Varying pump speed possible?
 

Hi all,

I've just put together my very first watercooling loop, consisting of a PA120.3, Iwaki MD20R and Storm and Maze4 blocks, and I'm extremely happy with the performance; under load my CPU temps, as reported by the onboard sensor, have dropped 20°c compared with the Thermalright XP120 heatsink I'd been using previously. :D

The only thing I'm not so happy with is the noise the pump makes. I'd picked my components thinking they'd be fairly quiet, and I was right with the fan/rad combination, however the Iwaki makes a noise comparable to a fairly powerful 80mm fan at full bore!

I was wondering what my options were with regards to varying the pump speed and hopefully quietening the pump at times when it's not essential that it's at full whack. With fans you have the ability to adjust the voltage and reduce the noise at the expense of reduced airflow - does a similar method exist for pumps?

Cheers,

GP

Re: Varying pump speed possible?
 

do you have the pump on spunge/other soft substance?

Re: Varying pump speed possible?
 

Well essentially yes; I enlarged the mounting holes slightly and inserted rubber grommets through which I have some M8 bolts. The pump and rad are mounted in a plywood radbox of sorts - its more of a cradle right now to be honest as it still lacks side panels :p

I don't think vibration is the problem because even holding the pump in my hand it makes just as much noise and at any rate the pump is so tightly bolted down that it cannot move at all. The radbox itself has felt feet applied so any vibration through the box is not being transmitted to the floor tiles.

This is a second hand pump so I guess it's possible that the noise is due to wear in the impellor bearings. Are these pumps normally quiet?

Re: Varying pump speed possible?
 

Hard to define quiet, but they are louder than an Eheim 1048, but quiteter than the fans that came with my Lian Li case

Re: Varying pump speed possible?
 

Hmmm ok. If I had to put a number on it I'd guess at my pump being around 45 dba, which may not seem a lot but the pc is in a bedroom so it makes for an uncomfortable background noise when you're trying to sleep :(

Back to my original query; is slowing the pump down possible?

Re: Varying pump speed possible?
 

http://www.variac.com/staco_Variable...former_Map.htm
http://www.action-electronics.com/variac.htm
A VERY little info plus links to a few products: http://www.elect-spec.com/variac_tutorial.htm

I think this sort of thing is the most elegant solution. There are cheaper ways but I'm not the expert so I'll not put foot in mouth.

Re: Varying pump speed possible?
 

Mikoto, thanks - this is interesting info, will read and digest :)

Re: Varying pump speed possible?
 

I use a fan rheostat panel to dial my pumps down a bit.

Re: Varying pump speed possible?
 

I'm currently running a couple of hydor L30s but I need something more powerfull.

I'm choosing between a Iwaki MD20R and two DDC-Ultras in series. With the barb mounted on the top the DDCs should give almost as good performance as the MD20R. The DDCs shouldn't be as loud as the MD20R and I could control the DDCs from windows with my T-balancer XL + miniNG. This way the pumps will always be as silent as possible. So at night when the cpu idle the computer would be as silent as possible.

I havn't found any Iwaki resellers in Sweden/europe yet but I've sent lots of mails. As soon as I know how much a MD20R would cost me I'll decide which pump to buy.

Re: Varying pump speed possible?
 

MX probably has the best idea. That MiniNG can handle 20W analogue or 50W PWM. I don't know how much wattage your pump draws, but you can look at the speed controller I made for my modified DDC.

http://forums.procooling.com/vbb/showthread.php?t=11909

Re: Varying pump speed possible?
 

Nice speed controller Hansfragger!

To bad that the MD20R runs on AC, it just makes everything much more expensive! A variable AC transformers are both big and expensive, and they might even emit a more annyoing sound than your pump!

I'm not an electrician but maybe you could use something less powerfull like a dimmer? Those variable AC transformers can handle loads of current, your pump draws maybe 0.5-1A? The easiest and cheapest solution would be to just pack lots of sound isolation around it. (Or do like me and buy some earplugs! :))

Btw. the miniNG has two channels and in analogue mode it can handle 20W on one channel or 25W if you combine both channels. In PWM mode you can use 50W per channel or 100W combined.

Re: Varying pump speed possible?
 

mx, explain why you need something more powerful?
According to my approximations with the PQ and CW charts of most pumps, I see the Iwaki 20 (I use RT20?) as the best pump there is.
Its better for a single Storm and MCR220QP than two Laing D4's or two MCP655's as the heat dump vs pressure is better.

But perhaps you have some special kit?
Or my approximations are out?

Re: Varying pump speed possible?
 

It's not just that I want a more powerfull pump, I want a more reliable one because I don't trust my L30s! I belive that a pump with more power than my current L30s should give me better temps as my rad is kinda big. As I see it there aren't too many pump setups that fit these criterias. A few Iwakis, 2x top barbed DDC-Plus, two 50Zs or maybe two D5s.

