Not so Good News for brandon and the Iwaki WMD30RZT:(

Well for all of those who are interested in using an Iwaki WMD30RZT I have some bad news. this thing dumps heat like there is no tommorow. My calculations show the heat dump to be a staggering 46.446watts. the test was run over a 1 hour period. well 1 hour and 2 minutes to be exact. the water temp went from 68.4F to 84.6F.

another tidbit of info here. after an hour or so the pumps case temp reads an astounding 82.78C!

to compare the WMD20RLT i have dumps 14.45watts based on the same test here.

through all of this there is SOME good knews. based on the pressure testing which showed with the RBX and maze4 (aluminum topped one) my flowrate was roughly 4gpm at 18.567ft of head!

here are some complimentary pictures. Brandon i was gonna get a video with some sound of this wierd sound the pump was making at first but it stopped before i could get the video going. its been running for over an hour since then without issue.


EDIT: I let it run for another hour for fun to see what would happen and the water temp is at 100F and the calculatiosn show around 40 watts of heat dump. im guessing its lower now due to the high heat differnetial. based on newtons law of heating, the higher the dT the faster heat will move. so the warmer the water gets the more heat it will loose through the insulation and the worse the reading.

anyway its time to stop running the pump as i dont want to boil water here.

Damnit MaxxxRacer this wasnt good news to wake up to. Oh well it only cost me 80 bucks.

-Brandon

ya its not so bad. i will post the pics now and the pump case temp.

Still bad at 50hz,MD version and a Monstercore? :confused:

50hz is worse. the heat output is almost the same with noticeable performance difference. but i still think this pump might be usable with a monstercore. I know cathar used one with dual 302's (i think they were 302's. they might have been 192's) and this temps were pretty good..

Yeah, my MD30RZ (not WMD) at 50Hz was dumping around 49-51W or so into the loop. Quite a staggering amount.

I was using twin Camry cores (effectively a Thermochill 120.2 each core) and they were still unable to handle the heat dump with respect to the CPU being warmer due to warmer temps because of the pump's heat.

Avoid the MD30's. The MD20R(Z) is the strongest pump from the Iwaki range that most anyone would want to use.

Quote:

---

Originally Posted by Cathar

Yeah, my MD30RZ (not WMD) at 50Hz was dumping around 49-51W or so into the loop. Quite a staggering amount.

I was using twin Camry cores (effectively a Thermochill 120.2 each core) and they were still unable to handle the heat dump with respect to the CPU being warmer due to warmer temps because of the pump's heat.

Avoid the MD30's. The MD20R(Z) is the strongest pump from the Iwaki range that most anyone would want to use.

---

:doh: Now I have to get me a MD20RZ and sell my MD30RZ

Its a pitty as the pump is amazingly powerful.

Hmm I guess its time to reinvest in either a Panworld, or the smaller Iwaki.

or start selling tea using the 30RZT to boil the water =p

Quote:

---

Originally Posted by Special_K

or start selling tea using the 30RZT to boil the water =p

---

Mmmmm good idea. NOT! Anyone want to buy a 115 volt portable water heater/pump.

lol...hmmm...watercool the watercooling pump!

Thats quite a hefty heat dump ..

BUT if you have a big rad you should easily be able to cope with a 50watt increase in load.

Can you give any approx numbers on CPU temp drop vs water temp increase max...
Would make for an interesting comparison.

The higher flowrate with a inpingement block might cancel out the water tmep increase.

Is this pump also noisy?

Quote:

---

Originally Posted by MaxxxRacer

Its a pitty as the pump is amazingly powerful.

---

Hmmm....might there be a link between pumping power and heat dump? :stick:

Quote:

---

Originally Posted by Holst

Thats quite a hefty heat dump ..

BUT if you have a big rad you should easily be able to cope with a 50watt increase in load.

Can you give any approx numbers on CPU temp drop vs water temp increase max...
Would make for an interesting comparison.

The higher flowrate with a inpingement block might cancel out the water tmep increase.

Is this pump also noisy?

