Laing DDC differences? (DangerDen, Swiftech, others?)

The alternative to direct air cooling

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
Mr_Smartepants
Posts: 539
Joined: Fri Apr 04, 2003 6:35 am
Location: Cambridgeshire, England

Laing DDC differences? (DangerDen, Swiftech, others?)

Post by Mr_Smartepants » Fri Dec 31, 2004 11:00 pm

Hi all,
I plan on purchasing a watercooling kit in the new year and I've settled on either the Swiftech H2O-120 rev-3 kit with the LaingDDC (MCP-350)pump or the equivalent kit from DangerDen (different WBs of course) :D .

Are there any owners out there of the respective DDC pumps that can share some insight as to the wiring differences between the two pumps (ie. 3-pin RPM motherboard connector, 4-pin Molex, wire lengths, etc.)
The DangerDen site claims to have both 4-pin and 3-pin connectors on their pumps but the Swiftech site only claims to have a 4-pin connector. The pumps appear to be from the same supplier so does DD modify theirs in some way?

In fact, if some lucky soul happens to have access to a pump from both companies, they can provide loads of info.

Just trying to be an informed consumer
Thanks for any help.

Edward Ng
SPCR Reviewer
Posts: 2696
Joined: Thu Dec 11, 2003 9:53 pm
Location: Scarsdale, NY
Contact:

Post by Edward Ng » Sat Jan 01, 2005 8:12 am

The Danger Den Laing DDC has molex for power, and then the 3-pin is only for reporting speed/activity info to your mainboard fan header. I haven't seen the Swiftech one yet, but I gather it's the same.

Mr_Smartepants
Posts: 539
Joined: Fri Apr 04, 2003 6:35 am
Location: Cambridgeshire, England

Post by Mr_Smartepants » Sat Jan 01, 2005 9:58 am

I thought they'd be the same too but a scrutinous eye of the below picture shows only two wires (black/yellow) coming from the pump. Any Swiftech pump owners out there who can verify the two-wire only pump?
The RPM sensing function would be ideal for a failsafe shutdown sensor in the case of pump failure.
Image

Edward Ng
SPCR Reviewer
Posts: 2696
Joined: Thu Dec 11, 2003 9:53 pm
Location: Scarsdale, NY
Contact:

Post by Edward Ng » Sat Jan 01, 2005 10:08 am

Looks like it's missing the activity report line.

Also, that pic shows some sort of screw-down lip that comes off the side; my DDC doesn't have that; it's just about square and the only thing hanging out the side are the two fittings.

EDIT: Actually the picture at the top shows the screw down, the picture at the bottom doesn't...

How odd.

-Ed

Mr_Smartepants
Posts: 539
Joined: Fri Apr 04, 2003 6:35 am
Location: Cambridgeshire, England

Post by Mr_Smartepants » Sat Jan 01, 2005 12:47 pm

Thanks Ed, :D
You've confirmed my suspicions that although the DangerDen and Swiftech pumps have the same manufacturer they have different "features".

So, the DangerDen kit it is then!

Mr_Smartepants
Posts: 539
Joined: Fri Apr 04, 2003 6:35 am
Location: Cambridgeshire, England

Post by Mr_Smartepants » Mon Jan 03, 2005 1:10 pm

Are there any Swiftech MCP-350 owners out there that can confirm that the MCP-350 does NOT have an RPM sensing 3-pin connector attached to the pump?
Anyone at all?

jfweaver
Posts: 30
Joined: Mon Jan 17, 2005 8:05 pm
Contact:

Post by jfweaver » Mon Jan 17, 2005 8:07 pm

I have that pump and can confirm that it doesn't have the RPM sending wire.

Mr_Smartepants
Posts: 539
Joined: Fri Apr 04, 2003 6:35 am
Location: Cambridgeshire, England

Post by Mr_Smartepants » Mon Jan 17, 2005 8:50 pm

pissboy wrote:I have that pump and can confirm that it doesn't have the RPM sending wire.
Bummer :cry:

Spoon Boy
Posts: 168
Joined: Sun Jan 23, 2005 11:44 pm
Location: London

Post by Spoon Boy » Thu Jan 27, 2005 10:10 am

Maybe if someone knows how to source Apple components

It should be no surprise to many that this version of the Laing DDC pump is based on the OEM Apple G5 2.5 GHz liquid cooling system that Delphi supplies to Apple. The main difference between the Apple G5 version of the pump and the Swiftech MCP350 is, no built-in volume compensator (required by the closed OEM liquid cooling system) and the 2-wire Molex connector. The Apple version supports both a tach signal output and variable speed control input, which allows the OS to dynamically control the pump speed based on thermal load.
http://www.systemcooling.com/swiftech_mcp350-03.html

someone know what a volume compensator is or does ?

pdf27
*Lifetime Patron*
Posts: 264
Joined: Wed Nov 12, 2003 12:59 pm
Contact:

Post by pdf27 » Thu Jan 27, 2005 10:24 am

I think it's equivalent to a reservoir, to allow for the fact that water will actually escape from a watercooling system over time (mainly through the walls of the tubing, at a very slow rate).

