Heathkit: Getting Closer This Time?

We’ve been following the Heathkit reboot for a while now, and it looks like the storied brand is finally getting a little closer to its glory days. I was thumbing through the new issue of QST magazine while I was listening in on a teleconference for the day job – hey, a guy can multitask, can’t he? – when I spied an ad for the Heathkit GC-1006 digital clock, which they brand the “Most Reliable Clock”. As soon as the meeting was over, I headed over to the Heathkit website to check out this latest offering.

I had cautiously high hopes. After the ridiculous, feature-poor, no-solder AM radio kit (although they sensibly followed up with a solder version of that kit) and an overpriced 2-meter ham antenna, I figured there was nowhere for Heathkit to go but up. And the fact that the new kit was a clock was encouraging. I have fond memories of Heathkit clocks from the 80s when I worked in a public service dispatch center; Heathkit clocks were about the only clocks you could get that would display 24-hour time. Could this actually be a kit worth building?

Alas, the advertisement was another one of those wall-of-text things that the new Heathkit seems so enamored of. And like the previous two kits offered, the ad copy is full of superlatives and cutesy little phrases that really turn me off. Then again, most advertising turns me off, so I’m probably not a good gauge of such things. Nor am I sure I’m in the target demographic for this product – in fact, I’m not even sure to whom this product is being marketed. Is it the younger crowd of the maker movement? Or is it the old-timers who want to relive the glory days of Heathkit builds? Given the $100 price, I’d have to say the nostalgia market is the most likely buyer of this one.

To be fair, $100 might not be that much to spend on a decent clock. I’m a bit of a clock snob, and I’ve gotten to the point where I can almost tell which chip is in a clock just by looking at the controls. The feature set of a modern digital clock has converged to a point where every clock has almost exactly the same deficiencies. The GC-1006 claims to address a few of my hot button issues, like not being able to set the time to the exact second – I hate that! An auto-dimming display is nice, as is a 12- or 24-hour display, a 10-minute timer (nice for hams, who are required to ID their station every 10 minutes), and a battery backup that claims to last for 4 weeks.

Is this worth buying? At this point, I’m on the fence. Looking at an unboxing video, it appears to be a high-quality kit, and it would be fun to build. But spending $100 on a clock might be a tough sell to my loan officer.

Still, I think I might take one for the team here so we have a first-hand report of what the new Heathkit is all about. And it would be nice to build another Heathkit product. I’ll let you know how it goes.

63 thoughts on “Heathkit: Getting Closer This Time?

  1. I bought one to relive my childhood. A few observations:

    1) The price is reasonable. Yes, it’s an expensive clock, but it’s not an expensive Heathkit. In the 1970s, a comparable Heathkit clock went for $70, so $100 today is not bad. You’re paying for the manual, the hobbyist-friendly design, and the kit-building experience.

    2) The manual is excellent and very similar to the HK manuals of old.

    3) I’m very impressed by the mechanical design. I’ve designed and built clocks and more complex electronics and to be honest, my approach to the finished project and enclosure is usually half-assed. In contrast, this kit has a sleek enclosure and well thought out mechanics.

    4) You know their motto, “we won’t let you fail”? Well, my assembled clock has a problem and after 3 weeks they haven’t even responded to my e-mail (beyond the automated response confirming they got it). One of my digits won’t light. I see on my oscilloscope that the common cathode isn’t being pulled all the way to zero volts – it’s only getting to 2.5V. Swapping LEDs doesn’t help, replacing the LED driver chip (I bought one from Digi-Key just for this) doesn’t help, and I haven’t been able to find any shorts. I know it’s real hard to debug something like this remotely, but it would’ve been nice if they at least responded to me.

    Despite my problem, I recommend this kit – I assume my problem is not widespread. I think it’s great for those who assembled Heathkits as a kid, even if you’ve since graduated to more advanced stuff. Whether this kit has a chance of diverting today’s teens, I just don’t know. Either they’re glued to their smartphones or they’re building robots in competitions: this kit is really a middle ground between the two.

    1. Have you put your scope on the input(s) of the LED driver chip? Just suggesting the next spot to check.

      Sorry to hear of the troubles, but your report on the kit is very helpful, I’m sure.