---

That is like another GPU on top of typical loop. Maybe if you have SLI to cool, additional 50watt vs restriction may cancel out on monster core.... but than, I don't really know if you want to put all those components in together which is not exactly cheap to get the performance which probably can be obtained cheaper using multi loop, etc. Maybe this will be good choice for peltier cooling crowd though. Additional 50W can't be too much on that scenario... like what... 5-10% of total heat output? :P

My point being that I can cool a pelt rig (500watts+) using only my dual heatercore with powerfull fans.

Im pretty sure that you guys with 120X3 rads can probably manage an extra 50watts with only 1*c increase in water temp, maybe less.

You may get 1*c lower temps due to the increased flow, maybe..

You might even be able to reduce this further by heatsinking and fanning the pump itself.

can you try putting a little airflow over the pump?

You guys are CRAZY I might try to hook it up just for giggles once I get it back from Maxxx. maybe if took both my caprice rads in series. with my maze4 and 6002 AHAHHAHAHHHA **evil**laugh.

B!

Quote:

---

Originally Posted by Holst

Is this pump also noisy?

---

Once isolated it is reletivly quiet (say mcp600 levels IIRC). Somwhat of a rumbling - not annoying at all IMO.

Quote:

---

Originally Posted by freecableguy

Hmmm....might there be a link between pumping power and heat dump?

---

More powerfull pumps have more powerfull motors.

But anyway, nice results Maxx, pretty much bang on what is written on the case of it (45W). Nice to see it confirmed :)

try watercooling your pump motor :stick:

:rolleyes:

P-H-A-S-E C-H-A-N-G-E

C

Quote:

---

Originally Posted by MaxxxRacer

the heat output is almost the same with noticeable performance difference

---

Any proof for this at all?


I know Iwaki gives the same 'Power Output'-values for both 50Hz and 60Hz, but for me it's still hard to imagine that simply raising 'Hz' increases power but not heat dump.

The more water the pump moves, the more heat goes into the water...

Moving water has kinetic energy, this is converted into heat through friction, this MUST go into the water.

So unless the pump manages to be more efficent at 60hz than at 50hz the power ratings must be a bit wrong (or the flowrates)

The accuracy of the numbers manufacturers supply are probably a bit dubious anyway.
Having found out how a fairly well known heatsink manufacturers test and validate fan specs and heatsink C/W ratings. (IE - very innacuratley indeed) it would not supprise me at all if other manufactuers specs are a bit wayward.

If you set it up in a exterior case in which the components are arranged so that the air flow from the monster core's 4 fans washes over the pump........

And perhaps add some heat sinks as someone else suggested to the pump it's self.....

Could tip the balance perhaps? Worth a shot before selling it I'd think....... As Holst points out, if it's possible to cool a 500W TEC'd rigs heat load with approx. 1C rise, then this should be workable as well. Fan noise may become another issue however in this approach although with the monster core and 4 fans you'd have a lot more surface area than a single dual fan rad so should be able to use a bit calmer fans........

I'd suggest you use a differant block than that Swifty 6002, the 6002 is geared toward lower flowing rigs. A TDX would be good, or perhaps the new Swifty Storm block. A overly restrictive block like the 6002 will compound the problem of the big pump by making it work even harder.

Best of luck with that beastie! Let us know how your testing in a actual rig goes. :toast:

well, if you dont want it I might.

You will have to use the right block to take advantage of the extra head/flow.

I think a swifty G4 is the way to go... but max will probably know more about this than me as ive not used any of these newer blocks.

Quote:

---

Originally Posted by Holst

The more water the pump moves, the more heat goes into the water...

Moving water has kinetic energy, this is converted into heat through friction, this MUST go into the water.

So unless the pump manages to be more efficent at 60hz than at 50hz the power ratings must be a bit wrong (or the flowrates)

The accuracy of the numbers manufacturers supply are probably a bit dubious anyway.
Having found out how a fairly well known heatsink manufacturers test and validate fan specs and heatsink C/W ratings. (IE - very innacuratley indeed) it would not supprise me at all if other manufactuers specs are a bit wayward.

---

If you seriously believe heat is created by friction of water inside a pump to a significant degree, I suggest some reading up on electric motors . While some friction related losses do occur, this is pump related. Have a quick read, you might find the losses and efficiency part useful.

Quote:

---

Originally Posted by phextwin

Once isolated it is reletivly quiet (say mcp600 levels IIRC). Somwhat of a rumbling - not annoying at all IMO.
More powerfull pumps have more powerfull motors.

