Maker Pro
Maker Pro

Help regarding possibility and planning

I came over this: https://www.tindie.com/products/akafugu/vfd-modular-clock-iv-18-smt-with-gps/
So this for me seems awesome, but it costs more than a pretty penny, so I wondered if this was possible to build for myself? It doesn't have to look like this, but I really like all these tube clocks I've seen.

I'm not exactly brilliant with a soldering iron, and I've got little knowledge of parts, but I will need to read up on the stuff needed to build something like this.

I need som advice on parts. I have absolutely no means of getting any parts except for incredibly basic switches and some resistors from a shop so everything will need to be ordered. I would also like to keep this as simple as possible so that I can actually do this by myself. It seems to me that some of the tubes used in some of the clocks I've seen are prone to heat development. Is this true? Is it possible to avoid components that are notorious for heat development? I also need some instructions on how to wire this up. I want to do this because I really like the look of these clocks and I would like to learn more about electronics.

-pusur
 
My only advice is learn to solder discrete components first.
Have you actually seen how small surface mount components are? (not my thumb):)

images


Martin
 
It is a resistor. Your circuit has lots of them too which need soldering.
It takes lots of practice and decent soldering tools to be able to do it.
Not to mention a decent magnifier with a light!.

Martin
 
Note that while we can't see the bottom side of the board in your link, none of the visible components are as small as in post #2. Surface mount (SMT) parts are fundamentally more difficult to handle/solder/unsolder than through hole (PTH), but there are degrees. The device in the link is not nearly as crazy as what goes on in phones and tablets.

ak
 
Bump.
Like said before, doesn't have to be/look like the one I linked to. I am more than willing to use modules that already exist for the purpose of keeping time or other aspects of this project.

-pusur
 
Bump? What is this? Oh, a reminder!!!
Don't bother bumping. Come back with another idea!
Show us what research you have done! What module (if any) you found!
Search Nixie tubes too.
Look for something SIMPLE for starters.
Oh, also, I mentioned about having certain tools for the job! Did you reply?
No..
I would also like to keep this as simple as possible so that I can actually do this by myself.
Here, you said it yourself!

Martin
 
Sorry if you felt my reply was acidic. It wasn't intended.
Sorry about this, intentions over text can be difficult. I have a 60W soldering iron with a basic pointed tip and a couple of different tweezers. I have a magnifying glass around but without a light and stand... I was reserving this for the part after I figured out what parts I needed. So, I think what I need is what will power the whole project, how will I drive the nixie tubes/VFD display and how will I actually keep the time. I think I would like to have it powered by cable. I suppose that can be done with either USB or a standard power connection. I'm not sure if I want to use nixie tubes (the heat "issue"?) but I'm confused, there seems to be VFD displays put in tubes and named "nixie" so I'm not sure. I suppose the time can be kept by an Arduino, but I'm not actually sure what an Arduino does except I know it can control electronics. I'm also not sure how I'm going to do that. Does it have to be Arduino + some module or just Arduino... and do I need programming skills and if so, how much?

-pusur
 
I do not think that heat would be a problem since a Nixie tube is quite large and dissipates little power, it does however need a high voltage, something over 100V. You will not get this from a USB supply so will need an invertor or voltage multiplyer.
There are various Nixie tubes with different pinouts so get a cosistent set.
You can get your frequency standard from off air fequency standards or from satellite. An alternative is to use a crystal oscillator which can be set to one part in a million or a clock/watch circuit. An Arduino uses a crystal frequency and could be used to drive the display but you will need to be careful driving the high voltage.

I have a Heathkit frequency counter using TTL logic, there are components much more recent than this.
 
I do not think that heat would be a problem since a Nixie tube is quite large and dissipates little power, it does however need a high voltage, something over 100V. You will not get this from a USB supply so will need an invertor or voltage multiplyer.
There are various Nixie tubes with different pinouts so get a cosistent set.
You can get your frequency standard from off air fequency standards or from satellite. An alternative is to use a crystal oscillator which can be set to one part in a million or a clock/watch circuit. An Arduino uses a crystal frequency and could be used to drive the display but you will need to be careful driving the high voltage.

