iCE40 Ring Oscillation Test Notes

Dependencies

• IceStorm tools
• yosys
• nextpnr

Install instructions for all of the above tools can be found on the IceStorm website.

Building

The default make target synthesizes the design and runs a timing analysis. The target prog programs an available iCEBreaker via USB.

Results

When clocking the FPGA at 24MHz, the resulting ring oscillation is still identifiable as a square wave.

By 111MHz, it's more of a triangle wave. But still stable and nearly rail to rail. This is the fastest clock speed that `icetime` approves.

Despite not passing the timing analysis, when the FPGA is clocked at 120MHz no problems seems to occur.

Even at 129MHz, stable ring oscillation can occur. I (Erik) thought that the first time I ran it I got a chaotic, aperiodic signal. But I haven't been able to reproduce this so I might have made a mistake. Still, I'll hold off on declaring this result official.

At faster clock speeds things start to go wrong. Usually there's still a periodic ring oscillation, just much slower than one would hope.