Two pumps config - serial or paralell? - Page 8

Puzzdre · 11-10-2002, 11:33 AM

That is the result for square pattern, channels made with 1.5 mm mill bit, no xy table, just hands...

Now, for circular pattern (I don't have the pic yet, I managed to do 1/3 of one half of the block today) it is much worse cos' you cannot go with straight moves and do lot's of holes, you can do only three holes at max in one straight cut. Other problem with doing this with mill bit of small dia is that you must do 3 or better 4 passes for every connection between two holes to remove all the remaining copper, mill bits I have have cutting flutes approx 1.5 mm high, so you first remove 1.5 mm height of copper connection, than lower the drill/mill and do another pass removing another 1.5 mm and in third pass you can remove all copper. That's just between TWO holes. Now imagine how many of them are here...

Yeah, you can do some holes in line like in square pattern, but connecting circular ones would be killer, even with dremel...

g.l.amour · 11-10-2002, 12:26 PM

why are u not happy with the result, as far as i can see the top piece looks great. man, u r almost there!!!

been looking around some
found a bench drill for 50€
1/4hp
220W
vice(100mm) for 16€

some design i came up with, but prolly way to complex to make...

what dia drill would be best, judging your xperience?

Puzzdre · 11-10-2002, 01:20 PM

I'm the kinda bloody perfectionist, well, almost...I always expect the result to look professional, which in case of making something for the first time, is hard to expect. Actually, tomorrow I'll be perfectly satisfied with the result, and keep on going happy. It's just me I guess, I must have one day of dissatisfaction for the happines of the other days. Today is The Day.

I'm pretty sure you'll be quite satisfied with the drill/vice with the price and the performance, at least for some time until something more complicated comes into the way. And the price is nice. I vote go for it!

Wow, design! Call it The Brain!

Maybe not so complicated as it looks, but something on the connections between the holes - #Rotor's design is all about water taking the least resistive way to go, all the channels equally open for flow in any possible direction. This one you made looks like it would be more restrictive IMO, and with not equally distributed flow. Maybe to widen the connections between the holes as they are further from the center/inlet, to equalise it, or to open all directions between the holes. Also, I would make those inlet/outlet holes slightly bigger, I'm thinking for mine to be 12 mm inlet, 10 mm outlets.

Ofcourse, let's make them and see how they perform!

Good thinking!

Btw, thx for bringing me back on the right (satisfied) track!

g.l.amour · 11-10-2002, 01:43 PM

well i mentioned the ID of the feed barb as 8
that gives u about 50mm² of water going through.

each side going to the outlet barbs has +/- 40mm² of openings for water to evacuate. so the outlet barbs won't be sucking at full capacity. but it does mean that the water hitting the chip area will flow +/- evenly in all directions.

rotor stated some times that for good performance it isn't always necessary to have low flow restriction. in this design i was hoping to tackle the lazy water phenomenon, and make the water go at all directions, instead of just going directly for the outlets.
someone else i heard state that it doesn't make sense to create a flow path with larger surface area than the surface area of the barb.

anyway, to come back to you being dissatisfied. is there anything you could have done better? holes u misdrilled, dremeled some wrong areas. i don't know, but i guess not. anyway, i wouldn't feel to bad about the dremeling being to difficult on your circular design.
when taking the easy/fast route;: if everything is finished and you mount the block, and you are sitting there enjoying your rebuilt pc some. u will start to thinkthinkthink whatif i spent some more sweat on working out a more difficult design, would i have gotten better performance? then before u know it, you will start working out that difficult design to c if it does perform better, and then u will have spent, double the amount of time for the same result. anyway, that how it mostly works for my twisted self.

im very curious to see how the circular part worked out. as i guess, your top isn't a mirrored version of the bottom part. so the heattransfer effect stated here, that lets the bottom pins transfer the heat to the top ones, as being almost made out of 1 pin; won't happen in your design. lot of turbulance i guess would come out of that situation.

anyway, forgot to ask, what dia drill would you use ideally (performance/ease of use)

cheerz puzzdre, and congrats again for having done what u did (+ draw me over the line to go the same route)

g.l.amour · 11-10-2002, 01:51 PM

btw , chk this here out, haven't deeply looked into it, but seems mighty promising

http://www.tapmatic.com/tech_manual/index.html

Puzzdre · 11-10-2002, 02:20 PM

LOL!

Quote:

when taking the easy/fast route;: if everything is finished and you mount the block, and you are sitting there enjoying your rebuilt pc some. u will start to thinkthinkthink whatif i spent some more sweat on working out a more difficult design, would i have gotten better performance? then before u know it, you will start working out that difficult design to c if it does perform better, and then u will have spent, double the amount of time for the same result. anyway, that how it mostly works for my twisted self.

