SALTechips / tubeclockdb.com giveaway

[Stuckey]

Hello -

SALTechips has donated a thermNeon kit to a giveaway via TubeClockDB.com. To enter, all you need to do is leave a post. 

http://www.tubeclockdb.com/forum/Notices/4889-Win-a-free-thermNeon-Kit-from-SALTechips%21.html

Disclosure - I own TubeClockDB.com

Brian

[michael gregg]

Neat. Thanks for the opportunity.

I saw the one that Dave at EEVblog received.

I like your design.

One question. How are you adjusting current in the bar graph? Are you
doing that with PWM?

Michael-

> --
> 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 an email to neoni...@googlegroups.com.
> To view this discussion on the web, visit
> https://groups.google.com/d/msg/neonixie-l/-/ugOHEpwZT1sJ.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

[AlexTsekenis]

Hi Michael,

Please allow me to briefly describe how that part of the circuit works.
I will be referring to the schematic diagram which you can find here.

As you know, the length of the illuminated bargraph is a function of the current flowing through it.
In the schematic area J-15, N1 is the IN-13 bargraph tube. The length is proportional to the current through terminals A(anode) and K(cathode).
Although the physics of the glow discharge inside the tube (primer here) will cause the tube to appear to function with a PWM signal, the error in displayed value will be unacceptable. Therefore the tube must be operated with a constant, or slowly varying current. Not PWM.

The thermNeon achieves this constant drive current by utilising an 'operational amplifier constant current sink'. The circuit is formed by IC3, Q3, R17 and P1. IC3 is trying to make the voltage on the top leg of R17 (emitter of Q3) equal to the voltage on its pin 3. By doing so, a current is allowed to flow through  R17, P1, Q3 and hence the cathode terminal. The value of the current is approximated by Ohm's law: I(tube)  ≈  V(pin 3) / (R17+P1). 

So far the circuit described takes a voltage on pin 3 of IC3 and turns it into a current through N1. The voltage on pin 3, which ultimately controls the length of the bargraph) is generated by the main microcontroller IC2, R14 and C10. IC2 generates a PWM signal on pin 13. The duty cycle of the is varied by the firmware according to the desired length of illumination. R14 and C10 form a low-pass filter, which effectively takes the average DC value 'contained' in the PWM signal output from IC2.

So there you have it, the bargraph length starts from a firmware instruction, turns into a PWM duty cycle, is averaged by a low-pass filter, drives a current sink which finally drives the IN-13 producing a beautiful neon glow :)

Hope this answers your question.

Alex

[mrLaw]

Hi all,

Thanks Brian for mentioning the competition to win a thermNeon kit which is currently running on TubeClockDB.

I thought I would take this opportunity to share some background behind the thermNeon and SALTechips.

Back in 2009 Alex Tsekenis built a thermometer using an IN-13 tube and in doing an internet search found that I had also put one together. Alex got in contact and we decided to collaborate on creating a new digital version and SALTechips was born!

We started to brainstorm what features we would like to see in the new thermometer and how they would be accessed. The menu system uses a combination of colour illumination and the scale centigrade values to show what option you are at and its value. Settings are stored in EEPROM memory.

When it came to choosing between the IN-9 and IN-13 tubes, the choice was clear. The IN-13 offers a larger bar-graph length, greatly reduced operating current, higher linearity and the auxiliary cathode. These qualities give the thermNeon an increased resolution and accuracy, reduced power consumption and self-heating and a reliable way to activate the tube. Despite its higher cost and reduced availability, the IN-13 is overall a more suitable tube.

Another subtlety of these bar-graph tubes is the loss of linearity as the glow approaches the top; doubling the input current will not double the bar-graph length. This is due to the overlap between the ‘normal’ and ‘abnormal’ glow operating modes of the tube, which can be modelled mathematically using exponential equations but it would require a lot of program memory to implement. The thermNeon elegantly accounts for this by using two different slope coefficients to drive the IN-13 tube. 

For more information on the theory and operation of the IN-13 tube, Alex has researched and written up an excellent document here.

Thinking about how we were going to mount the IN-13 tube with the scale, we decided to use a 12mm thick piece of acrylic, which would protect the tube from accidental knocks. A channel was machined out of the scale to slide the tube into, giving it a secure, but viewable place in the acrylic. The thick scale also makes reflections from the engraved gradations visible. We sanded the sides of the scale to give a frosted effect that smoothly diffuses the illumination.

Deciding on what temperature range to use, we looked at various temperatures around the world and decided to use 10 – 40 deg C (50 – 105 deg F) as that looked reasonable for an indoor room thermometer. If indoor room temperatures are below or above these temps, then more likely the user would be buying a heater or a fan instead of a nixie thermometer!

For the base design, we originally thought about using a solid piece of hardwood with most of its core routed out, but found the mechanics of the assembly difficult and it cost prohibitive. Looking at alternative ways of construction, we found using finger joints and 6mm wood, we could get the bases made up at a reasonable cost, be easy to assemble and look good! A stain and finish was used to highlight the grain and seal the pores of the wood to provide resistance from the elements and scratches.

Since we had built IN-13 thermometers before, we already had a very good idea of what would be involved with the electronics design for the new one. We started by interconnecting functional block diagrams of the different subcircuits. The schematic diagrams were then produced in EAGLE and simulated in PSpice, LTSpice and Mindi. A prototype was built to check the results from the simulators. Some of the firmware functionality such as the switch-mode controller was implemented in hardware until the firmware was ready. We decided to use mostly through hole components to make assembly more accessible.

The firmware was written in C in MPLAB X using the C18 compiler. About ¼ of the time went into documenting the C statements to improve readability and serviceability. The PIC18F1330 contains 3 hardware PWM channels, making it easier to control the number of PWM channels in the thermNeon (6 in total). The most challenging part of the firmware was the colour mixing for the scale illumination. Two RGB LEDs are assigned a 4-bit intensity value per die for a total of 4096 colours. The RGB values are calculated by a function that accepts an HSV colour representation, calculated depending on the read temperature, hot/cold user temperature setpoints and hot/cold user colour settings. Consequently these two functions have the largest Flash memory footprint. 

We have made the schematic, code and firmware of the themNeon freely available under CC BY-SA-NC and GPL. You can download them from our website here.

It has been a long journey for us (3 years!) from the original idea to the finished project. The electronics is only part of what one needs to consider and put in place before making the project available to others. We hope you enjoy the thermNeon as much as we did working on the project.

For more information, you can visit us at our website, facebook or youtube pages.

Cheers

Simon

[Dennis]

Thanks Simon. A great read and a really cool thermometer.

---

From: mrLaw <shk...@gmail.com>
To: neoni...@googlegroups.com
Sent: Tuesday, April 23, 2013 7:50 AM
Subject: [neonixie-l] Re: SALTechips / tubeclockdb.com giveaway

--

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 an email to neoni...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msg/neonixie-l/-/lkW6wzo4Ts4J.

[Stuckey]

Hi everyone -

Just a reminder that the drawing is tonight at midnight Eastern time. If you haven't entered yet, now is a good time to do so :)

http://www.tubeclockdb.com/forum/Notices/4889-Win-a-free-thermNeon-Kit-from-SALTechips%21.html

Brian