different pumps in same loop?

[iNoXiouZ]

Hey all, so quick question

I have a Laing DDC 355 18W Pump and a Laing DDC 350 10W Pump, can I run these two pumps in the same loop with this: http://www.scan.co.uk/products/ek-ddc-dual-top-v2-g1-4-white-acetal

thanks for your help

[brammers]

Not recommended.

The 18w pump will "drive" the 10w version harder than it should.

Perhaps mod the 10w to an 18w and enjoy the extra pumping power?

J

[iNoXiouZ]

alright, how do I mod the 10W pump?

Thanks for your reply

[justin150]

On restriction caused by blocks between one pump and the second pump.

You end up with one of two situations. The question is simple, if you remove the weaker pump from the loop what is the flow rate at the point the pump would be:

1. If lower than second pump's flow rate, then adding second pump is fine

2. If higher than (a) no point and (b) you will damage second pump

[iNoXiouZ]

Well the Top that I brought doesn't seperate the two pumps, so they both work together and at the same point in the loop, I would guess that if I put the weaker pump closest to the outlet then the stronger pump would force the weaker pump to go/work harder? And if I put the stronger pump closest to the outlet then there would be no point in having the weaker pump in the loop?

[Waterlogged]

Definitely a don't do.

As for the mod, it's going to depend on what the PCB looks like. If it has 5 copper pads with a wire joining 2 of them, it can be modded by connecting the 2 on the right hand side with a solder joint. If you have 4 copper pads and 1 is empty, that's the new non-modifiable PCB.

[iNoXiouZ]

Ok well that puts my plans out the window, So I have 2 questions (second question depends on the first questions answer).

1) Is my 18W pump strong enough to pump water through 2 x aquagraFX GTX580 blocks, A Heatkiller 3.0 LC, A XSPC RS360, A EK RAD XT 360 and a Aphlacool re-pack res with a half decent flow rate?

If the answer to question 1 is "no" then...

2)would my loop be effective like this... Res>18WPump>GPU2>GPU1>EK360>10WPump>XSPCRS360>CPU>Res? Thats with both pumps with EK Tops?

[Sideroxylon]

1. Yes

2. No

Different pumps in the same loop is never a good idea for the same reasons posted above.

[Holst]

Im not so sure about that.

As long as your not overspeeding the slower pump I dont think it will do any harm at all.

[iNoXiouZ]

Argh people saying conflicting things lol... Right so I've already placed my order and I'm at work from 9 - 5.30 so canceling my order and re-ordering will be very differcult, Could I not just have the faster pump closest to the outlet and have the slower pump feeding it? I'm pretty sure I wouldn't gain any benifits from the slower pump but atleast I could use my PC until I have some time to figure out an alternative.

[Martinm210]

Going against the grain here, but I still don't see a problem with mixing pumps as long as the resulting combined flow rate is at a reasonable point along the slowest pumps curve. If it's at all a higher restriction loop, it's pretty unlikely you'd be driving that slow pump any harder than a low restriction loop with the pump by itself.

http://www.swiftnets.com/assets/images/products/MCP35X/pq-curves-all-1200x870.png

The best operating point (http://www.engineeringtoolbox.com/pump-system-curves-d_635.html) is generally fairly close to the best efficiency point.

According to my old 3.1 test hthe BEP of a stock pumps was right at 1.0GPM, so I think as long as you're combine flow rate remains somewhere between .75GPM and 1.25GPM, you should be fine mixing the 350 with a stronger pump.

At least that's the theory we follow when designing for industrial pumps, never had any long term mixed pumps experience myself in a pc, but I know people do it and I have yet to hear about anyone killing a pump doing that. Vapor also tested his CPU blocks with mixed pump.

Also Laing's documentation specifically says the pump:



Motor:
Highly efficient and ultra quiet spherical ECM motor
design. Pump is designed for continuous operation. All
ratings are within working limits of the motor

http://lainginc.itt.com/files/LTSS-29.pdf


To me this say's it is designed to run continuous in all ranges up to 2.3GPM.

My suggestion, do a quick flow rate check, if you net less than 1.25-1.5GPM ish, don't worry about it.:up:

As always, I do think some extra air flow around these pumps is always a good idea especially the DDCs.

[Holst]

The single 355 will be fine on its own.

IMO You may get a small performance gain by running both pumps and the second pump will add some redundancy.

EDIT - Thanks martin for posting a better explanation than me.

As for where to put the pumps in the loop, it wont make any real difference (flowrate will be the same regardless of order)

[iNoXiouZ]

Thanks guys for your help, unfortunately I can't messure the flow rate of the loop I mentioned earlier as my order doesn't arrive until Tuesday which contains my aquagrafx waterblock, EK 360 Rad, EK dual top and bay res.

