How much pump is enough? How much is too much?

[RaptorRaider]

MaxxxRacer has posted some numbers about the input of his MD-20R; they're way lower than Iwaki's numbers. :shrug:

In a few months I should have my own numbers on the MD-20R (at 50Hz).

Cathar, can you tell us how you measured the heat dumped by the Iwaki pumps?

[Brians256]

I have to say that this thread is chock full of EXTREMELY good info. It does deserve sticky status (unless some good objection comes up).

[RADCOM]

Not sure if this is the right place for this query please move if inappropriate
I have this pump and I wondered if anyone had done a scientific comparison in relation to the pumps that normally make the shortlist?.
http://www.alphacool.de/perl/shop.pl...1&art_id=13106
and

[Razor6]

If the specs are believable, around D4 performance, but that 8W sounds too optimistic given the stated performance.

[IndianaCouple]

I know this may sound silly, but I'm in the process of making my own wc system. Wondering if a magnum 350 Here http://www.marineland.com/products/c...con_magnum.asp
will work. A lot of this forum seems to be over my head, but would be grateful for any ideas. Also wanted to use the evaporator from an air conditioner with fans to exchange heat.... Thanks, and great forum!

[Happy Hopping]

http://www.aqua-computer.de/download...040425engl.pdf

In this Aqua manual, they said on page 25 of the manual that "since the pump includes a synchronous motor, it generates a magnetic field as any electromagnetic motor. Also there is a rather powerful permanent magnet inside the pum. This magnet rotates and creates a changing magnetic field too.

"For this reason keep a safety distance of at least 0.5m to all sensible devices such as hard drive, screens and so on."

Now, eventhough Aqua is talking their own pump, surely the above applies to all pumps. But I notice w/ so many pic. posted from various users, that their pumps are quite close to other sensible devices, i.e., their hard drive.

Does anyone cares about loss of data?

[maxSaleen]

Indiana: you are looking at a canister filter. That won't be of much use to you in building a water cooling system. Sorry. Read through some more posts on this forum to get some general knowledge about water cooling.

[IndianaCouple]

Originally Posted by maxSaleen

Indiana: you are looking at a canister filter. That won't be of much use to you in building a water cooling system. Sorry. Read through some more posts on this forum to get some general knowledge about water cooling.

Yes, I agree... I will have to locate something with high positive pressure.

[pauldenton]

Originally Posted by Happy Hopping

http://www.aqua-computer.de/download...040425engl.pdf

In this Aqua manual, they said on page 25 of the manual that "since the pump includes a synchronous motor, it generates a magnetic field as any electromagnetic motor. Also there is a rather powerful permanent magnet inside the pum. This magnet rotates and creates a changing magnetic field too.

"For this reason keep a safety distance of at least 0.5m to all sensible devices such as hard drive, screens and so on."

Now, eventhough Aqua is talking their own pump, surely the above applies to all pumps. But I notice w/ so many pic. posted from various users, that their pumps are quite close to other sensible devices, i.e., their hard drive.

Does anyone cares about loss of data?

most pumps are shielded i believe ..... eheim 1048+ are (but 1046 isn't....)

[Long Haired Git]

Warnings exist to protect the company from lawsuits, and thus err on the side of caution.
DansData did some testing with a HDD and a "powerful" magnet, and couldn't get the HDD to fail.
Your results may differ, always read the label, the authors comments above are not a binding statement and merely an opinion. Consult your doctor if pain persists and read the license before opening the software.

[Strages]

I finally got my Storm G4 system setup in the beginning of June (well, I guess it's been a while ), anyhow, I was looking at the performance curves to approximate the flow rate of my system. I noticed, however, that the curves are based on a system that consists of a Thermochill HE120.2, where I have a 120.3. How much more resistance is there with an HE120.3 vs. a HE120.2?

[HammerSandwich]

Originally Posted by Strages

How much more resistance is there with an HE120.3 vs. a HE120.2?

Jump down to Graph 17.

[Strages]

Originally Posted by HammerSandwich

Jump down to Graph 17.

Thanks a lot! That helps out .

[blueline]

Cather i'm a newbie so just wanna check with u about the MCP600 and MCP650? it spec said both are a 12V pump but the conclusion u written 13.8v right? how u change it to 13.8v? and right now there are a MCP655 pump and does it perform better than MCP600? Thanks alot and sorry for the noob question.

[tigger995]

hello, i have an mcp350 running a swiftech 120mm rad,asetek antarctica cpu block,
asetek gpu block,with a swiftech 5.25" bay res.
my chip is a 64 3000+@2430mhz.my temps are usually around 25-30c for the cpu(depending on ambient temp)and 36-40c gpu.
is this pump good enough?.
thanks guys.
p.s. how can i get my pics in here?

[Marci]

@ blueline - MCP655 performs better but has higher heatdump than MCP600. Depends on your rad as to which is better for you. For 13.8v - independant PSU that can provide 13.8v instead of 12v - that's how.

@tigger - If temps are good enough for you, pump is good enough for you. Looks fine to me. Yes, could be better... but it can always be better. Bigger rad, more powerful fans on rad, bigger pump, different CPU Block with lower c/w rating (storm, mp-05-sp)... whether it's good enough is down to whether you consider your current temps to be good enough, and how much cash you're willing to dump into it to make it better, and how much noise you wanna put up with. More airflow always = better performance than less airflow, but results in more nosie.

ie: YOU create the definition of what is good enough for YOU.

How to post pics? Hit the "Reply" button instead of using quick-reply, and use the "Manage-Attachments" button to attach an image to a post, or publish from your own webspace...