        1. I haven’t looked at the kit, but someone mentioned the MAX7219 as LED driver.
          I’ve used them a few years ago and found them to be pretty sensitive.

          Try checking the filter caps connected to the MAX7219 and maybe compare the values with the chip datasheet. Otherwise check the soldering of the driver. These drivers are little bi*c** in my opinion :(

          On the other hand, if the problem occur only on the first or last digit it could be a problem with the datachaining. Either software side or the connection between multiple MAX7219s.

          But I’m just shooting from the hip. As I said, didn’t really look at the kit itself ;)

    2. The problem I had with my finished clock is that one of the colon LEDs just stays on full brightness. I emailed them and within 48 hours I received an email saying that a package had been sent. When it arrived there was a replacement LED driver (that I haven’t yet made time to install).

      I had expected/hoped for a conversation with suggestions and I had also probed around with my scope and suspected the driver chip. I’d rate their customer service decently from the one experience based on how fast they sent out parts.

      I do hope they’ll continue taking the old name in new directions. The AM radio offerings were pretty sad. I’d love to see them embrace the generation of makers and bring some of them into amateur radio.

  2. Can anyone put scans of the manual somewhere?
    I am curious how it keeps 0.5 sec/day accuracy. My guess would be software compensation, as ATtiny84 has built-in temperature sensor.

      1. Ah, I suppose in long term that’s gonna be accurate. In Europe, the grids are interconnected and in the central control place they issue corrections for mains frequency.
        Using MAX7219 doesn’t really make sense to me. It’s expensive and not easy to find in mom-and-pop stores should it fail. I could come up with half a dozen or so of cheaper methods, using vanilla logic for instance, which incidentally would have more educational value.

          1. It works tremendously well in other clocks I have. It was also the method of choice with old mechanical electric clocks that simply used synchronous motors. As long as you had 60Hz, you kept good time.

      2. I’m among the many folks who have analyzed the stability of the line frequency as we get it here in North America.

        My conclusion is that it has excellent long term stability – exactly what you want in a clock. But the short term stability is so poor that you should not bother to put a seconds display on it. It will easily swing +/- 5 seconds intra-day.

        For $100 they could have easily made a GPS clock. I’m tempted to make one in response and sell it on Tindie, but I don’t know if it’s worth the bother.

        1. Well, with GPS the problem is putting the antenna in the right place, which is only a meter or so from windows.
          Using ESP8266 and NTP would be more reliable.
          Another thing that nobody has done yet is using RDS. There are dirt cheap RDA5807M modules on Aliexpress (40c each). However, some stations broadcast a time that is quite off, not to mention the timezone/DST confusion. But, I suppose checking several stations and weeding out outliers could be viable.

          1. The ESP and NTP would be a lot harder to configure and install, though, and would probably actually work in fewer places (GPS works inside in our house – though so does WiFi – and at 40°N, 150°W, where there is likely no WiFi).

            In principle, a GPS clock could even infer timezone and DST from physical location. In practice, it’d either be imprecise or require a TON of memory for the map data.

          2. Why would it need map data to work out it’s timezone? You just work out which one of the 24 segments you’re in. They’re generally spread evenly, with a few exceptions and half-hour zones, so you’d need to store those, but it wouldn’t need much memory, certainly not a map.

        2. “For $100 they could have easily made a GPS clock”: Could they? What’s their current BOM cost, including a 60+ page printed manual, solid wood panels, and other unique pieces of hardware? How many hours did they put into design and writing? What’s their expected sales? What’s their profit margin?

          Sure, for $100 a hobbyist can throw together a GPS-driven clock for $100. But that’s very different from producing a kit and trying to make some money from it.

          1. There are plenty of clock kits out there. What is there no market for? A clock kit with a spouse-friendly appearance? A clock kit with GPS (maybe so – even commercial clocks don’t use GPS for setting time)?

          2. “The point is that a clock that just “displays” the time it receives from somewhere, be it the government, wifi, GPS, whatever: it’s just a display,”

            Not really. None of the “atomic clocks” I have ever had that receive WWV are “just displays”. Signal comes and signal goes. Actually, as far from Colorado as I live the signal mostly goes! My clocks have to keep their own time. An occasional bit of skip or some propogation anomaly gives them a chance to correct themselves keeping any error permanently below my threshold of noticing. However.. if the transmitters turned off today I would still make it to work in the morning for a very long time.

