Thermaltake Gets On The 3D Printing Bandwagon

We’re interested by a move from Thermaltake, a manufacturer of computer cases, fans, and power supplies. Thermaltake has released a computer case designed to be modded by those with a 3D printer. They released a set of models that fits the new case. These are all hosted on a service much like Thingiverse. So if you want a single SSD or a whole rack, print the model. Watercooling? There’s a model for that. In concept, it’s very cool.

We’re not certain how to feel about this. Our initial impression was that if Thermaltake is going to launch a case around 3D printing, they should at lease tune their printer and get some nice prints before they take the press photos. On our second pass we became intrigued. Is this a manufacturer cutting costs, crowd-sourcing design and engineering talent for free, or empowering the user? Arguably, a computer case is a great test bed for this kind of interaction.

Despite out skepticism, we’d like to see more manufacturers take this kind of contributing interest in 3d printing. If only to see where it goes. What other products do you think would benefit from this kind of, print the product you actually want model?

27 thoughts on “Thermaltake Gets On The 3D Printing Bandwagon

  1. A lot of computer case design is cargo-cult design. Vents and grilles and racks and cable throughs are thrown in without so much as a thought, even if and especially when the box costs hundreds of dollars.

    So you might as well leave it up to the buyer. They don’t know any better either, but at least there’s the Ikea effect where you think you’ve gotten your money worth because you’ve put your own effort into it.

    1. Also when it comes to liquid cooling, most people aren’t trying to pull massive OC out of their hardware – the get liquid cooling to make the machine quiet.

      There the cargo cult design becomes miserably obvious, because the biggest noisemakers in a computer aren’t the fans but the case. The liquid cooling often isn’t better – it’s simply heavy, which damps the vibrations from the fan motors and makes the noise profile go down. It also has unwanted side-effecs, like cooking the power regulators and chipset because there’s no more air flowing around at the motherboard surface, leading to erratic behaviour and crashing.

      The same effect could be had with a short plastic funnels over the heatsinks and 50 cent silicon rubber grommets. The effect is dramatically demonstrated by unscrewing the CPU fan and holding it in your hand in front of the heatsink – makes barely a whisper – but as soon as you mount it back on it’s making a hell of a racket.

      1. If you design it with some thought, you can use the fans to pull air from inside the case, have them well in the case (less noise) + everything looks neat on the outside AND use cardboard to direct airflow from outside over the VRM.
        This approach will cause “cold” air to blow over the VRM, not the warm/hot air from the heatsink, as most coolers blow air towards the motherboard.

        As for the noise – a lot of people forget that the fluid pump itself also makes noise, sometimes quite a lot (personal experience)…

        1. The problem is that most computer cases are so full of random holes and vents and gaps everywhere, that putting fans to “pull” won’t achieve the desired result.

          When people talk about “good airflow”, they think it means the computer case is made of chicken wire.

          Here in the Thermaltake case as well, it’s open on four sides and plastic on the fifth, which is completely unacceptable for both EMI/ESD and noise blocking, and the cooling will be sub-optimal because you can’t establish a proper through-flow.

        2. Considering that the power parts are usually rate much higher temperature than the rest of the system, they specifically done it that way. It make sense to cool the rest of the chips that are more heat sensitive *first* and leave the power supply etc near the exhaust. You don’t want to do it the other way around.

          Also note that there were a lot of research done on reliability and modeling done in the military and they recommend to keep junction temperatures below 105C (or a bit lower for the “IT” grade parts because of the aggressive way they spec their clock speeds). These type of temperature can still give you a long life time like 10+ (if not 20) years.
          105C is above boiling point of water.

          1. 1) 105°C is above BP of water only below a certain pressure, raise the pressure and the BP rises as well ;-)
            An alternative is to add stuff that drives the BP higer. Modern cars do a bit of both.
            2) Just because you can run a chip at 105°C junction temp, doesn’t mean you should. “Designing it that way” is just being lazy. Also, chips being that hot will inevitably result in the nearby electrolytic capacitors being hot, and those do not like heat at all.

