Why Cycle-Time Comparisons Fail: 75-Second Parts, 36-Second Pairs And 84-Second Baskets

Overview

Cleaning machine quotations are compared on cycle time more than on any other figure, and cycle time is the figure most often quoted on incompatible bases. One supplier states seconds per part, another seconds per basket, a third seconds per set of two components cleaned together. All three can be true, and comparing them directly produces a decision that looks arithmetic and is actually an error.

The confusion is not dishonesty. A basket machine genuinely cleans a basket, so the basket is the natural unit. A transfer line genuinely processes one workpiece per cycle, so the part is the natural unit. The error appears when the two numbers are placed in the same comparison column without conversion.

Big Bird Industrial (formerly Shimada Big Bird Industrial) has been building cleaning and deburring systems for automotive powertrain components since 2001, and the examples below are drawn from lines running both arrangements.

1. Three Units Sharing One Word

A cycle time is only meaningful together with the unit it refers to and the quantity processed in that unit. Stating one without the others leaves the number uninterpretable.

Part, Set And Basket

A per-part cycle processes one workpiece per indexing step. A per-set cycle processes a fixed group: an upper and lower cylinder block cleaned together on one pallet, for example, at 78 seconds per set. A per-basket cycle processes a basket of small components, which might be 20 gear shafts at once and might be a single large part.

The Same Machine Can Report All Three

A line cleaning cylinder heads two at a time at 36 seconds per cycle is a 36-second pair cycle and an 18-second part cycle. Both statements describe the same machine. A quotation that says only 36 seconds invites the reader to assume 36 seconds per part, which doubles the cost per unit in the comparison.

2. Why Baskets And Sets Exist At All

Basket and set arrangements are not shortcuts. They exist because the cleaning mechanism or the part geometry makes them the efficient choice.

Small Parts And Cavitation

Ultrasonic cleaning works by cavitation in a bath, which needs a volume of liquid larger than the parts. Small gear shafts are therefore cleaned in baskets, and the cycle is naturally per basket. Historical programmes on shaft and gear components show 30 seconds per basket for a CVT application, 60 seconds per basket for an AT application, and 84 seconds per basket for a later AT variant, all with particle limits around 600 micrometres.

Paired Components And Handling

Upper and lower cylinder blocks from the same engine are often machined and cleaned as a pair because they arrive on the same pallet and are assembled together. Cleaning them as a set removes a second handling cycle. The set cycle of 78 seconds covers two components, which is 39 seconds each, but the number that appears in the specification is 78.

3. Converting To A Common Basis

Conversion is simple once the quantities are known, and the quantities are the part of the quotation that suppliers most often omit.

The Arithmetic

Divide the cycle by the number of parts processed per cycle to get a per-part figure, then multiply by the annual volume to get machine hours. A basket cycle of 60 seconds holding 20 shafts is 3 seconds per part; a basket cycle of 60 seconds holding 4 larger shafts is 15 seconds per part. The cycle is identical and the capacity differs by a factor of five.

Where Conversion Stops Working

Conversion breaks when the loading time is not included in the cycle. A basket machine may quote the washing time and exclude the time to fill and empty the basket, which on a manual load is significant and on an automatic load is part of the station time. The comparable figure is the floor-to-floor time per part, including handling.

4. The Time That Does Not Appear

Two hidden times routinely change the ranking of two machines that appear close on paper.

Changeover

A line running several variants loses time at every changeover. If one machine changes over in minutes by loading a program and another requires fixture changes, the difference accumulates across a shift and can exceed the difference in cycle time.

Stoppages And Rework

A machine that meets its cycle but produces a higher false-reject rate spends the difference in re-cleaning or in re-inspection. The effective cycle is the quoted cycle divided by the fraction of parts that pass, which is a figure rarely quoted and always relevant.

5. What A Specification Should State

A cycle time specification that can be compared contains five items: the unit, the quantity per unit, whether handling is included, the changeover time and the expected first-pass acceptance rate.

With those five, two quotations become comparable by arithmetic. Without them, the comparison rewards whichever supplier chose the more flattering unit, which is not the same as choosing the better machine. A comparison that cannot be converted is not a comparison at all; it is a decision made on presentation, and the machine that wins may simply be the one that quoted in the most convenient unit.

6. Building A Comparison Table That Survives Scrutiny

The practical remedy is to build a comparison table in which every row has the same five columns filled, and to refuse to compare a row that is incomplete. That discipline is more useful than any single conversion formula, because it makes the missing information visible before the decision rather than after it.

A Worked Example

Take a machine quoting 36 seconds for a pair of cylinder heads against one quoting 18 seconds per single head. On the page the second looks identical. Filled in properly, the first is 36 seconds per pair, two parts, handling included, changeover by program, first-pass rate stated; the second is 18 seconds per part, one part, handling included, changeover requiring a fixture change. At equal volume the two are the same throughput, and the difference is entirely in changeover and in how the parts are presented.

Now compare a 75-second per-part shaft cycle against an 84-second per-basket cycle holding a basket of shafts. If the basket holds eight parts, the per-part equivalent is 10.5 seconds, which makes the basket machine look seven times faster. If it holds one large shaft, the per-part equivalent is 84 seconds and the ranking reverses. The basket content, not the cycle, decides the answer.

Where The Table Should Live

The comparison belongs in the technical specification, not in a purchasing spreadsheet, because the numbers have to survive the acceptance test. A cycle figure agreed in a spreadsheet and then renegotiated at commissioning is worse than no figure at all, since the machine has already been built around the wrong assumption. The same discipline applies to the cleanliness figure that accompanies it: a cycle and a milligrams limit written together can be verified together, while either one alone can be met by sacrificing the other.

Frequently Asked Questions

Why Do Suppliers Quote Per Basket Instead Of Per Part?

Because the machine genuinely processes a basket, and because the number of parts per basket depends on the customer part mix. Quoting per basket is honest as long as the basket capacity is stated.

Is A Faster Cycle Always Better?

No. A cycle that is short because a stage was removed will fail the cleanliness criterion, and the resulting rework costs more than the time saved. Cycle time is a constraint satisfied after the cleanliness target, not a target of its own.

How Do I Compare A 75-Second Part Cycle With An 84-Second Basket Cycle?

Multiply by the annual volume for the part cycle, and divide the basket cycle by the pieces per basket before multiplying by volume. Both then reduce to machine hours per year, which is comparable.

Does Handling Time Belong In The Cycle?

Yes, if the figure is meant to describe capacity. A wash-only figure is useful for comparing processes but not for comparing machines, and the two should not be mixed in one table.

Conclusion

Cycle time is a figure with a unit attached, and the unit is half the information. A 36-second cycle for two heads and a 36-second cycle for one head describe machines that differ by a factor of two, and the number alone cannot tell them apart.

The remedy is not a standard unit imposed on every supplier but a required set of five statements: unit, quantity, handling, changeover and first-pass rate. Once those are on the page, the arithmetic is trivial and the decision stops depending on which unit each supplier preferred.