Why OD, ID, and wall are not three independent decisions, what a tighter tolerance actually costs, and the specification most prints leave out entirely. From a precision micro extrusion team that has been in Jaffrey, NH for over 50 years. Part one of two.
Most prints are written before anyone asks the extruder
Most extrusion programs begin when a drawing arrives. The drawing is usually complete, carefully dimensioned, and missing the information that would have made it manufacturable.
That is not a criticism of design engineers. A tubing print has to carry knowledge across a boundary, from a team that knows what the device must do to a team that knows what the polymer will do. It is almost always written before the second conversation has happened, because that is the order the schedule imposes.
After 50+ years of receiving these documents, we can usually tell within a page which prints will scale cleanly and which will generate change orders. This two-part guide is what separates them. Part 1 covers the geometry. Part two will cover everything on the print that is not a dimension.
Two sentences of intent change what we recommend
The most valuable line on a tubing print is often a note, not a dimension.
A wall thickness tells us what to make. A sentence telling us the tube has to track through tortuous anatomy without prolapsing, or has to deliver a coil without binding, or has to sit against neonatal tissue without trauma, tells us what to protect when the tradeoffs arrive. And the tradeoffs always arrive.
When we know the intent, we can tell you which of your dimensions has margin and which one is load bearing. When we have only the numbers, we have to treat all of them as equally sacred, which usually means the design gets less than it could have.
State the clinical task. Name the anatomy. Say what the tube is competing against inside the device.
The tolerance arithmetic that rejects good tubing
Outer diameter, inner diameter, and wall thickness are not three independent specifications. They are three views of the same geometry, related by simple arithmetic: OD minus twice the wall equals the ID.
Specify all three with tight tolerances and you have written a print that can reject good tubing. A tube can sit comfortably inside the OD and wall tolerances and land outside the ID tolerance, not because anything went wrong on the line, but because the tolerance arithmetic never allowed all three to be satisfied at once.
The practical guidance is to decide which two matter to the device and let the third follow. If the lumen is what the design depends on, hold the ID and the wall and let the OD carry the accumulation. If the tube has to pass through a fixed constraint, hold the OD.
Then tell your extrusion partner which of the two is load bearing, so the process gets built around protecting it.
Concentricity belongs on the drawing
A tube can hit its ID and its OD and still be thicker on one side than the other.
That tube passes inspection and behaves in the device like two different tubes: stiffer in one plane, quicker to kink in another, uneven where the wall is asked to carry pressure or follow a bend. The catheter ends up with a bias nobody designed in, and the team chasing it usually spends weeks looking at the wrong variable, because every number on the report reads correctly.
An average wall hides this completely. A wall that measures 0.0008 inches on one side and 0.0011 inches on the other reports an average that looks fine on paper. The thin side still kinks first.
Concentricity is the dimension that is invisible on a print and obvious in a device, and it is the one most prints omit. Put it on the drawing, with a tolerance, and the conversation about how it will be held can happen before the first lot rather than after it.
What a tighter tolerance actually costs
Tighter is not free, and tighter is not always better.
Every micron of tolerance tightening has a cost somewhere: in yield, in lead time, in the amount of process characterization required to hold it lot after lot. Some of those costs are worth paying many times over. A tolerance the device genuinely needs is one of the best investments in a program. A tolerance that was tightened because it felt safer is a tax paid on every lot for the life of the product.
The question worth asking before the spec locks is simple. What happens in the device if this dimension lands at the edge of its tolerance band rather than at nominal? If the honest answer is nothing, the tolerance has room. If the honest answer is that the device stops working, you have found the dimension the whole program should be built around, and you should say so in exactly those words.
The drawing is the cheapest place to solve the problem
Every problem in a micro extrusion program can be solved at some stage. The cost of solving it multiplies at every stage it survives.
A tolerance conversation costs an afternoon before release. It costs a change order during development. It costs a lot on hold during validation. It costs a supply interruption after launch. Nothing about the underlying engineering changes. Only the price of addressing it.
That is the argument for having the geometry conversation with your extrusion partner while the drawing is still in revision.
Part two will cover the rest of the print: how the dimensions will be measured, how to specify material and layers by behavior, and what the quality section needs to say before a quality team asks.
If you are writing or revising a microcatheter tubing print, we would welcome the conversation.
