home / lasers

Coherent Avia Thor

The Avia family from Coherent consists of a variety of high power DPSS lasers, with output powers approaching 50W at 532nm, or 30W at 355nm. I managed to score this one on ebay, which was a 10w 355nm model, unfortunately the head was damaged in shipping so the original resonator had to be scrapped. All was not lost, the interesting bits (600w of 808nm pump diodes, yag rods, frequency doubler/tripler crystals) survived, and I have managed to nurse a fair amount of coherent photons out of it.

Don't Use UPS

The head was shipped to me via UPS, and was packaged with a healthy layer of foam around it. However, when I got the head I noted it rattled, and after further inspection things didn't look to0 promising.

The laser in its box, and the first sign of big problems

A few shots of the carnage, it takes a lot of force to break those solder joints!

The mount for the second harmonic generator crystal, and the pile of chips I found in the bottom of the laser

It appears that the forward mount for the plate with all of the optics (as can be seen in the image at the top of the page, all of the optics save for the pump modules are mounted on a separate plate inside the laser head, which floats on 3 mounts) broke, which resulted in the 1/2" thick plate flapping around inside the head--which promptly broke just about every piece of glass in the laser. Fortunately, UPS admitted fault and gave me a refund for the purchase price.

The Survivors

After sifting through the carnage, I managed to salvage most of the interesting bits of the laser, most notably the 2 pump modules, and the crystals.

Pump Modules

The pump modules consist of 3x 100W 808nm 'donut' modules, stacked on top of each other, with a yag rod running through the core. I took one apart and photographed it in the name of science.

A shot of the entire pump module, and with one of the end caps removed

A side on shot of the end with the rod/flow diffuser removed, and of the raw diode 'donut' stack with its 15 diode bars in it (5 per layer).

Between the rod and the water channels in the end caps there is a small diffuser, which is different on the inlet/exit ends. I can only assume someone at Coherent spent a lot of time thinking about that one!

The Nd:YAG rod, and its flowtube (note the coated areas on the flowtube, there is a highly reflecting coating at 808nm over most of the flowtube, with small slits that are aligned with the emitters to help couple the pump power into the rod)

I also decided to draw the assembly in SolidWorks, you can download the model here


The frequency doubler and tripler crystals are both mounted in heater assemblies, and from the outside they appear identical. However, the tripler crystal is mounted in a X/Y stage, which allows the crystal to be moved with respect to the beam, so that after it wears out you can move the crystal over a little bit and avoid replacing it. This is necessary because at the power levels in this laser any given spot in the crystal is only good for about 100 hours of use! The Q-switch operates at 80MHz/10W (and appears to be a Neos 33041), however it also was damaged in shipping as well (the transducer got chipped).

The tripler assembly and the q-switch

We Can Rebuild It

Not to be discouraged by the initial setbacks, I started thinking about the best way to rebuild the laser. First, it was decided that trying to use the original cavity would be impossible, due to the large number of damaged optics and no means to replace them in a cost effective manner. That in mind, I started drawing pieces in SolidWorks, and thinking about the best way to mount everything to a breadboard, and ultimately came up with a pretty solid design to attach the pump modules to an optical breadboard which would allow me to experiment with different cavity configurations and whatnot. A few hours in the machine shop later, I had something that was starting to look like a laser :)

The mounting plate I made, and after getting a module plumbed in.

Its alive! After initial alignment I measured just over 40W on the ophir thermopile sensor seen at the right of the first image, with a pump current of 25A (about 800w of electrical power, a measly 5% efficiency)


I also added a q-switch (not shown in any pictures), and despite it only being able to blank the output down to about 10w I was able to get a decent amount of peak power from it which made it quite an effective cutting/drilling tool. Furthermore, unlike things like diode arrays it has a very nice tem00 beam profile which can be focused down to small spot sizes quite easily.
The first victim was a standard razor blade, from which I extracted a nice square hole a few mm on a side:

The cutting process, and the result.

Macro shots of the cut from before, and a hole that was ablated at a ~15degree angle, which is about 250um in diameter and goes all of the way through the blade.

Further Experiments

This is still a work in progress, hopefully in the coming weeks I will find time to:

More Information

This page is being updated as I make progress on the laser, check back often!
I can be contacted at contact@krazerlasers.com

home / lasers