Itching to build...

[Terry S]

I know there's a plethora of IN-18 kits out there -- please tell me about your favorite. I've got tubes and I want to build my own case this time. If you're selling kits and want to blow your own horn I'm good with that as well. I'd design my own and have done so in the past, but I know there are already many choices.

 

I can do SMT work no problem, would like a highly featured clock with GPS sync capability, but it should be able to operate stand-alone as well.

 

Would consider either 4 or 6 tube, 6 preferred. LED lighting is unnecessary, colons are necessary.

 

Arduino based would be nice so I can play with the code a little easier, PIC ok as far as that goes too.

 

Compact form factor is a plus, as is low cost.

 

Please don't just throw out links.... I can find those. I'd rather hear about your actual experience with the kit.

 

Thanks in advance,

 

Terry

[gregebert]

I recommend building from scratch if you want to design your own case. That way you can showcase the tubes however you wish; with a kit you're pretty much limited by the board layout regarding where the tubes will be. If you want something that will look really nice (why not, IN-18's are nice tubes...) then it wont be low cost. I probably spent close to 1000 USD ($300 for 5 PCB sets, $300 for tubes, about $300 on remaining parts, $100 for the case) on my big clock, but it turned out so much better than I expected that my wife basically insisted I put it in the most prominent location in the house (above the fireplace).

I'm almost done my third built-from-scratch nixie project in 4 years and wouldn't do it any other way. There's lots of free high-quality design software for PCB layout, schematic capture, FPGA development, Verilog simulation, SPICE simulation that I have used. None of my boards have blue wires: 6 of 7 worked the first time (7th board is still in bringup, no blue wires yet).

After doing PC boards, I will never go back to doing circuits on a perfboard again. The amount of effort to do a PCB design is about the same as 1 breadboard, and from then-on you are ahead of the game for additional builds when using a PCB. Then there's the whole quality & neatness advantage of PCBs.

Likewise, I probably wont buy any PCB kits, either. Years ago I built several Heathkits, and they were a great learning tool. But they are gone, and with the free PCB tools and plenty of inexpensive manufacturers, I've moved on.

[Terry S]

I get your point, Greg, and I've designed and built several clocks, I know exactly how rewarding it is. I've been a PCB designer for over 20 years, so my point is, and it's not obvious I guess..... that the PCB design is not the challenge. I've got more PCB designs under my belt than most designers will ever tackle, including a couple of my own one-off nixie clocks. This time I want to buy a board, or a kit, and design a case around it. Yet I am fussy about the clock function, hence my list of requirements earlier. A few recent experiments into wood and metal working has me wanting to stretch my skills in that dimension. I see some IN-18 kits starting around $100 out there, so there is a certain logic in buying another's proven design. NIH (not invented here) is not an issue with me, my ego won't be bruised if I buy this time, rather than design.

[Jeff Walton]

I've built the IN-18 Blue Dream from Dieter Wachter (Nocrotec.com).  The kit (6 digit) is very easy for most as the SMT devices are already mounted so that all you are left with is the through hole items (relatively few).  Once upon a time I would have been comfortable with SMT but the shrinking geometries and my aging eyesight make it more of a challenge.  The kit does include a stainless steel base (which is gorgeous) but does not have any other cover.  I love the IN-18 tubes as they are among the most beautiful tubes ever made. I think that you can purchase the PCB and parts without the base.  GPS (optional) is easy and the clock operates flawlessly.  One thing I am doing with this clock is using a motion control to turn the tubes off when no one is around.  The kit already has programmable time of day brightness and adding the motion sensing will maximize the lifetime.  This clock is supposed to sustain all settings and features for 6 years without external power so there is no issue with turning off the external power for hours at a time.  There are no annoying start-up noises, flashes or loss of volatile settings when power is turned on/off.  The kit has colons in glass tubes and blue LED base lighting that can be turned on or off.  I'm not sure what you are looking for but the final product was excellent for me.  I get the impression that your skill level for assy and coding is probably far beyond the skill needed to handle this kit but you can decide that.

