DDC3.2 Pump Top Shootout! 10 tops tested..

[Meatpuppet]

They have a stand at performance PCS that will do just that, so I'm assuming so.

The pump stand/bracket is made for the non-res version of this pump. It should be fairly intuitive to see that the reservoir bleeds through the res cap and not the feed tube. As such, orienting the reservoir on it's side would present some very apparent difficulties.

[Martinm210]

Some folks were asking about the side inlet performances, so I've completed testing on that.



So in the 1 GPM or high restriction range, the ranking is Alphacool, then Stock top, then Danger Den, then Koolance. In the average restriction range of 1.5 they are ranking is Koolance, Alphacool, Stock, Danger Den. And the low restriction 2 GPM range, the ranking is Koolance, Alphacool, Stock, Danger Den. In the end, I can't help but strongly recommend against using the side inlets for these tops, you end up loosing most of the performance advatage over the stock top. The Koolance top is the only one to provide any performance benefit and it only occurs in a very low restriction loop. In a high restriction loop, the stock top outperforms the Koolance top. The Alphacool is the only top to pretty much match or very very slightly exceed performance of the stock top without loosing significant performance in one area or another. The side inlets loose much of their performance advatage due to sharp and abrupt changes in conduit diameter. There is typically a very sharp reduction in cross section to accomodate the G1/4 barb to the side inlet port and another sharp elbow type loss where this side inlet port enters the main chamber.

One interesting note is the Alphacool reservoir top had the same barb orientation, yet again we had better performance with the reservoir and it's ability to minimize negative pressure at the inlet side of the pump. Same occurrence with the XSPC reservoir top vs the standard top. The reservoir tops are showing their advantage in minimizing inlet negative pressure.

Bottom line...don't use the side inlets unless you absolutely have to.

[Xilikon]

Thank you a lot for doing this testing. This helped me decide that I will stick to the stock top since it's not worth the expense (the pump will be used in a mosfets/chipset loop only so it's not a biggie with about 1.27 GPM from the estimator including the SF800).

To get the benefit of a aftermarket top while being able to tube in a restricted height situation, I would need to get a aftermarket top with top inlet and a good elbow.

[Martinm210]

Thank you a lot for doing this testing. This helped me decide that I will stick to the stock top since it's not worth the expense (the pump will be used in a mosfets/chipset loop only so it's not a biggie with about 1.27 GPM from the estimator including the SF800).

To get the benefit of a aftermarket top while being able to tube in a restricted height situation, I would need to get a aftermarket top with top inlet and a good elbow.

I should probalby run one more test with some elbows mounted on the center inlet. I think a derlin elbow with a male to male fitting will do better than any of the side inlet setups.

That was a test case someone mentioned before I started all this testing anyhow. I need to work up a quick G1/4 coupler first and I'll test with a DD delrin 90. Maybe try a T experiment also...

[Statts]

By 'side inlet' you mean a 90deg elbow on the inlet? as opposed to straight barbs like most people use?

[Statts]

Oh I see what you mean, the tops listed have both top AND side inlet ports, got ya! Just ignore my posts

[Boogerlad]

is the ddc 3.2 at 12.95 volts more efficient than the alphacool ap1510 at 24 volts?

[Martinm210]

Curves are now added into the estimator:

http://www.xtremesystems.org/forums/...d.php?t=151627

You can try out the AP1510 at different voltages and these pump tops and see for yourself..

[Vargher Warg]

Another interesting update! thanks Martin!

Can't wait to see the results from a test with the Alphacool and XSPC tops at 10w (DDC 3.1)

[Martinm210]

Another interesting update! thanks Martin!

Can't wait to see the results from a test with the Alphacool and XSPC tops at 10w (DDC 3.1)

Thanks, I'll be working on that pretty soon. I have a few other parts I want to test, then I'll be back on the 3.1 tests.

I should also have a couple more tops to test and throw into the mix.

I figured since a large number of folks use these pumps, I'd take my time and eventually get it all tested.

[twwen2]

Curves are now added into the estimator:

http://www.xtremesystems.org/forums/...d.php?t=151627

You can try out the AP1510 at different voltages and these pump tops and see for yourself..

Thanks heaps Martin, i was just gonna ask for that! I think everyone really appreciates the hard work you do around here

[GAM]

More very interesting information regarding side inlets on these tops. Hmmm.

I don't know whether it has been mentioned, included or scheduled in the next round of tests, but I'd be interested to see how the DD Pump Top Reservoir performs. Danger Den brought one recently although I don't have any yet.

