MOVING A MACHINED BILL OF MATERIALS ONSHORE
By Joe Egbert, VP / General Manager, Libra Industries — Dayton, Ohio
Before we price a part, we read the drawing looking for a specific short list of things. It takes about 20 minutes, and it is the most useful 20 minutes in the whole quoting process, so it seems worth describing what is on the list.
We do a print review before quoting rather than after we have the work, because the same question costs almost nothing in a quote and a great deal after a purchase order is issued. At that point it is a change order, and everyone involved has to treat it like one.
What we look for
-
Requirements that migrated
A 32 micro-inch finish gets specified for a sealing bore, and by a later revision it sits in the general block and applies to the whole part. Same with a tolerance that was meant for one feature and has become the default for all of them. Nobody notices, because it changes nothing until it reaches a shop, at which point it can change the process, the tooling and occasionally the machine.
-
Positional tolerance on clearance holes
If the fastener has twenty thousandths of clearance, locating the hole to two thousandths is buying precision the assembly will never use — and it drives fixturing, which is where the cost actually lands. This one comes up more than anything else on the list.
-
Flatness on faces that get bolted down
A bolted joint pulls the face flat anyway. Specifying flatness as though the part sits unclamped pays for a condition that will not exist once it is installed.
-
Blanket tolerance blocks
The title block says plus or minus five thousandths on all three-place decimals because the drawing template has said so since it was made for a different part family. On a simple bracket that costs nothing. On a part with 40 features, it is real money, and usually 35 of those features would be fine much looser.
-
What is missing rather than what is there
Material condition left open. An inspection scope nobody named. Edge conditions, thread callouts, how the part should arrive. Those are the ones that turn up later as the conversation that starts with “these are technically to print, but”.
Why tolerance is the one worth talking about
This is the part that is hardest to see from outside a shop, and I did not really understand it myself until I was the one doing the estimating. Tolerance cost does not rise on a gradual curve. It moves in steps, and the steps are not evenly spaced.
Going from plus or minus five thousandths to one thousandth usually means more care on the same process. Better fixturing, a finishing pass, a slower feed, an added inspection step. Real cost, but incremental.
Going from one thousandth to two ten-thousandths usually means a different process altogether. Grinding rather than milling, a jig grinder rather than a machining center. It normally means temperature-controlled inspection, because at that scale the part changes size while you are handling it. The inspection can take longer than the cut, and the dimension may only be measurable in a climate-controlled room by someone who knows how to hold the part.
We have that capability here and we are glad to use it — jig and fine grinding, four CMMs in a climate-controlled lab, vision systems. It is worth having. It is also worth knowing that crossing that particular threshold is a step change rather than a small increment, because on a drawing the two numbers look about the same distance apart.
What happens when we ask
Often the answer is that the requirement has to stay, and there is a reason the drawing does not show — a downstream process, a customer specification, or a qualification baseline that cannot be reopened without a program-level decision. When we hear that, it is a complete answer, and we hold it without further discussion. That is a good outcome for the review, not a failed one.
Sometimes the answer is that nobody is quite sure where the number came from. That is also common and it is usually a drawing that has outlived the people who made the original decisions. It is worth knowing either way.
And occasionally the review costs us money, because a requirement comes off and the part gets cheaper. I would rather that than the alternative, which is winning the work and then having the conversation.
What it comes down to
At the end of the day, we just like getting it right. Some of that is pride, and some of it is that we’ve seen what happens when a shop guesses. Getting it right the first time is a lot cheaper for everybody than fixing it later. A wrong part costs you schedule and costs us credibility, and neither one is easy to get back.
So we ask. Most of the time it’s a quick answer, and we move on. Sometimes it saves you real money. Either way, you get a quote you can trust and a part that does what it’s supposed to do. Twenty minutes with a drawing is a pretty good trade for that.
Reach out via email to sales@libraind.com.
