The UltraMass 3" , Quad core monsterblock

[wittekakker]

He made it himself, and for a homemade block it doesn't look too bad...

Where did you read that

[Nate P.]

Where did you read that

Well then, who makes it?

[wittekakker]

Voila ben vandaag naar de machinist gegaan geluisterd wat kan en niet kan en heb nu een design uitgewerkt.

Heb gewoon iest zelfs verzonnen kwa desig om toch wat originele te zijn.



Oh witte das heel vriendelijk van jou

Translated to english, someone CNC'ed it for him, design was Troman's idea though.

[BeardyMan]

Translated to english, someone CNC'ed it for him, design was Troman's idea though.

Wrong.

edit*

lol pushed a bit to fast that reply button :p

It's not with cnc, none of the existing wc blocks are made with cnc. Impossibel to do.
This one is made with miling machine, the apogee and such are ponsed from a plate that's why there bases are that thin.

[Waterlogged]

It's not with cnc, none of the existing wc blocks are made with cnc. Impossibel to do.
This one is made with miling machine, the apogee and such are ponsed from a plate that's why there bases are that thin.

If that's what you think then your the one that's wrong. I'm not certain on the form of copper (bar stock/cast/forged) used for the newer GT/GTX blocks so I can't speak about them but the pin matrix can be machined on a CNC mill without a problem, and I know for a fact that it was machined into the Apogee Extreme. The Fuzion could also be machined from bar stock but it would be more expensive as well because the small end mills it would take to mill it would probably break somewhat often and they certainly aren't cheap. It seems most companies are getting away from bar stock in favor of forging, and with good reason. A forged copper block can be made much thinner and have much better transfer rate that bar stock due to the molecules being forced together under enormous pressure.

[YugenM]

How easy is it to machine a U-bend channel with a manual mill?

[Waterlogged]

How easy is it to machine a U-bend channel with a manual mill?

The way the ones on the block in this topic are, fairly easy. If you want a true U shape, that turns into a craps shoot very quickly. Human hands simply can't mimic a decent servo motor.

[BeardyMan]

If that's what you think then your the one that's wrong. I'm not certain on the form of copper (bar stock/cast/forged) used for the newer GT/GTX blocks so I can't speak about them but the pin matrix can be machined on a CNC mill without a problem, and I know for a fact that it was machined into the Apogee Extreme. The Fuzion could also be machined from bar stock but it would be more expensive as well because the small end mills it would take to mill it would probably break somewhat often and they certainly aren't cheap. It seems most companies are getting away from bar stock in favor of forging, and with good reason. A forged copper block can be made much thinner and have much better transfer rate that bar stock due to the molecules being forced together under enormous pressure.

I'm not the one that is wrong otherwise this block would have that design. I went to 4 machinist from big companys and all of them agree how it got made.

[K404]

Will you not have mounting issues for components around the mobo? Lookin forward to seeing installed pics and of course...temperatures

[Xilikon]

Anything can be made with a CNC machine, just think of the right steps. If it is impossible with CNC because you don't have the means to hold the block, just make it very thick, mill the channels and when it's done, mill the baseplate. However, it's not always cost-wise to make everything with a CNC/Mill and that's why forging or casting is more cost effective for big productions.

Indeed, true U turns can only be made with a CNC or a mill with servomotors. You need to be able to control 2 axis at the same time with the proper speed, hard to do manually without being jerky.

[tiro_uspsss]

so whats the deal/verdict? these are the 'U-turns' in question?

Lemme get this straight - some say only CNC can do, others... ?

[Xilikon]

This is manual. The round edge is because the mill bit is round A true U-turn would be a half-circle, not a square 90 -> 90 degrees turn.

[tiro_uspsss]

admittedly i have no idea about whats possible etc etc - but i am actually buying a mill-lathe (2 more payments to go!) when i get it i will be attempting a U-turn - a 'proper' one - i'm c o c k y'n'confident with my manual dexderity

[Xilikon]

It's doable to do a true U-Turn manually but it take lots of time. Do it in small steps (0.005 inch) alternating each axis till you are at the middle, then change one axis till you complete a half circle. Think of doing it like building a stair.

