Hey guys, I wanted to share another review which I felt would be useful to people. Introduction
The D5 pump from Laing/Xylem is easily the most popular pump in the DIY watercooling field, with nearly every brand selling their version of the pump (fixed 12V, 12 V + voltage control, Vario, PWM, 24 V + voltage control) but differentiating themselves with a sticker on the back and their custom tops (standalone or reservoir mount). Generally, however, the pump is identical among the various brands for the particular option used. A D5 PWM pump from, say, XSPC, is the same as a D5 PWM from Swiftech. Despite my preference for everything PWM generally, this is one case where I would go with a D5 Vario if I had the option. For whatever reason, Laing programmed the D5 PWM to run at an impeller speed corresponding to 60% PWM duty cycle when there is no PWM feedback signal provided, and it does not follow the PWM specifications Intel has provided for motherboard makers as part of the CPU_FAN and CPU_OPT PWM headers. The few PWM controllers that are out there also follow this generally, and this means that controlling a D5 PWM pump can be a hit or a miss.
EK sells the D5 PWM G2 by itself, and also the XTOP Revo standalone tops in acetal, plexi and a dual pump top version separately as well. They also sell a complete version with pump and top pre-installed and this is what I requested. As such, let’s take a look at the specs of the XTOP Revo D5 PWM (inc. pump) from the product page
: Technical specifications:
– Motor: Electronically commuted spherical motor
– Rated voltage: 12V DC
– Power consumption: 23W
– Maximum pressure head: 3.9m
– Maximum flow: 1500L/h
– Maximum system temperature: 60°C
– Materials: Stainless steel, PPS-GF40, EPDM O-rings, Aluminium oxide, hard coal
– Power connector: 4-Pin Molex- and 4-Pin PWM FAN connector Operational Regime:
– PWM duty cycle: ~ 20-100%
– Default behavior: Runs at 100% duty cycle when no PWM feedback signal is present Enclosed:
– EK-XTOP Revo D5 PWM (incl. EK-D5 PWM G2 pump motor)
– Mounting clip with rubber insert
– Self-adhesive mounting hole pattern sticker
– Mounting mechanism (incl. required Allen keys)
– Installation manual
Made in Slovenia – EU!
Pump motor made in Hungary – EU!
There are versions of this pump with the acetal top, plexi top and also the dual pump top version, and I got the acetal top version here. In hindsight the plexi top would have made more sense as we will find out soon, but functionally there is no difference between the acetal and plexi tops. The main change here compared to EK’s earlier D5 tops is a round top design compared to the square design that they and pretty much everyone else adopted. This means that the top can occupy less space relatively, and also there is now a smoother transition going from reservoirs with round tubes to a D5 pump with a round top as well. In fact, here’s a quick visual look at what I meant courtesy the EK XRES D5 reservoirs:
Back to the XTOP now. Another change is in the mounting where it is a lot smaller in footprint as well and this is something we will take a look at in more detail soon. The third big change is the pump comes with a “pump body” aftermarket part included with vibration dampening built-in whereas the previous such accessories from EK (and others) were more for aesthetics than anything else.
The D5 PWM G2 is, well, the same as a standard 12 V D5 PWM from Laing with some modifications to the PWM program to the motor. Instead of running at a 60% duty cycle equivalent with no PWM feedback signal (if you have not plugged in the PWM cable to an active PWM controller, for example), this runs at full speed which will make filling and bleeding a loop easier provided you have accounted for air in the loop to escape. Another change EK advertises is a larger range of RPM control with a different RPM response curve over the same PWM duty cycle of 20-100%, and of course this is something we will see for ourselves. I just happen to have an XSPC D5 PWM here for comparison.
Before we go for a detailed overview, a reminder that all EK products carry a limited 2-year manufacturer warranty including this one. Unboxing and Overview
EK likes to color code their product lineup, with a red-pink hybrid (I am not hue master!) reserved for the pumps and pump tops/accessories. You end up with a cube, the front of which has the company and product name, including the particular SKU (acetal version here, and there is no other option aside from black acetal for now). On the back are the specifications we saw earlier listed in multiple languages, and on two of the sides is contact information for the company. Yet another side has more information and confirms this is a pump top (black acetal) version with pump included. This side and the last one have seals which keeps the box inside secure with the outer sleeve, and also lets you know this has not been tampered with since shipment.
