Member #134773

Member Since: May 19, 2010

Country: United States


I started playing with electronics in the mid-1960s, and with computers shortly after Neil Armstrong took “one small step”. I got a degree in CS in 1980, and started working then as an engineer.

  • I was right! It IS entertaining! But regarding the IR Blaster… what I’d really like to see is one that interfaces with the venerable X-10® system.

  • When I click on the IR Blaster video, I’m getting a message that it’s been removed by the user! Bummer… SF videos are (almost) always quite entertaining!

  • Thanks for the post! I’ll have to check out the stuff you’ve suggested!

    I’ve been “doing” software since shortly after Neil Armstrong took his “one small step”! (At least we no longer have to punch paper cards for our programs…)

    I’ve always thought that the Arduino API is a real “blast from the past” – it has the feel and capabilities from APIs from the late 1970s, albeit using C++ (which is more recent). For something intended for “newbies” to use, it’s missing what I consider to be the most instructive feature – the ability to single-step through the program. (True, the ATmega328 lacks hardware support for single-stepping, but it is possible to support single-stepping purely in software, though it’s not easy to develop.) Single-stepping, especially when combined with “variable watch”, can be VERY instructive as to what’s going on in the program for the beginner. (It can also be very useful for the more experienced programmer, especially in tracking down those “dumb mistakes” that we all make!). The micro:bit’s simulator can be helpful in this regard, but simulators can have problems with hardware interfaces.

    The other problem I have with the Arduino IDE is that it doesn’t have an easy and obvious way to add “-D” options to the gcc command line. This can be very useful for things like ESP8266 programs where one might have, say, three or four processors for sensors that vary only in name (and/or IP address if, like me, you use fixed IPs for your “edge nodes”). At first blush, it looks like the CLI may work in this situation.

  • Great post! One minor point, I had to look up on Digi-Key to find out that the ADS7828 is an Analog to Digital Converter (ADC), that happens to have 8-channels, 12 bit resolution, and is capable of 50 kHz sample rate (with an I2C interface).

  • One of the things I like about the Teensy 3.5 is that it has a real RTCC built-in. Several of my projects need a (true) Real Time Clock Calendar for things like datalogging.

  • The RTC on the nRF52840 is actually a “real-time counter”, not a “real-time clock”. Please see my comment on the product page. (Not an attack, just a “correction”.)

  • “Real-time clock” is a bit of a stretch – it’s actually a “Real-time counter” that counts pulses from the 32.768 kHz crystal, rather than a full-blown Real Time Clock Calendar which has hardware that translates the 32.768 kHz crystal oscillations into seconds-minutes-hours-days-months (and usually years), and usually a CPU chip that has a RTCC has provision for keeping power to the RTCC (and the 32.768 kHz oscillator) from something like a “watch battery” even when all other power is off.

    The Nordic Semiconductor data sheet (available here) (551 pages!) is really pretty clear about it being a “real-time counter”.

    Although it has modes that draw less than 20 uA, if you’re relying on a 500 mAh LiPo, how long the counter will “keep time” depends on the state of charge of the LiPo when it goes into “deep sleep” and can also be limited by the self-discharge of the LiPo.

  • Since you mentioned JoAnne’s, I’ll give a reminder: Check the “junk mail” and/or the ads in the Sunday paper, as from time to time JoAnne’s runs coupons good for a substantial percentage off any one item (with restrictions). Michael’s runs them just about every week.

  • One additional thought: I think, though am not certain, that some Software Defined Radio (SDR) modules can be used as “spectrum analyzers”, although they have limitations.

  • First, I hope that you take all of this as I mean it, to wit, as constructive comments. I’ve been “doing” electronics for over half a century, and “doing” computers since shortly after Neil Armstrong took his “one small step” on the Moon. I worked as an engineer for over 20 years for a company that built testers for computer chips (we joked that they were multi-million-dollar voltmeters). I’ve also been involved in Amateur (“Ham”) Radio for more than a quarter century.

    I, too, have done many projects that use “breakout” boards, both from SparkFun and from Adafruit. I’ve been tinkering off and on for the past several months with Adafruit’s HUZZAH ESP8266 breakout combined with sensors from SparkFun, and a few miscellaneous parts from DigiKey.

    I’m more-or-less retired now, but at my last “paying” job (with a small start-up company - I was doing for them what we both do for a hobby, but the best part was that someone else was buying the “toys”!), I was forced to use [or as I would say, be abused by] Eagle. I’m not at all fond of Eagle. The SparkFun tutorials made it “do-able”, but still not enjoyable. I much prefer KiCad, or gEDA, as they are FAR more “intuitive” than Eagle (and they’re both free without the limitations of the “free” version of Eagle).

    Always, always, always verify every single “library” part for what ever CAD system you’re using, no matter what the source. I’ve learned that the hard way! I’ve even seen the wrong pinout for something as old as the 2N2222 transistor, and the wrong spacing for a “standard” castellated subassembly.

    I have a whole box full of adapter boards from SparkFun, Adafruit, and DigiKey. I just now did a search on DigiKey for “adapter boards” and found 927 results. (Adding “Normally Stocking” reduces it to 475.)

    From my days working on IC testers, I’m aware of how hard it is to measure currents down in the single-digit microamps or even nanoamps. One of the really big problems at those levels is RF noise – you really need to shield as much of the circuit as possible. (One of the things we did was to measure “quiescent current” in CMOS chips operating “at speed” – this involved measuring currents that were supposed to be in the high nanoamp range in time windows on the order of hundreds of picoseconds, as opposed to currents of many milliamps when the part was actually switching. You can guess why we called them multi-million-dollar voltmeters!)

    On the issue of spectrum analyzers, I’ve been involved for over 25 years with a local “club” that maintains several “repeater” sites that are on the 2m, 1.25m, and 70cm bands, and to do “site alignments” we have to have instruments that can work as spectrum analyzers. Being on the “tech committee” for many years, I can occasionally borrow these instruments to work on my own radios or other projects. Also, in the past few years, the “professional” instruments have become much more compact, lighter weight, and easier to use. This means that the older, bulkier, heavier, and harder-to-use ones have been showing up at larger “Ham fests” (Amateur Radio swap-meets) at much lower prices. So, if you’re not too picky about having an in-date calibration and not wanting to drag it around to remote sites, you might find something you can afford. (Look for something called a “station monitor”.)

    Two final thoughts – actually, “pet peeves”: It seems like the data sheet for every chip that can do I2C spells out what that interface is in painful detail, and most of them BURY the I2C addresses in it (though there are a few exceptions that mention it up at the top). And, most of the processors that are used in the “hobbyist” market don’t include Real Time Clock Calendars – there are many on the market, especially MicroChip’s PIC processors, that have RTCCs built-in, all you have to do is add a 32.768 kHz crystal and a battery. Most of my projects either have an absolute requirement for an RTCC or would benefit from having one. (Yes, there are a few, such as Teensy 3.5 that have a built-in RTCC, but not many!)

No public wish lists :(