Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

My Nixie clock project feels really close to working... but not quite

I've been iterating on this design off and on for quite a while, adding more and more complications as a way to learn different things.

Now I have a 12 Nixie display for date and time. I have a main controller board with a PIC MCU as the brains, which also controls a boost converter which generates the 180V supply for the Nixies. I'm getting 12V from a USB PD supply using a Diodes Incorporated 33772. A U-Blox Neo 6M GPS receiver provides the initial time source, and a DS3231 RTC keeps time from there. I also have a rotary encoder for input and mini OLED to provide status and diagnostics. Each individual Nixie has its own driver board with an MCU that controls the tube and is itself commanded over I2C.

PXL_20241014_190230716.jpg

PXL_20241014_190239413.jpg

PXL_20240717_012308850.jpg


This is all so close to working. The problem I have is that I can't get more than 3 or 4 Nixies to consistently light at once. I've done a bunch of debugging, and all I've really accomplished is eliminating possibilities.


I first thought it might be an I2C bus issue, like I have too much capacitance with all of the peripherals and controllers combined. I tested this by having the controllers echo back their commanded digit so I could confirm that they were all listening and hearing. This confirmed that all 12 controllers were up and responding.

Then I considered that maybe my boost converter wasn't supplying enough current. If that were the case, though, I would expect the voltage to crash, which it doesn't. It holds pretty steady at 180V.

My current suspect is my safety load circuit. At the interface to each Nixie controller, is a circuit that watches for return power from the Nixie tube and switches current to a dummy load if the Nixie isn't consuming power. This is to prevent the boost converter from destroying itself if the Nixie doesn't light for some reason. Here's a schematic:

Screenshot 2024-10-14 133309.png


NIX1:1 is a dedicated ground for the Nixie tube. If there is no current there, Q1 will pull the base of Q2 up, which enables the path to ground for the dummy resistors (and lights up the error LED).

I'm wondering if this circuit effectively latches to the "safe" state and sinks too much current, preventing more than a couple of Nixies from igniting. I had the brilliant idea of testing this by disabling all the dummy loads and seeing if all the Nixies would light up. Of course I didn't add any safeties to the control software so the boost converter promptly blew up (not really, it just kind of smoked serenely).

I'm not sure where to go from here. Is the design of the safety circuit bad? Is a safety circuit just a fundamentally flawed idea and I need to protect the boost converter in the control software? Does it sound like I'm barking up the wrong tree entirely and the problem is elsewhere?
 
I'd completely remove the current sensing circuit, there is no need for that at all on a tube-by-tube basis.

If you are concerned about the HT supply, add an overall shunt limiter to that directly, such as a zener controlling a transistor that switches a load resistor in.

Also, sensing anything in the ground return may cause intermittent problems in the communications between the PICs by causing glitches / spikes.

Make sure all things that are supposed to be ground are tied directly to the main board ground plane.
 
I'd completely remove the current sensing circuit, there is no need for that at all on a tube-by-tube basis.

If you are concerned about the HT supply, add an overall shunt limiter to that directly, such as a zener controlling a transistor that switches a load resistor in.

Also, sensing anything in the ground return may cause intermittent problems in the communications between the PICs by causing glitches / spikes.

Make sure all things that are supposed to be ground are tied directly to the main board ground plane.
What he said.
Also, if NIX1:1 and NIX1:2 are connected together internally in NIX1, and pin 2 is grounded, then the dummy load will always be connected. Can you look at LED DUM to see if this is occurring? Or am I misinterpreting your schematic?

 

New Articles From Microcontroller Tips

Back
Top