Pro/Forums - Another decent 12v pump

Pro/Forums (http://forums.procooling.com/vbb/index.php)

- General Liquid/Water Cooling Discussion (http://forums.procooling.com/vbb/forumdisplay.php?f=9)

- - Another decent 12v pump (http://forums.procooling.com/vbb/showthread.php?t=10467)

Sorry, I really wouldn't like to guess how loud it will be until I had one in my possession.

Of all the 12v pumps I've heard so far, the D4's are the loudest and most annoying, with the exception of pumps that used brushed motors - but nobody would buy one of these anyway for water-cooling use. The latest muffled D4's pretty much brings them in line with the rest of the brushless pack, but still not quiet, and still with an annoying (to my ear) whine.

Most people who say that the muffled D4's are quiet tend to have a few 35dBA fans running at the same time (and are used to such noise), and for them the perception of "quiet" tends to mean quite a different thing to those who are after truly "whisper" quiet.

Quote:

Originally Posted by Hansfragger

You are not kidding. Those chumps were going to stick an extra 75 bones in their pockets and laugh all the way to the bank.
In your opinion, how much quieter do you think it will be than the D4. The D4 is around 40+ DB at 12 volts. In comparison I have a Vantec Spectrum fan card in a pci slot blowing on my video card rated at 36.64 DB at 12V. It's noisy enough to be irritating if i'm not gaming. When turned down to 25 DB it's not bad at all. But you can really tell the difference on high. I don't mind losing the return shipping and 15 percent restocking fee if you feel the March pump will be substantially quieter. Thanks for the link- I owe you one.

hmm - i imagine you could sell it on here/some other forum without loosing too much more than that if you wanted to wait for cathar's review...

Thank you Cathar. What you said was exactly what I wanted to hear. The part I was concerned about was the whine. Many times the whine will end up grating on your nerves more than the overall noise level. I checked Depco's website and it states they are open M-Friday. I'm going to call them Monday. Things are looking up. Have a great one and thanks again!

Don't forget that they may have been clobbered by the latest hurricane - they're directly west of where it made landfall (about 150 miles).
The fact that their website is up doesn't mean their offices currently have electricity or phone service.
So... if you don't get an answer Monday, give 'em a couple of days.

Quote:

Originally Posted by pauldenton

hmm - i imagine you could sell it on here/some other forum without loosing too much more than that if you wanted to wait for cathar's review...

Good idea. Having a backup plan is a beautiful thing! Thanks.

Pump arrived today. Very fast shipping.

Indeed it is a small thing. Including fittings and mounting tabs:

11.5cm long x 7.5cm high x 6cm wide.

Will have a play with it sometime today as soon as my sick 4yo daughter stops pestering me every 5 minutes...

A rather mediocre picture of the March pump next to a Swiftech MCP650:

http://www.employees.org/~slf/12v/m893-s650.jpg

Can you be convinced to show us the impeller? (i.e. take the housing cover off?)

Your choice, of course...

Think it would probably be best for him to do whatever tests he planned to do with it first before taking it apart and putting it back together, removes another variable.

i agree with bale.

plus it is a $125 pump, I'd be a bit hesitent about taking it apart. That does look like a very impressive pump though.

This is a collection of posts I made over at OCAU:

Quote:

Originally Posted by Cathar

Hooked it up and ran it at 12v. Emits a non-quiet hum, and then started rattling, which I can only assume is the impeller. Put my finger over the end of the outlet and the rattling noise went away, so clearly it needs some back-pressure on it, which is hardly unsurprising as most of these higher pressure pumps aren't designed for no-load conditions.

Noise-wise I'd say its just a little quieter than an MCP600, but still noisier than an Eheim 1250. For the silence freaks, this is still perhaps a little disappointing. At 7V it is about as loud as an Eheim 1048, and pushes about as much flow as one as well. In wide-open mode (unrestricted) I measured 8LPM, which is about what I saw with the Eheim 1048 when I had one.

Hooking it up to the Cascade-without-base-plate jet test it was still pushing a very admirable amount of flow at 7V.

Vibration wise, at 12v, about as vibrational as an Eheim 1250, or an MCP600. At 7v, about as much vibration as an Eheim 1048. i.e. as awlays - if you stick the pump against something that's going to rattle and make noise, then it will be annoying, but this is true of every pump I've tried.
Will continue testing....