Exact description of me too. Congrat's!!!

For the circular pattern, I thought to make it mirror like (actually the two parts are the same), but I can also try to turn the top 180 deg and see how it performs not mirror like. I'll try it!

For the drill dia, for square pattern I used 4 mm HSS-super bit, and for circular 3.5 mm bit same quality. For 4 mm bit I went 4 mm deep, and for 3.5 mm bit slightly deeper, dunno for sure (accuracy of the drill press) something 4.25 to 4.5 mm plus the conical pit of the tip of the drill (I just measured the depth of several pits

)

One thing for punching the start pits, don't be afraid to punch them good, the drill really start easily when you got the pits good enough (not just lightly hit to mark the starter, but create a starting point of no return

).

Very interesting and informig site that tapmatic! Goota look into it a little deeper (and do saves to HDD)

Thx again!!

PS: just a little pic of the mill bit I used to cut the channels, bottom one (1.5 mm dia), compared to middle 1.0 mm dia, and top 0.5 mm dia...

g.l.amour · 11-10-2002, 02:29 PM

you didn't use dremel weels to cut out the chanels?

Puzzdre · 11-10-2002, 02:39 PM

No, I tried dremelling the chipset block to see how it goes, spent two discs on couple of rows, than saw the disc arbor is slightly bent and decided to try mill bits on this piece. Tomorrow I'm buying new arbor(s) for dremel, and than we'll see...

g.l.amour · 11-10-2002, 02:45 PM

the milling is way slower than dremel? the result sure looks nice

(btw, i was mistaken, thought your chipset piece was supposed to be the top of the circular design)

Puzzdre · 11-10-2002, 02:55 PM

Yeah, it's way slower, and what I did could hardly be called milling, I was just using the mill bit to cut the copper connections, (plexy shield on my face, gloves on hands etc...) and moving the vice under the mill bit to cut the channel.
I just had to try it. Without xy table, it's PITA.

g.l.amour · 11-11-2002, 07:31 AM

u need to explain me what pita means, read it everywhere, but haven't got a clue what it means.

puzzdre, the humanized cnc machine, lol

utabintarbo · 11-11-2002, 08:48 AM

Quote:

Originally posted by g.l.amour
u need to explain me what pita means, read it everywhere, but haven't got a clue what it means.

PITA = Pain In The @ss

Sorry to butt in!

Puzzdre · 11-11-2002, 08:52 AM

Quote:

puzzdre, the humanized cnc machine

YEEEEEEEEAAAAAAAHHHHHHHHHHHH!!!!!!!!
LOL LOL LOL LOL!!!!
Man, I like it!

Now to ask pretty pretty please pro/staff to put it instead my pro/uberGeek! Man!!!!

I just love it!!!

I'm pretty impressed. Gooooood thinking!!!!

PITA = pain in the a@@...

back to real:
I'we been doing some dremeling today, and man, it's way way easier than hand cnc'ing

. Bought new arbors and cutting discs (spent the discs already), but SATISFIED!!!! It turns out to be really good. Now I must go and buy more discs to finish the circular design, and then - straight to machine shop to barb those little gems! I'll be outta town at the end of the week, so I guess, the blocks should be ready for use next monday or so!

Btw, I'm still laughing!
Thx!!!

Puzzdre · 11-11-2002, 08:53 AM

Quote:

Sorry to butt in!

LOL! You're welcome!

g.l.amour · 11-11-2002, 08:54 AM

ahh thx ppl, i'm upto speed in the latest and greatest internet slang.

shit, puzzdre, so u mean to say u already burnt 5 discs at +/- 10€. thats fooking fast.

aren't u gonna tap the holes yourself?

edit: I WANT COPPER BARS NOW, can't wait to start, looks like endless and endless fun.

11-10-2002, 01:20 PM	#108
Puzzdre Cooling Savant Join Date: Aug 2002 Location: Croatia Posts: 969	I'm the kinda bloody perfectionist, well, almost...I always expect the result to look professional, which in case of making something for the first time, is hard to expect. Actually, tomorrow I'll be perfectly satisfied with the result, and keep on going happy. It's just me I guess, I must have one day of dissatisfaction for the happines of the other days. Today is The Day. I'm pretty sure you'll be quite satisfied with the drill/vice with the price and the performance, at least for some time until something more complicated comes into the way. And the price is nice. I vote go for it! Wow, design! Call it The Brain! Maybe not so complicated as it looks, but something on the connections between the holes - #Rotor's design is all about water taking the least resistive way to go, all the channels equally open for flow in any possible direction. This one you made looks like it would be more restrictive IMO, and with not equally distributed flow. Maybe to widen the connections between the holes as they are further from the center/inlet, to equalise it, or to open all directions between the holes. Also, I would make those inlet/outlet holes slightly bigger, I'm thinking for mine to be 12 mm inlet, 10 mm outlets. Ofcourse, let's make them and see how they perform! Good thinking! Btw, thx for bringing me back on the right (satisfied) track! __________________ 'Out of cheese error... ...please reboot the universe (press the GBL to continue)'

