DDC current draw

[Langer]

Sorry about the title typo...

Hey guys another query for you all.

I'm looking to determine the max current and wattage draws of a DDC pump.

I have a MeanWell S-240-12 PSU sitting here. It puts out 12VAD/DC @ 18amps w/ 240watts.

I want to use this PSU to drive 10xDDC pumps, will I be able to?

[Kayin]

I think the question needs to be, "are you insane?"

[Langer]

You have no idea...

Can I run 10pumps w/ 18amps?

[Baleful]

Title fixed.

[rubidium]

Sorry about the title typo...

Hey guys another query for you all.

I'm looking to determine the max current and wattage draws of a DDC pump.

I have a MeanWell S-240-12 PSU sitting here. It puts out 12VAD/DC @ 18amps w/ 240watts.

I want to use this PSU to drive 10xDDC pumps, will I be able to?

A stock DDC3.2 out of the box is rated 18W at 12V, which implies a 1.5 amp current draw. That would put you at 15 amps for 10 pumps. I've seen estimates for efficiency that range between 10% and 15%, so conservatively you might also want to plan for about 135W of heat dump.

rubidium

[Langer]

Thanks a lot rubidium, exactly the answer I was hoping for. Even at 15% a efficiency extreme that's only 17.25watts, so my PSU will work nicely.

The 10 pumps will be coupled with 10 360mm Rads spread across 5loops. So heat dump isn't a huge concern.

Thanks again!

[Snyxxx]

Langer,

If you look at Martin's graph for a DDC 3.2 (I assume that is what you are looking at buying), then you can back out the running amps drawn by dividing Power Consumption (Watts) by the relavent Voltage (12V, 10V, etc) for a particular flow rate.

My main concern would be the "in rush" current required at the start of the pump turning on. I thought I saw somewhere what this current demand was. I will keep looking.

The one thing I remember is one of the revisions made to the DDC 3.x series was a lower "in rush" current demand compared to a DDC-2.

This graph does not scale well in the post. Check out martin's pages in the watercooling section and look for the DDC 3.2 shootout test for bigger images of the graphs.

http://www.realredraider.com/vbulletin/attachment.php?attachmentid=346&stc=1&d=1233081396

[Petra]

The one thing I remember is one of the revisions made to the DDC 3.x series was a lower "in rush" current demand compared to a DDC-2.

In-rush current shouldn't be much of a concern with the DDC-3.2 at all. They exhibit a slow, steady ramp-up during the startup process (electronically controlled, no surprises). The old DDC-2, on the other hand, had that nasty 4A spike at startup...

[Snyxxx]

I searched around some and came up with a few things:

The in-rush current could be 105% of the operating current. This means each DDC could draw ~3 amps at the start for a short duration.

I sam a nice discussion on the history of the pumps which answers the initial delay of the pump start when doinf test with just a PSU atatched.

"When Laing first started to produce water pumps for use with watercooling applications, the DDC wouldnt start with older low power PSU's becuase the start up amps drawn by the whole system inc the pump was too great.. This is Version 1 - White Impellor"

"Version 2 followed with a software update and a new design impellor (Brown/Black) which allowed the pump to restart every 10 seconds if there was not enough power coming from the PSU (which in your case is not a problem) and also has a time delay start up so the Mobo/CPU/VGA/Ram all start up first then the DDC powers up.. Usually 5-15 seconds the delay is."

"Version 3 which laing now stock (we only have blue as well now) has a newly designed Blue impellor which is supposed to shift the water more efficiently and also has the same start up & auto restart software as Version 2"

Which fits into my 3-4 second delay as Ive got a version 2 pump and what the guys have been sayong over at bit-tech forums, hope this helps others.

I think I may try to measure this soon with a multimeter as I am now curious.

[Snyxxx]

In-rush current shouldn't be much of a concern with the DDC-3.2 at all. They exhibit a slow, steady ramp-up during the startup process (electronically controlled, no surprises). The old DDC-2, on the other hand, had that nasty 4A spike at startup...

Petra, you are probably right. Most info is old and based on the good ol' DDC-2.

BTW, have you measured the current draw for a 3.2? This would save me some time.

[rubidium]

Unfortunately, I don't have any data on pump startup current. Martin's curves do confirm that your PSU should support sustained running without any problems. You could always throw a quick circuit together to start up the pumps sequentially - say 100-200 milliseconds apart. This is what's typically done with big RAID arrays.

[Petra]

Snyxxx, that seems to be a little off because there were substantial software changes between the DDC-2 and DDC-3.2 ... nevermind that the motor (stator) was completely redesigned. Doesn't matter if they supposedly made the impeller more efficient, they crippled the DDC when they introduced that damn new motor.

As far as my old measurements go (need to get around to testing with the new setup), operating current for the 3.2 ranged from 0.9A-1.53A in stock trim and 0.91A-2.14A with the 01s top depending on where you were on the pump's PQ curve. I couldn't actually detect any startup spike with the DDC-3.2, unlike the DDC-2. I'd probably need to borrow a nice, datalogging lab/bench ammeter to pick anything up (if there is anything substantial to pick up).

[Langer]

I was thinking exactly that Rubidium, make a little arduino circuit to power on the pumps sequentially and run them at minimum speed for ~5secs before ramping up.

This should give me lots of overhead

I just found another PSU from meanwell, it offers 12V @ 42A w/ 396W.
The USP-500-12 - PDF clicky (http://www.meanwelldirect.co.uk/public/ranges/pdfs%5Cr881%5Cr881_3.pdf).

I don't want to spend $280 for it though, I already have the S-240-12.



New question, what's the best DDC revision: 1, 2 or 3?
What are the differences?

[Petra]

New question, what's the best DDC revision: 1, 2 or 3?
What are the differences?


DDC-2's are going to be the most powerful option, hands down... but they're also less reliable (by a fair margin) and you'd have to cope with their troublesome 4A startup spike.

DDC-3.2's (with tops) are still a tiny bit better than the D5 as far as performance goes but are a far cry from the DDC-2. Pluses are vastly improved reliability, the elimination of the DDC-2's infamous startup spike, and, uhh... something else... I'm tired.

DDC-1's aren't worth your time if you're looking for performance.

*edit* Also, the DDC-1 and DDC-2 are no longer in production... the changeover to the DDC-3.1 and DDC-3.2 happened early last year (the DDC-3.1 and 3.2 are the same pump, 'tis just a bridged solder pad which determines the operating speed)

[Septim]

well from all this, i assume that its better to keep that ddc-2 running 24/7 instead of power on/off the PC... good enough for 24/7 crunching...