But anyway, nice results Maxx, pretty much bang on what is written on the case of it (45W). Nice to see it confirmed :)

---

OMG, did anyone not understand my sarcasm in this statement? I am a nuclear engineer by trade and don't need this explained to me. :stick: :p:

ok thats alot of posts to respond to.

holst I can try hooking the thing up to my rig right now and giving you temp readings with the heatercore hooked up, but I doubt it will be of much use as my cpu reads wacko temps and I do not consider motherboard/videocard onboard sensors to be useful in any testing. but with that said i will still give it a a shot and see what happens.

freecable guy, the performance increase from the 30 going from the 20 isnt 3x. BUT the heat dump IS 3x. so the two are NOT linear and thus this is a poor choice of pump.

Brandon, your 6002 will NOT like this pump. after 2.5gpm the gains has by increasing flow on the 6002 are abou ziltch. as others have said this pump is really only even remotely useful with a G4.

I do agree that there is a point where there would be enough RAD power to cool the water, but IMO its just not worth it. to cool the water enough you would proll need about 2.5-3x the rad power that you would fora MD20 or other 12volt pumps. and that is JUST to cool the water. you would also need some way to vent the hot air away from the pumps casing, which would in turn heat up the room. so there is really no way you could have this pump in your case, even with a large custom made case.

about the 50hz and 60hz thing. both pumps are using the same input voltage and wahtnot. just the pumps speed is based NOT on volts but entirely on the hz. why, i have no idea as im not an EE. so at 60hz it pumps more than at 50hz but the only increase in heat from the 60hz one is a slightly higher power draw and a little more friction. according to what cathar calculated the lower heat ouput from the 50hz (its a small amount) does not make up for the loss in flow.

Quote:

---

freecable guy, the performance increase from the 30 going from the 20 isnt 3x. BUT the heat dump IS 3x. so the two are NOT linear and thus this is a poor choice of pump.

---

Ahh!!! Stop. Thanks. :D

Edit: You will actually find (discluding pump efficiency) that double the pump head requires 2.8x the power (cube-root)....

This should help...
http://www.engineersedge.com/fluid_flow/pump_laws.htm

Quote:

---

Originally Posted by MaxxxRacer

about the 50hz and 60hz thing. both pumps are using the same input voltage and wahtnot. just the pumps speed is based NOT on volts but entirely on the hz. why, i have no idea as im not an EE. so at 60hz it pumps more than at 50hz but the only increase in heat from the 60hz one is a slightly higher power draw and a little more friction. according to what cathar calculated the lower heat ouput from the 50hz (its a small amount) does not make up for the loss in flow.

---

Why the pump speed is entirely based on speed instead of voltage?
I'd say that's because internal electrics convert different amounts of voltage in the same amount of power.

Though why 60Hz and 50Hz dump the same amount of wattage confuses me.
While I'm probably completely wrong, if I'd have to make a simple guess I'd say that both input and output would increase from 50Hz to 60Hz by 60/50=1.2.
Iwaki's specs confirm my error. While some of the input voltages' ratio are very close to 1.2 (like the MD-15 and MD-20 series), too many aren't.

Quote:

---

Why the pump speed is entirely based on speed instead of voltage?
I'd say that's because internal electrics convert different amounts of voltage in the same amount of power.

---

Ahhh! (again) Speed of the pump is based on frequency. Pumping power is based on voltage (actually, this is a measurement of apparent power). Seriously though, read the URL that I posted...it covers the *basics*.

-FCG

Quote:

---

Originally Posted by freecableguy

Ahhh! (again) Speed of the pump is based on frequency. Pumping power is based on voltage (actually, this is a measurement of apparent power). Seriously though, read the URL that I posted...it covers the *basics*.

-FCG

---

I know that. F=1/T.


But you're right, I said it wrong.
I mean I think the internal electronics prevent the pump from running faster at higher voltages.



Edit: Though I will read that link when I have time. I think it does contain a lot of information I do not know yet.

Yes, frequency is the inverse of the period...this is just a mathematical definition.

