Heatkiller 3.0 LC and Heatkiller 3.0 LT Installment
October 3, 2009
Intro The Watercool Heatkiller 3.0 LT is regarded as the king of the mountain right now. And for good reason, it's a low restriction block that has tremendous popularity due to its thermal performance. Though the Heatkillers have long used channel based cooling (akin to the Supremes and Whitewater and others), the newest flagships, 3.0 LT and Cu, take it to a whole new level. There's more channels than ever providing more low-restriction surface area within just 2mm of the IHS than ever. It's less complicated than it sounds really--the base is 2mm thick and over the width of a typical IHS, it has ~52 microchannels that are 1.5mm deep into the base. What does that mean for you? You have water flowing within .5mm of the very bottom of the base and have a lot of surface area really, really close to the heatsource, your CPU. In addition to that, you also have an impingement plate meant to distribute flow evenly through the channels. It's a winning combination. It should be no surprise that the Heatkiller 3.0 LT provides the best as-tested performance of any block I've tested so far (though the review is of course worth a read--I show you how to get an even better performance below!). The Watercool Heatkiller 3.0 LC is the low cost sibling of the LT/Cu. Coming in some twenty dollars less expensive, it's an interesting block. It provides all the features of the LT/Cu, but does it with less intricacy and flair. There's no metal accent piece at the nozzles, a minor aesthetic difference--but there's also more important differences to performance. The number of microchannels is down roughly 1/3rd (reducing surface area by roughly 25%) and the impingement plate is down. What results is an even lower restriction block with the same fundamental performance as the LT/Cu, but just a tick or two hotter and tick or two less expensive. This test will focus on the performance of the blocks in general and over a large flowrate spectrum. Results from the installments of Roundup #2 will be compiled, as they're posted, into an Overall Comparison page. |
|
Thermal Testing Methodology/Specification Methodology My waterblock testing methodology has evolved over the past few months and I think it's finally at a resting point where I can start piling up test results rather than tweak the methodology (and thus preventing cross-comparisons). I use Dallas One Wire DS18B20 temperature probes at various points through my watercooling loop and at the air intake to measure temperatures, I've isolated the radiators so that the flowrate through them never changes, I use six different pump settings for each block, and use good testing practice by performing 5 mounts. Where applicable, I will also test various modifications to the blocks. These include testing various orientations and removing/adding various midplates, nozzles, dividers, etc. In some cases I will also modify the mounting system and present results from increased mounting pressure. For my waterblock tests, I'll perform 5 mounts of each configuration for every waterblock. The best configuration will then go on to be tested through the full flowrate spectrum. Specification
|
Thermal Test Results Now finally some results! First up, the individual configurations testing. |
 Here we can both the LC and LT prefering the Vertical orientation (barbs parallel to the socket latch) and the LT benefiting from having the plastic divider at the inlet removd (not the metal plate! Do not remove that!). Overall, the LT being roughly 1C better than the LC at this pumping power. In my observation of the blocks, I actually noticed a design flaw. The base, while it sits flush with the internals when you just place one on the other, does not sit flush with the internals once you tighten it down and form the bow. Basically, when you force the base to bend outward at the center, you're causing a gap between the injectors and the base. That means flow 'escapes' over the microchannels. While that lowers restriciton, it's also less efficient thermally. I decided to close up that gap with inexpensive, completely waterproof, removable, and easy-to-use 100% silicone caulk.  It doesn't have to be perfect, but using Q-Tips and silicone caulk, I formed a compresible and resilient 1mm layer on top of the impingement plate. That's what it looked like before I installed it. What's really interesting is what the performance looks like once it's installed! From here on out, I'll call it the LT+ (you'll see why).  I've included performance from three different baselines: 1) the best-as-tested LC config, 2) the best-as-tested LT config, 3) the LT's base with the LC's top (Frankenblock!). The showstopper is just how much performance improved from adding that little bit of silicone, that's the "LT+" on the chart. We took what is largely considered the best block on the market (and the best I've tested so far), and made it better, noticeably better. Flowrate went down, as predicted, but not by much--from 1.77GPM to 1.55GPM (more flow than the Supreme LT at the same pumping power). |
Specific Pumping Power Now that we've figured out what the best configuration is for each block, let's chart its performance over the entire flowrate spectrum.
|
Other Graphs
Here we can see both blocks showing benefit only up until dual DDCs ("High" pumping power) on my testbed. This remains mostly true even on a setup with 1/3rd the radiator power, with the results splitting between favoring single or dual DDC. Overall, the Heatkillers just don't benefit from lots of pumping power--both in terms of thermal performance and flowrates. There are still other benefits of running multiple pumps, namely redundancy, but doing so with Heatkillers just won't provide tangible performance improvements. |
Conclusion This is another data-centric comparison between blocks. The LC and the LT are fundamentally the same block, just the LC does a few things differently for the sake of reduced cost. In turn, you get reduced thermal performance (though a surprisingly small bump backward) and slightly higher flowrates. You also get a solid black delrin top, missing the metal accent piece. Like I did with the EK Supreme and Supreme LT, I want to point out the poor mounting system. This one is even more dinosauric than the EK, but at least doesn't use painful thumbnuts. That said, it is a worse mounting system than the EK when you consider how the mounting plates attach to the top (it's a pin-based holding mechanism that requires you to fully open up the block to swap mounting plates). Maybe they think it's clever or the best option aesthetically, but it's a nuisance. The overall mounting mechanism is the biggest drawback of the block and would be just about any block, even a poor performing block. This is something that needs to be updated in 3.5 or 4.0 or whatever comes next. The option of doing the silicone mod is really enticing as well--it's easy and nets a real performance boost that's awesome to see. It makes a great performing block even better. I'd suspect that doing it on the LC would result in a similar gain (bringing performance ahead of the stock LT!) and that presents a great value. Overall, both these blocks are winners--their performance is great and they're low restriction. They've earned their reputation.
|