        3. There is talk of reducing the accuracy of the power grid, since it isn’t used for timekeeping as much as it used to be, so this clock (and many others) may not be as accurate in the future.

        4. The entire point of a clock is to count time. If it can’t even do that on its own (and needs a GPS or wifi for this), it almost loses its entire reason to exist. Also suddenly you rely on a government to keep your time, rather than relying on a simple crystal. Unfortunately this particular clock relies on the line frequency, which isn’t much better, but it’s still a lot more elegant than the Rube Goldberg machine that is a GPS assisted, or wifi enabled clock. As an engineer you should strive for simplicity and elegance. If you need accuracy, just use a DS3231 or something similar and keep it local. It will then have a good chance to still work 50 years from now. Try that with wifi or GPS.

          1. maybe you don’t understand that the whole point in knowing what time it is is that everyone else knows what time it is too. relying on the government to keep time isn’t a bad thing if it means everyone has the same time.
            if everyone had a clock that had a different time then it would be utterly pointless.

          2. Pff.. I’m afraid you missed my point. The point is that a clock that just “displays” the time it receives from somewhere, be it the government, wifi, GPS, whatever: it’s just a display, and a very inelegant way to show time. If I build a clock myself, I want to build a clock… not a display. And it should count time, not just fetch time from somewhere and show it. And other than these “emotional” reasons, I also gave you more objective reasons, i.e., a clock that needs GPS or wifi may not work anymore 50 years from now or even 20 years from now. Not to mention the obvious drawbacks that it needs an antenna, and sit close to the window (in case of GPS) or have a cumbersome way to set the wifi details, maybe through some app that won’t be available anymore 5 years from now.

          3. “you rely on a government to keep your time, rather than relying on a simple crystal. Unfortunately this particular clock relies on the line frequency, which isn’t much better, but it’s still a lot more elegant than the Rube Goldberg machine that is a GPS assisted, or wifi enabled clock”

            I also am reluctant to trust government, but when you have your completely independent crystal based clock, how do you set it? You still go to the government (at some level) to get the standard time.

          4. To get the ACTUAL time, you do what we’ve always done. Look at the Sun. When it’s highest in the sky, it’s noon. You can bother with a sundial if you like, there was one on HAD years ago that was digital! It used light pipes, shadows, and some other stuff, to suck in sunlight from particular angles. Depending on exactly which, it displayed the time, I think accurage to 5 minutes.

            You can buy them now, the geek in question started a company. They’re not cheap, they’re corporate showoff-garden money. But the company will come out and install it for you exactly, which needs doing, since it’s setup depends on exactly where you are in the world. After that, I suppose it’s mainenance free forever, up until the Earth’s orbit decays enough to send it 5 minutes out of whack.

            With a compass, some maths, and some bits of cardboard with holes in, you could probably knock up a “noon detector” fairly easily”. You could average out the whole time the thing is illuminated, and assume the centre of that period is actual noon. Perhaps adjust it, whenever you like, to match local time from other sources. Once it’s set up, it can never drift. Well, except continental drift, that’ll put it off accuracy-wise as well.

          5. This goes all the way back to the 19th century. Back then, all time was local. Then came the railroads and the lack of coordinated time across large areas caused people to ACTUALLY DIE because trains would crash into each other because the clocks weren’t coordinated.

            I’m all in favor of small government, but there actually *are* appropriate jobs for government to do, and one of them is to define and promulgate coordinated time. Accuracy demands between various users will vary, certainly, but they still exist.

            If you don’t think so, then you’ll probably really have some head scratching to do when your boss fires you for being a half hour late to work every day.

  3. My recommendation to Heathkit would be to make a line of inexpensive test equipment kits, similar to the ones that they were famous for in their glory days. I understand that you can get decent Chinese test gear these days, but it doesn’t even come close to the feeling you get when you build it yourself.
    That is what I would like to see them do, but I doubt it will happen.

