The issue behind PR #713: CAM refuses to post a tap because CAD models the hole the normal way, at tap-drill size, while the gouge check sweeps the tap at full major diameter. The only way through was to remodel the hole at the thread's major diameter, which makes the CAD wrong.
Recurring agent failure issue in Camcraft (NeoTech)
What engineers report
A recurring agent failure problem in Camcraft (NeoTech) that mostly hits the mechanical designer.
Failure pattern: validation / failure
Analyst note
From the Grok pain-radar digests, Nov 10-11 2026 (Nov 10 8:41 AM ET through Nov 11 7:41 AM ET digests plus the Haitch editions). Verified live 2026-11-11: issue/PR read; title and state confirmed.
What it costs a team
Modeled annual cost
$30,000
For a team of 5
Prototype delay
3 days
Modeled schedule slip per occurrence
Time lost
240 min
Per engineer per month, about 1 incidents
How the cost is modeled
240 min/month × 5 engineers × 12 months = 240 hours × $125/hour = $30,000
A modeled estimate with medium confidence. Swap in your own team size and loaded rate to size it for your team.
Engineering tools that address it
No engineering tool profile lists this error yet. Browse tools that connect to Camcraft (NeoTech).
Evidence
More recurring errors in Camcraft (NeoTech)
- Contour a 40 x 40 x 30 box with a 4 mm cutter that has 12 mm flutes and a 6 mm shank. verify returns 'ok': true with 84 findings, every one marked shank_hit and blocks_post. The Fanuc post then refuses with 84 warnings. A 3D finish on a test STEP gives ok: true with 3,232 blocking findings. The hits are real, the summary flag is wrong.MCAD · impact 74
- The STEP importer never reads LENGTH_UNIT, so a metre file imports 1000 times too small and an inch file 25.4 times too small, with no message either way. Such a part generates, verifies and posts a correct-looking CAM program at the wrong size, and no import scale command exists to fix it after the fact.MCAD · impact 74
- A plain cylinder exported from CAD was refused as if a feature crossed the rotary seam because a refinement leaned the axis by tens of microradians on rounding noise.MCAD · impact 66
- CAD models a tapped hole at tap-drill size while the CAM sim swept the tap at full major diameter, so every normal tapped hole failed the post gate.MCAD · impact 66
- In a CAM tool, a STEP model imported coarser than the importer's 0.005 mm promise only left a line on stderr. Its facets sit inside the true surface on convex walls, so a real gouge up to the mesh error could pass verification. The fix flags the model, tightens gouge checks by the excess, and refuses operations once the excess eats the whole tolerance.MCAD · impact 66
- The MCP feature listing reports both flat caps of a cylinder as facing 'up'. Agents pick faces by that field for mates, lay-on-points and work offsets, so a bottom face read as 'up' can put a part on the wrong side with no refusal. A test had to pick the bottom cap by Z height to get around it.MCAD · impact 66
Sourced from public community reports. Costs and delays are modeled estimates, not measurements. Know a better source or a fix? Tell us at hello@productflo.io.