16 Segment Starburst smart socket driver

[Richard Scales]

I have long been inspired by 'word clocks' as well as the 'smart socket' concept which was used to drive things like the Burroughs B-7971 displays.

Having some time on my hands recently I decided to create a 'smart socket' to drive 2.3" Starburst LED displays. I have now got to the point where I actually have something to show as the proof of concept would appear to be successful!

 

The original smart socket code which was created by Chris Barron ran on a PIC16F690 which did not have enough ports to drive all 16 segments plus the decimal point so I moved over to a PIC16F886 (I am no PIC expert so the choice was based on what my development environment and programmer would support).

At the same time a colleague of mine was hinting that I really should have a go at surface mount components. The result is a single board with the PIC and some Darlington Drivers and associated series resistors. The 2.3" LED display plugs in to the back.  As many sockets as you desire (1-255) then link together and by sending simple ASCII commands, any required data can be displayed.

Multiple transitions as well as user defined characters are supported.

I re-purposed the original 'underscore' command (the B-7971 has an underscore segment) to control the decimal point.

That's pretty much it.

I already have a 'controller' which handles all the WiFi, NTP Time setting and fancy display controls including motion detector activation etc. I just need to make a version specifically for these displays which will provide optional PWM control for the 12V LED power and we're done.

I plan to make a 10 segment version, the case for which is currently a concept in my head - so it could take a while to see the light of day.

I plan to publish full details over at the Smart-Sockets group when I get things finally buttoned down.

- Richard

[Michail Wilson]

Again, simply awesome.

 

Did you make the files public?    I am interested in making up a set myself.

 

Great Job.   That was fast.

 

Michail Wilson

206-920-6312

 

From: neoni...@googlegroups.com <neoni...@googlegroups.com> On Behalf Of Richard Scales
Sent: Tuesday, July 7, 2020 1:53 AM
To: neonixie-l <neoni...@googlegroups.com>
Subject: [neonixie-l] 16 Segment Starburst smart socket driver

 

I have long been inspired by 'word clocks' as well as the 'smart socket' concept which was used to drive things like the Burroughs B-7971 displays.

 

Having some time on my hands recently I decided to create a 'smart socket' to drive 2.3" Starburst LED displays. I have now got to the point where I actually have something to show as the proof of concept would appear to be successful!

 

 

 

 

The original smart socket code which was created by Chris Barron ran on a PIC16F690 which did not have enough ports to drive all 16 segments plus the decimal point so I moved over to a PIC16F886 (I am no PIC expert so the choice was based on what my development environment and programmer would support).

At the same time a colleague of mine was hinting that I really should have a go at surface mount components. The result is a single board with the PIC and some Darlington Drivers and associated series resistors. The 2.3" LED display plugs in to the back.  As many sockets as you desire (1-255) then link together and by sending simple ASCII commands, any required data can be displayed.

 

Multiple transitions as well as user defined characters are supported.

 

I re-purposed the original 'underscore' command (the B-7971 has an underscore segment) to control the decimal point.

 

That's pretty much it.

 

I already have a 'controller' which handles all the WiFi, NTP Time setting and fancy display controls including motion detector activation etc. I just need to make a version specifically for these displays which will provide optional PWM control for the 12V LED power and we're done.

 

I plan to make a 10 segment version, the case for which is currently a concept in my head - so it could take a while to see the light of day.

 

I plan to publish full details over at the Smart-Sockets group when I get things finally buttoned down.

 

- Richard

 

 

 

--
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 view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/a3baf95d-4dd1-4844-9e1f-861e192b62f1o%40googlegroups.com.

[gregebert]

Will LEDs ever have the intrigue or nostalgic glory that nixies have ? I sometime wonder.

[Richard Scales]

I would say almost certainly not! For me it's all about the smart socket.

The next version will be a 21 segment neon based display, watch this space...

Sent from my Huawei phone

--
You received this message because you are subscribed to a topic in the Google Groups "neonixie-l" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/neonixie-l/h3R5pqcfP8Y/unsubscribe.
To unsubscribe from this group and all its topics, send an email to neonixie-l+...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/c87d30fc-03e7-4e5d-b367-e9711c9604a9o%40googlegroups.com.

[newxito]

Very nice! I really like these types of sockets. 

I have been busy for the past 2 months, no time for the hobby. But now I’m looking forward to building a smart “multi-socket” socket (5 x IV-26, 7x5 matrix, 1 digit). I hope I can start next week with the design. Currently I have no idea how to drive these tubes, must do some research.

[Terry Kennedy]

On Tuesday, July 7, 2020 at 4:53:23 AM UTC-4, Richard Scales wrote:

The original smart socket code which was created by Chris Barron ran on a PIC16F690 which did not have enough ports to drive all 16 segments plus the decimal point so I moved over to a PIC16F886 (I am no PIC expert so the choice was based on what my development environment and programmer would support).

There was an IV-17 smartsocket project which exposed very serious problems in how the sockets enumerated on power-up. Those were 4-tube boards and it was basically impossible to guarantee that all 4 boards of a 16-tube display would all be visible and in the correct sequence. I have a 4-board chain and a pile-o-PICs in the basement from that era. AFAIK, the issue was never resolved. It may have been because it was a 4 tube per board design, but it is something to watch out for with smartsockets. 

[Richard Scales]

I have personally only ever used single socket board designs and have not seen that issue (yet!). I have also only used a maximum of 16 sockets in any single chain. 

I just had a quick look at the IV-17 files for the first time and I can see that it's a little more sophisticated than the single display version and it includes some commands to re-enumerate the displays (software version 235), is that the version you were using and did you try the commands to re-enumerate the tube numbers?

- Richard

[gregebert]

If the smart-sockets use a serial interface, my number 1 suspect of erratic/unexpected behavior is hold-time violations on the serial-data signal vs the clock signal.

[Richard Scales]

I've finally got to the point of displaying meaningful data on the Smart-Socket driven 16 segment LED's.

Lots of work to do on the software but a basic clock showing time, day, date, month, year, temperature, pressure, wind speed and direction and a random word is working. Lots of display choreography to be sorted out and a case - but it's getting there.

 - Richard

[Nicholas Stock]

Nicely done!!

Nick

Sent from my iPhone

On Sep 4, 2020, at 04:15, Richard Scales <ric...@scalesweb.co.uk> wrote:



I've finally got to the point of displaying meaningful data on the Smart-Socket driven 16 segment LED's.

Lots of work to do on the software but a basic clock showing time, day, date, month, year, temperature, pressure, wind speed and direction and a random word is working. Lots of display choreography to be sorted out and a case - but it's getting there.

 - Richard

<pic1.jpg>

<pic3.jpg>

<pic2.jpg>

<pic4.jpg>

<pic5.jpg>

<pic6.jpg>

<pic7.jpg>

<pic8.jpg>

<pic9.jpg>

On Monday, 13 July 2020 at 06:49:41 UTC+1 gregebert wrote:

If the smart-sockets use a serial interface, my number 1 suspect of erratic/unexpected behavior is hold-time violations on the serial-data signal vs the clock signal.

--

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 view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/9b3e1594-c39f-4436-924a-63e6347ddd41n%40googlegroups.com.

<pic7.jpg>

<pic1.jpg>

<pic4.jpg>

<pic6.jpg>

<pic5.jpg>

<pic2.jpg>

<pic8.jpg>

<pic3.jpg>

<pic9.jpg>