This thread has sparked some decent discussion; it's certainly caused me to think again about the pump/system interface. However, Snutten has just shot his argument dead.

He started with:
And more recently posted:
Think I'll reread BillA's second post a few dozen times.

Hello ProCooling,
just registered here to see what's going on.

Too bad I'm already shot dead here...

With "obvious improvement" I mean going from really poor flow to OK. Anybody who questions this should try hooking up a 1048 with two blocks and two rads and you'll see what kind of Niagara flow I'm talking about.

Anyways, I won't argue my case but I'm happy to answer any questions and learn. (Just remeber I'm into w/c solely for silence so hunting for temp decimals doesn't do it for me. Stuff like flow through rads won't apply because my rad fans spin less than 400 rpm, and so on...)

sorry, bad manners: welcome snutten

one thing can be stipulated
with a fan speed "below 400 rpm" its a whole different world

did you actually measure the fan speed ?

Hey snutten - welcome here!

I suspect a good part of this thread came from my mis-understanding your "empirical" to mean "measured". Wish I'd thought to ask you "how had you measured the flow, in serial and parallel loops?" right after I'd re-stated what I thought you were saying. Sigh...
This is not to say you were wrong - just that I wouldn't have mentioned it here at procooling, implying that it was measured. (Flow, temp, and energy measurement is a big deal here - the same way dB measurement is a big deal over at silent PC review.)
Hope you stick around anyway...

Welcome, snutten! Hopefully, you won't catch too much flack from the old-timers here. There's a lot of good info in these forums (as in SPCR) and some good brains.

I run my fans from an Innovatek Fan-O-Matic which reads the tach from the fans. My rad fans are two Papst 4412 FGL @ 3.5 Volts.

I even undervolt my DDC, how's that measuring up on the Corny-scale huh?? It only pumps my secondary (separate) loop, cooling the Raptor which is also totally wrapped up in sound isolation. This system cools passively by passing copper plates glued to the outer rims of my alu-case. Undervolting the pump makes it easier to silence it. Big flow is moot for a HDD. Reason for separate loop is that temps must be kept significantly lower for HDD than CPU - GPU.


I've heard you guys are really mean bastards so this warm welcome is kind of disappointing...

we are only 'mean' to the persistently ignorant, lazy, and anyone who questions us
j/k
still no real idea of your system (and I'm not inquiring), but yours is not amenable to 'normal' analysis given the extremes you are operating at
- jeez, I've never even tried 4412-FGLs at 3.5 V

I've been testing my Yate Loon D12SL-12s, and most of them start consistently at 3.5V. Should be about 400rpm, but I cannot measure that low. Such speeds are definitely an option with the right fans. Did you ever get any YLs, Bill? PM me if not.

I've also played with undervolting my DDC but didn't find worthwhile gains. Mine's sitting on ~3/4" of melamine foam inside a heavily damped/foamed case, and that's enough to keep it quieter than my Super Silencer (with noisy stock fan).

Snutten, did you ever post any pics of your rig?

Bill, you should join us in SPCR's forums. You'd probably find some product opportunities, and the WC forum could use some shaking up.

HS
yea, got a half dozen sitting on the bench - I'm REALLY frustrated by not having any LOW noise measurment capability
what am I testing if I cannot measure it ?

I read SPCR, not sure I need more flack as a 'shaker'

Hi snutten,
Welcome to the forum too (though I'm a newcomer myself).
hope you didn't mind my theorising about your loops and possible improvements. Just noticed that you said you had your DDC undervolted running your secondary loop. Have you swapped your pumps about in the different loops? With a block and rad in series , then having two of these combos running in parallel I wonder what would happen if you had your DDc on your primary loop and your 1048 in your secondary loop. If you've tried this I'd be interested to know what you found?

Trying that would require extensive rebuilding. You wouldn't believe what my case looks like. The 1048 is totally suspended in elastic cords, the DDC packed up in thick heavy sound isolation. The outer passive loop, double lines, split main line, shrouded soft-mounted rads, cooled isolated Raptor, automatic fan controller, side-vented modded PSU, blue glow-stripe sides with spring connectors, dust filters, 2 extra suspended HDDs, dvd and burner - all this is squeezed into a sound isolated AMS g-Tower midi tower. I need Area-51 type fingers to change anything at all.

I have some pics on a server somewhere, perhaps I can find them and post. You might get a good laugh at my expense if you enjoy Jack-ass type humour. (Me casted as a Jack-ass.) But: the system's loudest part is the Nexus "realsilentcasefan" in the Seasonic SS-400 with cool air intake. PSU review at SPCR

I'd imagine the DDC wouldn't benefit from a split line anyway since it has excellent head and poor flow. I bought the DDC one and a half year after the main line was set up, to create the secondary HDD loop.

I hope you find the pictures as I like 'busy' cases.

How well is the DDC running undervolted>

Re: "parallel flow best with 1046" - hunh?
 

So, did anybody try it out or what? Still new age mumbo-jumbo to split lines of big systems with small pumps? Curious to find out if it was a frek accident, not to mention downright impossible as stated by some?