      1. I run a website and every now and then I discover that someone, in some forum somewhere, has written that I should change something or other on my site. And I wonder, why did they post that in a forum instead of telling me directly? I have a prominent link on my site to a Contact Us form.

        So if you have advice for Heathkit, don’t post it here. Go to their website or Facebook page and put it there.

  4. Lets us recall the way Heathkit started. We had a large number of electronic enthusiasts created by the transistor’s discovery, plus the war vets going into ham radio and the old way of point to point wiring by hand that made equipment very costly to buy. There were builders, the ham magazines had plans to build all sorts of pieces of radio and electronic gear. Of 100 started, perhaps 20-25 were ever finished, what with drilling front panels, painting dial numbers, lettering meters etc, there were many barriers to completion. So Heath came up with his concept, a bag of parts, and very detailed part by part assembly manual, stamped, painted chassis etc to make a finished built item that they sold in 2 ways – kit or built, and they offered a fixit service = success.

    As time went by point to point wiring ceased and printed circuit boards emerged, then wiring harnesses, and flat cable snap in and finally we went to auto assembled, auto tested surface mounted parts. Almost all the labor was engineered out. Away went the Heathkit business model – they died.

    In much the same way we have old railroad and old car clubs – fueled by enthusiasm – but not by economics, we can have old Heathkit nostalgia, it will never be economically viable, the margin it once lived on is gone.

    Much like the last old WW1 soldier died in 2012, and soon enough we will run out of WW2 warriors. (http://www.nationalww2museum.org/honor/wwii-veterans-statistics.html), one day we will run of of Heathkitters, while memory serves.

    We now have thousands of hacker spaces and maker groups that serve a similar need.

    1. HealthKit could adapt to serve that maker community – like AdaFruit and Sparkfun do today. Both companies sell breakout boards for SMD components that make new technology accessible to makers and learners without the learning curve of PCB design and SMD soldering.

      My strong suggestion to HeathKit is/was to sell a hobbyist reflow oven. All of the other accoutrement for SMD work is available, but it remains a right of passage for SMD makes to make their own reflow oven much like Jedi making their own light saber. It doesn’t HAVE to be that way.

      1. The world doesn’t need another AdaFruit or Sparkfun. Heathkit has a different niche. If there’s a market for sleek, nicely done kits, then the Heathkit name helps. If that market no longer exists, there’s no point in using the Heathkit brand name while selling something else. It would be like using the name “Polaroid” to sell crappy digital cameras.

        1. “The world doesn’t need another AdaFruit or Sparkfun.”

          I totally disagree with that. Maybe it doesn’t have to have the HeathKit name, and maybe it doesn’t have to be in the U.S., but the world could use dozens more outfits that are as dedicated and useful to the maker community as those two companies are.

          1. I don’t understand why people talk about this as if it was the original Heathkit company. It is just some guy who bought the trademarks when the old company went bust.

    2. “Away went the Heathkit business model – they died.”

      No, that’s not what I remember. What I remember is Heathkit undergoing a change of management. The new management decided that their was a better future in selling courses to schools than kits to individuals. A lot of people complained that they were wrong to shut down the kit business and pointed out that it was still turning a profit.

      Make no mistake, this was a time where home DIY electronics were definitely in a decline. There was a feeling that it was no longer relevant because of things like SMD and ICs that contain pretty much all of a device’s functionality. Heathkit like many others bought into all that pesimism and didn’t forsee anything like the ‘maker movement’ that we have today.

      They were still profitable, they would have certainly had to survive some hard times to get here and it is no gaurantee that they would have but it wouldn’t have been impossible. As for their educational business.. well, we all see where that went. I can’t understand what they thought they were doing with that. They aimed at both High Schools and Colleges. How many High Schools do you know that teach any meaningful electronics? I wish they would but I certainly wouldn’t bet my livelihood trying to sell soldering irons to them! As for colleges? Sure, there are plenty of colleges with EE programs but are they really shopping around for materials? Professors either choose their books from the established textbook companies or write their own. They don’t want Heathkit.