Let’s begin the discussion by describing what causes an AC induction motor to run at a particular speed. Every AC induction motor has poles, just like a magnet. Unlike a simple magnet, these poles are formed by bundles of magnet wire (called windings) wound together in slots of the stator core. In most cases, you can look inside the motor and count the number of poles in the winding: they are distinct bundles of wire, evenly spaced around the stator core.

The number of poles, combined with the alternating current line frequency (HZ), are all that determine the no-load revolutions per minute (RPM) of the motor. So all four-pole motors will run at the same speed under no-load conditions, all six-pole motors will run at the same speed, and so on.

The mathematical formula to remember in helping make these calculation is the number of cycles (HZ) times 60 (for seconds in a minute) times two (for the positive and negative pulses in the cycle) divided by the number of poles.

f*120/n = speed of the AC motor (the pump)

So, we can see that a 50Hz motor will spin at a lower speed than a 60hz motor, once again, assuming no load.

I'm gonna try putting heatsinks all over my pump and I mean ALL OVER. :D

Sorry if I'm making a stupid comment, but that doesn't explain my earlier reaction in which I said it seems impossible to dump the same amount of heat at different voltages, right?

One critical relationship between line frequency and magnetic devices is efficiency. The physics of electric circuits tells us that AC magnetic devices increase in efficiency as line frequency increases. Sounds simple, doesn’t it? Just build your power systems and devices to the highest frequency possible. Not so fast. Another physical characteristic to keep in mind is that current flow in a conductor tends to be closer to the surface of the conductor as frequency increases. So as the frequency goes up, solid conductors begin to resemble hollow pipes as the electrons making up the current flow migrate to the outer surfaces of the conductors.

At these higher frequencies, the energy of the electrons has a tendency to actually leave the surface of the conductor. A common example of this principle in action is radio transmission. As the frequencies get higher, all of the energy can be made to leave the conductor in a form of energy called radio waves. This also helps explain why overhead power lines tend to interfere with radio reception (the annoying 60-cycle hum). What you are hearing is energy loss from the power lines becoming a radio wave that is intercepted by the radio receiver.

Consequently, the designers of electrical devices must strike a balance: the desire to use higher frequencies to improve the efficiency of converting electrical power to mechanical work and the need for lower frequencies to keep power from escaping the conductor as radio waves.

So the evolution of 50 and 60 hertz systems developed as a result of this need for balance, with additional influences coming from politics and geographic considerations. North America and other regions struck the balance at 60 hertz, while Europe settled on 50 hertz. Who’s right?

You probably realize by now that frequency is a parameter for motor selection and application. What you may not realize is that a motor designed to operate at the lower 50 hertz frequency may operate in a satisfactory manner at 60 hertz, at least from an efficiency and heat loss standpoint. This is because a 50 hertz motor contains added material to make up for the modestly decreased efficiency found at the lower line frequency. But keep in mind that when going from 50 to 60 hertz, the motor’s speed will increase proportionately to the increase in frequency (60/50=6/5=120 percent).

now there is a man who knows his motors!

thanks for the lesson FCG.

thanks for the link too. helpful info.

Quote:

---

One critical relationship between line frequency and magnetic devices is efficiency. The physics of electric circuits tells us that AC magnetic devices increase in efficiency as line frequency increases. Sounds simple, doesn’t it? Just build your power systems and devices to the highest frequency possible. Not so fast. Another physical characteristic to keep in mind is that current flow in a conductor tends to be closer to the surface of the conductor as frequency increases. So as the frequency goes up, solid conductors begin to resemble hollow pipes as the electrons making up the current flow migrate to the outer surfaces of the conductors.

---

Its called the skin effect. Its pretty much irrelevent though for power circuits. You have to be in the khz range before its even noticable.

Quote:

---

At these higher frequencies, the energy of the electrons has a tendency to actually leave the surface of the conductor. A common example of this principle in action is radio transmission. As the frequencies get higher, all of the energy can be made to leave the conductor in a form of energy called radio waves. This also helps explain why overhead power lines tend to interfere with radio reception (the annoying 60-cycle hum). What you are hearing is energy loss from the power lines becoming a radio wave that is intercepted by the radio receiver.

