Cookie Jar's profile
Specialize in designing sensors and data acquisition systems that work in severe environments with minimal power.
Cookie Jar's contributions
- Goodbye 70-Watt ceramic metal halide lamps
- Metal halide bulbs have a smooth spectral response approaching that of an ideal black body giving them a CRI (Color Rendition Index) of 100. On the other hand, CFLs and LEDs generally have narrow and spikey spectral response which takes a great deal of proprietary calculations to come up with their advertised CRI figures. In my experience many bulbs with advertised CRIs of over 90 still muddy some colors to my eyes as well as the results from a camera when compared to hot body light sources like incandescent and metal halide. P.S. sorry about my inadvertent previous post.
- Cluttered desk, creative mind
- Suzanne, you do not have a cluttered desk. The little clutter you have could be tidied up in a few minutes. Personally, I just don't have the time left in my lifetime to clean up all my stuff. I must admit, that when I do clean up a small corner, to make passage less hazardous, it does feel good - memories and all. But in a few days, it's all back to normal. All the time spent cleaning up didn't make a dint.
- Overvoltage transients: The silent killer
- MOVs are less than ideal for protecting analog sensor lines coming into buildings. MOV breakpoints are not particularly sharp, resulting in significant voltage rise on a transient which can damage your other components as well as the MOVs themselves due to their now much higher power dissipation. The result is often leaky MOVs that mess up sensor values and are a headache to troubleshoot. In comparison, besides lower capacitance and faster speed, TVS diodes have a much sharper breakpoint giving much better component protection. The sharper breakpoint also results in less power dissipation than a comparable MOV during a transient. If severely overloaded a TVS diode will short, rather than leak. A direct lightning strike to your sensor will blow the TVS and much else apart. Both failure conditions are easy to troubleshoot. For added transient surge capacity, spark gaps are hard to beat and can effectively be used at building entrances, the wiring inductance reducing the transient power reaching the TVS. The third and most important additional defense is to keep your equipment within the cone of protection of a properly installed lightning rod system, so your components do not take a direct lightning hit. If the entire system is properly designed, not only will your system survive a direct lightning strike, but the data will be uncorrupted.
- Negative feedback in audio amplifiers: Why there is no such thing as too much
- Often an amplifier with more measurable distortion can sound better than one with less measurable distortion. The problem isn't with the hearing system but rather with the measuring technique. The lowest distortion figures are always achieved with (out of this world) pure resistive loads and these are the published specifications. When driving a speaker, an amplifier deals with reactive loads, generator effects as well as capacitance in crossovers and resonances. Feedback can aggravate the performance under reactive loads. Many is the speaker with a reputation for being hard on amplifiers. Feedback amplifiers as a rule hate capacitive loads and many power amplifiers become unstable with highly capacitive electrostatic speakers as a load. With the large variety of speaker characteristics, it's small wonder that amplifiers can sound so different. I have found that if distortion is measured with the actual speaker load used for listening comparisons, there is little disagreement between measured and heard. Many a low feedback tube amplifier outperforms a high feedback solid state amplifier when driving an actual speaker. With a few amplifiers I've measured a couple of orders higher distortion with a speaker load than with a resistive load. My favorite power amplifier shows little difference in distortion between a resistive load and an electrostatic speaker load. It pays to test an amplifier to ensure it behaves properly with the full range of reactive and resistive loads, right down to a short as some speakers can show very low impedances. It appears that very few are tested this way let alone optimized - too bad.
- Is China destroying the solar energy industry?
- There are some economists who would say that a product that's guaranteed for 1 year, failing after 2 years is a good thing. Keeping product going out the door to willing customers is the only way a manufacturer can survive. Making products that survive too long will kill your manufacturing business once your customer base is saturated. Through the years there have been a number of panty-hose manufacturers that produced panty-hose that would last a couple of years rather than the usual week at best. Consumer Reports liked their products. Unfortunately, none of them survived more than a year. Another way around this dilemma is to keep bringing out new product that makes your older product obsolete. This works for fashion and did for a long time with yearly car style changes. Our consumer electronics industry appears to be reaching the saturation point for new "features" that consumers are willing to spend their hard earned money on. PC sales are going down and it looks like smart phones are soon to follow. Once this happens, product becomes commodity and profitability goes down unless held up by combines or monopolies. Don't despair, electronics is getting into everything now. My 2007 Buick has no less than 108 fuses and circuit breakers. My 1962 Volvo had a total of six. Mind you, 100 years from now you'll be able to get your 1962 Volvo to run, but the Buick with all its processors will be a total loss when all the programs stored in 10 year Flash are corrupted. So we're all responsible for making products guaranteed to die, never to be repaired. It's basic economics.
