Photos Part Two: The Accessories and PWM Adapter

Inside the center box we find a variety of things:

Inside the accessories sack we find...

The PWM adaptor, fan splitter, six fan mounts, a sack of screws and loopholes for the fan mounts, the hold down plate for mounting the heatsink, a thumbscrew to hold it to the heatsink, some diamond-like paste and six silicone strips.

You get a bunch of rubber standoffs in the smaller sack too, I'm not sure what to use them for.

The silicone strips go on the heatsink like so:

Next up is the PWM adapter, the first thing on my list was to see if it works. It does work, quite nicely in fact. The second thing was to put it on the oscilloscope and see how it works. Here are the results, first with a 25kHz PWM speed, second with a 3kHz PWM speed. Top graph is the PWM signal, bottom is the output to the fans. Both graphs are set to two volts per divider, the fan output graph's "zero" voltage level is well off the bottom of the screen.

Interesting business. The multiple traces on the bottom show different voltage curves depending on where in the power cycle the fan's motor was when the trace was laid down. Across the top you can see a couple where neither fan's motor fired during a full cycle and the voltage stays up. On the bottom you can see one PWM cycle where both motors fired, and even a dip right when the second motor fired.

The PWM adapter does a good job converting the PWM signal to something that the non-PWM fans can deal with, there aren't any big spikes or large dropouts to cause issues, and there were no unpleasant noises. The maximum fan speed does drop, just like the specs show. This implies a BJT type transistor is involved rather than a MOSFET. One kind of fun thing is that at a zero duty cycle the fans stop, this is rare even among PWM fans. They start up at around 3% duty cycle.

Now that I have tested the PWM adapter I'll tear it apart...

Nothing special here, just two PNP BJ-Transistors, four things calling themselves resistors and a small capacitor. That matches what we saw on the scope. The traces are fairly large so you could probably get away with using it for beefier fans, though I wouldn't recommend it without contacting Phanteks to ask about it first. Here are the ICs:

Now that I've torn that poor thing to shreds, let's look at the cooler.