Break open the seals, and the sleeve cover slides off the inner box, which has a fairly simple design including the large EK logo. Opening the box, we see a hard copy of the installation manual, an online copy of which can be found here
Next to the manual is the accessory pouch that has the mounting hardware needed, and then some. All you need are the two Allen keys (2 and 2.5 mm), 4 M4x10 screws with the 2.5 mm hex head, 4 plastic washers and 2 M3x12 screws with the 2 mm hex head. The rest are extra thrown in, and I have no idea why that short screw is even there. Packaging error most likely. You also get a template for drilling the 4 screws that secure the pump to wherever you wish it to be installed. This template is on a self adhesive sticker so you can just stick it on, drill the 4 holes as marked and remove it.
A reminder that EK makes other products also, and then we get to the meat of things. The pump itself pre-assembled, and the optional rubber damper is bundled in with recently purchased pumps and henceforth. This damper has rubber pyramids, if you will, that hold the pump in place and do a good job dampening down any vibrations from the pump to the case or vice versa. The stock damper comes pre-installed, and we will take a look at that soon.
The pump comes in a plastic sheet with an EK seal on it, for further reassurance that you are getting a new product and also to keep any dust outside. Let’s remove the plastic now:
It’s a cute little thing if I say so myself. Coming in at 68.5 x 91.5 x 78 mm (LxBxH), it is a lot easier to fit in compared to most other standalone D5 pump setups. This has a tradeoff though- there is a single dedicated inlet and outlet port each whereas many aftermarket D5 pump tops offer options for one or both. Both ports were threaded perfectly and follow the de facto BSP G1/4″ standard as can be seen above. There are also 3 EK badges on this, with a protective cover to be peeled off as desired. To give a visual comparison with other tops/body covers, let’s take a look at the XTOP Revo D5 in the middle with the stock Laing D5 pump top on the left and the EK XTOP D5 clean plexi w/black cover on the right:
This is why I was saying that the XTOP Revo Plexi would have been better to do a like to like comparison with its predecessor, the XTOP D5 clean plexi top. Also, while not a direct comparison, here’s how the clean plexi top w/pump cover looks next to the Bitspower mod top and cover for the D5:
So we can see that the new XTOP Revo is smaller in most dimensions, has mounting a lot closer to the pump body and also occupies a smaller footprint, has a functional body cover as opposed to a purely aesthetic one, and has no optional inlet or outlet ports. Let’s take the pump top off:
The volute is similar in design to a lot of other D5 tops I have seen, but has subtle changes where the outlet neck is located- even compared to the older design:
So there may well be a slight performance change also here. What about the rest of the unit?
