How Many Setups? The Number That Breaks Most Quotes
Material you can weigh. Cycle time you can model. Setup count is the one number on a quote that comes from judgement, and it is the one that quietly decides whether the job makes money.
A part Zenith got badly wrong
A shop ran one of their own parts through Zenith's engine and then told me what it actually took on their floor. I am not naming them, but the numbers are theirs:
| What Zenith said | What the shop measured | |
|---|---|---|
| Setups | 1 | 5 separate milling operations |
| Setup time | 1.5 hours | 15 hours |
| Cycle time | 4.3 minutes a piece | 110 minutes a piece |
Ten times on setup. Twenty five times on cycle. If a shop had bid Zenith's number and won the job, they would have lost money on every piece, and they would not have found out until the second setup.
Why setup count is so hard to read off a print
A 2D drawing tells you the geometry. It does not tell you how the part gets held. Two parts with identical features can need one setup or four depending on whether the work is reachable from one orientation, whether there is a face square enough to clamp on, and what the shop has in the way of fixturing and fourth axis.
The honest signals a print does give you are narrow. It names a second face when a callout says FROM BOTTOM or FAR SIDE or BOTH SIDES. A genuinely turned part with holes across its axis needs a lathe and then a mill, because a lathe cannot drill across its own centreline. Beyond that, a program is guessing, and most of them guess one setup because one setup makes the quote look competitive.
What I changed
I did not pretend to fix the count. Fixing it properly needs measured setup data from more than one shop, and so far there is one. What changed is the silence.
A quote from Zenith now says on its face when the setup count is a default rather than a reading. If a part lands on one setup, the print never names a second face, and there are more than eight machined features on it, the sheet carries a line that says so in plain words: setup count is a default, not a reading.
On a simple plate bracket whose print puts every feature on the top face, that line does not appear, because there one setup is a conclusion. Same number, two different levels of confidence, and now the sheet tells them apart.
What to ask any quoting tool
- How many setups did you assume, and why? If it cannot answer the second half, the first half is a guess.
- What does it do when it is unsure? A tool that always returns a confident number is not more accurate. It is less honest.
- Can you check the arithmetic? Setup hours times a rate should be a line you can multiply yourself.
I would rather hand you a number with its weakest assumption labelled than a cleaner number you cannot audit. On a job you are about to bid, the assumption you cannot see is the one that costs you.
Send me a part you already ran
If you have a part on the floor with real setup and cycle numbers behind it, that is the most useful thing you can send me. I will run it, show you Zenith's number broken out line by line, and you tell me where it is wrong. That is the calibration pilot, and your first 10 prints are free.