Mazatrol Linked Mode on the Multiplex: What It Is, and How Shops Get Stuck in It

The Multiplex is a different animal from a single-turret lathe: two spindles, two turrets, built to machine both ends of a part in one setup. I spent 19 years at Mazak Corporation as an Applications Engineer working on machines like this, and the years since doing the same work independently. The feature that makes a Multiplex productive is the same one that confuses shops new to the machine: linked mode. Get a Multiplex running linked and balanced and it's a money-maker. Get stuck in linked mode at the wrong moment and the machine feels possessed.

What linked mode actually is

In linked mode, the two sides of the machine — left and right spindle, head 1 and head 2 — run the same program in lockstep. One program drives both heads, and the control keeps the two sides synchronized. That is the correct way to run balanced two-sided production: identical parts, both spindles turning, cycle times matched, no idle spindle sitting there costing you money.

It is also, by design, the wrong mode for almost everything that happens before production and everything that happens after something goes wrong. Setup is a one-side-at-a-time activity. So is proving out a program change, running a single-sided job, and recovering from an alarm or a crash. A shop that doesn't understand linked mode will fight the machine through all four of those, because the machine keeps insisting on doing everything twice, together.

How shops get stuck

The pattern I see is consistent. The machine ran production linked, something interrupted it mid-cycle — an alarm, an operator stop, a crash, a power event — and now the two sides are in different states. One side finished a cycle; the other didn't. One chuck has a part; the other doesn't. The control, still in linked mode, wants both sides to do the same next thing, and there is no "same next thing" that is correct for both.

The other way shops get stuck is simpler: they inherit the machine. Used Multiplexes arrive in shops with linked mode active from the last owner's production job, and the new shop can't even do basic single-side setup work until they understand what the machine is trying to do and how to get it out of that state deliberately.

The path out, described the way it actually works

I'm not going to give you a keystroke sequence here, and you should be suspicious of anyone who does without asking which control is on your machine. The reset path differs between control generations — a T-32-era Multiplex and a Matrix machine do not get there the same way, and a wrong sequence on a machine whose two sides are in mismatched states can move iron you didn't intend to move.

The logic, though, is constant:

1. Establish the true state of both sides first. Which spindle has a part, where each turret actually is, what each side was doing when it stopped. Not what the program says — what the machine says.
2. Get both sides to a known, safe, matching reference state in manual operation, one side at a time. Linked mode can't be exited cleanly from an unknown state; it can only be exited from a state you've verified.
3. Then take the machine out of linked operation using the control's mode/selection path for your generation, and confirm each side responds independently before you trust it.
4. Re-establish your cycle start conditions — program position, parts counters, chuck states — before resuming anything automatic.

A related twin-spindle quirk worth knowing while you're in there: shops running EIA/ISO programs on these machines run into the parts counter behavior driven by M198/M199, which forum threads on Multiplex and SQT machines document in detail. It's a small thing until your counter and your actual part count disagree at the end of a shift.

Where to stop and call

Here is the honest line. Exiting linked mode on a healthy machine in a known state is a procedure. Exiting it after a crash — a turret into a chuck, a transfer collision — is a recovery, and the difference matters. After a collision you don't know that turret alignment, spindle alignment, or transfer position are still true, and forcing the machine back into coordinated two-sided operation on unverified geometry can turn one crash into a worse one. If the machine stopped because metal hit metal, stop troubleshooting the mode and get someone who knows the machine to assess it. That includes me: I'll tell you plainly when what you have is beyond a programming and process problem.

If the machine is mechanically fine and simply stuck in linked operation at the wrong time, that's a training and familiarity gap, and it's fixable in a day on your floor.

When something on a Mazak lathe isn't working

I do this for a living. If you have a Multiplex stuck in linked mode, a twin-spindle machine your shop inherited and nobody fully understands, or a transfer process you don't trust, send me the part, the machine, and the control version. I'll tell you what's possible.

Thank you, Tom Herzog