I have a Heathkit frequency counter using TTL logic, there are components much more recent than this.
Alright, I didn't have much knowledge about what you are saying here, but I took a look at it. I assume you're talking about an inverter (I'm not trying to sound pompous by correcting you, just need to be sure we are talking about the same thing.) The wiki page I found talks about AC to DC, doesn't it need to be DC? So I assume there are inverters for that. This can be incorporated into the circuit with the use of a "wall-wart" that converts to the correct specifications, correct? If I choose to use a voltage multiplier could I use a power source such as USB and used diodes to up the voltage, would that be possible? Might create a cool effect. Or would I need 95 diodes for this? Then it's undesirable for obvious reasons...

Crystal Oscillators I do know about and was once explained to me like quartz vibrates a precise amount of oscillations per second and is therefore used to measure time as you count the number of oscillations and that reaches the correct amount of times, it will register as a second. Is this correct? Won't I still need something that will change the numbers around on the nixie tubes according to the input from the CO? How would it work with the off-air frequencies and satellites?

I looked at your frequency counter and I assume this is the part that will actually count the oscillations, but that is about the size I would like for the whole project, I assume there are better solutions than this?

-pusur
 

hevans1944

Hop - AC8NS
I came over this: https://www.tindie.com/products/akafugu/vfd-modular-clock-iv-18-smt-with-gps/
So this for me seems awesome, but it costs more than a pretty penny, so I wondered if this was possible to build for myself? It doesn't have to look like this, but I really like all these tube clocks I've seen.

I'm not exactly brilliant with a soldering iron, and I've got little knowledge of parts, but I will need to read up on the stuff needed to build something like this.

I need som advice on parts. I have absolutely no means of getting any parts except for incredibly basic switches and some resistors from a shop so everything will need to be ordered. I would also like to keep this as simple as possible so that I can actually do this by myself. It seems to me that some of the tubes used in some of the clocks I've seen are prone to heat development. Is this true? Is it possible to avoid components that are notorious for heat development? I also need some instructions on how to wire this up. I want to do this because I really like the look of these clocks and I would like to learn more about electronics.

-pusur
Pretty nice clock. If you want to build something similar you will need to divide the project into digestible pieces.

First thing to do is decide what type of digit display you want: VFD or Nixie. Then find a source for your selection and buy twice as many as you will need for one clock so you will have spares. Get four digits minimum for hours and minutes, more if you also want to display seconds and the date. Digits don't have to all be in the same envelope, but that is waaay kwel! Here is a link to a source of Russian VFD displays. If you decide you want the warm orange glow of Nixie tubes, Google for those. Either way, get the parts on hand and learn how to use them. We can help with that.

Second, you need to decide on a time base. I would start with the AC power line, either 50 or 60 Hz depending on where you live. It is easy to make a circuit that detects zero-crossings of the power line voltage using a small step-down transformer and a couple of transistors. We can help with that.

Third, you need to decide what logic you will use to convert clock pulses to digit displays. There are integrated circuits available that will do that with minimal fuss. We can help with that. You will also pick a human-machine interface (HMI) to the logic to define how you set and control the clock. We can help you with that too.

Finally, you need an interface circuit between the logic and the display digits. This will depend on what type of display you choose. After you pick one and play with it for awhile the interface circuit can be designed. We can help with that.

So draw up a block diagram of what I've described above and we will try to fill in the blanks with circuits and components.

Oh, get a smaller pencil iron. About 30 watts maximum with a small conical tip. A temperature-controlled iron is nice but not absolutely necessary. If you don't already know how, learn how to solder with tin/lead 60/40 solder. Avoid no-lead solder like the plague it is. You aren't making this to satisfy some Nanny State regulators are you? Most of all, have fun and enjoy playing with and learning about electronics. This is a pretty good first project.

Hop
 
Pusur.
Thank you for corektin my spelin.:) I can never remember whether such words should end with or or er. Have you got a simple rule? You missed another mistake.

The frequency counter is simpler than a clock since it has to divide by ten whereas a clock has more complicated logic. This can be done with counter chips but it would be easier to use a microcomputer. I have not played with such a beast in the last forty years. The computer could also decode the count to drive the display.

A crystal vibrates as you say, there are various modes of vibration, usually by shear. I have a 100kHz crystal which is a bar in a glass vacuum envelope, this works in flexure.

It depends where you are as to what frequency standards are available off air. In the UK we used to have GBR from Rugby but it has now moved to Anthorn so its callsign may have changed. Many of the clocks available here use a German standard. If the standard gives hours, minutes and seconds, then the decoding becomes simpler. You have kept your location secret so it is impossible to comment on the standards available to you.