---

The 60Hz hum is NOT radio. Its simple induction. A 60Hz radio wave would have a wavelength of about 5000km. Thats a substantial fraction of the Earth's width. You're not going to generate meaningful radio energy at anything within a few orders of magnitude of a 60Hz circuit. Thats why most stuff is at Mega and GigaHertz (though AM is in the upper KHz range).

Quote:

---

Consequently, the designers of electrical devices must strike a balance: the desire to use higher frequencies to improve the efficiency of converting electrical power to mechanical work and the need for lower frequencies to keep power from escaping the conductor as radio waves.

So the evolution of 50 and 60 hertz systems developed as a result of this need for balance, with additional influences coming from politics and geographic considerations. North America and other regions struck the balance at 60 hertz, while Europe settled on 50 hertz. Who’s right?

---

Thats not right. 60Hz was chosen by tesla because it was the lowest frequency that did not flicker visably in an incandsant bulb (think 60 fps in games). No one knows where 50Hz came from, but probably much the same reasoning as 60Hz. The frequency does not effect the effciency of the power distribution system (well unless you're at extremes), although low frequencies are a little easier to work with since induction is less of a problem at low frequencies. If they'd put it too high then you'd lowpass out your AC signal due to LR filter effects in the lines. OTOH go too low and transformers become more expensive. None of this effects effciency though.

More info: http://en.wikipedia.org/wiki/Alternating_current

Quote:

---

None of this effects effciency though.

---

Try again.

Well out of curiosity i stuck the pump on my test setup.

the radiator outlet temp with the 50Z was 78.5F and with the MD30 it was 83.2F with all other things being equal.

I should also note that the MD30 raised the ambient temp some. not sure on the exact numbers as i didnt document it, but it dumps a noticeable amount of heat into the air through the casing.

the radiator i used was the PA160 with a TT thunderblade fan at 12volts. I am gonna retest the md30 with both a BIX2 and 2-199 heatercore with denkis at 12volts on both to see if i cant get the temps back down again. I will wait a while though as i need to get the room temp back down a C or so.

We haven't even gotten to talking about voltage, current, armature/field current and resistance, slippage and torque...of course all of this leads to what you really care about, work, power, energy and heat output (overall efficiency).

another minor update here. it seems the water temp is varying between the previosuly said 83.2 and 84.3. it slowly swings back and forth between them. might have somethign to do with the cpu, but im not sure. i checked ambient between the two temps and it was within .3F so there is no issue there.


wel right now the ambient temp is 72.1F and the water temp is at 75.6F right now. here is how i did it.

i need help you say? ya ur probably right ;)
http://www.xtremesystems.org/forums/...id=34228&stc=1

lol... you know you can avoid all the radiators and fans and stuff just by hooking up your loop to your faucet and putting the output hose into the sink... a lot of apartment/condo's have WATER BILL built into the rent... check it out ;)
Tap water runs a constant 15C in most areas!!

C

insulation then comes to mind....

no insulation neded FGC. my room temp is around 20-22C.

charlie i thought of that. the head pressure on the fauciet is amazing too. you would get around 4-5GPM with it. but the problem arrises of draping large 5/8ID tubing from the faucet to my room and my mom killing me. or in your case your wife killing you.

insulation need, i mean, if you use 15C faucet water...

i know what u meant. but the water temp would be a feeble 5-7C below ambient. with the relatively how humidity in los angeles it wouldnt get condensation.

mount a series of small aluminum heatsinks (like the cheap ones from radio shack) to the pump housing with thermal epoxy after etching the finish off of those sections of the pump housing -- then, add a low airflow (close to silent) 92mm or 120mm fan directly behind the pump. The temp difference might surprise you. ;)

it has worked with other pumps of a like design anyway...

ya its something to consider rod, but even if you do that you need to remove the heat out of the case...

we've been doing that for a long time, dont think its going to be a problem max ;)

A length of copper corrugated pipe on the outlet of the pump may make up for some of the increased heat.

a little bit of copper pipe wont do much imo..

i think i am officially done with this pump for now.time to put the 50Z back in. not nearly as much perfomrance, but i can deal.. way less noise and 1/5 the heat dump.

WEll. we all learned something from this. HEHHEE, I guess I'll be ordering a 20RLT next week sometime.


I'd like to Thank MaxxxRacer for all his work in getting this done and agreeing to do it in the first place.

