标签: vaccum

This is just a quick update with regard to my Vetinari Clock project . One rather tasty feature of the clock is an interestingly-shaped vacuum tube mounted on the top. Sitting just under this tube I have a 16-element NeoPixel Ring from Adafruit.     These things are great. Each element contains super-bright red, green, and blue LEDs, and each LED has an associated 8-bit pulse-width modulated (PWM) controller. The elements are daisy-chained together, and you can control the whole thing using a single pin from your microcotroller.   If you take a peek at this video , you'll see some of the lighting effects I've been playing with:   iframe width="400" height="360" src="https://www.youtube.com/embed/Ls5OjYr8J-s?feature=player_embedded" frameborder="0" allowfullscreen/iframe   Unfortunately, my camera really doesn’t do this justice, possibly because the LEDs are so bright. The problems with the camera give you an appreciation for the tremendous range of brightness that can be accommodated by the human eye.   We start with a simple "breathing" effect in which all of the pixels fade up and down together. I'm currently using 20 steps from fully off to fully on, and vice versa. Also I'm using linear increments between steps, with a 0.1 second pause in the middle of the cycle (fully on) and again at the end of the cycle (fully off). If I decide to use this effect, I'll time it such that it takes exactly one second to perform a complete cycle, but at the moment I'm just approximating things.   As I say, I'm currently using a linear ramp as depicted by trace (a) in the illustration below. I've been wondering if other waveform envelopes would look any better, such as the ones shown in (b) and (c) below. I'm sure a lot of research has been done in this area. If you have any thoughts on this, please share them in the comments below.   I really wish you could see this in real life. The effect with two white pixels chasing each other around on a red background (i.e., all the other pixels are red) is pretty impressive. I also tried two red pixels on a bright white (full on) background, but the red ones were washed out. However, this does look good if I dim the white background LEDs down to a dim glow.   Another effect that looks very tasty is having a bunch of four white LEDs racing round, with the lead LED being fully on and the other three of diminishing rightness fading to a dim pink background.   The way this ring is positioned, there are two LEDs either side of the 3 O'clock, 6 O'clock, 9 O'clock, and 12 O'clock positions (with 6 O'clock being the one closest to the observer at the front of the clock.   As you'll see in the video, one effect starts with two white LEDs at the 6 O'clock position, then we illuminate additional LEDs to the left and right (clockwise and anticlockwise) until we reach the 12 O'clock position, then we turn them back off again until we end up back at our 6 O'clock starting point.   I also tried stating with two LEDs lighted either side of the 3 O'clock and 9 O'clock positions, and then turning additional LEDs on until we reach the 6 O'clock and 12 O'clock positions, but we would really need more LEDs to make this look effective.   So, that's the current state of play. As I say, this looks pretty awesome in the real world; I only wish I could capture a better video. I will certainly try to do so. In the meantime, can you think of any other effects I should try?

I'm now cursing myself for being a fool. I have let chances and opportunities slip between my fingers. I deserve to be berated soundly. But what is the cause of this gnashing of teeth and rending of garb? Well, let me explain …   As you may recall, my Inamorata Prognostication Engine is to be mounted in an antique (circa 1929) wooden radio cabinet. On top of this will be a box containing my Ultra-Macho Prognostication Engine . And on top of this will be an array of vacuum tubes.     The five tubes I've decided to use are incredibly tasty and rather large as these things go. They currently reside in a box on the floor of my office as shown below. Just to provide a sense of scale, the largest tube in the upper left-hand corner is about 12" (30 cm) tall.     In the fullness of time, I will be lighting these tubes using tri-colored LEDs. Do you recall this video showing me holding the large tube in my hands while a band of Adafruit's NeoPixels illuminate the tube from its base? I think you'll agree that the end result is rather spectacular. Now imagine all five of the above tubes lit up in the same fashion.   One of the problems I've left on the back-burner is how I'm going to attach these tubes to the brass panel that will be mounted on top of the Ultra-Macho Prognostication Engine. I knew at the back of my mind that there must once have been proper holders for the tubes, but I never really thought about it all that much. And, if I had thought about it, I would probably have assumed that it would be difficult to find a bunch of different holders.   This is the point when I start to slap myself on the head, because I never really looked closely at the bases of my tubes. I'd assumed that they would all be different on the basis that the tubes themselves look so different. The end result is that I only just discovered that the bases of four of the tubes are identical -- only the little rascal shown in the upper right-hand corner of the box has an alternative configuration.   So how did I come to this realization? Well, while I was at the Huntsville Hamfest earlier this year, I picked up all sorts of interesting things. The photo below shows my treasures spread out across our dining table.     In particular, note the three tubes in the far right-hand corner. These little scamps are currently sitting on the book shelves in the Pleasure Dome (my office) awaiting a suitable project. Now, these tubes are smaller than the ones I'll be using for Ultra-Macho Prognostication Engine, so I'd not really given them much thought. However, I happened to glance at them the other day as I was ambling back to my desk with a cup of coffee, and my attention was caught by the ceramic mounts holding these tubes.       I realized the tubes were held in place by a sort of bayonet fitting. When I took one of the tubes out, I saw it had four pins on the bottom. As near as I can "eyeball it" here in the office, these pins are each 3/16" in diameter, and the main metal collar on the base has an internal diameter of 1 13/16".   When I peeked at the tubes I'm going to use for the Ultra-Macho Prognostication Engine, I realized they had the same four pins. "Could it be?" I thought to myself. Well, blow me down, all four of my big tubes work in this holder, so ideally I need four of these bases -- all identical. And then there's the final tube as shown below.       In this case we have four pins, each 5/16" diameter, arranged in a sort of diamond formation (with an additional glass protuberance in the middle). The frustrating thing is that -- now I know what I'm looking for -- I bet I could have picked these tube holders up at the Huntsville Hamfest.   If all else fails, I'll have to wait until our EETimes Road Trip to Hamvention, but that's not until May 2015 and I'm an impatient man. So, do you have any ideas where one can find these sorts of tube holders?