X-bit labs

Water-Cooling Systems Roundup: 16 Systems Reviewed! (page 3)

Category: Coolers

by Kirill "ALT-F13" Balalin

[ 03/30/2006 | 08:48 AM ]

Testbed Characteristics

• Any amount of heat is generated within a range of 10 to 300 (350) watts
• Consumed power is controlled with digital instruments
• The die temperature was monitored for different tests

We quickly discovered some considerable defects like the overall fragility of the testbed, inconvenient fastenings and other drawbacks. Some results we got during our experiments were downright incorrect and were not repeatable in a series of subsequent tests. The testbed was dismantled and the defects and errors were accounted for when we were designing the next version of the testbed.

Based on the components from the previous version, the Mark Evolution testbed was created. It is mainly intended to carry out experiments with a heater and measurement instruments. The testbed is universal.

Here are the points where the Evo-1 differs from the Mark 300:

• Higher wattage (600 watts against 300 watts)
• Wide power adjustment range; efficient programmable overheat protection scheme; LED-based temperature indication on a big display
• Digital serial connection to the sensor
• 0.5°C accuracy with a measurement step of 0.1°C
• The heater element has been redesigned; it is now based on two transistors, has better manageability and higher overall reliability
• A protection circuit was implemented to protect the heater’s transistors from damage in emergency situations like unallowable current in the base circuit, unallowable voltage in the base, disruption of the base, static potentials, a rupture in the control elements; it also filters random impulse noise
• High robustness of the fastenings of the testbed
• The dimensions of the heater’s sole are 35x35 millimeters
• The heater’s sole is polished and finished on a machine
• The temperature sensor is located in the center and deep in the sole
• The tested sample can be easily installed on the testbed; the fastenings are universal
• The testbed has its own power source for coolers and pumps with the standard voltages of 5 and 12 volts with switching capability
• The measurement tools have their own stable power sources and do not need batteries
• Inexpensive, but reliable measurement tools from UNI-T are in use
• A toroidal-core Tornado transformer ensures high current density and a high power reserve; it does not heat up in peak operating modes
• The testbed features a gaudy, cannibalism-style appearance that may be shocking to the public at closer examination.

The noise from water-cooling systems is evaluated by the ear: silent, low level, average level or noisy. You should be aware that all water-cooling kits include components with moving parts, the pump motor at least, so there will always be some noise, but we consider a system silent if we can’t hear its noise at a distance of half a meter.