-Brandon

Quote:

---

Originally Posted by MaxxxRacer

ya its something to consider rod, but even if you do that you need to remove the heat out of the case...

---

muahahaha. have you forgotten that I am a mod-everything psychopath??

what is the diameter of the pump housing? How much does it weigh? That biatch can most certainly be weaponized. The members of this forums have figured out how to dump massive amounts of heat when we only have a few scant square millimeters of contact area to work with -- surely, we shall not run into a pump with awesome performance and be stifled by a little heat dumping problem when we have a lot of housing area to work upon...

Initial ideas:
1. slot the pump housing and add a small grill so that the heat will have a less restricted route out of the pump. If EMI is a problem, it can have a raised copper cover and copper mesh grills.

2. ventilation fans. Back in the days when I used nothing but air cooling, I managed to add fans to almost everything. My cable modem has a hard-wired 5v cooling fan and heatsink (wired directly into the 5v supply on the modem)and it is the only one of its model that can hack a 5 meg connection without burning up.

3. heatsinks added to the housing - every little bit helps. I did this on a pump that I used to have that had a bad heat problem. The difference was amazing.

send me that bastage.

Here's what I have always wanted to do. Take, say a 3.5" round piece of aluminum barstock and use a rotor to drill out an inner circumference that is exactly the size of the motor housing. Then (somehow) use a CNC machine to work the surface of the heatsink so that there are ridges for optimal ariflow. Then slater with AS Epoxy and slide onto the motor housing. I bet that would cool the son 'um ':banana::banana::banana::banana::banana: down.

Quote:

---

Originally Posted by freecableguy

Here's what I have always wanted to do. Take, say a 3.5" round piece of aluminum barstock and use a rotor to drill out an inner circumference that is exactly the size of the motor housing. Then (somehow) use a CNC machine to work the surface of the heatsink so that there are ridges for optimal ariflow. Then slater with AS Epoxy and slide onto the motor housing. I bet that would cool the son 'um ':banana::banana::banana::banana::banana: down.

---

Im gonna try to do a similar thing to an old heatsink I have laying around.

It's gonna look a little like this:

may god have mercy on the iwaki.

Quote:

---

Originally Posted by Pessimisten

It's gonna look a little like this:

---

You're going to paint it that ugly green color too? It's going to be quite the punk-pump. ;)

its already an ugly arse green. its ugly but it makes up for it in spades with character.

Quote:

---

Originally Posted by MaxxxRacer

its already an ugly arse green. its ugly but it makes up for it in spades with character.

---

hmmm...2 h311 with character...if I touch it, it will be stealth black before it is done. heh.

if ya wanna argue, I will be in CS under the name weapon.

:p:

lol.. ive been playing BF2 recently. I own in the demo.. 3 to 1 minimum kill ratio..

rod, you should ask brandon if he is interested. if so I could ship the pump to you for the modfest.

btw if u want to paint the plastic ull need to do a minor acid etching or sanding.

Heat sinks on motors work pretty well. Did one a few years ago to cool a motor in a non-computer app. Fortunately I was able to find a finned aluminum oil filter housing that was almost the correct i.d. and length. Cut off the bottom and top then took the remainder to an engine rebuider and had them hone the id out til it was a snug fit over the motor's housing.

It left about 1/2" at each end of the housing uncovered. Before, with a fan blowing on the motor, those areas were still noticeably warm. With the cover in place and the same fan those areas weren't warm at all. Temp must have dropped at least 30 - 35 F.

As an alternative - think I might try making one from soft, thin copper as it's easie to bend than aluminum. Bend some zig-zags like the filler in corrugated cardboard, snip the ends off a few of the corrugations to leave some tabs so it can be tie wrapped to the motor. Probably work better with fewer zig-zags and more flat sections to pick up the motor's heat.

bfx

All I can say is you guys ARE extreme! slotted covers and heatsinks. However that doesnt correct the heat dump factor, which is why I'll be hmm "promoting" on ebay, as so as I get her back from Maxxx.

Quote:

---

Originally Posted by Weapon

muahahaha. have you forgotten that I am a mod-everything psychopath??

what is the diameter of the pump housing? How much does it weigh? That biatch can most certainly be weaponized. The members of this forums have figured out how to dump massive amounts of heat when we only have a few scant square millimeters of contact area to work with -- surely, we shall not run into a pump with awesome performance and be stifled by a little heat dumping problem when we have a lot of housing area to work upon...

