I have a project in which I wanted the ability to make it modular. So, I use TLC5916 per tube (IV-9); I have about 6000+ tubes (lifetime supply). A station where I could use 1-8 modules (although I normally have all 8 modules).
The project is enclosed in glass. No heat dissipation, and the entire project does get warm or even very warm, but Not hot. Michail From: [email protected] <[email protected]> On Behalf Of Mark Moulding Sent: Sunday, May 10, 2026 5:06 PM To: neonixie-l <[email protected]> Subject: Re: [neonixie-l] Re: Do IV-9 tubes get very hot? I've found that the larger DR2000-type tubes do run stone cold, but the IV-9 and IV-16s will heat up a bit (warm to the touch) in continuous use. I've only made devices with a few digits - up to six. Sixteen could dissipate a reasonable amount of heat, but consider that there also must be a fair viewing area for that many tubes, even if there is a bezel with a thin lens, through which heat can escape. Twelve watts shouldn't be too bad at that size - perhaps 1" x 6" (25mm x 150mm) or so? especially if there are cooling vents somewhere in the bottom and top of the enclosure. ~~ Mark On Saturday, May 9, 2026 at 12:09:57 AM UTC-7 newxito wrote: To keep the segments warm, my plan was to use just PWM. Maybe it would be easier to add just a bias resistor for each segment. On the other hand, there are 128 segments, so it wouldn't be that simple after all. To limit the current I will configure the TLC5921 to deliver 19mA on the outputs. For the 4.5V 3A power supply I've found the AP62301WU-7. It needs just a few external components and is cheap, small and easy to solder. gregebert schrieb am Samstag, 9. Mai 2026 um 01:56:23 UTC+2: I also recommend current-limiting so that you dont get surge-current on cold filaments. I put preheat hardware and software in place; during preheat, all segments get just enough current to make them warm, but not glowing. This reduces the thermal shock when turning them on as well as off. I dont have enough numitrons to generate any lifetime data, and I dont run those clocks very often...only to gawk at them every few months. Then back to standby. On Friday, May 8, 2026 at 9:02:34 AM UTC-7 newxito wrote: Good point. I will add the thermal monitoring to the firmware. I will use the same firmware as for the nixie calculators, I just have to add a numitron display driver. The idea is to also reuse the same case, which already has ventilation slots at the top and bottom. This is my first numitron project. I'm now designing a 3A 4.5V power supply that will replace the HV PSU and a new driver board with 8 x TLC5921. I'll upload a thermal image once I've finished the project. -- 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 [email protected]<mailto:[email protected]>. To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/645f0464-9d38-4351-af98-8d2c827226efn%40googlegroups.com<https://groups.google.com/d/msgid/neonixie-l/645f0464-9d38-4351-af98-8d2c827226efn%40googlegroups.com?utm_medium=email&utm_source=footer>. -- 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 [email protected]. To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/DM6PR01MB514673FA806888D3CBFADBC282392%40DM6PR01MB5146.prod.exchangelabs.com.