Quote:

Originally Posted by Cathar

Very rough 12v flow test would seem to indicate that the rated peak flow is perhaps a little pessimistic. Was easily seeing around the 13LPM mark. Will need to test this again, but it is definitely significantly higher than the stated 10.5LPM peak.

Just letting the pump run for a while and it is quietening down. Obviously just needs a little time to "bed in".

Quote:

Originally Posted by Cathar

I pulled the pump apart about an hour ago. Internally it's the same sort of design as a Swiftech MCP600, with the spinning magnetic armature. I ran the armature motor without the wetted-end body, and by itself it is very quiet. Apparantly spins at around 4000RPM, and has a LOT of gyroscopic torque.

The magnetic armature and impeller magnet are tiny though. Perhaps around 2/3's the overall size of the MCP600, with the impeller magnet being about 1cm in diameter, and the magnet armature cavity not even large enough to fit my little finger inside it.

The impeller itself is basically a near exact look-alike of an Eheim 1048's impeller. A basic 4-post straight impeller design. One can only imagine the pump's pressure capability with a closed impeller design.

Measuring power-draw, very, very impressive low power draw.

At 12.00v, acting against a moderate pressure drop as would be somewhat typical of a full cooling loop, I measured a current draw flipping between 0.66-0.67A, or about bang on 8.0W.

Am steadily growing more impressed with this pump. Quieter than an MCP600 (and much quieter than an MCP650). 8W vs 12W power draw in comparison to the MCP600. Sealed and submersible for those who want that.

Oh, I just realised why this pump here is pushing more flow than as rated. These pumps are meant to have 3/8" OD outlets on them, and the pump I have is a 1/2" OD outlet, being significantly less restrictive, hence the much higher unloaded flow rates.

I now need to measure the peak static pressure head, and once done, I'll start gathering system flow-rate data.

Well the pump is good, but it is not quite as strong as rated in pressure terms.

Static pressure head test at 12.00v is pretty much bang on 2.80m, which is quite a bit lower than the rated 3.20m pressure head.

I've managed to develop a working PQ curve for the pump, which is not final, but will do until I finish more testing.

I also tested at various voltages. To my total surprise the pump actually ran (and started) all the down to 2.80v, which is the lowest I was able to test at. Needless to say, the pressure here is absolutely pathetic, at around 0.09m of pressure head. Flow is a meek trickle that I didn't measure. I measured up to 14v, which is the peak recommended voltage for the electronics inside the pump. Here's the static pressure head chart. Pressures are rounded to the closest 5cm mark.

3.0v => 0.10m
4.0v => 0.22m
5.0v => 0.40m
6.0v => 0.65m
7.0v => 0.90m
8.0v => 1.20m
9.0v => 1.55m
10.0v => 1.95m
11.0v => 2.40m
12.0v => 2.80m
13.0v => 3.25m
13.5v => 3.45m
14.0v => 3.70m

http://www.employees.org/~slf/curves/pumps/12vpumps.png

interesting investigation
Thanks Cathar

.5LPM difference between this and a 650 at cascadeload?

interesting

even more so if price comes down.

Quote:

Originally Posted by greenman100

.5LPM difference between this and a 650 at cascade load?

interesting

even more so if price comes down.

If one factors in in-line pump heat, it does get interesting. 16W of in-line heat with a 650 vs (my estimate) of 6W with the March (based on what the Iwaki's and MCP600's do with a similar internal design). I'll measure the actual in-line heat sometime soon.

With a 10W pump heat difference, and a radiator efficiency of, say, 0.03C/W, the actual difference at the CPU is likely to be zero.

seems as though the march pump is suffering from its relatively "simple" impeller design?

Quote:

Originally Posted by Althornin

seems as though the march pump is suffering from its relatively "simple" impeller design?

I have a theory that it may be due to the 1/2" OD outlet. Because the main impeller housing is different to cater for the 1/2" OD outlet, as opposed to the normal 3/8" OD outlets, I suspect that the clearances are a little larger between the different models, resulting in less pressure.

My impeller also seems to be slightly off-center, and I suspect that this is also the cause of the noise that I'm hearing.

Tomorrow I'll be heading around to the local distributor to grab a new impeller and the 3/8" OD outlet housing.

With two impellers, I'll look at hacking up a closed impeller modification to one of the impellers and see what difference that makes (if any).