I'm currently running this rad and a storm G4. I use six of those fans. They're NOT quiet but I can stand them on low speed. Maybe I'll get new fans some day...
I have plans for more blocks and some peltiercooling.
http://72.29.82.190/~susanne/annat/jonas/cu_rad_01s.jpg

Two L30s need quite a bit of room, and one Iwaki, two 50Zs or two D5s wouldn't be much smaller. Two DDCs wouldn't require as much space. Controlling two DDCs with a miniNG manually from win or automaticly depending on the cpu temp would be sweet! And probably very silent too. A couple of DDC-Ultras in series with top mounted barbs (article showing a regular DDC not the new more powerfull one) should be quite a bit more powerfull than two standard DDCs. I havn't seen any data on this setup so I can't know for sure, but that just makes it more fun to try! :)

It's very possible that Iwaki 20 or 30 would dump less heat and be more powerfull than two DDC-Ultras with top mounted barbs, but hopefully not by too much. And with the miniNG, space and noise arguments two DDCs feels more tempting!

Re: Varying pump speed possible?
 

what type rad is that?

Re: Varying pump speed possible?
 

It's a copper rad from a car. I don't know which car model.

Re: Varying pump speed possible?
 

looks interesting to mount on side , passive.

Re: Varying pump speed possible?
 

Just ordered a couple of DDC-Ultras and a miniNG. Hopefully I'll have them next week.

Re: Varying pump speed possible?
 

See SystemCooling for the details of four square solder pads on one edge of the pump's circuit board. One is for pump voltage speed control. Or, you can run your own voltage control via the pumps regular two power wires.

Bottom post and into the next page ...
http://www.systemcooling.com/forums/...pic=3099&st=30

Even though the pump is rated to run between 6~14V, I would only run it as low as 7.5V in my system. Many variables like coolant mixture and flow resistance will determine your actual low speed of operation.

Hope this helps!

Re: Varying pump speed possible?
 

Thanks stev!


Currently I have the first two solder pads connected to my miniNG.

1) Gnd = Ground
2) Variable motor voltage (Vm) (6 to 12 V, approx 1.0 A max.)

The third pad is connected to my T-balancers analoug sensor hubs flowmeter ports.

3) TACH

I don't really understand how to use the fourth solder pad. :doh:
Would there be any advantages using it?

4) Vp = +12 V (runs control circuit, approx., 50 mA max.)

Re: Varying pump speed possible?
 

miniNG + 2 DDC combo - that would be slick. mx, are the DDC pumps you got the 18W or the 10W ones? Would be cool to hook both up to a single miniNG ... but it looks like the miniNG can only handle a single 18W DDC in analogue mode (max 25 watts both channels, max 20 watts single channel) but could probably handle both 18W DDCs in PWM mode (max 100 watt both channels, max 50 watts single channel). You think it is safe to run two 18W DDCs on PWM .. ?

I think the fourth solder pad interfaces with macintosh G5 cooling hardware, software .. thats what I gathered from the 2 systemcooling articles and forum posts over there ..

Re: Varying pump speed possible?
 

I don't think you should run pumps on PWM. No idéa why but someone told me not to on some forum. :)

I didn't even know there where an 18W version? I got mine from alphacool and they don't mention how many watts the pump use. So I read over at vcore.dk that the Ultra version uses 10W. My DDCs are the 1Plus version /w plexi top called DDC-Ultra so they are probably the 18W ones? I thought thought they used a bit too much power but I assumed it was because I ran them on 13.5V at the moment.

The miniNGs got over 100W peak in analogue mode, the only reason you can't use all that power 24/7 is because of the cooling. Hence if you cool it better you can use more power. It's got some kind of safety function so that if you overload it in analogue mode it switches to PWM mode. I've run my two DDC-Ultras from the miniNG in analogue mode at 12V for a few hours w/o any problems. I had a fan blowing directly at it. My rig isn't very organized yet so I run them of my external PSU untill my chassi is done. I guess I should look for some better cooling for the miniNG. :)

Re: Varying pump speed possible?
 

actually it's the other way, pwm is great to control motors.
pumps included. simple and an efficient way, and keeps them running even with a low setting.

Re: Varying pump speed possible?
 

Really?
I'll have to try that when I've got my stationary computer up and running. Upgrading from maze4-1 to a Silverprop Cyclone FusionSLT (Low Profile, Thermoelectric) /w Swiftech MC14 RAMsinks and moving it to a modded P180 SPCR.

Re: Varying pump speed possible?
 

That depends upon the control circuitry though. You have to be careful about the frequency of the PWM. It's better if the motor controller has speed input, because it can vary the time and/or current sent through the coils.

Re: Varying pump speed possible?
 

the 18W version is the DDC-1+ 18W .. I guess DD carries it .. that's what I gathered anyway from a thread over at bit-tech ..

.. but I don't think it will work with PWM .. would have to try it out and see ..

would be cool to run two of the 18W DDCs off the miniNG in analogue mode .. but that would exceed the 25W 2 channel max .. but two of the 10W pumps should work in analogue mode though .. still, a single 18W or 10W pump controlled from the miniNG is pretty cool ..

I wonder if you can undervolt the D5 with the miniNG in analogue mode .. instead of the manual rheo .. then you could control via t-ban software config instead ..

mx, you said you had both DDCs hooked up .. are you sure they are the 18W .. cause that would be 36W .. also, you said you connected the third pad on the DDC to one of the analog sensor hubs flow meter pins .. how come not to the RPM signal on the miniNG pins .. ?