Initial ideas:
1. slot the pump housing and add a small grill so that the heat will have a less restricted route out of the pump. If EMI is a problem, it can have a raised copper cover and copper mesh grills.

2. ventilation fans. Back in the days when I used nothing but air cooling, I managed to add fans to almost everything. My cable modem has a hard-wired 5v cooling fan and heatsink (wired directly into the 5v supply on the modem)and it is the only one of its model that can hack a 5 meg connection without burning up.

3. heatsinks added to the housing - every little bit helps. I did this on a pump that I used to have that had a bad heat problem. The difference was amazing.

send me that bastage.

---

Weapon you should get an award for being most creative and Extreme!
LOL

-Brandon

brandon it will go out tommorow. I will take some pictures for your ebay auction tonight.

weapon, if you can find a way to make this pump useable, I will officially love you. I must have one of these, and i WILL find a way to make it work, even if it means epoxying heatsinks all over it (and, of course, repainting it). Also, for my loop, a meager extra 30 watts or whatever won't make that big of a difference... 100+ watts from CPU, probably something like 400+ from GPU/TEC, ??? from bridge rectifiers... whats a measly extra 30? :)
I'm starting to feel as though dual 2-302s won't be enough. *shakes his head*

bloody, if you are running TEC's, this pump will be hardely noticed temps wise. especially if u got 400watts coming off the tec from the gpu.

Brandon, here are the pictures of the pump. they are high quality to say the least. I blurredo ut the serial number so no one tries to rip you off on ebay and use that serial for RMA purposes.

http://images5.theimagehosting.com/MD30_Front.jpg

http://images5.theimagehosting.com/MD30_ISO.jpg

http://images5.theimagehosting.com/MD30_Side.jpg

http://images5.theimagehosting.com/MD30_Tag.jpg

see you mentioning epoxying...
slight offtopic:
if i was gonna use epoxy + thermal grease on a chip for alu heatsinks, what would be the best ratio to use them in?
epoxy:grease = 1:1 or more grease than epoxy???
(asking for mem chips on vga card and large alu heatsinks)

thanx!

1 parts epoxy to 0 parts thermal interface material.

This assumes you are applying the heatsink pretty much permanently and are applying it correctly (10+ lb's force during set time, and both surfaces must touch well..).

Quote:

---

Originally Posted by STEvil

1 parts epoxy to 0 parts thermal interface material.

---

didn't quite get what u mean by this. :(

use no TIM and just use epoxy is what he is saying. dont particulary agree with this at all though.

50/50 AS5 & AS epoxy seems to hold well & i removed 1 easily.

assuming you isntall it correctly you shouldnt need to use a TIM.

TIM is for when the contact is bad really.. and some TIM's dont spread out well or work correctly when mixed with Epoxy and will instead become a barrier to heat.

Straight epoxy is the best if you can make the surfaces contact each other well enough. If not then you wont be gaining much from using the heatsinks.

omfg....i juz purchase a resun md40 ( cloned iwaki md30RX ) ....coz it's really cheap at my place....RX hv 2.9/4.1 only (m)max head,and RZ is about 11 (m) max head....

im affraid tat RX series oso hv this kind of problem ...

Maxxx this pump looks promising ! :D

http://www.drsfostersmith.com/produc...188x006&ctt=64

could you run some tests with it ? LOL I was thinking maybe 12 in parallel ?

12 in parallel.. intersting major... if i thought u were serious for atleast one second id need to bring out the psychiatric ward.. but im gonna give u the benefit of the doubt and say ur not.

but if you are, send 12 my way and ill test em. just make sure to include a packing slip to send em back to you like brandon did (thankyou btw brandon [well brandons wife], it made my life alot easier as im totaly broke)

I bet 12 Mag's would do better. :p:

I bet 12 MD100RZT;s would do better... mauahahahh :lol:

Does anybody have a picture how the pump(MD20/MD30) looks when the back casing of the pump with the motor is removed?. I can't seem to get the back casing off, do I have to remove the Drive Magnet unit to get it off?