Keep up the excellent work!

Cheers
GAM

[alo]

Thanks a lot Matrin!!!

You are doing an invaluable job! Once again thanks!

I'm joining to the waiting for a DDC3.1 results!

[Martinm210]

grat job Martin

what can you tell me about oclabs xptop? i'm trying to know the real power of this top compare to the others

Sorry I missed your question earlier.

I have been in contact with OClabs for a sample.

Early last week they said they were currently out of stock for about a week and would be sending me a sample of their new black color top.

Bitspower is also sending me one to test, and the good folks at http://www.watercoolingshop.com/ are sending me some G 3/8" barbs so I can finally get that EK G3/8" top tested.

So if all goes well, I'll have 3-4 more tops to add into this..

[Jupiler]

A quick note from me regarding the XSPC tops (both normal and with reservoir) and the barbs used.

I ordered 2 normal XSPC tops (8.50€/each) and the res top and got them yesterday.

As I have 6x 1/2" EK and 6x 1/2" DD barbs laying around, I wanted to check if they fitted nicely in the G 1/4" threads. Normally they should (at least, that's what I expected).

Unfortunately, I ran into problems when I tried the barbs.

Dangerden barbs :

I bought about 12 pieces from PTS last year and 6 of them are in use in my several WC rigs.
When I tried to twist the first barb into the thread, it blocked after only 1 full turn. I found this weird, took the barb back out and tried again. Same thing, wouldn't turn in.
Tried it in the other thread of the top but same problem, could do 2 turns and then it blocked.
Same thing with 4 other barbs and the 2 tops and res top, they simply wouldn't fit.

Only 1 barb would turn in into the threads of the 3 tops.
One other barb would fit into 1 top but not in the 2 others.
Weird, if you ask me.

EK barbs :

No problems with all 6 of them on the 3 tops, they fitted nicely.

I then decided to check if the DD barbs would fit into the Alphacool and PTS tops.
No problems whatsoever with all threads of the tops.

I don't know what to think of this.
Were the G1/4" threads on the tops not properly tapped in by XSPC, so they might give problems with certain barbs?

Or is there something wrong with the G1/4" thread from the DD fittings?
I don't see why there should be something wrong with the DD barbs, as they fit without problems in the Alphacool and PTS tops.

Think I'll be using the EK fittings with the tops.

[Martinm210]

A quick note from me regarding the XSPC tops (both normal and with reservoir) and the barbs used.

I ordered 2 normal XSPC tops (8.50€/each) and the res top and got them yesterday.

As I have 6x 1/2" EK and 6x 1/2" DD barbs laying around, I wanted to check if they fitted nicely in the G 1/4" threads. Normally they should (at least, that's what I expected).

Unfortunately, I ran into problems when I tried the barbs.

Dangerden barbs :

I bought about 12 pieces from PTS last year and 6 of them are in use in my several WC rigs.
When I tried to twist the first barb into the thread, it blocked after only 1 full turn. I found this weird, took the barb back out and tried again. Same thing, wouldn't turn in.
Tried it in the other thread of the top but same problem, could do 2 turns and then it blocked.
Same thing with 4 other barbs and the 2 tops and res top, they simply wouldn't fit.

Only 1 barb would turn in into the threads of the 3 tops.
One other barb would fit into 1 top but not in the 2 others.
Weird, if you ask me.

EK barbs :

No problems with all 6 of them on the 3 tops, they fitted nicely.

I then decided to check if the DD barbs would fit into the Alphacool and PTS tops.
No problems whatsoever with all threads of the tops.

I don't know what to think of this.
Were the G1/4" threads on the tops not properly tapped in by XSPC, so they might give problems with certain barbs?

Or is there something wrong with the G1/4" thread from the DD fittings?
I don't see why there should be something wrong with the DD barbs, as they fit without problems in the Alphacool and PTS tops.

Think I'll be using the EK fittings with the tops.

That is wierd , I had no trouble with the DD barbs because they have shorter threads. The EK barbs are longer so that doesn't quite make sense.

The only barbs I had trouble with was the D-tek barbs in the inlet, they were just a touch too long. I could have fixed that by sanding down the barb a few mm though.


BTW, I ran into the same barb length issue with the Koolance side inlet ports. Barbs that were too long conflicted with the plugs that came down from the top.

It's a good thing we have lots of barbs to choose from I guess

[Jupiler]

I see what the problem is.