11-10-2002, 02:39 PM	#113
Puzzdre Cooling Savant Join Date: Aug 2002 Location: Croatia Posts: 969	No, I tried dremelling the chipset block to see how it goes, spent two discs on couple of rows, than saw the disc arbor is slightly bent and decided to try mill bits on this piece. Tomorrow I'm buying new arbor(s) for dremel, and than we'll see... __________________ 'Out of cheese error... ...please reboot the universe (press the GBL to continue)'

11-10-2002, 02:55 PM	#115
Puzzdre Cooling Savant Join Date: Aug 2002 Location: Croatia Posts: 969	Yeah, it's way slower, and what I did could hardly be called milling, I was just using the mill bit to cut the copper connections, (plexy shield on my face, gloves on hands etc...) and moving the vice under the mill bit to cut the channel. I just had to try it. Without xy table, it's PITA. __________________ 'Out of cheese error... ...please reboot the universe (press the GBL to continue)'

11-10-2002, 01:43 PM	#109
g.l.amour Cooling Savant Join Date: Mar 2002 Location: on da case Posts: 933	well i mentioned the ID of the feed barb as 8 that gives u about 50mm² of water going through. each side going to the outlet barbs has +/- 40mm² of openings for water to evacuate. so the outlet barbs won't be sucking at full capacity. but it does mean that the water hitting the chip area will flow +/- evenly in all directions. rotor stated some times that for good performance it isn't always necessary to have low flow restriction. in this design i was hoping to tackle the lazy water phenomenon, and make the water go at all directions, instead of just going directly for the outlets. someone else i heard state that it doesn't make sense to create a flow path with larger surface area than the surface area of the barb. anyway, to come back to you being dissatisfied. is there anything you could have done better? holes u misdrilled, dremeled some wrong areas. i don't know, but i guess not. anyway, i wouldn't feel to bad about the dremeling being to difficult on your circular design. when taking the easy/fast route;: if everything is finished and you mount the block, and you are sitting there enjoying your rebuilt pc some. u will start to thinkthinkthink whatif i spent some more sweat on working out a more difficult design, would i have gotten better performance? then before u know it, you will start working out that difficult design to c if it does perform better, and then u will have spent, double the amount of time for the same result. anyway, that how it mostly works for my twisted self. im very curious to see how the circular part worked out. as i guess, your top isn't a mirrored version of the bottom part. so the heattransfer effect stated here, that lets the bottom pins transfer the heat to the top ones, as being almost made out of 1 pin; won't happen in your design. lot of turbulance i guess would come out of that situation. anyway, forgot to ask, what dia drill would you use ideally (performance/ease of use) cheerz puzzdre, and congrats again for having done what u did (+ draw me over the line to go the same route)

11-10-2002, 01:51 PM	#110
g.l.amour Cooling Savant Join Date: Mar 2002 Location: on da case Posts: 933	btw , chk this here out, haven't deeply looked into it, but seems mighty promising http://www.tapmatic.com/tech_manual/index.html

11-10-2002, 02:29 PM	#112
g.l.amour Cooling Savant Join Date: Mar 2002 Location: on da case Posts: 933	you didn't use dremel weels to cut out the chanels?

11-10-2002, 02:45 PM	#114
g.l.amour Cooling Savant Join Date: Mar 2002 Location: on da case Posts: 933	the milling is way slower than dremel? the result sure looks nice (btw, i was mistaken, thought your chipset piece was supposed to be the top of the circular design)

11-11-2002, 07:31 AM	#116
g.l.amour Cooling Savant Join Date: Mar 2002 Location: on da case Posts: 933	u need to explain me what pita means, read it everywhere, but haven't got a clue what it means. puzzdre, the humanized cnc machine, lol

11-11-2002, 08:54 AM	#120
g.l.amour Cooling Savant Join Date: Mar 2002 Location: on da case Posts: 933	ahh thx ppl, i'm upto speed in the latest and greatest internet slang. shit, puzzdre, so u mean to say u already burnt 5 discs at +/- 10€. thats fooking fast. aren't u gonna tap the holes yourself? edit: I WANT COPPER BARS NOW, can't wait to start, looks like endless and endless fun.

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