Do you guys think having the weaker pump feeding the stronger pump with the EK dual Top will be fine? That would be the easiest solution atm.

[Martinm210]

Thanks guys for your help, unfortunately I can't messure the flow rate of the loop I mentioned earlier as my order doesn't arrive until Tuesday which contains my aquagrafx waterblock, EK 360 Rad, EK dual top and bay res.

Do you guys think having the weaker pump feeding the stronger pump with the EK dual Top will be fine? That would be the easiest solution atm.

Doesn't have to be fancy testing, just grab a 5 gallon bucket and time how long it takes to fill it up. If you want more precision, use an old milk jug to mark out exactly 5 gallons.

I personally wouldn't bother with dual pumps unless you have less than 1 GPM with a single pump. Heck I'm even running my current CPU only loop at .9GPM. It just means more heat and more parts wearing. Put the pump in the second pump in a drawer and save it as a spare.:up:

[Holst]

Doesn't have to be fancy testing, just grab a 5 gallon bucket and time how long it takes to fill it up. If you want more precision, use an old milk jug to mark out exactly 5 gallons.

I personally wouldn't bother with dual pumps unless you have less than 1 GPM with a single pump. Heck I'm even running my current CPU only loop at .9GPM. It just means more heat and more parts wearing. Put the pump in the second pump in a drawer and save it as a spare.:up:

If it were me I would test both options and see what difference it made.

With your components out of the PC its easy to test the flowrate and it means you can compare your flowrate with skinnee's graphs and see how well your blocks will work :)

I like to use a 2L coke bottle to test flowrate, its pretty accurate and easy to do.

[coolmiester]

I’m in total agreement with Martin on this one.

Sure if both pumps were hooked up with some kind of fixed bicycle chain and the 18watt pump was doing 100RPM against the 10watt trying to do 50RPM then yeah something would give and try to slip a cog but you’re talking about two pumps running totally independently with only liquid as the obstruction.

Maybe the 18watt pump will slow down a little (as it will anyway with multiple blocks and rads) and the 10watt might speed up slightly but its not like either pump is going to go into melt down as both are well capable of doing way more RPM than what they will be doing anyway:shrug:

[iNoXiouZ]

So it is technically possible and won't do any harm to either pump

I'll test flow rates of both options (single 18 pump and 10W>18> + EK Dual Top) and see which is best on Tuesday. Whats the best way to test flow exactly, I've never done it before.

[Martinm210]

Whats the best way to test flow exactly, I've never done it before.

Best if you have something like two 5 gallon buckets at roughly the same level. Just time how long it takes to draw from one filling the other. If it takes 5 minutes, that's 5 gallons/5 minutes (5/5=1) gallon per minute. If it takes 3.5 minutes it's 5/3.5 or 1.42 GPM...etc.

The only thing to watch is make sure the pump is fully primed (no air) and also try to keep the in/out levels roughly the same to minimize the pressure gains/losses from the open loop. A flow meter is of coarse easiest, but for simple checks, no need to waste money (or add restriction) on a meter that's just going to sit there constant.:up:

[coolmiester]

Just grab a large container and a measuring jug – fill the container with 5ltr or 1gallon of liquid and mark the level with a marker which gives you the volume.

Now all you need is a stop watch to time how long it takes to fill the container

Note: 1 Imperial gallon = 1.20095042 US gallons so you got to be careful to clarify which gallon you go for and would recommend going with 5litres

[Holst]

Gallons i very annoying for those of us in Europe, I can never understand what you yanks are on about.
Litres is much easier to understand :D

[coolmiester]

Gallons i very annoying for those of us in Europe, I can never understand what you yanks are on about.
Litres is much easier to understand :D

LOL, don't get me started on all their other crazy measuring idiosyncrasy's :ROTF:

Anyone remember how we measured barbs before the european G1/4 came into effect :shakes:

[Holst]

LOL, don't get me started on all their other crazy measuring idiosyncrasy's :ROTF:

Anyone remember how we measured barbs before the european G1/4 came into effect :shakes:

I thought you just bought random barbs then gave up on finding the correct thread and PTFE taped them till they stopped leaking :D

[NaeKuh]

LOL, don't get me started on all their other crazy measuring idiosyncrasy's :ROTF:


i prefer the metric system.

its so easy, its in multipuls of 10.

Unlike the messy customary, where its like ounces pt, qt, gal...

But like everything else in the world us Americans think we need to rule everything and wont adapt. :rofl:

[coolmiester]

Sorry to go off topic but while we're on the subject....