I measured the thickness of the G1/4" threads of the barbs :

all EK barbs : between 12.94-12.98mm
4 DD barbs : between 13.15-13.19mm

1 DD barb had the same measurements like the EK, so 12.98mm, 1 other barb was 13.03mm (so tight one).

So it explains why 4 of the DD wouldn't fit.

[Martinm210]

I see what the problem is.

I measured the thickness of the G1/4" threads of the barbs :

all EK barbs : between 12.94-12.98mm
4 DD barbs : between 13.15-13.19mm

1 DD barb had the same measurements like the EK, so 12.98mm, 1 other barb was 13.03mm (so tight one).

So it explains why 4 of the DD wouldn't fit.

Interesting...I went ahead and measured what I had on hand and saw some similar variability although my DD barbs were all normal.

The DD original barbs I have here are 12.80mm,
DD FatBoy = 12.80-12.85
XSPC barbs = 12.70-12.75mm
EK = 12.80-12.85
DTek=13.05

So most of mine are in the 12.85mm range with the D-tek being a little bit bigger.

Anyhow, looks like you figured it out. There is some variability in the barb thread Outiside diameter.. The XSPC barbs fit fine, but they are a touch on the smaller side.

I know when threading with a tapping die, you have control over how tight the threading is, but I suspect any of these mass produced barbs are done on a swiss screw machine, so they can dial in whatever tolerance they want and it probably loosens up a touch with tooling wear. Anyhow I guess one G1/4 (BSPP 1/4"-19) is not necessarily equal to another..

[Jupiler]

Thanks Martin.

I still have some older EK barbs, with the smaller O-ring (the new ones have a bigger one).
I added another O-ring to them, so they fit nicely into the thread of the tops.

I don't have the expert tools for measuring the flow rate of the DDC tops.
I'm using a Digmesa FHKUC 70 flow meter (installed right before the PA120.3 inlet), and which is attached to an Aquaero.

WC loop :

Swiftech Micro Res > DDC 3.2 with PTS top > DTek Fuzion > EK8800GT block > PA120.3 rad > Res.

Flow rate, according to the Digmesa is around 5.07-5.10 liters/minute.


BTW :

Thread is "stuck".

[Xilikon]

Thanks Juplier, the crowd need this kind of informations to make a good decision.

[Jupiler]

No problem.

Well,
I just installed the XSPC top on my DDC3.2 (known as Laing Ultra here) and I must say that I'm impressed by it's performance.


Petras DDCT-01 Top : 5.07-5.10 litres/minute
XSPC Top : 5.68-5.71 litres/minute

Quite a difference, if you ask me.
In which way the flow rate, measured by the Digmesa, is accurate, I don't know.
But I notice that the water movement in the Micro Res is much higher with the XSPC top than with Petras DDCT top.

[InfamousChu]

Martin i know this is a ddc3.2 top comparison.. but what do you think about adding a D5/MCP655 to the charts.. just for comparison?

[Martinm210]

Martin i know this is a ddc3.2 top comparison.. but what do you think about adding a D5/MCP655 to the charts.. just for comparison?

Oh yeah...sure. Getting a little busy, but here that is with the D5 in white.

Here are some general flow differences you might expect:
D5 To DDC3.2 w/ XSPC Res Top Delta
High Restriction = .1 GPM (13% difference)
Average Restriction = .25 GPM (15% difference)
Low Restriction = .12 GPM (5% difference)

So the DDC 3.2 with XSPC Reservoir top is capable of producing 5 to 15% more flow rate than a stock D5.

[ArtosDracon]

Wow, truely impressed here martin. Have you considered designing a top of your own. I think you've got soem great ideas for improving the design. Especially the idea of the angled/tapered inlet. All of my flow experience is in airflow but, the theories are quite similar. If the inlet was tapered from 1/2"ID to match the internal diameter of the inlet barb down to the impeller inlet with a slight radius to get the water moving into the farthest point in the pump from the outlet it should maximize efficiency and increase the velocity of the the flow.

Heck, maybe I'll make one

[Sparda]

Sold the moded alpahcool top and it crappy reservoir at ebay.

Sayonara to modded Alphacool top and it crappy reservoir.


Sold one got two

modded inlet of Alphacool top. Believe it or not I use sand paper for metal with grid 40 to sand down those hole. Just bought my dremel yesterday.


normal xspc top inlet. The top impeller chamber has a rough quality built and has no xspc label since it an older revision.

This sucker keep stop randomly several month ago. I soo piss off and resolder all the coil , check the molex , shake/slap the pump...blablabla but it keep stop randomly. I slap those heatsink then suddenly they dont stop randomly anymore until today