    1. That nixie clock comes without any enclosure and without a power adapter, so you need to add those in. It uses a crystal oscillator, so the time will be less accurate. I looked at its manual and, IMO, the Heathkit manual is much better. Also, the Heathkit manual is printed out and bound for you. You might not consider that to be valuable, but it IS an expense that Heathkit has and the nixie kit does not.

      1. But there was that whole period when digital clock building was common. Endless paces selling a blank board, a clock IC, the readouts and sometimes the rest of the parts. Minimal instructions, the schematic probably was the datasheet for the IC. These followed the expensive clock kits of a few years earlier, but one of the selling points was that you could put it inside of some other project or build your own case. There was a kid at school who was good at woodworking but not so much at electronics, he liked those kits.

        A few years later the minimal kits were replaced by clock modules, from National and maybe others, small boards not much bigger than the display, various options through jumpers, you just needed to add switches, and a power transformer.

        And then it all faded out, since like calculators before them, you could buy really cheap digital clocks off the shelf. The only reason to build one was for something exotic.

        As for crystal time bases, a good clock would include one so if power goes off, the clock can still keep time. There was the classic National IC which used a color subcarrier crystal, common and cheap, to generate the needed pulses for digital clocks. How accurate the oscillator was depended on the effort out into it. You could adjust it, you could add temperature compensation, you could pick a crystal frequency that divided down properly but which was more stable to begin with. Or put the crystal in an oven so its temperature was constant. Some people were very interested in that, still are. I gather there is or was a mailing list for the topic of very accurate time. The same issue came with frequency counters, which were common projects at the same time as the wave of clock projects, indeed some built up a very good frequency reference (and made the effort to set it precisely) and then with divider chis used it as a time base for both a frequency counter and a clock.

        A partial kit is more open to customizing than a full kit. And a Nixie clock at least offers a different readout from the average ten dollar clock. One is paying a lot here for the sake of putting the kit together, and gets nothing special for the results. It’s long after 1971 when electronic digital clocks were really expensive, and building a kit was the only way for the rest of us to have one.

        Mchael

    2. The price is a little high, but if somebody likes it…

      When I built my first LED clock (because I saw an offer for a clock chip) it was in the end much more expensive (with transformer, LED-7seg displays, enclosure) than the chip alone and it was more expensive than a ready made radio-alarm-clock. Without a radio.

      But I built it myself and I learned something about electronics and reading datasheets. Had also to design a multi voltage PSU for the 1980 something PMOS Chip :-)

      But I don’t know, why I should buy a nixie clock. Why should I buy something with such an ancient, outdated and probably quite power hungry display? I much prefer the appearance of LEDs.

  5. Back in the day (in UK, Europe), I built several clocks, and added a radio time code interface. Heathkits were some of my first experience of electronics, for which I was very greatfu I have a collection of their test gear from the 70s that I refurbish and love. A fellow ham builds HK style SSB transceivers..

    1. Surely it’s a play on “the most accurate clock”, which the old Heathkit did once sell. But that one synced to WWV, and even made adjustments to its internal crystal oscillator to get the clock more accurate if it couldn’t receive WWV. That was before cheap WWVB clocks, so the promotion did mean something, and you did get something better than the average clock.

      Michael

  6. I’ve scanned through the comments and much of it is with respect to accuracy of the time displayed. How freakin accurate do you need? I completed the build of this clock on Oct 22, 2016 and the time has never varied more than a few seconds from other sources. I have not had to reset the time either. I use mine to tell time of day and to log my ham radio contacts. I can’t think of any use for an atomic clock’s accuracy. I paid $100 and had a very nice, enjoyable afternoon assembling my clock. One of the brackets factory glued to the top of my clock was unglued upon arrival. Heathkit had one out to me the same day and I received it the next day. I’m totally pleased with my well functioning attractive clock.

    Tom.

  7. 1) if your day is affected by +/- 5 seconds a day then you need to change your life bro!
    2) if you reeeally need +/- 5 seconds a day accuracy, then you wont be building a kit, unless what your after is a tablet with a wall-mount.
    3) the only market for +/-5 second a day are people doing cyrptography, plant automation, security, infrastructure, banking (back-end), or just anything that does NOT involve LEDs sitting on a normal desk, anything THAT critical will be made using LCD so it can run for a long time on a coin-cell.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s