you are viewing a single comment's thread
view the rest of the comments
[–] 34 points 7 months ago (45 children)

Looks nice. Why they don't sell PCs with cooling like that? What are the downsides?

  • source
  • hideshow 45 child comments
  • [–] 95 points 7 months ago (13 children)

    I would guess that the low surface area would lead to problems. At first it would cool very well because of the huge thermal mass, but once it reaches thermal equilibrium the cooling would be quite weak.

  • source
  • parent
  • hideshow 13 child comments
  • [–] [S] 66 points 7 months ago (4 children)

    I'd also think moving your PC will rip your CPU right off the motherboard

  • source
  • parent
  • hideshow 4 child comments
  • [–] 13 points 7 months ago (2 children)

    So, you're saying that putting blocks of copper on everything in a PC will automatically shed unnecessary parts, building a more efficient system?

  • source
  • parent
  • hideshow 2 child comments
  • [–] 4 points 7 months ago (3 children)

    how long till it reaches thermal equilibrium? maybe it can endure a full load for an hour

  • source
  • parent
  • hideshow 3 child comments
  • [–] 10 points 7 months ago

    You're looking at about a half hour per kilogram of copper to raise it by 50 °C with 100W of heat.

    Actual delta from ambient to thermal limit will typically be a little higher than that, but so is processor wattage on mid-to-high performance CPUs, so I'm happy enough with that as a ballpark estimate.

    Someone else estimated that block as 4.5kg, so you've got something close to two and a half hours of cooling from an ambient start.

  • source
  • parent
  • [–] 43 points 7 months ago (13 children)

    Do you have any idea how expensive a solid block of copper that big is?

  • source
  • parent
  • hideshow 13 child comments
  • [–] 40 points 7 months ago (2 children)

    If that block is roughly 4.5cm x 4.5cm x 25cm then the volume of it is about 500cm³ which translates to 4.5kg of copper. At 11€/kg that makes about 50 euros.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 6 points 7 months ago (3 children)

    Copper isn't that bad. It's not cheap exactly, but it's probably going to cost what an expensive CPU cooler already would.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 17 points 7 months ago* (last edited 7 months ago) (2 children)

    I have a micro ATX case that itself is the cooler. Heatpipes transport the heat to the case walls and they have fins to increase surface area. It can handle up to 65 watt CPUs.

    It's not produced anymore. But with all the talk of the Gabecube I've been itching to make a new build with it. Unfortunately I have neither the money or the energy.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 14 points 7 months ago* (last edited 7 months ago)

    Weight, cost, and it's probably not effective for the long haul. The mass of a copper ingot like that will work like a heatsink, but it has a very low surface area for the energy it can absorb. So it'll heat up to a point that is uncomfortable for the CPU, then fail to radiate that energy out to the air effectively.

    As a test-bench temporary heatsink, this is actually kind of inspired. No fans, to fussy clips, just stack a copper brick on the CPU, run some benchmarks, and then turn it all off.

  • source
  • parent
  • [–] 12 points 7 months ago (7 children)

    Incredibly unwieldy. Real quick estimate of volume puts that at around 1.75kg of copper, so it wouldn't be possible to mount in a vertical PC case orientation (ie the majority of consumer PC cases) without significant (expensive) modifications to both the mobo socket mount and the case, else its weight would snap the motherboard, or just slowly flex it until traces failed.

    It may not even be able to be used vertically like that for very long or it will compress and damage the CPU / socket / mobo. Just as an example, the weight limit of the thermal solution (HSF/water chamber heatsink/etc) for socket LGA 1700 is 950g.

  • source
  • parent
  • hideshow 7 child comments
  • [–] 6 points 7 months ago* (last edited 7 months ago) (6 children)

    Real quick estimate of volume puts that at around 1.75kg of copper

    I assume it's at least ~5 cm × 5 cm × 15 cm. Given the mass density of copper, 8.96 g/cm^3 , its mass is at least 3.36 kg.

  • source
  • parent
  • hideshow 6 child comments
  • [–] 2 points 7 months ago (5 children)

    i find your estimate to be overed

  • source
  • parent
  • hideshow 5 child comments
  • [–] 1 point 7 months ago* (4 children)

    It begins with the question: How wide is the cpu?

    Based on that dimension, it's approximately 3× as tall.

  • source
  • parent
  • hideshow 4 child comments
  • [–] 3 points 7 months ago (2 children)

    (if you notice the space bar, i think it’s a tiny computer)

  • source
  • parent
  • hideshow 2 child comments
  • [–] 10 points 7 months ago (1 child)

    My guess is that will only work until it saturates with heat. Some liquid cooling setups are also like that, where the rad isn't capable of dissipating heat fast enough to prevent the whole thing from overheating, but it'll work fine for a while because the loop itself can absorb a bunch of heat before it stops being able to take any more. Then they probably blame the chip maker for running too hot even with liquid cooling when their liquid cooling setup is actually less effective than the stock cooler or their case has horrible airflow and would choke any size or number of rads. But their reservoir acts as a heat buffer, so it takes 30 mins to even realize that, but they've already concluded it works.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 3 points 7 months ago* (last edited 7 months ago)

    Copper is actually ~25-250X leas efficient at transferring heat than a heat pipe and convection is hundreds of times more efficient than radiation at transferring heat and the fins on a heat sink would have hundreds of times more surface area for dissipating heat all that is to say this might work but it would be orders of magnitude less efficient than a standard heat sink.

  • source
  • parent
  • [–] 2 points 3 months ago

    Despite the size i would be very surprised if this was even equivalent to a moderate performing air cooler. The reason normal coolers have all the fins is for more surface area and youre able to easily and effectively push lots of air through, pullung heat off those surfaces.

  • source
  • parent