Thinking_J

's profile
image
VP

Design products for many industries


Thinking_J

's contributions
  • 08.23.2016
  • Tesla's Autopilot: setting expectations ... or not
  • Brian, All this over the use of "auto" (autopilot)? Rehashing Consumer Reports attacks on using this terminology ? .. you can do better. A managed expectation cannot hope to revolve around the usage of one word. No matter what name given the technology, someone will be attempt to demand a standard of performance that cannot be achieved -ever-.. allowing some lawyer to "test" their word usage in the courts. I understand the value of managing expectations. But that is a goal that can only be applied to small groups or individuals.. not a realistic goal in large groups. Certainly you can bet Tesla's company didn't I certainly wouldn't assume Mr Brown died because of the term "autopilot" was used. Why would anyone think buying a Tesla would equate to 100% safety? This would be hubris. This doesn't happen on trains or airliners - with a LOT more effort, in a much more controlled environment and with far more resources available to prevent accidents.. why would anyone think it is possible to prevent this in any individually owned car? Of course there will be deaths in autonomous or semi-autonomous cars. No matter how far the technology advances. It will just be fewer deaths than would happen with those that can't understand the meaning of words - in the driver's seat.
  • 07.19.2016
  • What is the resonant frequency of a cavity?: Rule of Thumb #30
  • A (via) stitch in time (or power planes) saves.. what? time? (or was that a lot of pain?) A direct variation from the examples are the created cavities in a metal enclosures and shielding. At that point you should be aware of the impact of micro deflections (mechanical vibration) impacting the cavity resonance and another issue - modulation via stray capacitance between the cavity/shielding surfaces and the circuitry. (and the fact ceramic caps are piezoelectric in nature) Helicopters... with all their vibrations.. a great environment for re-discovering this stuff.
  • 07.08.2016
  • The first killer app for 5G wireless may not even be mobile
  • Larry, While I agree with the concerns you have expressed. I agree the frequencies in question do make sense for fixed applications not mobile. I question how our industry is using term "5G" (not a reflection on you). "5G" is a term which evolved from cell phone marketing hype (no legal definition to "Generation" available in the US for 1,2,3G).. except what was defined "after the fact". (2, 2.5, 3, 3.5, 3.75, 4 now 5G?) Marketing people love purposely vague terms. (new and improved!) I certainly don't see "generations" being used (or defined) for Wifi, VHF, microwave, etc.. communications. At least not on any comparable scale. As a consequence, "5G" (or any "G") for most of the populace, is still equated to a mobile cell generation of radio technology. Regardless of recently defined standards within the electronics industry. Witness how Apple was sued by the Australian gov for using the term "4G" (different definitions for "4G" around the world) In other words.. applying these frequencies/channels/ranges (5G cell standards) to fixed internet access.. equates to new alternatives to Wifi, microwave links, etc... but it is no longer a valid "mobile cell" communications standard to be described in "generations". Words are important. As leaders in technology, we should be mindful of the path we take in selecting descriptors, and of likely confusion it may create later. An additional responsibility for engineers in an age where tech terms we create are quickly adopted (and misused) by marketers and the general public. I always thought we (engineers) should have never adopted "Generations" for describing anything. The term is simply to vague when used by competing companies. When a incremental improvement deserving of a "generation" title?
  • 06.06.2016
  • A plea for one connector, one cable, one display
  • I am impressed you were able to get this setup to work. But to your main argument... I don't see anyone expecting their 1992 Honda to park itself. Yet it is not an uncommon feature on many new cars. I don't see car standards that make it easy for me to know how to work everything in the car when I rent one during my travels. I have to read the car's manual before I can leave the parking lot (or risk an accident when it starts to rain as I leave the airport). It seems like a reasonable request. Standards for how things are used by humans. When the rate of changes in technology continue to be high, I don't think it is going to happen. It comes down to allowing innovation to prove itself in the market place vs Confusing and frustrating people that were perfectly happy with prior interface. (why do we need another GUI or a different way to turn on the windshield wipers?) I am sympathetic to the frustrations of poor documentation or worse. However, I seriously doubt our concerns will lead to any improvements. Only the voting dollars by the masses will be taken seriously. But we can hope someone is listening. I think most will be happy with 2 external monitors and a single display port connection without requiring a docking station. I am waiting for 4K and WITH additional color resolution bits (HDR).. I hate the "banding" I see in scenes with fog or clouds (film noir) where the extra resolution of color is required to keep the banding from becoming distracting. What are the chances of this being supported with your present setup? Slim.