Trevor
 
Thank you both very much for your reply. I think it would be the simplest if I go for this: http://ussr-tubes.com/index.php?main_page=product_info&cPath=98_170&products_id=1150 I wasn't planning on using seconds as well but I don't mind. I can probably do the format like this XX-XX-XX since it's 8 digits. I do need to learn how to use them... Oh and I use 24-hour time format

I live in Norway, so for me its 230 volts and 50Hz I would absolutely like to use an integrated circuit for time keeping, It would be awesome if as duke37 talks about to be able to get the correct time from radio/satelite/(GPS?) but if that presents problems, then I really don't need it. For the interface, I would like something sleek but perhaps retro, perhaps not. I plan on building the whole thing into some kind of wooden enclosure. so either sleek or retro suits me. I like the look of these: http://media.conrad.com/medias/global/ce/7000_7999/7000/7010/7013/701308_LB_02_FB.EPS_1000.jpg As stated I need to order all parts so, I'd like them cheap-ish and preferably on eBay or some Chinese online retailer.

I do have some concerns about playing with mains power though. It doesn't go clear of your post and with fear of sounding incredibly stupid, doesn't these components work on DC? If so I suppose som kind of commercially available wall wart might do the trick? I can get the tools you state here, but do I need to take and respiratory precautions when working with lead solder? Why should I not use no-lead solder?

This diagram is probably the crudest one you've ever seen. That isn't my usual handwriting, but if I used that you wouldn't be able to read anything at all.

-pusur
 

Attachments

  • IMG_20160110_0001_NEW.pdf
    825.8 KB · Views: 167
Last edited:

hevans1944

Hop - AC8NS
@pusur44017: I like the Russian VFD you have selected. It was once used in an Adafruit clock kit, now discontinued. To save you some time, I have added the Adafruit schematic below:

icetubeschemv1.1.png


You can also go to this link (http://www.ladyada.net/media/icetube/icetubeschemv1.1.png) to download a fresh copy that may be easier to read.

Looking at the number of wires coming out the end, it is apparent that the anode segments will need to be multiplexed with a high-voltage (50 V DC) driver. Some research on the world wide web could be required to see how others have successfully done this. I hate re-inventing wheels if I can find something that already works, and at least one has been successfully demonstrated. So you go look, and I will too as time permits.

The Adafruit design uses a MAX6921 driver, so that is the first place I would start. There may have been problems getting reliable parts for their kit to explain why it was discontinued, but if you can find a copy of the firmware used in the ATMEGA 168V you may be home free... build from the above schematic, download firmware. Voila! A pretty nice clock. Or you can "roll your own" design and learn a LOT of electronics along the way.:cool:

If you use the 50 Hz mains for your initial time base, you will need a small transformer to reduce the 230 VAC down to a few volts. Just about any voltage less than twenty volts or so, and greater than six volts or so, will work. Center-tapped 6.3 VAC filament transformers work well. The center-tap is connected to your circuit common. The two remaining leads are connected to the bases of two NPN small-signal transistors through a 1 kΩ to 10 kΩ 1/4 watt carbon film or metal film resistor. The two transistor emitters are connected to circuit common. The two transistor collectors are connected in parallel and then to a single 1 kΩ 1/4 watt resistor connected to a positive logic supply level (5 V for TTL or 10 to 12 V for CMOS). Zero-crossing pulses at 100 Hz will appear on the transistor collectors and should be passed to a Schmitt-trigger input logic device to produce fast rise and fall times. The pulses will bracket the zero-crossing, occurring slightly before and ending slightly after the zero crossing. Small-signal diodes should be placed across the base-emitter junctions, cathode to base and anode to common, to prevent reverse biasing those junctions. This is also a convenient way to trigger SCRs and TRIACs at the zero-crossing of the line voltage for reduced RFI with incandescent and resistive loads... not so much with inductive loads like motors though.

How it works: the transistors are alternately forward biased, 180 degrees out of phase, by positive excursions of the transformer voltage, causing their paralleled collectors to be saturated near common potential most of the time. As the transformer voltage approaches the zero crossing, both transistors turn off and a positive pulse appears on their collectors. The Schmitt trigger that follows converts these pulses into fast logic-compatible pulses. Pick a logic family before proceeding.

