• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Custom manifolds arrived!

M

Mysterae

Gawd
Joined
Sep 11, 2005
Messages
648
Look at these puppies!

manifolds_01a.jpg


The mouse is included in the pic to give some scale.

The manifolds need a little fine sanding to make them more clear and purty, but I'm well happy with how they have turned out.

In case you are wondering what the hell they are for, against all that is taught, I'm going parrallel baby! Here's some history.
 
They turned out really nice, they should look great when the system is up and running.............. :D
 
Thanks BH, just need to build the other bits now and perhaps test the flow of these manifolds and see how restrictive they may be :eek:
 
holy balls! that's wild. really good idea, whether it works out or not is another story though.

what kind of pumps are you gonna have pushing them? your goal of having a quiet system is gonna depend on you being able to get quiet pumps that can push a lot of water...
 
You should have incorporated an O-ring into the design to make a watertight seal...how do you plan on making the seal watertight?
 
BH - The centre hole is for a 1/2 BSPP fitting and the six outer holes are for 1/4 BSPP fittings. I wanted the centre fitting reasonably big to get enough water to the other ports, to downsize the pipework going around the case.

DJ - I'm going to use a D5 pump, with the possibility to add another one if the circuit should need it. The manifold may be restrictive, but hopefully having the loops in parallel should ease it somewhat. Here's hoping!
 
Yes Erasmus, I had thought about a large o-ring between the body and the lid, but couldn't find one large enough. Instead I'll use a silicone based sealer I have that will bond the gap between the curved part of the lid and the body. That and the screws should ensure it's water tight. I'll be testing my theory when my fittings arrive.
 
I dunno where all you looked... but I know for sure you can get o-rings beyond 3" diameter. I used some good 3"OD o-rings on a potato gun project about a year ago.
 
I don't understand - how is that setup not identical to this one using T-fittings?



It seems like these manifolds would just cost you extra money (T's are cheap, I assume cheaper than manifolds) and force you to use extra tubing. Is it because of the high pressure between the pump and first device? Anyway, this would just cause me a nightmare, as you'd have to have blocks for different things with very alike pressure drops. Otherwise, if one thing drops twice the pressure, it's going to get half the flow rate.
 
decker, with that T fitting design, there will be little to no water flow by the end of the chain, it will all be going through the first couple of blocks. A manifold with correct water restricion will allow even flow to all the blocks.

I kept up with the original thread and am very excited to see how this works out.
 
I'm not sure what I was thinking, what I drew is totally different! Pardon my ignorance, it's been a long day.
 
As Silent Assasin has said, it's all about equal flow to each block. Having said that, the return manifold (or hot as I call it) will have a mini-ball valve on each port, with the two cpu returns as the exception. The cpu must have as much water flow as possible.

The valves will be used to tweak the flows of the other blocks, to ensure the more important blocks get the required flow. This is going to be the trial and error part, as without fitting flow sensors (more connections!) it'll be difficult to judge as you can't see the water flowing. I'll use heat as my indicator for which and by how much the valves will be adjusted.

If this were a utopian build, and I had the money and the knowledge, it would all be processor controlled! For example, using servo valves or some such, a processor would monitor the flows and heat and adjust each sv accordingly. Overkill for a home pc, and it probably wouldn't be worth it, as once you find the correct ratio of flow for each block, I doubt you'd want to adjust it, imo.
 
Mysterae said:
If this were a utopian build, and I had the money and the knowledge, it would all be processor controlled! For example, using servo valves or some such, a processor would monitor the flows and heat and adjust each sv accordingly. Overkill for a home pc, and it probably wouldn't be worth it, as once you find the correct ratio of flow for each block, I doubt you'd want to adjust it, imo.
Click to expand...

Now THAT would be a kickass mod...if I ever won the lottery (say the 300million$$ powerball) I would finish school and then design all the necessary parts to do that, even a microprocessor to monitor the temps and adjust the servos. Heh, would be lotsa fun to have the money to actually manufacture it.
 
I know Erasmus! At the start of this when someone mentioned some blocks getting more flow than others, I thought about valves, and then servo valves. When I researched flow control, servo valves, flow monitors etc, I quickly noticed it would have added another grand to the build cost! Would be sweet though, and CAN be done!

Guess I gotta do it manually for now :p
 
The fittings I ordered to test fit and measure arrived today:

assortment_01.jpg


There is a small problem though. The 1/2" fitting doesn't fit! It threads in about 1/4 - 1/2 a turn in and jams.

duff_bspp_01.jpg


This is with both manifolds. I thought the bloke who made the manifold had cocked up and threaded it tapered not parrallel, or that the thread was crap. Called said bloke and he tried to ensure me that the tap was a brand new 1/2" BSPP. But my fitting doesn't fit I said! Send it back and I'll re-thread it he said, 'cause he's a real nice bloke.

Then some one at work gave me a radiator valve with fittings so I could groove it and perhaps make the thread good. Only this fitting was a perfect fit!

temp_bspp_01.jpg


So, are radiator fittings threads parrallel? Got a debate going about which was parrallel and which tapered. Consensus says that the fitting the supplier sent me was BSPT, although I ordered BSPP and the paperwork says BSPP! Someone's parts bin maybe a bit mixed up? Someones got it wrong :mad:

threads_01.jpg

It's a difficult pic to judge I know, can anyone confirm the above?