          2. If you’re running your power regulators at 100 C, you’re cooking the electrolytics to hell. Even if the MTBF would suggest they last 6 years in those conditions, in reality about half the parts fail well before.

            (or 37%, or whatever cut-off point they’ve decided to use)

          3. Typically for a air cooled system at high load, you get +20C rise in temperatures from intake to exhaust. For ambient of 30-40C, that’s about 50-60C. That’s should still buy you a long time of operating life for the average 105C electrolytic caps. Most of the motherboards have switched over to solid state caps mostly due to newer power requirements for low voltage/high surge currents around the processor these days.

            I am simply stating that what the computer user thinks is hot is not necessarily mean it is same for components. It is silly to try to cool stuff to below 30-40C.

          4. “that’s about 50-60C. That’s should still buy you a long time of operating life for the average 105C electrolytic caps.”

            The 50-60 C is the ambient temperature around the component. The internal temperature under load is higher than that, and that’s the temperature that matters.

      2. A proper watercooling system is superior to a fan/heatsink combination and the fans (and HDDs if someone uses those anymore) _are_ the main noisemakers! In no water-cooled system I’ve made have the weight of the system anything to do with the silencing effect, the use of efficient and slow spinning large fans combined with self circulation of heated air is one important part of it, the shared cooling of all important components is another. And even when I started with water cooling experiments in the mid 90s making sure there were enough airflow for non-watercooled components was well known, the Japanese hobbyists were in the forefront then but the information were available.

        Silent fan/heatsinks have used your rubber grommets for at least 15 years BTW.

  2. I’d think that they didn’t get a higher-quality print for the photo because they want to show a typical result, not due to lazy assembly–or someone messed up and used an internal photo for press.

  3. The computer I’m using right now began life as an eMachines ET1161-05. Pretty much all that’s left original is the case. The cooling capability wasn’t up to handling a Phenom II 550 AM3 dual core, despite the large exhaust fan in the back.

    So I cut a hole in the side panel to mount a 4″ fan with a chrome wire grille salvaged off a dead PS. It blows right onto the CPU fan. Stays much cooler.

    What I see so often with case modding is people cutting *round* holes for the fans. Unless the fan has a simple cylindrical shroud, that will not allow maximum flow with minimal noise.

    I always carve the holes out to exactly fit the shape of the edge of the tapered shroud opening so the air can draw into the fan without restriction.

    3D printed case parts would make doing such matching fan hole shapes easier.

  4. Is this a manufacturer cutting costs, crowd-sourcing design and engineering talent for free, or empowering the user? None of the above, it’s just about getting some buzz. And it works, since you’re writing about it…

  5. I’m a little surprised that hackaday is not applauding them for encouraging moding of their cases in a new way. Seems a little cynical. Hack a day should reach out to them and offer some help to really make their kit ‘hackable’.

  6. Ah, as if these types of clowns haven’t taken all the fun out of case modding in the first place. How about making stuff for building smaller more efficient cases with no moving parts, longer lifespan and a sensible efficiency to watt ratio?

    1. You’ll need something thin and strong like sheet metal for those expansion slots area. Printed plastic that is thick enough for the strength would interfere with plug in cables etc. The metal case acting as ground and shield also helps for EMC.

      1. I’m not sure the part have to be exceptionally strong but using proper engineering using T bars between slots and a extra thick rigid casing above and below the cards should be strong enough as long as one is careful when working on the machine…
        To reduce electromagnetic radiation there are conductive graphite and copper sprays that works on most plastics. The Copper variant tend to stink though, not suitable for using inside a living space.

    2. Because I dont want a plastic computer?

      But having reference files available for drive bay covers? Or fan mounts? Or even for the manufacturer to offer ‘watercooling compatibility design files’ that can be modified for the customers particular components? This is why home 3d printing has a future.

Leave a Reply to Richard Cancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.