Exploring Layer Failure Modes With Wrong Starting Height

This image shows a 1000μm (1mm) long, 5-step staircase. The treads are 200μm square. Each step is 5μm (notionally) taller than the step to the right. Top right above the staircase is a series of 10 dots. These were used in an attempt to determine the probe height empirically:

As you can see (now I have my gimbal slide holder and the new microscope imager going) it has gone all blobby. Now, when initially trying to determine the probe height manually, I ran some GCODE that did 10 dots, each one 1μm lower than the previous one. Through the USB microscope, I could only make out the last 7 dots and so lowered the probe an extra 3μm, but as you can see I was in fact already in contact. This put the probe at least 3μm too low when commencing the first square, and things got horribly chewed up and uneven by the time we got to the 20μm tall square.

If you look at the top edge of the staircase, you'll see an indication of the change in step height. The random small blobs scattered around are where the probe lifted from the surface to recharge the probe. These may decrease if I get the height right.

Anyway, if I can't show you something going right, I can at least show you where I'm going wrong. If I write it down, it's science...

I'm going to have to take a break now to prepare for a μRepRap presentation at Everything Open in a week or so.

FYI Probe 9, 2.3μm layer height, 34μm line width, solid infill, lines drawn as segments not dots, 150 x 15μm segments (2.25mm total linear distance) deposited per probe "dip".