Our beloved laser cutter (affectionately called Mongo due to its immense size) has been a steady workhorse for completing member projects at Nova Labs. It is really a model Charley, named after an actual hurricane like all of Hurricane Lasers’ models.

One day recently, Mongo began tracing the outline that should have been cut instead of actually cutting. No smoke, no cutting. The red spotting LED was visible, but that really isn’t useful because it doesn’t cut anything.

We began troubleshooting with the great folks at Hurricane. There are evidently 3 different power supplies within Mongo, and the one dedicated to powering the 100 WATT tube wasn’t operational – no LEDs, no fan running. We removed the power supply and took it apart, finding a crispy thermistor.

Power supply with toasted thermistor.

Burnt components are always a bad thing. Hurricane promptly sent us a replacement, so we desoldered the leads and installed a new one.

Putting power supply back together after soldering in new thermistor. Hand model is Leon.

Mongo worked briefly, but stopped lasing after a few minutes. We suspected that more internal components in the power supply were damaged than were apparent from a visual inspection. Hurricane shipped us a new power supply, while we shipped this one back for repair. Once the new one was installed, we could see the appropriate LEDs lit in operation. Seeing the red LED and observing the fan running was the first step. Turning on the water chiller (it cools the laser tube while in operation) triggers a flow sensor and the yellow LED.

Laser power supply with top red LED lit (powered on) and bottom yellow LED (Chiller operation) lit.

The ultimate test was a cutting job. It was continuous enough that we could capture the green LED, which lights up when the laser is actually cutting.

In addition to power and chiller status LEDs, this shows the bottom green (Laser) LED, which is lit when the laser fires.

During the cutting, we saw listless airflow in the chamber. This led to a filter-cleaning exercise and later to mirror focus calibration and cleaning. Since cleaning the mirrors, we’ve been able to back down the power settings to almost HALF of what they were – the steady accumulation of slight amounts of deposited soot cut back the effective power.

Mongo was briefly not working the next day, so of course we wondered if we needed to start all over. It turns out the safety switch for detecting the state of the cabinet door is a hall-effect (magnetic) switch. It sometimes gets out of alignment, so was adjusted. We will be replacing this with a standard switch soon, so we can better tell when it contacts.

Back to work!