- Rich voltage, poor voltage: My incandescent tale
- Your 240 volt supply comes from a center tapped transformer, with the center tap being neutral with black and red each supplying 120V relative to neutral (and 240 between black and red incoming). A good electrician will approximately split the loads evenly to both 120V feeds (which all feed your outlets coded black as hot, white as neutral and green as ground). If the neutral connection, which is also bonded to ground gets loose, things will stay near normal as long as the loads of both halves are in balance. But if there is a heavier load on say black, then the red to neutral voltage will rise just like a voltage divider and can burn out a lot of devices from bulbs to motors to electronics, while the black half suffers low voltage. Copper has the unnerving tendency to creep and loosen under a once tight screw. Repeated tightenings a few weeks apart will show just how much it creeps as the formerly tight screw needs to be advanced some more each time to tighten it. Eventually the work hardening of the copper reduces this creep, though connections will still tend to loosen through the years. I found that 20 years after installing our mains switch with 3 initial tightenings 2 weeks apart, I could read a 30C temperature rise using my new IR temperature sensor at one of the terminals, necessitating another tightening ritual. Warning, the incoming wires are live and can only be disconnected via the meter or pole switch. It's best to hire an electrician to check them. So you cannot always blame the utility.
- Airport body scanners: Are they hazardous?
- All new technology is deemed safe until eventually after many decades it is proved otherwise. We cannot expect a bottom line driven business to express anything but total confidence in the safety of their product. To have any reservations would be suicide from a business standpoint. Where the public has reservations, the industry will sponsor research, whose aim of course is to verify safety any way it can. He who pays the piper picks the tune. If given the option, many industries will try to keep under the radar. One example is asbestos reinforced concrete water mains whose use is widespread, as they do not rust. In most locations a water distiller will show significant asbestos fibers in the residue after distillation. Another example of secrecy over safety issues is the highly corrosive properties of the newer formulation of green pressure treated lumber. Fastener failure is rampant as decks collapse under human loads, but the industry and its watchdogs have been silent for over a decade. Of course in these paranoid times, security from terrorism is the trump card - safety be damned - "damn the torpedoes, full steam ahead."
- Inexpensive 3-D printer kit creates plastic components
- One question I would like to be addressed is the high cost of 3D printer consumables. It seems that many 3D printer companies use the same business model as the paper printer manufacturers who keep the cost of the printers low and the cost of ink as high as possible.
- Wise words from Einstein, Tesla, Spock, and others
- My first engineering boss, Jay Dickson used to say, "How come we never have enough time to do it right the first time, but we always have enough time to do it over again?"
- Audiophile heaven
- I'd like to put in my two bits on the solid state versus tube sound. Many people say a good tube amplifier sounds better than most solid state amplifiers. Then the engineers chime in saying it's ridiculous, just compare the harmonic and I.M. distortion figures. Figures never lie. The big mistake the engineers make is using a dummy resistor for a load. When hooked up to a real live speaker, the dozens of tube amps that I tested all showed remarkably less distortion than the transistorized amps at the speaker terminals. After all, a speaker is a motor generator set that easily makes the high gain, high feedback solid state amplifiers misbehave compared to low feedback tube amplifiers. The extreme, though real world load example is an electrostatic speaker with a highly capacitive load. While almost every tube amplifier can handle the the electrostatic speaker load, 95% of solid state amplifiers will go wildly unstable. Mind you there are a few well designed solid state amplifiers that can handle an electrostatic speaker well, but they are the rare exception, rather than the rule. Engineers generally design power amps with a pure resistive load, it's far simpler that way. Of course, using a pure resistive load for testing an amplifier yields orders better figures than using a real speaker. And for sales, figures are the name of the game. You can hardly expect an amplifier manufacturer to advertise real speaker load figures for a variety of speakers as they will inevitably be compared to resistive load figures. So unfortunately, you cannot judge an amplifier by its specs. To be accurate you have to test it with your actual speaker load. Instruments will give you a more accurate reading than your ears in this way.
Page 1 of 2Next >