No pics but try this ......

http://www.iwakiwalchem.com/iwaki/Li...W)MDmanual.pdf


Page 16 has an exploded view.

bfx

pessimisten if you want a pic of the WMD I can help you out, but no joy on the MD.

Quote:

---

Originally Posted by MaxxxRacer

pessimisten if you want a pic of the WMD I can help you out, but no joy on the MD.

---

That would be great if you could do that for me :)

Quote:

---

Originally Posted by bfx

No pics but try this ......
http://www.iwakiwalchem.com/iwaki/L...(W)MDmanual.pdf

Page 16 has an exploded view.

bfx

---

Yea I already have looked at those but the engine isn't shown on any picture, maybe industry secret?

well you cant see the motor from the pics I will be able to show you. just a fyi.

Were you after pics of the coil? The armeture?

I can get MD ones if you wish. They probably look the same as the WMD ones though.

Quote:

---

Originally Posted by phextwin

Were you after pics of the coil? The armeture?

I can get MD ones if you wish. They probably look the same as the WMD ones though.

---

Yea, I can only get i off like 1 cm and all I see is a copper coil maybe you need to remove the Drive magnet unit to get the casing off completely? I need a picture of the back without the casing to see how and where the heat is lead out to the casing. Need that info for planning some coolingmods to it :D

Are these what you are looking for?

Am unsure how to take the coil out at this stage.

wow.. nice pics. that thing is crazy.. god only knows how many pole motor it has.

Is it possible to remove the armeture with the WMD?

With the MD there are 2 bolts holding it on that can be acessed once you remove the impellor assembly. It's fairly straigt forward. The coil doesnt seem to have any obvious way to remove it, which is probably a good thing, broken wires etc.

Thanks for the pics.

According to the pictures the copper coils have direct contact with the casing and that's why it gets so damn hot if I understand the pics right. So if you could cool down the casing a little, less heat would be transmitted to the water or isn't that simple?

pessimisten, thankfully its exactly that simple. u just need to channel the airflow properly so that it doesnt warm up ur case

Do a volenti and zip tie some heatsinks to the side.

http://forums.overclockers.com.au/sh...enti+heatsinks

Here for the pic if you havn't or CBF signing up to OCAU.

Quote:

---

Originally Posted by phextwin

Do a volenti and zip tie some heatsinks to the side.

http://forums.overclockers.com.au/sh...enti+heatsinks

Here for the pic if you havn't or CBF signing up to OCAU.

---

I already have cut me some heatsinks for it and they are alot bigger than his and alot more :) If that don't work watercooling is the next stop :D

watercool the water pump.. now thats a new one!

Just make sure you use a seperate loop to watercool the pump, otherwise all the work will be for naught.

Haha only at XS would someone watercool the pump used for watercooling the cpu.

Quote:

---

Originally Posted by phextwin

Just make sure you use a seperate loop to watercool the pump, otherwise all the work will be for naught.

Haha only at XS would someone watercool the pump used for watercooling the cpu.

---

Why would cooling the pump casing (in some inventive fashion) with the same loop thats cooling your CPU be for nothing if your radiator can handle the extra load? Would be a pump that dumped 100% of its load into the loop.

which is worse than where we started, with only dumping a smaller % into the water.
The coolest thing to do would be to braze long copper nails to every square mm of that baby and put a massive fan over it :)

Quote:

---

Originally Posted by Bloody_Sorcerer

which is worse than where we started, with only dumping a smaller % into the water.
The coolest thing to do would be to braze long copper nails to every square mm of that baby and put a massive fan over it :)

---

Lol a large spiked mace.

I dont see how its worse off if you have a giant radiator that can easily handle the extra 50watts... Wouldnt keeping the pump at a close to room temp help its efficiency at all?

Holly cow!!! You guys are real pros. I mean, water cooling is something but some of you got some serious plumbing action going on in your system.
And I haven't been able to get a standart watercooling system running yet. Damn!

Just think guys its sitting in my closet collecting dust. :D

-Brandon

brandon.. either get a monstorecore and mod some cooing fins on that thing or sell it and get a MD20... no point for it to be doing nothing!

Yeah I know.. I'm just toooo busy today was my first real day off in 11 days. I'll get it all squared once I get thru the next month or so.

-B!