William, that was an instructive story. It is often difficult to get the test setup correct, and when others are doing it, they may not have all the information they need, as you discovered. Thanks for the detailed account.
So very true, I often see people hit the "auto" button on a scope rather than configuring for what they need. I had another Engineer that would use single sweep over and over rather than using normal triggering. It was so frustrating to watch I wanted to push him aside and just reselect the trigger.
Your experience is exactly why my policy is to involve me in testing all power amps on the kw output range. I had a brilliant (and I do mean very talented) design engineer plug our brand new HP RF Network Analyzer directly into the output of a 800w rf amp. The analyzer input was 27dBm (1/2w) max! A careless moment can cost thousands of dollars and weeks of lost time.
My rules are: "Don't just do something. Just stand there! (And think.)", and "Double check your math before hitting power on, not after the smoke comes out".
Another, luckily somewhat uncommon, lapse is the use of an underwattage attenuators. Think! If you need to pad down a 1000w amp output by 3dB, why would you use a 1watt, 3dB attenuator? The answer is always, "I didn't think about that!" By they way, an exploding attenuator teaches a great lesson by itself.
I have long claimed that if I could get a nickle a pound for all the stuff my guys have destroyed I could be a retired millionaire today.
Have to agree with you. Too many engineers really do not know how to operate instruments. Ever watch a guy using Auto Trigger trying to capture an event or leaving the trigger set to down slope while looking for an event transitioning to high? Mant (most) have no clue about using A delayed B triggering, or even how to look at differential signals. Some days I think the technology is doomed!
I was testing a newly purchased speaker for the low cutoff. Speaker was for a folded corner horn I designed. I was not getting any signal out so I kept increasing the input. Then I thought to check for loose connections. Well the cone did not fly across the room as I have heard but but one thump and that was it for that speaker. It hurt in the wallet. I did successfully test its mate and also obtained a new mate. The folded corner horn which I found documented a magazine and referenced in my report got a B.
The one thing that I have found to be worse than those who don't have "a grasp on the obvious" is those who don't understrand about how much they don't have a grasp on. The fact is that most folks do not learn as much in six months as a good engineer learns in five years of school plus ten years of experience.
Of course it would have made a lot of sense to specify an input signal level for the testing. I did have some techs who seemed to delight in asking about t test conditions whenever I forgot to specify something. I always made a big deal of thanking them for catching my error, since that usually prevented disasters, they wound up making all of us look good.
He didn't say the test gear's default setting was maximum output... just that it was set to +20 dBm [100 mW] (Its maximum output? Perhaps the value required the last time it was used?) whereas what he needed was -20 dBm [0.01 mW].
His bad for not specifying test input power (not that I've *EVER* ommitted a key spec or dimension <grin>)! Four orders of magnitude generally will make a differance!
The entry on the corner horn speaker reminds me of a trip to a stereo store about 40 years ago. I set up an amp to drive a 1 watt signal into a (made-in-USA) famous folded horn speaker, and proved to my satisfaction that 98 or so dB/w figure was likely under-rated. Nearly put the front windows of the store into the parking lot. I'm not sure I've ever seen a stereo store owner move so fast in my life: likely approached Mach 1. I think he could tell who caused the problem, as I was the guy with the eyes like pie plates. Probably not going to let that happen again. The speaker survived, so far as I can tell. Lesson learned.
I smoked one of those HP rotary step attenuators with about 100 watts of VHF. I was lab testing a radio for spurs and had a coupler, a notch filter and this external attenuator hooked up. The radio was supposed to transmit through the coupler to the load, the coupler feed a notch filter, the attenuator protect the spectrum analyser input. I mixed my cables up. I noticed as I adjusted the attenuator to increase signal, it instead was going down. I lifted PTT just in time for "Ode of Ohmite". A $500 attenuator was smoked but the analyser was saved. I haven't blown up too much, save the expensive rare switching power supply I once was trying to fix went "Chernobyl" on me.
Are they robots or androids? We're not exactly sure. Each talking, gesturing Geminoid looks exactly like a real individual, starting with their creator, professor Hiroshi Ishiguro of Osaka University in Japan.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.