DDC ULTRA test @ H2okoeling.dk

Printable View

Show 100 post(s) from this thread on one page

08-17-2006, 05:34 AM
rosco

Visibly, you don't know how to really measure static pressure because your setup is a bit wrong. No parts must be between pump outlet and pressure point reference. The barbs, rotary sluice and T you put just after outlet (on the left on the setup pic) induce a undesirable pressure drop and you don't take in account when flow is not null, this is why all your graphs are underestimated. This setup don't measure true static pressure at outlet. I see you don't have an inlet pressure sensor but are you sure than pressure due to heights don't play a role here ? Reservoir and pressure line seems to be at a same level to annihilate it, but we have to be sure (this gives an offset).

Moreover, I suspect a bad calibration or a flawed setup because you only get 3.34 mH2O max for a normal DDC (~3.9 mH2O normally) and a true DDC+ is nearer of 6 mh2O than 4.5 m2HO (there are 4.7 and 6.1 mH2O DDC versions)... Your T seems to be far to large too and not adapted, static pressure must be taken by the smallest orifice possible without perturbation just before and after (or induced by the pressure connector itself), sluice must be after pressure point measure and not before it. Finally, your 12 V is just at the power supply outlet, not at the molex pump and with the length and ~2 A drawn, there's a small voltage drop between them. 12 V have to be at connector, not before.

Make your articles lighter & clearer, lots of graphs are not useful ;) (Testsektion #3 isn't really interesting, it refers to a particular setup, not to tubing only)
08-17-2006, 06:55 AM
nikhsub1

Quote:

Originally Posted by rosco

(there are 4.7 and 6.1 mH2O DDC versions)...

Hi rosco, good comments as usual. So not to confuse, the 4.7 version is NOT the 18w version correct? I would then assume that all 18w versions are 6.1M and all 10w are 4.7M? Or is there more to it than that?
08-17-2006, 07:44 AM
MaGiX

Quote:

Originally Posted by rosco

Visibly, you don't know how to really measure static pressure because your setup is a bit wrong. No parts must be between pump outlet and pressure point reference. The barbs, rotary sluice and T you put just after outlet (on the left on the setup pic) induce a undesirable pressure drop and you don't take in account when flow is not null, this is why all your graphs are underestimated. This setup don't measure true static pressure at outlet. I see you don't have an inlet pressure sensor but are you sure than pressure due to heights don't play a role here ? Reservoir and pressure line seems to be at a same level to annihilate it, but we have to be sure (this gives an offset).

Moreover, I suspect a bad calibration or a flawed setup because you only get 3.34 mH2O max for a normal DDC (~3.9 mH2O normally) and a true DDC+ is nearer of 6 mh2O than 4.5 m2HO (there are 4.7 and 6.1 mH2O DDC versions)... Your T seems to be far to large too and not adapted, static pressure must be taken by the smallest orifice possible without perturbation just before and after (or induced by the pressure connector itself), sluice must be after pressure point measure and not before it. Finally, your 12 V is just at the power supply outlet, not at the molex pump and with the length and ~2 A drawn, there's a small voltage drop between them. 12 V have to be at connector, not before.

Make your articles lighter & clearer, lots of graphs are not useful ;) (Testsektion #3 isn't really interesting, it refers to a particular setup, not to tubing only)

When I first started dreaming about making a test section for watercooling products, I was a nOOb into the technical stuff surrounding such a test section, so I did my best to find a reference testbed that we could make our own version out of, and in that case I found the Systemcooling.com testbed for pumps, and did all that I could to take the good stuff from that, and copy it into our own, but using some other instruments.

http://www.systemcooling.com/images/...image15big.jpg

The best instruments we were able to find (the price taken into account) was a Rotameter from Aalborg.com and a digital manometer from Keller. The accuracy for both of these is pretty accurate, but the lifting height for the rotameter is its biggest problem, and at the same time it’s a bit restrictive (only 11 mm ID) but we could live with that, the price taken into account.

Our main objective though, was not to build a testbed that was in no way restrictive or pressure reducing, but to make a testbed that would give the same results every time, and that I believe we accomplished to a good point. Therefore, every test we make is not comparable to any other tests around the world, only to compare two pumps against each other is the any good.

