4. Damage Caused by Heavy Torque

(article continued from previous page)

As you have seen, I selected fairly thick workpieces to build the custom components for the PCB shear motor attachment. And, I used high-quality stainless steel screws and bolts.

Have you guessed the points of failure in the motor-driven shear? As you’re about to read, it ends in disaster.


Before building the foot-switch control circuit, I wired up the motor directly to a bench power supply for a few test runs. This gave me complete control over voltage and current limits, as well as an instant-off switch. The test runs consisted of:

In each case, I noted performance, noise, speed, and current consumption. It looked like a winner!

Although current consumption rose from 50 mA to 250 mA when testing it on a short-length of breadboard, the motorized shear appeared to handle it satisfactorily. It seemed ready to cut an actual panelized circuit board with no holes and a length of 2.8 inches.

The machine started out just fine. The PCB flexed slightly (as expected) when the blades made contact. The current climbed to 800 mA @ 24V. The motor slowed, but didn’t completely stall. The noise seemed to change pitch and I could sense forces building up until...


There was this horrible noise and the motor case started rotating. Clearly, something had broken, but I couldn’t tell what.

Incomplete Shear

A partially sheared PCB.

A partially sheared PCB.

The PCB shows only the beginning of shearing. It got about halfway across. That’s really lame.

Broken Screws

Motor mount separated from planetary gearbox.

Motor mount separated from planetary gearbox.

The motor case was rotating because it disconnected from the motor mount. Rather than the motor shaft rotating, the motor case rotated and the shaft stayed fixed.

Four 2 mm screws sheared off the motor mount.

Four 2 mm screws sheared off the motor mount.

The motor mount is held to the gearbox case with four tiny little 2 mm diameter screws. All four screws were sheared flush.

This is awful. Who makes a 2178:1 gearhead that is held together by four tiny screws? I mean, the motor mount bolt holes are twice that diameter. If anything should be designed to break, it should be the externally accessible bolts.

Obviously I used this motor beyond its specifications. But, it was being driven at the rated voltage and it didn’t stall -- it self destructed.

At that point in the inspection, it appeared that the motor could be made to work again if the screws were replaced. Sadly, screw extractors don’t seem to be available in such small sizes. And I can’t get a grip on the screws because they’re flush and slightly contaminated with gearbox grease. I dare not drill them out for fear of getting metal shavings into the gears.

I suppose I could simply epoxy the motor mount onto the gearbox case. But, that seems like such an amateurish repair.

Oh no! Twisted Shaft

At least I could take comfort in knowing that none of the parts I machined were damaged. That is, until I tried to remove the coupler from the gearbox shaft to make motor repairs.

A 0.050-inch hex key with a twisted end.

A 0.050-inch hex key with a twisted end.

My first clue that other damage had occurred was the set screws refused to budge -- even after applying enough force to twist the hex key. Of course, eventually the set screws became stripped from repeated attempts to loosen them.

Motor shaft flat twisted away from set screws.

Motor shaft flat twisted away from set screws.

Looking into the coupler, I saw that the flat portion of the motor shaft no longer lined up with the set screws that originally pressed against it. Were the set screws forced back into the brass coupler? Are the ends of the set screws sheared or smeared inside the hole?

It’s worse than that, it’s dead Jim!

Even after drilling out the set screws, the shaft still needed to be pounded out with a hammer and steel rod.

Damaged motor shaft twisted and grooved.

Damaged motor shaft twisted and grooved.

I could hardly believe my eyes. The stainless steel shaft is torsionally twisted with deep gouges from the set screws. Although the motor still functions, it is beyond my ability to restore to factory condition.

Kirk Charles points out that the setscrew gouges could have been avoided by using a clamp-style coupler (see McMaster-Carr #3084K31 or #9800T1). I have a somewhat similar example on this website, but the clamp holds the motor rather than the shaft. Still, the principle is the same. Depending on the situation, clamping can be more effective and less damaging than set screws.

Another way that the gouges could have been avoided would have been to use larger set screws. I was lazy when I selected #4-40 screws, instead of going upstairs to find that I had #10-32 set screws available.


Fortunately, there is no apparent damage to the Micro-Mark miniature shear. This is not surprising, given that the shear is capable of cutting through a 1/16 inch PCB when driven using the manual lever.

If I ever try this again, clearly it will require a larger motor with thicker output shaft. However, at this point, I’m going to shelve the project.