I dunno who's fault it was but how the hell did we get barbs measured in both - G1/4 to 1/2" WTF is that about but it seems to be the standard for the time being :rofl:

[Church]

I find converting values very easy / very fast .. by using google's calculator.
Just type in search something like "16 gallons per mile in litres per 100 kilometers", or "25 inches in mm", or "35 ounces in grams" or "1 gallon per minute in litres per hour", and voala, you get your conversion with just browser around (which probably you have, if you post in this forum :) ).
Of course, i would prefer metric system everywhere, but world is not ideal place and yankees probably will be stubborn enough to drag imperial measurement system for few more centuries along, no matter how much problems it introduces. Even at USA in several areas there are used mostly some very weird mixup of both imperial and metric systems with ratio depending on traditions on regulations pushed from above.

[thegcpu]

Going against the grain here, but I still don't see a problem with mixing pumps as long as the resulting combined flow rate is at a reasonable point along the slowest pumps curve. If it's at all a higher restriction loop, it's pretty unlikely you'd be driving that slow pump any harder than a low restriction loop with the pump by itself.

http://www.swiftnets.com/assets/images/products/MCP35X/pq-curves-all-1200x870.png

The best operating point (http://www.engineeringtoolbox.com/pump-system-curves-d_635.html) is generally fairly close to the best efficiency point.

According to my old 3.1 test hthe BEP of a stock pumps was right at 1.0GPM, so I think as long as you're combine flow rate remains somewhere between .75GPM and 1.25GPM, you should be fine mixing the 350 with a stronger pump.

At least that's the theory we follow when designing for industrial pumps, never had any long term mixed pumps experience myself in a pc, but I know people do it and I have yet to hear about anyone killing a pump doing that. Vapor also tested his CPU blocks with mixed pump.

Also Laing's documentation specifically says the pump:



To me this say's it is designed to run continuous in all ranges up to 2.3GPM.

My suggestion, do a quick flow rate check, if you net less than 1.25-1.5GPM ish, don't worry about it.:up:

As always, I do think some extra air flow around these pumps is always a good idea especially the DDCs.

Is the MCP35X really that better?

[jadawgis732]

I posted this same question elsewhere but the pumps were a MCP655b and a X2O 750 and was told it was fine. Here is the thread: http://hardforum.com/showthread.php?t=1576792 Do you guys disagree with the responses?

[Church]

thegcpu: Compared to MCP355 with stock top - yes. Compared to MCP355 (or even DDC 3.25) with good aftermarket top like those of EK or XSPC - not that much. But nevertheless, 35X is very good overall pump, a bit on pricy side though.

[thegcpu]

Thanks as I need a new DDC. I currently have a new EK V2 top and I was thinking of going with the 35x. But I'll stick with the 355.

[Martinm210]

I'm not sure on the 35x, that's coming from swiftech's site. I haven't done any testing myself, so I'm not sure. I think skinnee has some results in his pump thread you can check though.

I was just trying to show that even a stock 350 still has a good max flow rate. And considering they have been the same pump (Same PCB just wired slightly differently) as the 355, I'd have a hard time believing mixing would put any more stress on the pump if used in an restrictive loop.


And don't get me started on units. I worked for many years in English units, started a new job where they were using metric, learned metric after adjusting for 2-3 years, then they switched back to English. Finally another 5 years later I feel like I'm in a groove again.

At this point I could care less what units I work in, as long as it stays the same!...:)

[Alexandr0s]

A while ago I used Martin's pump top test and superimposed the results of the best pump top of that test (XSPC Res Top) onto the graph above. The MCP35X with Swiftech top (which is a pretty good top) is quite similar to a standard MCP355 with XSPC Res top.

http://i54.tinypic.com/a0f05g.png

[thegcpu]

OK..
Anyone want to buy a EK V2 top.. lol

[Alexandr0s]

I really doubt you'll see a performance difference. Unless you really want (and/or need) PWM, I'd just stick with your MCP355 ;).

[Waterlogged]

http://www.swiftnets.com/assets/images/products/MCP35X/pq-curves-all-1200x870.png

For some reason, I just paid really close attention to that graph and the numbers for the 355 looked familiar to me. It seems that graph kind of validates the testing I did as the numbers fall right in line for the 355/3.2 and 350/3.1. :eek:

http://www.xtremesystems.org/forums/showthread.php?t=245795

I think I keep buying 3.25's thank you. :p:;)

[jadawgis732]

ok..
Anyone want to buy a ek v2 top.. Lol

ygpm

[cdawall]

i am running to different pumps laing D4 and laing D5 i have tested myself not actual flow but what you could see in the loop D4 vs D5 by themselves you couldn't see much of a difference but when you flip both on there is a good flow difference.

[Red M]

i prefer the metric system.

+1 soooo much simpler I wish everyone would use it

[Fatfool]

Been there, done that

http://img.photobucket.com/albums/v140/Fatfool/P1000031s.jpg

The more powerful pump is obvious. I guess the smaller pump was the only thing supplying it with enough liqquid since the MCP355 is supposed to have an inlet of 3/8" at least. (I use 1/4")