EMC test passed in the first go
.
Officially compliant to EN55022 A, measured at TÜV-Nord.
That's not easily achieved with a naked board with huge dc/dc converters like this one.
Those who ever had to pass an emission tests can testify that.
Test receiver:
Chamber:
Test Setup, ATX power supply is in that box:
More detailed:
[
And i have some more overclocking figures for you guys.
Note: I expect that future versions will be more stable due to a few changes.
341 Mhz 1.2V 57.5W - passed long term tests with 0.2% hw-errors.
409 Mhz 1.2V 68.04W - hw-errors increase to 3%
448 Mhz 1.2V 74.4W - not stable
448 Mhz 1.25V 81.6W - unstable 50% hw-errors
448 Mhz 1.28V 84.016W - unstable 50% hw-errors
Burnin when you say "future versions", are you referring to the hardware or software?
I personally could wait for revised hardware that could run at a stable 448Mhz.
When i said future versions I meant the production version, done some improvements to the design based on the prototype.
If that enables us to have them running at 450mhz - more test required.
I will do more extensive tests tomorrow, when I've fitted the production heat sink (which arrived yesterday).
News for the Watercooling guys:The boards can now also be ordered with Toolheads Watercooling solution fitted.
Do you have an idea what causes the extra power consumption?
When chip gets hot it starts consuming more at the same frequency and voltage. Semiconductor basics.
Thank you for your attempt at contributing something useful. Please allow me to brush up
your knowledge on semiconductor basics a little:
The microscopic conductivity of a material, sigma, can be described in terms of the motion of electrons (or other charge carriers like holes):
sigma = n*e^2*tau/m'
where
n = carrier concentration
e = electron charge
tau = relaxation time (time between collisions)
m' = effective mass of the electron
In semiconductors, the carrier concentration increases with temperature, due to excitations across the band gap Eg. So n is proportional to exp(-Eg/2kT). The relaxation time tau is inversely proportional to temperature. As the exponential dependence of n dominates,
conductivity increases with temperature in semiconductors.
The conductivity of
metals however, linearly decreases with temperature. But that effect is not sufficient to explain the almost doubling of the power draw at 448MHz, compared to 282MHz - because that would imply that the chips were running twice as hot (not in Celsius, but in terms of absolute temperature K), so that would be 500+°C.
For more details, please refer to Charles Kittel,"Introduction to solid state physics", or any other book on basic solid state physics.
So, any more explanation attempts from your side?
Nice explanation, for
static semiconductor losses.
We're talking CMOS logic here, the
power increases linear with frequency and quadratic with voltage.
And the efficiency of the dc/dc converter decreases with increased load.