You should either use a fly-back boost converter for the VFD anode supply (as in the Adafruit schematic), running it from your low-voltage logic supply, or buy a "filament" transformer that also provides another secondary winding providing twenty to a hundred volts of so. You rectify and filter and possibly voltage-regulate the AC output of this winding (possibly with a voltage doubler or voltage tripler circuit if the secondary voltage is too low) to provide about 50 VDC anode power. The current requirements are modest, about 100 mA or so. I would buy the transformer, some rectifiers and filter capacitors, and perhaps a three-terminal voltage regulator just to learn about how to build a power supply. You can always switch over later to the Adafruit boost converter.

If you use the Adafruit design you can power everything from a low-voltage DC wall-wart, no transformers necessary. Timing is derived from the crystal oscillator servicing the microprocessor. You can use any solder you want, but lead-free melts at a higher temperature and is a little harder to use. It is only available because of Nanny State concerns about lead in the environment, especially landfills where consumer electronics are dumped. I have been using lead-tin solder since the 1950s. The lead melts but does liberate any significant lead fumes in the process. It is only a problem when dumped into the environment or absorbed by ingestion. Wash your hands after and don't smoke during soldering.

I thank you for the "block" diagram, handicapped as it is by your having to hand-annotate in English. Now just overlay it with the Adafruit schematic for a complete design.:D

On a personal note, I am handicapped by being too lazy to draw schematics (most of the time). If one picture is worth (at least) a thousand words, I opt for the words every time.:) C'est la vie! Below is my unit's patch from when I served with the Strategic Air Command as a defensive fire-control systems mechanic. The bulldog represents the Gunner and the crow represents the Electronic Warfare Officer or EWO. The bulldog is holding and firing a 20mm Gatling Gun. This aft-mounted defensive armament was a radar-directed, remote-controlled, fire-control system. The crow dispensed chaff, rope, flares, and jamming signals. If that didn't work, the bulldog finished business for the day.

In the B-52H heavy bomber, both Gunner and EWO sat side-by-side facing to the rear in the upper-level flight deck, EWO on the left and Gunner on the right (facing backwards). Every other member of the six-man crew faced forward and had some sort of view of the surroundings, although there isn't much to see at 50,000+ feet. Pilot and co-pilot sat forward at the aircraft controls and the radar-navigator and bombardier were side-by-side on the lower flight deck with both a radar and an optical bomb-sight view. There was also a fold-down seat on the upper deck for an "observer" but I never got to sit there in flight. I was not exactly a "company man" during my four-year hitch in the Air Force, so no invites to fly. The B-52H is still operational, but the Gunner position has been eliminated. The Air Force now puts its faith in electronics and air superiority to protect the BUFF from danger. So far, so good.

Crowdog.gif


Hop
 
Last edited:

I found adafruit project while researching this a bit but it was discontinued so I moved on. Luckily the whole project is open source so I found this https://learn.adafruit.com/ice-tube-clock-kit/downloads there is also a detailed parts list and instructions. I need to know how to load the firmware into the chip and all.

I think I will go for the adafruit solution as it's "proven" and the least risk for someone like me. An oversimplification of it is to run wires through the components in the schematic. Adafruit uses a PCB. I won't since I don't have equipment to make one, so any advice on how to fix the parts to something? I suppose with all the wires from the VFD I should build a spacer... does any parts have to be in close proximity to the crystal?

Is the schematic repeated anywhere? Does it show connections from different "perspectives"?

-pusur
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
If I was going to build something like this I would use an arduino, an lcd display and maybe a gps module. I'm pretty sure I could just grab the time from the GPS.

The amount of soldering would be minimal, and no need for high voltages.
 
If I was going to build something like this I would use an arduino, an lcd display and maybe a gps module. I'm pretty sure I could just grab the time from the GPS.

The amount of soldering would be minimal, and no need for high voltages.
Although that might be simple, That's not what I'm going for here :)
-pusur
 
A bit of research shows that buying all the parts separate costs quite a lot, anyone know a reliable, cheap online parts retailer I can contact to make me an offer?

-pusur
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
A bit of research shows that buying all the parts separate costs quite a lot, anyone know a reliable, cheap online parts retailer I can contact to make me an offer?

How many thousand units per month will you be purchasing? (That will be their first question)

With the quantity you're buying the chance of you negotiating on price are slim (unless you can get a "sample"). The best you'll likely do is save on freight by purchasing from a single source.
 
Well the way it stands now, it costs me way more than the kit I linked in the first place. I'm not giving up, just need to get the costs down..

-pusur
 
Top