Printable View
Nice vid WL!
I am finding that all pumps are not created equal, D5's have the same variability, I don't have an RPM sensor...so just amp draw, flow and pressure. I don't want to completely derail Malik's thread, so I'll start my own on my findings. It was something I always figured, but now I have numbers for it.
Thanks skinnee, it was a pretty simple test, so I figured "why not?".
OK, I finally got to testing (a bit later than I wanted thanks to Mother Nature dumping some of that white crap) all my 3.x's trying to verify a RPM range....turns out they are remarkably consistent. The 3.2 (in the video) ran 4350-4380 with a quick spike to 4410 when flow was completely cut off. The 2 solder modded 3.1's ran 4320-4350 and spiked @ 4380 when the flow was cut off. Of course, this was all done with the same PSU so that might have some kind of bearing on the numbers, so I decided to do a single test with a Antec Neopower 650 I have laying around and one of the solder modded pumps, and the RPM's went up to 4350-4380 with a spike to 4410 with no flow. Looks like amps might play some kind of role in determining RPM's for the DDC3.x's, as suggested by Vapor.Quote:
Originally Posted by Waterlogged
i just saw the vid excellt stuff there...
Thanks Hc. :up:Quote:
Originally Posted by Hondacity
Quote:
Originally Posted by Malik
Again, I hate to harp on the results of Malik's gracious testing but, this statement and graph were bugging me a bit. With all things being equal....there's no way the pump will spin below what it did in my test unless the voltage was altered. If the tests were conducted at a reduced voltage, it could very well be hurting the results gathered. We know for a fact that the GTZ requires very good pressure to pull top numbers, reducing volts to the pump will hurt the GTZ's performance, possibly quite badly. We've also seen in my test that higher restriction equates to more RPM's with a DDC-2. If the volts were the same for both tests in the graph above, the Enzo is even more restrictive than the GTZ. I've been tinkering around with the testing setup some more and added an original Apogee to the DDC-2, and the RPMs went up ~90-120 RPM's, I then added the POS GT Stealth I have and it went up yet another ~90-120 RPM's. Switching to the 3.2, the RPM's only go up ~60 max.
If the testing wasn't done at the same volts....well...I'm not prepared to go down that dark, dark path at this point.
Sad, Malik was good enough to share his results with us and all you can do is pick it to bits and basically call him a liar. He may well think twice before sharing again in future and I cant say I blame him.Quote:
Originally Posted by Waterlogged
Exactly.Quote:
Originally Posted by mcoffey
I would think and hope that anyone who posts test results is willing to have them studied, dissected and picked to pieces by the others here. Not in a bad way, but all in the name of accurate and consistent testing.
Malik we thank you for your testing and i think we're interested in investigating the RPM difference too. :)
I sincerely thank Malik for doing the test. It's just that I saw something that looked like an inconsistency and decided to "follow my nose". What I found intrigued me so I dug deeper. That led to more and more questions that, as of yet, are unanswered. Those unanswered questions could help explain the results which no one was really expecting. If you read Malik's topic in the Log build sub forum, ppl are believing the results without a second thought. I'm not saying he's lying, he's just not disclosing all the facts. If his MCP355 is a 3.2 then there is some FUD going on with the voltage, if the MCP355 is actually a DDC-2, then the Luna is a very restrictive block that flows less GPM than the GTZ (which is opposite of his RPM chart), this is what I'm trying to get to the bottom of...nothing else. :up:Quote:
Originally Posted by Fragger
Sorry guys but this is some stupid talk about me. I don't lie with this what i have done - about test. I only show the difference between this two blocks in my loop. I'm not a pro tester like other here.
So for the last time, i use the same voltage for cpu, the same voltage for pump ( ver. 3.2 - info for WL ), the same thermal compound, the same liquid etc. etc. I made all the SAME for this two blocks.
RPM's i watched in aqua-computer poweradjust software which control my pump. I show you on chart max value of rpm each block - i can check this like this:
http://www.abload.de/img/rpmx0lc.jpg
That is sad...hear something. Maybe next time i just wait for some PRO test.
Now i sitting on laptop so no more test, maybe that is good for me because some here on forum, pro lc people, will not call me a layer.
That's all from me.
No Malik, I don't think anyone is calling you a liar, I just think many people like to look at other's results and see if they can improve the test results data. I personally like seeing "real world" user tests.Quote:
Originally Posted by Malik
The one thing that is tough to understand and gets lost in translation without emotional expressions in forums (and Instant Messenger for that matter) is the intent and context of a user's posting is sometimes misunderstood. Some read the reply and think "hey, never thought of it this way", while others read it and think its a personal attack. Its really tough to convey what the poster meant. I didn't read this like an attack. It just read like statistical data that, while conflicting with your own results, just shows a different point of view. And really, isn't that what we all do? Give our points of view based on our own experimentation and or experience?