[tiro_uspsss]

It's doable to do a true U-Turn manually but it take lots of time. Do it in small steps (0.005 inch) alternating each axis till you are at the middle, then change one axis till you complete a half circle. Think of doing it like building a stair.

meh.. no disrespect or anything; u'll think i'm a smart-arse (probably am ) - i'll just grab a piece of scrap metal & try it on the fly - i'll give it a few goes, if it doesnt wanna play nice, *then* i'll try it like u suggested

[Xilikon]

LOL, doing pratice is also doable as well

My own milling/lathing expertise is from my former airplane building classes (milling, lathing and CNC programming). It's been 15 years ago already so I'm sure the technology would have improved by then but nothing beat the old mill in the fun factor

[YugenM]

It's doable to do a true U-Turn manually but it take lots of time. Do it in small steps (0.005 inch) alternating each axis till you are at the middle, then change one axis till you complete a half circle. Think of doing it like building a stair.

aww crap

even with the sharp 90-degree angle in the inner turn of the U (as is the block in this case), will it still be appropriate for an O-ring, or do I have to file it down or whatever?

[BeardyMan]

Will you not have mounting issues for components around the mobo? Lookin forward to seeing installed pics and of course...temperatures

AM2 has allot of space, s775 doesnt

I'm leak testing it now , i had to use the force in order to make it leakfree

the force,

[Waterlogged]

I'm not the one that is wrong otherwise this block would have that design. I went to 4 machinist from big companys and all of them agree how it got made.

Then they have no business being machinists. I should know, I was a machinist for 12 years. I was one of two ppl responsible for programming/setup/operating the three 5 axis swiss machines the the company I worked for had. I was also the sole program/setup man for the milling dept for the last 3 years I worked for that same company.

[Xilikon]

I agree with Waterlogged on this and that is based with my small experience machining airplane components. Anything can be done if given enough time and budget for it. For mass production, forging or casting is better since you are basically repeating the same pattern over and over.

[Waterlogged]

Yep, cost is usually the deciding factor for what parts get made in a machine.

We actually lost a contract once, because it was cheaper for the company we made the parts for to have a mold made for $10,000 for injection molding, than to have us make the three pieces that made up that part and assemble it. I'm positive they had more than one mold made too.

[BeardyMan]

Then they have no business being machinists. I should know, I was a machinist for 12 years. I was one of two ppl responsible for programming/setup/operating the three 5 axis swiss machines the the company I worked for had. I was also the sole program/setup man for the milling dept for the last 3 years I worked for that same company.

Then it would be cool if you pm'ed hwo to do it, so i can put it under their nose

[ColonelCain]

Holy block batman!

It could go a little bigger, I guess...

But, nice job on a homemade block! :cheers:

[Waterlogged]

Then it would be cool if you pm'ed hwo to do it, so i can put it under their nose

Why bother with PM's? I just do it here.

Honestly, that's about as simple as you can get for a CNC mill.

It's been 4 years since I've done any CNC programming and I've forgotten pretty much everything in regards to G and M codes, on top of that, almost every machine manufacturer has their own version of control code. There are also too many mill manufacturers out there, many I've never heard of. Without knowing what make mill these incompetent fools you've asked have access to, it would be impossible for me to do it right.

A basic machine path though would go something like this.


[Example]

Block size: 3" x 3"
End Mill size: 3/8" Dia.
Internal wall thickness: .025"



Position 1 is dead center over corner of block . We shall call this X0, Y0.
Pos 2 X .500, Y .500
Pos 3 X 2.500, Y .500
Pos 4 X 2.500, Y 2.100
Pos 5 X 2.100, Y 2.100
Pos 6 X 2.100, Y .900
Pos 7 X 1.700, Y .900
Pos 8 X 1.700, Y 2.100
Pos 9 X 1.300, Y 2.100
Pos 10 X 1.300, Y .900
Pos 11 X .900, Y .900
Pos 12 X .900, Y 2.100
Pos 13 X .900, Y 1.5
Pos 14 X .500, Y 1.5
Pos 15 X .500, Y .900
Pos 16 X .500, Y 2.5
Pos 17 X 2.500, Y 2.5

[/Example]

That's a oversimplification, as there normally would be a bunch of G and M code mixed in with F's (feed) and S (spindle speed) as well as the co-ordinates I've posted. Then you would switch tools and tell the machine where to put the small round recesses inside the channel in the same manner. Then mill recess for o-ring or gasket, then drill and tap threads and it's done on one side.

[BeardyMan]

i was talking about the apogee block not mine ^^

here are some pics mounted
no problems with mounting due size = as am2 bracket