[img]http://the.web.address/of/your/image.jpg[/img]

[ColdFlo]

I know there is a 12volt version, but for a cheap power supply for the RD-30 couldn't you use the +12 and -12 on a cpu psu?

[BGP Spook]

The -12v line is very weak, usually little more than 1 amp.

[muffy]

i think i'll refute the bases for this post on how much heat is created by the pump . sure you will loss output power than the input power and a % of it will be loosest as heat but this is not the biggest factor. Most of it is loses in the inefficiencies of how a motor works .. A motor works because a current is passed though a wire around the Satator which creates a magnetic field which then attracts or repels the magnets in the rotor which creates a rotation . The losses involved in this process is where MOST of the electricity is lost not to heat . i believe that this post shouldn't be a sticky cause it's not based on fact . and therefore should be "DE-stickyed" as it's putting alot of people wrong. i agree with the general idea of the post but may it should be re-written with theoretical pumps creating theatrical heat.

[murray13]

Conservation of energy. You put into a motor a set amount of electrical energy. You get out an amount of physical work. Those two are not equal. The rest of the input energy IS output as heat. The only discussion that can be made is for an inline pump and how much of this heat is transfered to the water.

I'm not even going into the whole moving the water creates heat discussion...just talking about the pump itself.

[muffy]

Conservation of energy. You put into a motor a set amount of electrical energy. You get out an amount of physical work. Those two are not equal. The rest of the input energy IS output as heat. The only discussion that can be made is for an inline pump and how much of this heat is transfered to the water.

I'm not even going into the whole moving the water creates heat discussion...just talking about the pump itself.


totally agree the dedate is how much of it is turned into heat ... and i believe this numbers are very wrong

and yeah there is friction and sound created that i agree aren't worth going into

[muffy]

eee

[bobo5195]

Originally Posted by muffy

i think i'll refute the bases for this post on how much heat is created by the pump . sure you will loss output power than the input power and a % of it will be loosest as heat but this is not the biggest factor. Most of it is loses in the inefficiencies of how a motor works .. A motor works because a current is passed though a wire around the Satator which creates a magnetic field which then attracts or repels the magnets in the rotor which creates a rotation . The losses involved in this process is where MOST of the electricity is lost not to heat . i believe that this post shouldn't be a sticky cause it's not based on fact . and therefore should be "DE-stickyed" as it's putting alot of people wrong. i agree with the general idea of the post but may it should be re-written with theoretical pumps creating theatrical heat.

Motors are indeed inefficient and how inefficient they are could easily be determined by graphs of performance in operation (or getting graphs for similar motors if these are not available). Although any graph won't tell where the heat goes which is the question you are shooting for.

Pumps tend to have a large amount of sealing apparatus on moving components and this will contribute a large amount of friction which I believe you are ignoring. Heat generating by this is very much more likely to enter the water than heat from the motor. Therefore this is likely to be what is warming up the water in the pump.

I'm not exactly sure that sound and vibration can be completely ignored in terms of losses. Doesn't sit right but that maybe cos im eating a burger.

Either way i think i may have said this before in a thread somewhere but a pump is not a right fit (including if it is too big):
- operating in a weird part of its range i.e. running two fast or two slow.
- if you are pumping too much / too little mass flow of water compared to air going through your fans. A little bit on the air excess side is better thermodynamically. A little more water is better noise wise.

Has anyone tried putting a pump in a sealed chamber and letting it get very hot. A insulated pump is not transferring heat to the air. So heat is forced to travel to the water. Compare insulated and uninsulated cases and you've got some nice numbers to work off. For extra science points put thermocouples on bits of the pump. the bits that get relatively hotter may indicate what is putting heat where.

[muffy]

[quote=

Has anyone tried putting a pump in a sealed chamber and letting it get very hot. A insulated pump is not transferring heat to the air. So heat is forced to travel to the water. Compare insulated and uninsulated cases and you've got some nice numbers to work off. For extra science points put thermocouples on bits of the pump. the bits that get relatively hotter may indicate what is putting heat where.[/QUOTE]

UMM yes .... well close enough

i disbelieved this guys numbers so much that i spent a few days creating this rest rig .. What you can see in this pic is my modded 655 pump which has max 78watts at 12.24 volts and 2 resisters two mounted to a heat sink and other in the water. Three were three needed to get the wattage the same as what the pump would be put into the water. I am basing my numbers on the authors numbers which i believe is 15 watts of heat for 24watts of electricity. The resister is adding 37.4 watts of heat to the loop and with the pump set to minimum speed its adding 5.1watts. So there is a total of 42.5 watts being added to the loop .For the second test i didn't run any power through the resisters i ran the pump at a higher wattage instead, of 68watts of electricity. This should mean it would make 42.5 watts of heat
.
i then ran these two test systems for an hour and measured the water temp

To my surprise after taking ambient air temp into account the temperatures were closes enough to say they were the same of 30 degrees C over ambient..

So I’ll have to take back what I said a few posts ago .. and say even thought the high inefficiency of a pump sounds crazy it seems to be correct

[bobo5195]

off the top of my head and really badly remembered and probably wrong:

A really bad pump as a peak eff of about 9%
A normal cheaply made pump eff of 30%
a well made cheap pump with 2D rotor is 40%
To get over 60% you need a 3D CNCed machine piece of kit.

My old lecturer used to have a turbo compressor in his office ( 1/4 million plus RPM for pumping air) i think he said that it was 96% eff but it looked like it cost at least $100 to machine and air is alot easier to move than water.

It is easy to make a cheap pump that performs good enough with injection moulding. The complex shaped air turbos are hard as nails to make.