k'er . PpS. do you put 'locator pins' in there to allign the threads?(barbs), when the two halves are assembled?. I could forsee some nasty thread stripping from someone who tries dis/re~assembling them thereselves , iff they don't line them up perfect...

10-07-2002, 05:29 AM	#11
MadDogMe Thermophile Join Date: Aug 2002 Location: Just shut up ;) ... Posts: 1,068	I can think of a way of retaining it, I don't know if it's feasable though. have two M4 holes drilled each end of the block over the three lugs at each end, have them 'like' countersunk to accomadate springs to push the M4 bolts upwards. have two peices of metal (call them retention straps) that fits under the three lugs at each end with two M4 threaded holes in each one for the bolts(either side of the middle lug). position the block over the top and thread the bolts into the 'retention straps', the springs should act as a buffer to avoid overtightening. it would be fidley but doable. you could probly make the retention straps into a one~peice square plate assembly that fits around the socket. to aid fitting/hole alignment and make it more secure(stop it slipping over the lugs!), make one end so it slides in under the lugs for fitting and slides back out for removal. use springs or tiny finger nuts for tightening/fixin in closed position. :[X]: for fitting :[ X]: for removal X= socket, : = holes for mounting, [ = retention straps. Doh!, thats what the top pic is ain't it?. I thought that 'bottom bit' was holes you were gonna put in the mobo!!, not a peice under the lugs [hides head in shame] oh well great minds (and fools) think alike . Last edited by MadDogMe; 10-07-2002 at 05:42 AM.

10-07-2002, 10:17 PM	#12
LiquidRulez Cooling Savant Join Date: Sep 2002 Location: In Hell Posts: 322	wheres the o ring groove? or are you soldering the 2 halves together?? Oh and the pics in the first post arent showing up....id like to see them too. you got a link??

10-08-2002, 07:34 PM	#13
#Rotor Cooling Savant Join Date: Feb 2002 Location: Dione, sector 4s1256 Posts: 852	I do not solder or use o-ring... there can only be GOOP well none of my customers are morons so I'm not worried about them screwing up something like that... I am however taking care of that possibility that the two halves might miss- align, by being very precise in my drilling... the holes align perfectly, and since the bolts are counter-sunk, they re-center the top block as you tighten them... Works really good actually.. a milling machine would indeed be very preferable to actually be ably to have each hole overlap, with the next, and in doing so, nullify the need for all the Dremelling... why you wana donate me a mill? Maybe?? Maybe? I dono.... the links on the first post seem to be in perfect order... can not see why they would not work.... MDM your idea sound prety close to what I had going in that rendered picture of the block... __________________ There is no Spoon....

10-11-2002, 09:22 AM	#15
#Rotor Cooling Savant Join Date: Feb 2002 Location: Dione, sector 4s1256 Posts: 852	the bolts are (SS 8-32nc CS) which means they go in tight through a 11/64" hole . which is what I'm using in the top plate. obviously the bottom plate has the thread cut into it... the tolerance for radial movement of the two plate is in order of less than 1sec and at the radius on which the threads are, from the furthers pivot (bolt) in the block, that constitutes to less than 1 "thow" (0.001") I think that's good enough... what I'm saying in Shade tree terms... them two plates don't move when the bolts are in and threaded for re assembly, it is recommended too, that before tightening the block down finally, to screw in the two fittings, then to tighten the block with them in place.... It will insure a perfect seat of all components involved... access sealant will fill up any microscopic imperfections as it gets tightened. I am after about 3 years of doing it this way, still 100% leak free..... and I am proud of that.... but I should watch my mouth, and not say that so loud ... there is always a first time.... __________________ There is no Spoon....

10-12-2002, 04:56 AM	#16
MadDogMe Thermophile Join Date: Aug 2002 Location: Just shut up ;) ... Posts: 1,068	I did'nt think about putting the barbs in loosly before final tightening . That was my only worry ...

10-16-2002, 10:57 PM	#17
FRAGN'STIEN Cooling Savant Join Date: May 2002 Location: Rathdrum, ID Posts: 380	Well #Rotor I have to say you have inspired me the build my blocks using the crosshatch pattern. I ordered some copper from online metals monday and it arrived here today. WOW that was fast, but seeing that I only live about 6 hours away from Seattle... I'm going to use a 3 barb setup, one 1/2" center inlet and 2 1/4" outlets. The stock I bought is 2"x3.1875x1/4". I bought 6 pieces so I have 3 shots at it. I hope I get it right the first time. Also what type of GOOP do you use? I was thinking of soldering mine, but I have a set of taps so I may just go that route. __________________ This space for rent

10-17-2002, 03:57 AM	#18
MadDogMe Thermophile Join Date: Aug 2002 Location: Just shut up ;) ... Posts: 1,068	It's got to be best to be able to dismantle and clean a block has'nt it?. especialy one with lots of surface area...

10-18-2002, 10:40 PM	#19
#Rotor Cooling Savant Join Date: Feb 2002 Location: Dione, sector 4s1256 Posts: 852	I use Marine Goop.... soldering does work for when you have the fittings in from the top.... but from the side, like that... the taper of the fitting, tends to sheer the solder, as it is very soft and pliable... now I suggest you take your time, with marking out the grid... the more accurate you are with that first step, the easier the rest will be, choose the drill-bit you want to use for the grid, then design the spacing between the holes so they just just do not cut into each other, believe me, you will have a hard time to drill straight with ordinary bit's , if half of it is going through air, and the other half through copper.... oh and for tapping the assembly holes, don't go with too thin, Copper is just as hard on taps, as on anything else... I use 8nc32 CS-SS bolts for it... works gr8. __________________ There is no Spoon....

10-19-2002, 12:52 PM	#20
FRAGN'STIEN Cooling Savant Join Date: May 2002 Location: Rathdrum, ID Posts: 380	Great. Thanks for the reply #Rotor. I'm going to use a 7/32 bit with a 143° tip. It's only black oxide, but should be OK as long as I keep it lubed durring cutting. Oh and I'll be starting it here in just a couple minutes so I'll keep you guys posted __________________ This space for rent

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)