If negative pressure (-mbar) is happening, then the manometer will also tell us that (goes from -1 bar to +3 bar) and every time we test a pump, it is first zeroed, then the pump is started, turned off, zeroed again, and so forth, until the manometer after each on/off of the pump shows 0 mbar, and therefore is in absolute zero. First then is the test started.

But I’m not sure I get what you are saying about the tubing to the manometer, did I get it wrong that you mean the tube will be far better in a small size than the 12 mm ID we use at this point? Did some search, talked to some people before we used those, and got the reply it didn’t matter if it was large or small. On the new testbed I’m building at the moment (testbed for blocks and radiators) I use 8 mm ID tubing though, but that is pretty much for the convenience of the build…

I must say though, to my own defence, that all the fittings on this testbed, have been bored up to 12,0 mm ID, to reduce flow drop, and the T line on the top of the rotameter is going to be changed to another fitting in a couple of days, to reduce the flow and pressure drop of the rotameter.

Also, the meaning of the whole project was to make some good Danish tests and articles, had never dreamt that our work would be spread so far around the world as it happens to be now :)

But again, this is a hobby project, I never dreamt about getting to XS.org and similar pages, and all in all, we did our best to make the best testbed we could with the money available (its more or less totally paid for by myself alone,) but all comments, ideas and thoughts is more than welcome to how we can make our test beds even better.

MaGiX
08-17-2006, 08:18 AM
Radical_53

@magix: Great to hear that. If you are interested, I do have two Dual-DDC heads here. Both the tops from Alphacool and Watercool. If you'd like to use them for a test, just drop me a mail :)
08-17-2006, 09:40 AM
rosco

1 Attachment(s)

Quote:

Originally Posted by nikhsub1

Hi rosco, good comments as usual. So not to confuse, the 4.7 version is NOT the 18w version correct? I would then assume that all 18w versions are 6.1M and all 10w are 4.7M? Or is there more to it than that?

Hi nik. Here in France, we got a 16 W version just before 18 W but it was quickly replaced by 18 W version. Laing is a bit confusing in their descriptions and models because they introduce variants without changing case reference for ex. Laing USA and Laing Germany don't do exactly the same... In fact, there's 4 versions of the DDC if we take in account the original 10W, the modded 10W for starting problems and no more current limiter, the 16W and the 18W. Remember the pressure values for max head measured by a forumer here 1 or 2 weeks ago, he found ~6 mH2O for DDC 1+ (normally called DDC-2 internally at Laing...) which is OK, others pump values were OK too. This is why I suspect a little mistake or an offset in calibration in the setup presented here, or the pump is an old DDC, I don't know :( .

MaGiX > The Lee's test bench is well designed, like mine (we've got similar results doing our respective tests & reviews). You could see well how different it is from yours in the parts layout so just copy it exactly. It's not a problem of instruments, they are good enough, it's more how to use/assemble them correctly to get what you want ;) . The problem is principally about the placement of your T to go to manometer (just after outlet pump, not at 3km with ten things between outlet and T) and the T geometry. We want only static pressure and not static pressure + a bit of dynamic pressure because flow could induce some flow disruptions due to a big diameter (see paint pic attached). A 1mm hole is far enough to catch static pressure without disturbing water flow. Tubing diameter to go to manometer isn't a problem, there's no flow movement inside, pressure is only transmitted by contact and it don't depend of diameter and length, you're right. It's just useless to get such big tubing for that part (I use 4/6 mm tubing).

Seeing your relative "low" values each time, you should normally ask yourself some questions about them and try to compare to manufacturer datasheets (Laing data are well made) or even to others technical review. Good luck to improve it :)
08-17-2006, 11:04 AM
MaGiX

Quote:

Originally Posted by rosco

MaGiX > The Lee's test bench is well designed, like mine (we've got similar results doing our respective tests & reviews). You could see well how different it is from yours in the parts layout so just copy it exactly. It's not a problem of instruments, they are good enough, it's more how to use/assemble them correctly to get what you want ;) . The problem is principally about the placement of your T to go to manometer (just after outlet pump, not at 3km with ten things between outlet and T) and the T geometry. We want only static pressure and not static pressure + a bit of dynamic pressure because flow could induce some flow disruptions due to a big diameter (see paint pic attached). A 1mm hole is far enough to catch static pressure without disturbing water flow. Tubing diameter to go to manometer isn't a problem, there's no flow movement inside, pressure is only transmitted by contact and it don't depend of diameter and length, you're right. It's just useless to get such big tubing for that part (I use 4/6 mm tubing).