The pump innards are no different than a standard D5, and no complaints here. The magnetic rotor and the ceramic bearing are tried and tested, and performance at 12 V and full speed is no different than any other 12 V D5 in the same ~4500-4800 RPM range. Unsleeved PWM cable and full size The new thing here is the rubber damper and holder, and the “older” one is pre-installed out of the box. The holder is like an open clip which secures in via 2 screws on one side, and this is where the M3x12 screws come in. Once you are ready to screw this into a case, be sure to screw the holder as well. Open it to help release the pump and change the damper if need be. Speaking of which, let’s take a look at the new damper vs the older one:
The new damper is supposed to have a slightly worse grip on the pump, and less contact with the holder, but provides more vibration dampening. This is also the recommended damper if you are using the pump vertically as described in the manual. Just be sure to have the pyramids on the surface sticking up and towards the locking nut:
Do note, however, that it is best to use the regular/old damper if shipping a PC with this pump so you have the best grip on the pump and it does not loosen. The XTOP Revo can be used without the holder or damper, as the O-ring and the locking nut are all you need to secure the pump top in place:
This is useful if you are hooking up the pump to a reservoir directly via fittings, but in case you are concerned about the dead weight it is also possible to use the EK-UNI pump bracket for a 120 mm fan horizontal (140 mm version is in the plans too):
or the EK-UNI pump bracket (120 mm fan) vertical:
Whatever method of installation you choose, just make sure the inlet port is not upside down and coolant entering the pump is not fighting gravity. That’s basically it! Let’s take a look at how the pump top and the new PWM G2 pump perform. Performance Testing
Let’s examine the new PWM motor first as it is the easiest to do. I have here an XSPC D5 PWM (a G1, if you will) and I used the same XTOP Revo setup to test the RPM profile of this pump vs the new EK D5 PWM G2 pump as a function of PWM signal, and also when there is no PWM signal provided. I used an Aquacomputer Aquaero 6 XT (one that actually works with the Laing D5 PWM, and does not have the same issues as some Intel socket motherboards) for PWM control (each step was averaged for 10 minutes) and RPM readout of the pumps. Direct power from the PSU via MOLEX cables helped run the pumps. A Frozen-Q 400 mL cylinder reservoir and distilled water + quick disconnects helped for a quick swap of pumps, and there was nothing else in the loop.
EK has a chart up on the product page of the pump showing the RPM to PWM response curve, and my findings were pretty close to theirs- well within the +/- 10% standard deviation I allow at each data point. Both D5 PWM implementations begin around 4800 RPM (the numbers themselves can vary due to the top and restriction in the loop slightly, and also from sample to sample) at 100% PWM and are similar till about 80% PWM signal after which the EK D5 PWM G2 spins slower and slower till it hit 792 RPM at 10% PWM here compared to 1451 RPM with the XSPC D5 PWM. Both have a PWM duty cycle of 10-100% thus, but the EK D5 PWM G2 just has a larger range of RPM control as advertised which is nice to see. Do not be surprised if you find the stock Laing 5 PWM switching off completely at < 10% PWM signal though as this is intentional. Unlike most other PWM controlled devices that have a minimum operating speed and stick to it even till 1% signal, the Laing D5 was designed to switch off completely at low PWM signals so customers could save money on electronic relays that do the same when using these for water circulation in larger industrial loops. Bear in mind that these were never designed for PC watercooling as a primary purpose. The EK D5 PWM G2 on the other hand does not turn off, and runs at ~800 RPM all the way to 1% PWM.
Both RPM response curves are fairly linear, but the lower operating speeds on the EK D5 PWM G2 is not a major selling point for me. The D5 is quiet enough at ~2000-2500 RPM to the point where I don’t even bother slowing it down further. The main selling point for me here, however, is the behavior when there is no PWM feedback signal provided. I removed the PWM cable from the 4 pin PWM header, leaving behind the RPM sensor only. Powering on the pumps, there is no PWM signal provided and the XSPC D5 PWM ran at 2541 RPM- close to the 2562 I measured above corresponding to the 60% PWM duty cycle number. The EK D5 PWM G2 ran initially very slowly for 10 seconds or so (1200 RPM or so) but very quickly ramped up to 4792 RPM (full speed, corresponding to 100% PWM duty cycle). So yes, the stock Laing D5 PWM does spin slower with no PWM signal provided and the EK D5 PWM G2 takes care of that issue and makes it spin at full speed as I would expect PWM devices to behave. For this alone, I will recommend the EK D5 PWM G2 over other D5 PWM pumps and the linear control plus no more hassling over a dial on the back means I finally have a reason to not go with D5 Vario pumps.
