Why an average diameter always undercounts
The face area of a log end is proportional to the square of its diameter. Double the diameter and you cover four times the wall. That squaring is what makes an average useless for counting a mixed pile: the small ends cover very little face each, so it takes a great many of them to fill their share of the wall, and the average diameter has already thrown away the information that they are small.
The calculator does both counts and puts them side by side. It takes each diameter band, works out the area that band covers, divides by the area of one end at that diameter, and adds up. Then it takes the same bands, works out an average diameter weighted by area share, and counts the whole wall as if every end were that size. The band count is never lower than the average count, and on a genuinely mixed pile it is meaningfully higher. When all your wood is one size the two agree exactly, which is the sanity check on the method.
The practical version: measure your pile before you order or cut anything. Sort into a handful of diameter bands, estimate what share of a laid-up square yard each band would cover, and use those. It takes twenty minutes and it changes the number of ends you need to buck.
Packing is a measured quantity, not a guess
The share of the wall face that is wood rather than mortar is the other lever, and it is worth measuring rather than assuming. Lay out a square yard of ends on the ground the way you intend to lay them up, with the bead spacing you plan, and work out how much of that square yard is wood. A tight pack with mixed sizes filling the gaps between the big rounds behaves very differently from a loose pack of uniform ends, and everything downstream — the log end count, the mortar, the cavity fill and the cords of wood — runs off that one percentage.
Stacked cords and solid wood are different volumes
A cord is 128 cubic feet of stack, and a stack is wood plus air. The wall, by contrast, is solid wood where the ends are. So the round wood you have to buy is the solid volume divided by whatever share of a stack is actually wood, and that share depends on how straight and uniform your pieces are. Crooked mixed hardwood stacks loose; straight uniform softwood stacks tight. It is an input here for the same reason as everything else on the page: it is a property of your material.
Beads and cavity are one decision
The space between the ends is divided between two mortar beads, one at each face, and a cavity between them. Widen the beads and the cavity narrows by exactly the same amount, because the wall thickness is fixed by the length of the ends. The calculator derives the cavity as the remainder for that reason: it cannot get out of step with the beads, and the two always add back to the total.
What this page will not tell you
No R-value, no species advice, no bead width, no moisture target, no verdict on the section. Cordwood masonry sits outside the prescriptive path in most adopted codes and normally reaches a permit through an alternative materials review, often with an engineer involved. Ask the building department before you buck several cords of wood to a fixed length.
The failure worth naming is shrinkage. Log ends put into a wall wetter than the wall will eventually be shrink away from the mortar as they dry, and the resulting gaps are both an air leak and a path for water. That is why drying the ends first is the thing every experienced builder talks about first, and it is also why this page will not tell you how dry is dry enough — that depends on your species, your climate and your assembly.
Questions people ask
How many log ends does a cordwood wall take?
It depends on the diameter mix far more than on anything else. On the defaults here — a 20 by 8 foot face with two small openings, 55 percent wood, and bands at 4, 6, 9 and 13 inches — the band count comes out well above the count you would get from the area-weighted average diameter. Both figures are printed side by side so you can see the size of the effect on your own pile.
Why does counting off an average diameter give the wrong answer?
Because end area goes with the square of the diameter. Small ends cover very little face each, so filling their share of the wall takes a great many of them, and an average diameter has already discarded the fact that they are small. The band count is never lower than the average count, and on a mixed pile it is noticeably higher. Where the pile is genuinely one size the two agree, which is the check that the method is behaving.
How much of a cordwood wall face is wood?
That is an input here, because it is a property of how you lay up and what sizes you have. The way to settle it is to lay a square yard of ends on the ground with the bead spacing you intend and measure how much of it is wood. Everything else on the page scales off that figure, so a guess in that field is a guess in every result.
How many cords of wood do I need to buck?
The solid wood in the wall is the wood face area times the length of the ends, which is the wall thickness. Round wood is sold stacked, and a stack is part air, so the cords you buy are that solid volume divided by whatever share of a stacked cord is really wood. That share is an input because it depends on how straight and uniform your pieces are, and the page prints the solid and stacked figures separately so the difference is visible.
Do the log ends need to be dry before they go in the wall?
Drying is what experienced cordwood builders talk about first, and this page has no schedule to offer because it depends on species, size, climate and the assembly. What is worth stating plainly is the failure mode: ends laid up wetter than the wall will eventually be shrink away from the mortar as they dry, which leaves gaps that are both an air path and a water path. How dry, and how long that takes, is a question for someone who has built in your species and your climate.