One thing I would love to find is a IN-18 clock that uses a remote control unit like the ones that YanZeyuan (Nixieclock.org) did for his smaller clocks (IN-8-2 & VFD)  The remote functions are so much easier than the multi-button approach on the clocks.  I don't know if YanZeyuan has a kit for IN-18 but if he did, I'd look at it. 

Jeff 

[Terry S]

Ok, so I went ahead and started designed my own clock, borrowing heavily from an existing open source design.

Here's what it entails. I have IN-18's and love those tubes, so naturally it uses them. I designed a main module that has the Arduino interface, power supply, control logic, 2 anode drivers, and 2 IN-18 tube socket footprints. At the end of that module there is a 20 pin connector to mate with a 2nd board I designed, identical in shape and size to the main module, but only has the 2 IN-18 socket footprints and the corresponding anode drivers. This slave module also has the 20 pin connector at far end.

The result is that a clock can be configured using 1, 2, or 3 modules. If someone wanted, they could build a 2 digit clock (not sure why you would do that) or a 4 digit (my preference) or a 6 digit clock.

I will be including provisions for under-tube LED lighting, even though I dislike it, and colons, which I do like. The colons will be controllable. I prefer a flashing colon on 4 digit clocks, and a static colon on 6 digit clocks.

Since it's Arduino driven, the end user can do whatever he/she likes as far as timekeeping, (GPS, etc) features, etc. I will be producing a number of boards excess to my needs, so let me know if you are interested. I will NOT be supporting software.... that will be up to you. The board is designed such that the open-source software should work without modification, or with very little, for basic 4 or 6 digit clock functionality.

The boards will have both thru-hole and surface mount parts. You will need to be able to solder your own SMT components.

Any interest? As far as pricing, I'm only going to be looking to recover my blank board costs.

Terry

On Wednesday, January 21, 2015 at 10:09:35 AM UTC-6, Terry S wrote:

[Nicholas Stock]

I'd be interested Terry.

Nick

Sent from my iPhone

--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.
To post to this group, send email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/36f2d014-2779-4a81-b6c1-85b2d05621b9%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Mich...@aol.com]

Terry,

 

I am interested.

 

Michail Wilson
206-920-6312

[Dylan Distasio]

I'm interested.

--

[blave]

Sorry to throw out a link, but I wanted to make sure you're aware of this nixie shield for the Arduino: http://arduinix.com/

I have built the kit (v1); it went together fine and I ended up debugging some of the "crossfade" code for the 6-digit daughtercard. I also used the Arduinix for a non-clock application: https://www.youtube.com/watch?v=9kbrIwxNr9g

cheers,

Dave B.

[Matt]

There are three things that are stopping me from building my own clock:

1) I only like simple shinny metal and glass, neither of which I can build or design on my own.

2) I only would consider direct drive circuits. If I needed to reduce the brightness, I would PWM it myself, but I do not want the
multiplex circuit limiting the maximum brightness.

3) Time. Creating a direct drive PCB from scratch is a bit much.

I would prefer a simple Arduino shield or daughter board, but it is difficult to find something that is direct drive and in a form
factor that is appropriate for a clock. I would prefer something using https://pinocc.io/ so that I can get time from NTP instead
of GPS.

> --
> You received this message because you are subscribed to the Google Groups "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com

> <mailto:neonixie-l+...@googlegroups.com>.
> To post to this group, send email to neoni...@googlegroups.com <mailto:neoni...@googlegroups.com>.

> To view this discussion on the web, visit

> https://groups.google.com/d/msgid/neonixie-l/7d120ea6-4086-4a8a-aca1-7f0f293c7162%40googlegroups.com
> <https://groups.google.com/d/msgid/neonixie-l/7d120ea6-4086-4a8a-aca1-7f0f293c7162%40googlegroups.com?utm_medium=email&utm_source=footer>.