Now for the XTOP Revo D5 pump top. I used the same reservoir as previously mentioned, but removed the quick disconnects and added in a Dwyer 490-1 wet-wet digital manometer across the pump inlet and outlet. I also added a calibrated flow meter in the loop, and a gate valve to add/remove flow restriction to the loop. All measurements were done with the loop filled and bled of air, and this was an identical setup to what I had used before when measuring the stock Laing D5 top and the XSPC D5 dual bay reservoir, so I did the same for the EK -XTOP Revo D5 and the EK-XTOP D5 (clean plexi) tops using the same XSPC D5 PWM pump used for the earlier measurements. So here is the updated P-Q comparison plot:
Excuse the ugly plot, poor grad students can’t afford expensive, fancy software generally. Focus on that blue plot- see how it starts off pretty well, and above the predecessor EK D5 top? It remains above till ~0.8 GPM after which the older top performs better. This is a bit of a mixed bag in that EK never promised better performance than the previous generation, and even so the XTOP Revo does perform better- especially in higher restriction loops- but when you consider that there is an error margin of 0.075 PSI from the manometer, and a ~0.05 GPM error margin from the flow meter then those differences don’t seem off much. In fact, in a typical loop, you really won’t notice a big gap between these two or even the other 2 tops compared. Only in very low restriction loops with the stock top show its weakness.
What about noise? I am glad you asked. I measured noise levels at full speed only as this is where any difference would be most noticeable, if any. Measurements were done in a 19 dBA anechoic chamber with the probe 6″ away from the pump and taken after air was bled reasonably enough (micro bubbles can take over a week to be fully gone, but they do not affect noise). The XSPC unit is a bay reservoir so it is not an apples to apples comparison here. Among the others that employed the same cylinder reservoir, the D5 on the Laing stock top came in at 42.1 dBA, the EK-XTOP D5 clean plexi came in at 42.4 dBA and the EK-XTOP Revo D5 came in at 40.6 dBA
. I did not notice a significant difference between the two dampers either (in a normal, horizontal layout as you see it in the pictures generally) but your mileage may vary. Also note that I did not use any decoupling nor did I screw this down to the chamber for obvious reasons so there was minor vibrations felt throughout which of course was less with the XTOP Revo featuring rubber dampers. If you use something like an Aquacomputer Shoggy Sandwich or similar other external decoupling for vibration dampening, the built-in dampers in the XTOP Revo may not be so effective. Either way, it is good to see that the dampers do work. Conclusions
The EK-XTOP Revo D5 PWM (inc. pump) costs $119.99 from the EK webstore to US customers and £80.99 (inc. VAT) from Overclockers UK to UK customers as of the date of this article. For those interested in the pump motor alone the EK D5 PWM G2 costs $89.99/£55.99, and for those interested in the top/holder alone the acetal and plexi versions both cost $43.99/£29.99 each. The pump/top/holder combination thus makes the most sense economically for those who want a complete working solution, and is reasonably priced given the competition and the features it offers. The pump by itself runs $10-15 more than a standard Laing D5 PWM sold by other brands, and comes in the same price territory as the Aquacomputer D5 Aquabus edition which is a popular choice for their Aquaero owners. This offers higher compatibility out of the box though, and works with Intel spec’d PWM fan headers also. The top on the other hand is a more interesting beast. Aesthetics are personal so I am fairly sure not everyone will like how this looks, and EK went with function as a priority here. The smaller footprint occupied overall will also make it more applicable relative to other D5 pumps in situation where people went with a DDC style pump before. The included dampers do a good job, and may well save you the cost and hassle of 3rd party decoupling tools while giving the option of fan/radiator mounting using their UNI mounting brackets (optional purchase).
Not everything is rosy though. Packaging could be improved as everything was squished together and there was no real protection to the pump on the sides. It relies on reseller packaging and we have seen before already how this can have a serious impact on moving parts. There is a single dedicated inlet port and outlet port, and it would have been nice to have a second outlet port. But the motor can be rotated easily to have the outlet port face the direction desired so this is not a deal breaker. The top performance was slightly lacking relative to the top it directly replaced, although this is not really going to have a significant practical effect.
The D5 PWM G2 pump by itself is a definite recommendation from me, and so much so that I will likely get one for my personal use as well. The XTOP Revo D5 + pump is also an attractive package overall. The top by itself is probably the part where people will have more reservations, especially those who like the Bitspower style mod tops and covers. Aesthetics being a personal thing, I leave it to you to decide if the top is worth it. The package as a whole get a recommendation from me.