First off, I'd like to thank you for your patience, and thoughtful discussion. On Most boards, I'd have been flamed down. This is why I love CD and FIRST so much. We can disagree, but we do so with facts and respect.
Now back to our regularly schedualed discussion
Quote:
Originally Posted by Al Skierkiewicz
Daniel,
I have to caution you, you are mixing apples and oranges here. Wattage is not the issue. 680 volts to exposed skin on students, volunteers and mentors is the issue as well as 680 volts to robot frame when the lamp breaks and the cathode touches the frame. I will grant you it is low current but I bet you wouldn't like it if you touched the exposed voltage. There is a difference between safe primary wiring practices and exposed high voltage.
|
I do not believe I am mixing apples and oranges here. The 12V directly from the battery has the power to weld it's wires together. Power is power, wattage is wattage. When you setp up the voltage, you step down the current. In the case above the step up transformer is a 50:1.
So lets look at the circuit current at the 2 different voltage levels. Lets assume it's a 12 watt lamp (for ease of figures)
When looking for current Ohms law is I=P/E
Thus I = 12watts/12volts ----- or 1 Amp
Now at the 680 volt level
I=12Watts/680Volts ----- or 0.0176 Amps
In a short circuit The wattage and the current both climb until the current passes the reset point of the breaker. If it shorted to the chassis could it hurt ... yes ... and it could hurt you just as much as the 12 volt battery shorted directly to the chassis. Since your breaker is on the 12V input to the device the 680V would never pass 1 amp because the 12 volt supply would have already passed 20 amps.
Quote:
Originally Posted by Al Skierkiewicz
On the camera, I am speaking from my everyday television experience. Small bright light sources are especially nasty to CCD pickups because the the AGC circuits and other electronics tricks used to get a nice looking picture are confused by bright spots in a normally lit frame of video. There are no optical filters before the pickup so all light reaches the face of the pickup where there are individual pixels for the three colors. Most light sources are fairly broad in their spectrum but LEDs can have some rather narrow bandwidths. A white super bright LED for instance has a rather high peak in the blue spectrum and a rather broad peak in the green. That being said, a small bright light produces a sharp image on the face of the pickup which for a variety of reasons, leaks light through the surface of the sensor and also reflects light on the back of the lens which then in turn sends it back to adjacent areas of the pick up. Besides the effect it has on the pickup, the circuitry used for making video outputs and the adjacent leakage paths, the signal then has a significant impact on white balance and local heating on the face of the sensor. All of these things aggravate the ability of the sensor to determine the presence of the correct color light. Add to that the 60 degree field of view of the lens and you are now producing problems that most teams will not be able to overcome.
|
OK. I'll propose a test. Take your digital camera (which has a CCD array just like the CMU camera) and find something in your house that has a bright LED (Your PC hard drive is a good example, but many household electronics have LEDs). Take a picture of it from beyond 1'. Did it wash out your digital camera? I'll bet you some donuts that it won't. Thats because the AGC in cameras works for each set of RGB CCD sets. So while 1 spot may be 'washed out' the rest of the picture is still clear. This would show up as a very small target with a very low confidence level.