Hehe. Well, only because we have been testing, and doing experiments with ESD for more than six months and we know that if ESD was to come in contact with a laser diode directly, the results would be catastrophic.Originally Posted by uk-laser
Saw right through me
We also know that what you showed here would not be possible with just plain old ESD and plain old laser diode. When we use plain old ESD and a plain old laser diode, I can "see the light" coming right out of a laser diode when we hit it with ESD, but I doubt it would be possible to "see the light" coming out of a collimated package projected on a target as you did. When using plain old ESD, the amount of light we see is really tiny.
What you have provided there (unwittingly or otherwise) is a rectifier and filter. So, basically you named your post correctly. This is really an ESD-powered laser system. The ESD itself is rectified, and then filtered by the capacitor. The capacitor served to store enough charge to stretch the pulse out for a relatively long time, thus allowing enough light to be generated and then projected.
There are other things that I can "see" with your video that other people might overlook. For example, even calling this ESD is a misnomer. What you have there is a spark generator, nothing more. But there is nothing that tells us the quality of that spark -- in other words, the voltage or impedance of the output or its character. I strongly suspect that what you have there is rather an AC generator, and I seriously doubt that the rise time is on the order of 0.7nS, as would be the case for "real" ESD.
What has saved your bacon is not only the 12 ohm resistor, TVS, and capacitor, but the relatively long red and black wires that lead to the laser. The inductance of that system is substantial enough that all of this "works" in favor of the laser diode.
Another thing we don't know is the latent damage to the laser itself. Just because it makes light, doesn't mean that it will keep making light for the rest of its prescribed lifetime...
It takes expensive instrumentation in order to characterize all of these things. Nevertheless, I will fully admit that this video raised my eyebrows, as I am sure it would for many people! My eyebrows are not raised often and for that, I commend you.
I think it's 20 or 30 volts. I can't remember the exact figure... Definitely good enough to protect a series string of laser diodes, but our philosophy is that ESD will come in direct contact with the leads of a laser diode itself and, as such, it is best to protect each laser diode with a separate LASORB.
I can say "yes and no". It would depend on HOW the laser diode driver were damaged and HOW tie "excessive current" were being applied. For example, as we showed in our video, and we also discussed in a powerpoint, there is a REALLY FAMOUS LASER COMPANY that produces a lot of lasers, but whose drivers are total crap. You can see the spikes if you have proper instrumentation and, this famous laser compay's laser diodes always die prematurely. If we put LASORB on their driver -- problem solved.
In any event, one thing I have been telling people is -- send me your driver and we will fully-characterize it using our equipment. If there are remaining problems we will let you know and prescribe solutions.
Not meaning to brag (Doc) but we have a 10GHz sampling scope, two separate real ESD guns, a Spiricon beam profiler and a whole lot of other equipment that most people don't have, so it's easy for us to perform high quality tests and really fully-qualify products. For best results, I would suggest everyone do this who wants to apply LASORB. So far the two companies who are using it have done just that.
Hehe. People who know us know that we are in it for long term relationships. Generally we are more concerned with customer support and satisfaction, than we are "making the sale". While it is true that we could sell more LASORB if we tell people to put them everywhere, the reality is that I doubt it would be needed. As I wrote in a previous post, the reason why laser diodes themselves are succeptible to ESD is because laser diodes themselves are really fast. The video that is on our web site now is actually an excerpt of a longer video we made for a different audience. Within the longer video we show the datasheets of some laser diodes. One laser diode we showed can react in 2 nanoseconds, and another one can react in 30 picoseconds!!! This means that the laser diode itself and "feel" and react to the ESD. Most semiconductor products (for example the op-amps and power transistors that drive the laser) can't do that...
Think of it this way. I am not going to mention any names, but I know someone very well who drives an SUV. When this unnamed personsees a speed bump in the road, he actually speeds up so he can't feel it. (Hehe, I laugh when everyone else slows down.) The reason he can't feel the speed bump is because he goes so fast, and the SUV is so heavy, it just can't react that quickly. This would be the case with the support circuitry in laser diode drivers or other equipment. By contrast, a laser diode is like a pebble meeting the speed bump. Going faster and faster won't help. The pebble will always feel the full brunt of the speed bump. So the laser diode can ramp up enough light to (what we call) "evaporate" the mirrors. This can actually be seen in the video we made where the mode profile was changed by the ESD.
Incidentally, a few more points. The video shows our ESD coming in contact with the laser diode. Probably most people will not recognize that the yellow clip-lead going to the other side of the diode is from the other side of the ESD gun, so yes indeed we are putting ESD directly across the laser diode itself. I think people might tend to overlook this, and assume that we were only shocking the laser diode driver package itself, and possibly overlook when the ESD sparks are going directly to the laser diode pins themselves (later in the video).
Also, all of our ESD guns are for the human body model that has 150pF and 330 ohm resistor (hence nearly 50 amps at 15kV), and not the Human Body Model that Steve mentioned in a few of his posts, which only put a mere 10 amps at 15kV... We always develop things to do more than they need to, and test to higher and more stringent standards.
Bill
Reply With Quote
Got to follow this idea for a moment, if no further... What if the cheap one had a Vf slightly higher than the main one, allowing normal operation? Parallel connection ought to work, no? So long as the cheap one takes all the flak, anyway.. (And for the moment I'm ignoring reverse bias protection).