Please, don't stop posting your results. I like have multiple data points to compare to. It helps average out of whats expected against only one point of view. Keep it up! :up:
Again, I've never said you lied, but we do seem to have one hell of a conundrum here...don't we. You have a test showing a "spike" differential of 236 RPM's between the block (supposedly due to loss of flow/restriction), and I have a test verifying that the DDC 3.x RPM's don't fluctuate due to restriction. Is there any particular reason you chose to run the test with the pump at less than full power? This could be the curve ball that's screwing everything up in your test. I also don't know if I like the way that RPM's oscillates like that, doesn't seem very steady for testing purposes.Quote:
Originally Posted by Malik
Malik, I may not like the way you change your rig more than a woman changes her mind but, I've always acknowledged your skill and keen eye for style.:up:
@ everyone thinking about doing testing:
There are plenty of ppl that have posted tests here that have been kind enough to list how they do it properly, it would be nice if everyone that does testing would learn from them, so things like this don't happen in the future. If a test can't be done right or withstand the slightest scrutiny, it probably shouldn't be posted.:2cents:
italian review. the cooler has 4 different midplates. the site is usually a good one for reviews.
flow rate:
http://www.nexthardware.com/recensio...a/142_1179.htm
performance:
http://www.nexthardware.com/recensio...a/142_1180.htm
I can back that up. I've had a DDC3.2 for about a year now. In my first loop (EK Supreme - CPU only) the rpms read about 4550~4600. Now it's running in a 3-block loop (EK Supreme, NB S-Max, FC-4870) and the rpms are the same as before. I saw no change in rpms between loops of very different restriction. I have the DDC hooked up to my Scythe Kaze meter.Quote:
Originally Posted by Waterlogged
Thanks for the links Odin...performance drops in roughly where it was believed to be. :yepp: I'm not crazy about the rad they use but that pump is a little beast. :eek:Quote:
Originally Posted by OdinEidolon
twwen, looks like you have a pretty high amp PSU for the 3.2 to run @ those RPM's. :up:Quote:
Originally Posted by twwen2
First chance I get, I'll be taking the testing of the DDC-2 and DDC 3.2 to the next level with more data. I just need to get my hands on some extra gear first. I'll start a new topic for that so keep an eye open.;)
yeah the sanso PDH054 is one of the best pumps around... real 9 meters head pressure! :shocked:Quote:
Originally Posted by Waterlogged
and its dimensions arent too great, its smaller that a mcp655. It costs around 110€ which isnt too much. and 12V, no need for a meanwell 24V dedicated PSU
i'd really like to see that against an IwakiRD-30
here is a good review on the sanso, from the same site: http://www.nexthardware.com/recensioni/scheda/93.htm
odin,
any retailer of that pump? watercoolinguk does not carry it
thanks
I wouldn't mind surrendering a very small portion of that pressure for a bit more flow...hmm, wonder how hard it'd be to make a new top for that.:slobber: Hehe...oh Alex....Quoc...any chance of etting some of these pumps?:DQuote:
Originally Posted by OdinEidolon
Sanso PDH-054 is Iwaki killer!!!:yepp::up:Quote:
Originally Posted by OdinEidolon
ybris cooling:yepp:Quote:
Originally Posted by Hondacity
the only retailer i have foud till now is ybris cooling, in Italy. i'm pretty sure they ship internationally tho.Quote:
Originally Posted by Hondacity
more powerful sisters:
http://www.ybris-cooling.it/images/p...so/pd31-06.jpg
EDIT: shouldnt we stop the OT? maybe a new thread would be a good idea
HX620 baby! :cool:Quote:
Originally Posted by Waterlogged
So Waterlogged.. where are your results for the Luna that you say you tested and that others are referencing? I asked for them like 2 pages ago. Please, post them.
Wrong. RPMs on the DDC do fluctuate! Its due to the nature of the ddc, being a microcontroller driven pump with a hallsensor, that rpm vary! Its somewhat comparable to microcontrollers of the better (3-phase) fans. Their rpms fluctuate too depending on the backpressure u have.Quote:
Originally Posted by Waterlogged
Its only eheim pumps with only two phases, that sometimes use a fixed frequency, but usually the more advanced 12V versions of these pumps use a hallsensor too.
The Hallsensor inside the DDC is what drives the microcontroller and the current on each phase, it is perfectly accurate. And this Hallsensor is what is read by the Aquacomputer poweradjust, its as precise as it can be.
So dont tell me a swissflow or other flowmeter is more precise. After all they are just flow measuring tools with a lot of inaccuracy, whereas the poweradjust shows the actual rpm of the pump.
He might not know what a hallsensor is therefore it's best to show it in an image:Quote:
Originally Posted by davidzo
http://upload.wikimedia.org/wikipedi...ensor_tach.gif
This is turning into a pissing match though. All in all Malik's testing methods were perfect. Anyone claiming a flow meter to be more accurate than a hallsensor needs to go back to engineering school.
It is possible though, that since the GTZ seems to be more restrictive, that there was a lack of flow in the loop to allow it to operate properly. This has been mentioned before. The only other possibility is that the Luna is a better block than the GTZ.
Now all I want is to see waterlogged's data everyone keeps talking about but which I can't find anywhere. this would help greatly in concluding this absurd pissing match affair.
Yes, me too. I just had a nice tall, stiff mug of Boca Java organic Peruvian-- great way to start a day. :DQuote:
Originally Posted by mcoffey