An Interview with mCubed

The Silent Front
Viewing page 2 of 2 pages. Previous 1 2

A recent analysis of your Borg Heatpipes right here on SPCR showed that the 30W energy transfer per pipe that the specs indicate as their capability are a bit on the high side under normal circumstances. Under what test conditions did you arrive at that 30W figure?

The 30W figure is part of the specifications for the heatpipes that we currently carry. As your article indicated, the effectiveness of heatpipes always depends on the specific setup. More specifically the delta between the two ends of a heatpipe is one of the most important factors in understanding how a heatpipe works. Assuming the ambient temperature within a room is 25°C it makes a lot of difference whether the other end of my heatpipe is at a hard-drive (which shouldn't become warmer than 50°C to 55° C), a CPU (~80° C max.) or a GPU (often 110° C). The maximum delta in these cases is 25° C, 55°C and 85° C respectively.

Of course other factors such as the diameter of the heatpipe (6mm is better than 5mm) and the location of the cold end of the heatpipes (it's better to evenly distribute them over the enclosure) can also have an impact on the heatpipe's effectiveness.

An idea that we have been toying around with in the past months is to introduce a "power-factor" to better describe the effectiveness of heatpipe-based cooling solutions. As mentioned above a single value such as watts might be quite easy to derive but it can also be misleading. Any useful "power-factor" would include more parameters which should lead to a more objective measure of heatpipe effectiveness.

Also in some cases heatpipes are simply bent too much or at times even cut open (and resealed) which of course renders the heatpipe useless.

What does the internal structure of your heatpipes look like?

In general there are two types of heatpipes, the ones that can be installed horizontally and the ones that can be used vertically. This depends on the form and shape of the internal structure. In our case we're again focusing on flexibility as our heatpipes use an internal mesh which allows both kinds of mounting orientations.

In terms of performance, how much variance is there among the heatpipes? How does your quality assurance process work?

As with all mass-produced articles there's a certain variance amongst them. Our heatpipes are specified to be within a 10% to 15% margin from our requirements. This is also a value that fan manufacturers are, for example, using to specify the RPMs of their products.

With this being the case we rather prefer to be safe instead of sorry which is also the reason why we're always using 4-heatpipes to cool components such as the processor and GPU. Therefore even if all of the heatpipes were to exhibit an effectiveness of 20% below our requirements and we would fail to realize that during quality assurance then the system would still work safely. So you can consider this to be somewhat of a safety buffer.

We hand-test each and every heatpipe before it is installed in a system. Our tests are aimed at recognizing broken heatpipes and therefore we always check whether a heatpipe works or not. It is simply not feasable to test whether each and everyone lies within the specified margins. This is also the reason why we add that safety buffer as detailed above. However it is also vital to realize that we have never received a single RMA complaint where a potential malfunction could not be traced back to a wrong installation.

Do you manufacture the heatpipes yourself?

Our heatpipes are currently manufactured in Asia, according to our specific requirements and specifications. One of them is that all heatpipes offer a 90° bend which make them much easier to install in most enclosures. Additionally they're relatively easy to bend which also gives our customers an added degree of flexibility.

Which single component do you consider to be the toughest one to cool silently?

Each component within a modern computer system poses its unique challenges to cool it silently. I think in general we have seen a lot of progress, also from the big players, with regards to processors, graphics cards and hard-drives. For a long time optical drives have been amongst the biggest sources of noise within PCs as spinning a 12cm disc at >10000RPMs can't really be done quietly. With the recent rise of notebook sales people have started to somewhat focus on slimline-optical drives which require significantly less power and also tend to be quieter than their 5.25" cousins. This is also the main reason why we developed a silencing solution for slimline-opticals for our "HFX mini" enclosures. What we're providing is basically a vibration absorbing box which ensures a very quiet operation. Thanks to copper plates connected to heatsinks it also effectively dissipates the heat produced by those drives.

Close-up of a hard disk silencer box.

Given that I'm personally into Mini-ITX systems, can we expect to see a smaller version of the HFX mini that's catered at that market?

Well, I'm not really ready to discuss that topic right now. It's certainly something that we have thought about but we're not sure whether we'll go ahead and produce an enclosure aimed at that market.

However we're very close to releasing a new power-solution which we think will be very interesting to both silencing and Mini-ITX enthusiasts. We all know that the power-supply is one of the most difficult components to cool silently. This is also the reason why people have been focusing on these two-component solutions with the external brick-style AC-DC adapter and the internal DC-DC board in past years. The problem with these solutions is the fact that external bricks currently are currently maxed out at ~150W. In fact, most solutions available today only provide 80W to 120W of continuous power. So while the internal DC-DC adapters have been capable of significantly more than that (we're currently aware of several 200W solutions) users haven't been able to fully utilize these products due to the lack of potent external power-bricks.

Prototype of mCubed's 210W external power-brick

This directly leads me to our yet-to-be-named power-brick which will be able to provide a max. of 210W of continuous power! As you can see the unit is a bit bigger than most current AC-DC adapters and more importantly the wire connecting the unit to the internal DC-DC adaper (which carries ~17 Amps of 12V DC current) is significantly thicker than what you're seeing today.

We're currently planning to fully launch the final product within the next couple of weeks and rest assured that the readers on SPCR will be the first ones to get all the details.

Isn't the fact that the latest processors from Intel (I'm mainly thinking about the Core Duo here) use significantly less power that earlier products hurting your market?

Well, no, I'd actually say the exact opposite. Up to now most of our solutions have been based on AMD Athlon64 platforms due to the favourable performance-per-watt characteristic. With Intel now entering the market with a lot of momentum by offering extremely potent processors within a low thermal-budget we're suddenly able to more that double our potential market. From my point of view Intel introduced the perfect product at a perfect moment in time!

Where do you see the (after-)market for silent products in 3 years from now?

Phew, that's a tough one… I think in general the biggest difference will be that more people are going to be aware of the implications of silent computing. With computer systems such as HTPCs moving out of the office and into living rooms combined with the fact that running them 24/7 will be normal is going to change people's expectations towards them. While you may accept a relatively loud system while you're surfing the web and chatting it's going to be a lot more annoying during that oh-so-romantic scene in the latest Orlando Bloom movie.

Moreover the whole computer industry will be looking significantly more into power-saving features. Power bills will undoubtedly rise and become a bigger factor once people have multiple 24/7 sitting in their homes. So these days everyone is aiming at low-power solutions and I think we'll really some amazing benefits of the current research in the years to come. It's really all about consumer-awareness which is also the reason why we're pushing our dealers to add power-consumption figures (both in idle and load modes) to all of the systems which they are selling.

In general these developments will of course make the quest for silent computing easier for all of us. It's really a win-win situation on both the thermal and acoustical front.

Thanks again for taking the time out of your busy schedule to do this interview!

* * *

Discuss this article in the SPCR Forums.

Previous 1 2

The Silent Front - Article Index
Help support this site, buy from one of our affiliate retailers!