Semi-log graph paper
A logarithmic vertical axis against a linear horizontal one. Exponential growth and decay plot as straight lines.
Opens with these settings already loaded. Change anything you like, then download as PDF, PNG or SVG.
Why the lines bunch up
Within each decade the horizontal lines are spaced by the logarithm of 1 through 10, so the gap between 1 and 2 is much larger than between 9 and 10. That compression is the whole point: it lets a single sheet span several orders of magnitude, and it turns any exponential relationship into a straight line whose slope is the growth rate.
Choosing the number of decades
Count the orders of magnitude your data spans. Bacterial counts from 10² to 10⁸ need six decades; a capacitor discharge from 10 V to 0.1 V needs three. Using more decades than your data requires wastes the sheet and flattens the curve; using fewer means running off the top.
Semi-log versus log-log
Semi-log — log on one axis — straightens exponentials of the form y = a·bˣ. Log-log, with both axes logarithmic, straightens power laws of the form y = a·xⁿ. If your data is straight on semi-log it is exponential; if it is straight on log-log it is a power law. That diagnostic is often more useful than the plot itself.
Printing it accurately
This only measures correctly if your printer doesn't rescale the page. In the print dialog set Scale: 100% — not "Fit to page" or "Shrink oversized pages". If you're not sure it worked, print the calibration page, measure the ruler on it with a real ruler, and it will tell you exactly what your printer is doing to the page.