BillA
-- Vendor --
Posts: 10
Joined: Sun Aug 24, 2003 10:04 am

MCP350 / Laing DDC

Post by BillA » Thu Feb 17, 2005 1:32 pm

a couple of notes of clarification
all these pumps are made by Laing, and they are the same if the part numbers are the same
this may, or not, be the case in comparing 'different' DDC pumps

DD apparently elected to start selling the DDC with the RPM lead even though it outputted 10 pulses per revolution
Swiftech elected to wait until the circuitry was changed to output 1 pulse per revolution
- different companies have different philosophies
all MCP350 pumps shipped for the last month or so have the RPM lead and it outputs 1 pulse per revolution
- you may have to ask the distributor which version they have ? (no new part # as it is the same pump)

as Laing sells these pumps direct, I cannot comment on what else may be around
- but I believe only DD was peddling the pumps with the 10 pulse output

the DDC with the compensator is only for pressurized systems with EXTREMELY LOW permeation losses, i.e. metal tubing or similar

Edward Ng
SPCR Reviewer
Posts: 2696
Joined: Thu Dec 11, 2003 9:53 pm
Location: Scarsdale, NY
Contact:

Post by Edward Ng » Thu Feb 17, 2005 4:35 pm

*points* BillA is here!!! 8)

Mr_Smartepants
Posts: 539
Joined: Fri Apr 04, 2003 6:35 am
Location: Cambridgeshire, England

Post by Mr_Smartepants » Thu Feb 17, 2005 11:00 pm

BillA, thanks for the clear/concise info....um...are you a vendor?

HammerSandwich
*Lifetime Patron*
Posts: 1288
Joined: Sat Oct 25, 2003 3:21 pm
Location: 15143, USA
Contact:

Post by HammerSandwich » Fri Feb 18, 2005 2:13 pm

Bill has been Swiftech's design engineer for almost 2 years. Looks like he's been hiding here for a while, too...

Mr_Smartepants
Posts: 539
Joined: Fri Apr 04, 2003 6:35 am
Location: Cambridgeshire, England

Post by Mr_Smartepants » Fri Feb 18, 2005 11:11 pm

DOH! :oops:
I guess I should pay more attention.

Bat
Posts: 279
Joined: Sun Jan 05, 2003 12:09 pm
Location: U.K.

Post by Bat » Mon Feb 21, 2005 11:23 am

I would rather have ten pulses per revolution anyway. Why throw away some of the detail by only reporting every tenth pulse? If necessary this can be done at a later stage by whatever software or circuit is making use of the data.

The ten pulses could be particularly useful if you're wanting to make circuitry to automatically set the pump to a particular speed.

BillA
-- Vendor --
Posts: 10
Joined: Sun Aug 24, 2003 10:04 am

Post by BillA » Mon Feb 21, 2005 11:31 am

Bat wrote:I would rather have ten pulses per revolution anyway. Why throw away some of the detail by only reporting every tenth pulse? If necessary this can be done at a later stage by whatever software or circuit is making use of the data.

The ten pulses could be particularly useful if you're wanting to make circuitry to automatically set the pump to a particular speed.
what software would you use to do this ?
are you suggesting that 0.1RPM is useful information ?

Jam
Posts: 1
Joined: Thu Feb 24, 2005 5:27 am
Location: Staten Island, New York

Post by Jam » Thu Feb 24, 2005 5:42 am

Greetings BillA and the rest of you lot,

:lol: :lol: ,s

I did not know about the difference in the pulse repetition rate from the RPM sensor output on older pumps.

Recently, I bought three of these pumps, one from Swiftech via NewEgg - the last to arrive and with a blue RPM sensor wire.

I previously bought one from FrozenCPU.com with a Swiftech label and no RPM sensor.

I bought the third from Danger Den also w/o the sensor wire. These purchases were transacted within the past month.

I came across and article by Lee "Robotech" Garbutt over at www.systemcooling.com that described a mod that would allow these pumps to perform very close to the MCP-650.

In the process of mod'ding these three pumps for 1/2" inlet barbs, I discovered the following:

Image

I reasoned that the pumps were likely to be the same internally, so I added an RPM sensor wire as indicated above.

I then set the pumps up on an AeroCool GateWatch fan controller and power supply and they all indicated the same RPM +/- 100.

{I ran them dry for only a few moments and obtained an indicated RPM of around 4,700, no load. I have another couple weeks work before the system under construction will be complete.}

BillA?

Any idea for what the fourth PCB tab is intended?

Could it be that moving the wire over one tab would produce a fractional rate?

I did not want to attempt measurements running the pumps dry for more than a couple of moments.

Cheers,

Joe


Image


PS . . .

Naturally, one has to be willing to void tha manufacturer's warranty to tinker in such a fashion.

Post Reply