Some other bits to give you an idea of the build..

bvalve_01.jpg


bvalve_02.jpg
 
looks as though someones parts bin is very messed up...btw nice looking design..i look forward to seeing it when completed :)
 
Yeah, your doing it!!!! I love it!! I hope i had a little to do with your inspiration..

man those manifolds look sweet. I love it! Its going to look awesome.
 
Mysterae said:
The fittings I ordered to test fit and measure arrived today:
Click to expand...

where did you get those?

the manifolds are polycarbonate? look real nice ;)
 
Memory Remains - yeah I agree, and thanks!

bradyapda - yeah I'm doing it dude! And you gave the nessasary encouragement which is always a good thing! Doubt I'll be finished before your build, looking good!

Ice Czar - the fittings are from Airlines Pneumatics, based in Glasgow, Scotland (where I'm originaly from ;) ) so might be a bit far for most of your readers. They sell practically everything you'd want fitting wise, and their website makes it real easy to pick what you want. Order online and you get a 5% discount to boot. Have I plugged them enough? :p ! Yeah, the maifolds were made from poly, and turned out better than the drawings. I just hope they perform as well as they look!
 
From what I can make out in the pictures your manifold is threaded with standard “machine” threads, as is your ball valve fitting.

The rest look like a standard “Pipe thread” fitting which is tapered.

I don’t think I’d try to re-tap the manifold at this point too much chance of messing it up and damn, it does look pretty. Nice work. ;) :D
 
One of the problems associated with tapping synthetics is that it takes a special geometry on the tap to cut correctly. Another issue is that the taps wear very fast as acrylic is abrasive though they should have lasted easily for the small amount of holes they tapped. Perhaps they had a old tap that had worn out? Usually when you tap acrylic's you want to do it as cool as possible and cut slightly oversize as the material tends to close up resulting in a smaller hole after the tap is removed. In any case you should send the part back as it is scrap. But it sure does look nice. :D
 
BillR - The 6 (countersunk) threads on the top of the manifold (to close the lid) are M6. The 6 ports are 1/4" BSPP, and the centre port is 1/2" BSPP for sure. This should explain it. The mini ball-valves are 1/4" and a perfect fit, but lack the sealing o-ring. The 1/4" fitting have the o-ring but is difficult to see in the pic I posted earlier. That's another clue that the 1/2" fitting I was sent was tapered, it didn't have the o-ring or the 'shoulder' to seat one. BSPP fittings don't seal on the thread so need the o-ring, but don't need the messy plumbers tape or goop. Be careful judging thread sizes from a picture, especially through a poly block that distorts the view somewhat!

Top Nurse - Don't say that, it ain't scrap :eek: ! The bloke who did the machining assured me that it's 1/2" BSPP and sounded hurt when I insinuated he got it wrong. He even said he had to buy a new tap for the job. It's more likely the supplier sent me the wrong type of fitting.

SomeoneWhoIsntMe - Nice name, can you explain the reasons for calling it "Tritium"? Is it because I'm "Tri'in-sUM" thing odd :D and probably pointless?
 
Tritium
a radioactive isotope of Hydrogen
Tritium combines with oxygen to form a liquid called tritiated water (T2O or partially tritiated THO), somewhat like heavy water
Click to expand...
light that puppy up and nuclear it will be :p
 
Mysterae said:
SomeoneWhoIsntMe - Nice name, can you explain the reasons for calling it "Tritium"? Is it because I'm "Tri'in-sUM" thing odd :D and probably pointless?
Click to expand...

You wanted to name your machine H3, and Tritium is the name of 3H.
 
Cool name, good reason. I'm not that clever though, to have thought of that... I'll name it soon, when some more parts arrive and I can start a work log, say a week or so!
 
in your diagram that is a bad idea to be sucking water out of a res. to go into the loop, it should have the rad sucking into the loop, and the res in the hot water part.
 
DaCoOlNeSs said:
in your diagram that is a bad idea to be sucking water out of a res. to go into the loop, it should have the rad sucking into the loop, and the res in the hot water part.
Click to expand...

Actually it is just fine where it was...what I would change though is move both of the pumps to where pump 2 is, just have one after the other.
 
Ice Czar said:
Tritium
a radioactive isotope of Hydrogen

light that puppy up and nuclear it will be :p
Click to expand...
'fraid not. to get all thermonuclear you need deuterium and tritium, in a gas phase, with lithium to produce additional tritium as the reaction proceeds, as well as any number of other special conditions.

adding all of that oxygen into the mix (tritium oxide and deuterium oxide, tritiated heavy water) would mean that nothing at all would happen. it takes too long to strip the chemical bonds apart to keep a reaction cascade going.

CANDU: the way of the present. canadian reactors ROCK! :)
 
add light and dye

if it glows it must be radioactive\nuclear :p

a wasn't being literal, the FBI might object :p
 
Ice Czar said:
a wasn't being literal, the FBI might object :p
Click to expand...
oh my, i'm scared now. :)

seriously, you can learn that and more in an undergrad level course in nuclear tech. in point of fact, i did just that two years ago. :D

awsome course, by the way. university of waterloo SCI 270: regarding nuclear technologies. general interest course aside, it's interesting. offered via distance ed.
 
All this talk of nuclear reactions; there's some clever people on this forum. Once I'm finished this pc build, I hoping for it NOT be radioactive, only look like it ;)

On the subject of pump placement, I'm at a turning point. So many ideas of how many pumps and placed where keep coming my way. I don't have enough experience in water cooling, or plumbing for that matter, to decide. So I'm going to start a poll to see what everyone thinks. Up for it?
 
You must log in or register to reply here.
Back
Top