  • 05.23.2016
  • Teardown: A tiny drone and its controller
  • instructions on these are typically.. dismal. The pushing in of the joysticks typically changes the sensitivity of the controls and sometime the range of control (sometimes the same thing). The video subsystems I have seen on these very small products are often extracted from rejects in cell phone manufacturing. camera, video encoder, micro SD card interface in one piece. Like used in the video key fobs .. ( $5 each) Large percentage of these units are broken from the start. Binding motors is the most common issue. Who is going to bother to demand a replacement? We ended up buying a bag of replacement motors..
  • 12.15.2009
  • Burn-in, burn-in: dc inferno
  • Still, to be fair, I remember a number of projects with crazy schedules that were put in place for this "a time of war", and we were all ready to commit to what ever it took to get the job done. Sad that it takes a war to get that level of unity and commitment.
  • 12.15.2009
  • Burn-in, burn-in: dc inferno
  • The moral of the story? I can think of a couple: - No substitute for extended testing of the final product in critical applications. (We used 8 weeks at elevated temps on samples from every production lot) - "haste makes waste" (3 months of 80 hour weeks?!).. that is just asking for trouble. This story was as much a management failure, as a engineering one.
  • 12.15.2009
  • Burn-in, burn-in: dc inferno
  • Manufacturer didn't specify exclusively for AC applications either..... or short duration DC applications..
  • 12.15.2009
  • Burn-in, burn-in: dc inferno
  • Ouch. I haven't seen a spec on optocoupler for addressing this issue either. Isolation resistance.. is often spec'd.. with a given test voltage. From there you can determine some level of leakage - that may help/apply. With each component there are often dozens (if not more) characteristics that are not spec'd. If anyone thinks they can eliminate ALL assumptions .. they are only fooling themselves. Spec sheets are a starting point... sure.. and one shouldn't make any assumptions beyond the specs. But often one cannot get the additional info regardless of the effort applied. I often come across the need to specify a characteristic .. with no one at the factory aware of their product's performance on the item in question. Even if you go to the trouble of characterizing the component. The manufacturer will not guarantee the characteristic in question will not change dramatically with the next production lot. Result: All of us are forced to make assumptions ...... And sometimes they bite us. Ever try to spec the level of piezoelectric effect created by your ceramic capacitors during vibration of the product? ( creating another noise source in the signal chain). Or try to get specs on the expected value shifts created by ESD discharges..on your resistors? Unless you have no budgetary and time restrictions.. you will be making some assumptions. Still, it possible .. we should try - if possible - to allow failures to be less catastrophic (loss of feed back - leading to smoke)? yikes!
  • 05.18.2016
  • Wet and dry switching
  • I have wondered about this.. the vague threshold defining "wet" and "dry". I was under the impression .. contacts designed for inductive or capacitive (not resistive) loading also played a part in the definition. A bit redundant , generally larger loads are AC and have some inductive component. These terms appears to be kinda .. mmm.... anachronistic? So many issues that an engineer needs to be aware of .. thermal noise of contacts (an issue for switching thermocouples) , intended type of loading, level of loading , etc.. Too many variables to be addressed with any simplified term. A relay story: When I first got out of school , I was performing maintenance on a test panel used for testing very large AC motors ( 5,000 hp, 4160v 3ph... 15 feet tall -vertical mounting- .. used for pumping - flooding valleys) You could drive a small car through these motors when the rotor was removed. We had to perform full voltage @ rated speed (generally 4 pole=1760 rpm) REVERSALS! All the testing had to be performed at night.. .only time it the electrical consumption costs were reasonable.. and the grid could handle it (we had our own substation for just the test rig). The relays used? WWII submarine units .. 12 contacts, each @ 1.5 inch in diam with the entire assembly submerged in oil (contained the arcing and provided cooling of the contacts I assumed). I was asked to replace the contacts (~$500 each in the early 1970s) after a modest level of testing cycles. And sometimes replacing the oil ( ~30 gallons?). I did this multiple times. "wet" ? I think so... under every possible definition applied to relays (physically and loading) Side note: if things went "wrong" during testing .. we could take out the power for hundreds of miles around us. If (when) the motor broke loose during the tremendous forces being applied... well, everyone could break their own sprinting records getting out of the building! Very exciting... seeing one of these monsters run amok!