Picked up new impeller today, and the 3/8" outlet housing. Block's pressure has improved from quick testing. Need to investigate more. A little quieter too, but still somewhat rattly. Definitely an improvement. Perhaps the 3/8" config is the better option, and just step up to 1/2" outside of the block, rather than getting the 1/2" option by default.

Suspect the rattling is due to 1/2 turn of slack in the rotor/impeller coupling? (As seems to be typical on mag drives)
Must be a cure for that...

Hey guys, I know March, and they make some damn fine pumps !

Pump choice is not easy, you have price, size, flow, EM, power use, replacement part costs, looks, etc to think about.

But in my opinion, even though we are some what there competitor (not really given low PC liquid cooling numbers), that you should feel secure with one of there brushless units. :)

Don't worry about them only rating the motor, there are SO MANY factors effecting pump life that it's almost impossible to predict for every application.

"March has an impeller spinning on a shaft
C-Systems has a steel shaft with a seal
- both weak links if going for 5 yrs service"

Sorry Bill, but must say something about this.

First your right, both designs can fail. But a good designer considers there will be failures and what to do about it. Assuming even a Laing unit will not ever fail is wrong.

One answer is monitoring, and low cost replacement parts. With AC pumps you have this now, as the impeller will make noise and a replacement is cheap. Not sure what cost of impeller is for the March unit, but would be nice if low.

But the "ideal" solution is mag coupling, impeller speed monitoring, low costs, small size, cheap replacement parts, long lasting in fluid bearings,and low EM / power use /noise.

But that pump does not exsist, and likley never will, well maybe with future superconductors :)

So whats the final prognosis doc cathar? And have you tried pairing the 3/8" pump with 7/16 hose? I am very interested in this as I want to ditch this noisey arse mag3 with it's 12' python cord.

Quote:

Originally Posted by psychofunk

So whats the final prognosis doc cathar? And have you tried pairing the 3/8" pump with 7/16 hose? I am very interested in this as I want to ditch this noisey arse mag3 with it's 12' python cord.

Haven't had much of a chance to play with the pump since I got the 3/8" OD outlet.

A quick pressure test did seem to show that the pump's static pressure head was better than with the 1/2" OD outlet, however overall flow rates did not improve. Pump was a touch quieter but still has a pinging/rattling sound that will annoy the truly quiet conscious. The pump would be as quiet as an Eheim 1250 if I could only
get this faint pinging noise to go away. There seems to be something faintly touching the back of the sealed motor section which I can't get to.

No, not a Mag 3 beater, or even matcher. Power-wise it's mid-way between an Eheim 1250 and an MCP600. Still good - just not special.

Quote:

Originally Posted by Cathar

No, not a Mag 3 beater, or even matcher. Power-wise it's mid-way between an Eheim 1250 and an MCP600. Still good - just not special.

Any 12v pump you recommend for a quiet concious type of guy looking to beat or match a mag3. I was looking at the mcp650 but the stuff I have read about noise has me questioning it. I am not against spending money but I want it all. Performance, compact size and quiet.

Quote:

Originally Posted by psychofunk

As soon as I find one, I'll let you know about it.

Perhaps getting one of the Iwaki RD20 12v pumps at a rated 20dBA would be the best bet, but you're talking mega-bucks (~$300US at the least). Performance would kick a Mag3 square in the hiney. I'm a little crazy about the water-cooling hobby with a nice salary to support the obsession, but even I balk at that sort of price-tag.

At that site they do say 24v, however you can get them for 12v operation from Iwaki if you special order it.

Niiiiccccceeeee!!!! Only prob for me is that it's a little too big.

What are your thoughts on the Iwaki RD-05H? Maybe even two of them?

Quote:

Originally Posted by psychofunk

Niiiiccccceeeee!!!! Only prob for me is that it's a little too big.

What are your thoughts on the Iwaki RD-05H? Maybe even two of them?

Probably not a good idea. Rated only for intermittent use.

Ahhh, nice catch, I missed that, thanks.

Yeah, that woulda been a shame to kill a pump like that. Shame is that it has that almost perfect curve too heh.

Quote:

Originally Posted by psychofunk

Niiiiccccceeeee!!!! Only prob for me is that it's a little too big.

Too big? The RD-20 is about the same size as a Laing D4.

Quote:

Originally Posted by Cathar

Too big? The RD-20 is about the same size as a Laing D4.

I was going by the spec sheet which says it's the same size as the RD30 which is 106x108x110, is that not correct?