Seeing your relative "low" values each time, you should normally ask yourself some questions about them and try to compare to manufacturer datasheets (Laing data are well made) or even to others technical review. Good luck to improve it :)

Point taken, I will in the next couple of days se how I can make some changes, without disrupting all the tests results made before this date :slapass:

Regarding the pressure, yes I did get some “dynamic” data, the pressure did jump a couple of mbar up and down, but to counter that, I noted the highest and lowest data on each setting, and took the middle ground (within 8 mbar typically)

The pressure measurements I got from each test (max pressure) actually did line up pretty accurately with the pressures given by manufactures in the long run (swiftech and Laing among others), the data I could not get to line up with manufacture data, was the flow, but that’s an old story which many other sites have also reported.

But thanks for your help, I had never thought about the pressure being so dynamic, because of the tubing, that fault will be dealt with shortly. Drawing says it all :)

Do you suppose 8/10 mm tube will do, or do I have to go all the way down to 6/8 or below? The systemcooling.com tube seems to be 8/10 mm?

Also, the valves just before the manometers on the Systemcooling.com rig, are they any problems? The ones we use are 1/2" ball valves, with a 11,8 mm ID hole through, and those I would love to hang on to, but if they have to go, then…

MaGiX
08-17-2006, 12:09 PM
rosco

Tubing diameter depends on your ability to link manometer to the T. Lee's tubing seems to be 6/8 IMO, not important because he has small barbs to achieve a good attach. Valves at manometers are not a problem, there are open and in fact useless in the process (no movement in the pressure line). It's just to purge air from line for example or remove manometer without purge all system.

Pay attention that dynamic (due to fluid movement, like the wind on your face) and static pressure (pressure measured in a closed container is a pure static pressure, no fluid movement) are 2 parameters that define total pressure in the fluid (the sum of both). I didn't say that dynamic pressure induces dynamic behaviour on your manometer. This dynamic change is probably because of flow disruption into your T or elsewhere in the rig, there's a bit of turbulence and pressure may vary locally. 8 mbar is nearly 82 mmH2O, it's quite important for a variation (depends of the manometer resolution obviously). Doing my measures, I have less than 5 mmH2O variation on any pump due to rotor (flow delivered by a pump isn't really stable and continuous, but a bit cyclic because impeller have a finite number of blades. So, we always get a small pressure variation according to response time of the sensor).
09-03-2006, 08:43 PM
krylon

Is this the DDC Ultra?

http://www.dangerdenstore.com/produc...5&cat=4&page=1
09-03-2006, 08:52 PM
nikhsub1

Quote:

Originally Posted by krylon

Is this the DDC Ultra?

http://www.dangerdenstore.com/produc...5&cat=4&page=1

That is ONE part of the ultra... the ultra consists of the DDC+ (which that link above is) in ADDITION to the alphacool top.

http://www.performance-pcs.com/catal...07201d0469f56b
09-03-2006, 08:58 PM
krylon

Thanks. Just wanted to make sure the actual pump was the right version. Ordering this and the alphacool topper! Hope to see an improvement from my D5.
09-04-2006, 05:12 AM
nealh

Quote:

Originally Posted by krylon

Thanks. Just wanted to make sure the actual pump was the right version. Ordering this and the alphacool topper! Hope to see an improvement from my D5.

you can get everything from Jab tech as well..they ship fast and have excellent CS..same prices..usually very fair shipping..and I have never seen any CS issues about them like performance pcs(one user reported getting a ?? refurb or used DDC ultra as new)
http://www.jab-tech.com/Pumps-c-56.html

http://www.jab-tech.com/Pumps-c-56.html
09-04-2006, 06:08 AM
SiGfever

Quote:

Originally Posted by krylon

Thanks. Just wanted to make sure the actual pump was the right version. Ordering this and the alphacool topper! Hope to see an improvement from my D5.

I verified with Laing that the DDC-2TPMP (which is what DD sells) is the 18w high output version. See my thread on mounting the DDC-2TPMP...

http://www.xtremesystems.org/forums/...d.php?t=114009

I removed the blue heatsink and put two larger versions, one on each side for a total of four.

Show 100 post(s) from this thread on one page