I have friends who I built Tubehobby clocks for over 10 years ago (multiplexed design) that are still going strong with both IN14 and IN18 tubes....I think the concerns about multiplexing aren't worth worrying too much about (IMHO).
Nick Sent from my iPhone > On Jul 16, 2021, at 16:01, ZY <zongyiy...@gmail.com> wrote: > > Do you use PWM at all for dimming, or do you control the tube voltage? I do > find that my IN-18s sing a bit at certain PWM settings so I've wondered too > if that would have a negative effect. > > In terms of motion sensing, I did add that feature to a recent build I did > but I was curious if the thermal cycling from tubes turning on/off would be > more harmful than just keeping them constantly on. I mean the tubes do get a > tiny bit more warm than ambient when on. > >> On Thursday, 15 July 2021 at 11:47:49 UTC-4 gregebert wrote: >> From what I've read, the peak current is what leads to tube wearout, and the >> effect is exponential: double the current, and the lifetime will be less >> than half the normal expectancy. So when a tube is pulsed, such as with >> multiplexing, the perceived brightness will always be lower compared to >> static operation and the only way to compensate for that is higher current. >> If you have a single tube you can control with a microcontroller, you can >> easily emulate how it will appear when multiplexed. >> >> I have limited data on perceived brightness vs current. When I was taking >> extensive data on my b7971 and RZ568m tubes, I found that once you get to >> about 75% of normal operating current, further increases in current dont >> give as noticeable increase in brightness. Of course, this is subjective >> because I didn't use a photometer to measure the optical output. I never >> attempted to get PWM data, because I only use direct-drive in my designs. >> >> Multiplexing is tricky because you have to tradeoff between peak current, >> duty-cycle, frequency, dead-time (to minimize ghosting), ambient lighting. >> Ultimately, it amounts to a tradeoff between brightness and tube life. Some >> folks have said you can actually hear tubes making noise when multiplexed or >> PWM-dimmed, and I've often wondered if the mechanical stresses involved in >> producing the noise affect the tube's life. >> >> The best way to prolong tube life is to make sure you never exceed the >> maximum current rating in the datasheet. Better manufacturers will provide >> both min and max values, and if they dont provide a typical value, then use >> the average of min & max to select the operating current. >> >> Next, use a cheap passive- infrared motion sensor (PIR) to shut off the >> display when nobody is around, or when you are asleep. This will greatly >> extend the life of your tubes. >> >> Unless the tube is specifically designed for multiplexed operation, such as >> multi-digit displays that have shared cathodes, your best option is to use >> direct-drive to maximize lifetime. Electronic parts are inexpensive and >> readily available; nixie tubes are expensive and the finite supply is >> rapidly shrinking. >> >>> On Thursday, July 15, 2021 at 1:33:49 AM UTC-7 ZY wrote: >>> >>> Ooops, sorry Gregebert, I'm not experienced with google groups and I didn't >>> know the difference between reply to group vs. reply to author. >>> >>> I'm going to repeat my question so it's group visible: >>> >>> Basically I was wondering for multiplexed vs direct drive, if given the >>> same voltage and current limiting resistor, would the max brightness of the >>> multiplexed tubes be basically like the direct drive tubes operating at 1/6 >>> pwm ratio? It might not be 1/6 brightness due to non-linearity, and maybe >>> additional effects based on frequency of the pwm or multiplexing. But in >>> that case, there shouldn't be any additional wear on the tubes right? The >>> wear only comes from if you want to increase the multiplexed tube's >>> brightness even more, to match a direct drive with a higher pwm ratio? >>> >>> I was wondering because I thought about going multiplexed to save on board >>> space and soldering time. >>>> On Tuesday, 13 July 2021 at 05:26:50 UTC-4 andybiker wrote: >>>> It's also worth mentioning that with a multiplexed design it's easy to dim >>>> the tubes for a dark room and obviously extend the life. >>>> I have a bedroom arduino-based clock that has 4x NL5780 multiplexed. >>>> The brightness auto-dims so that it's nice at night (and at a level that >>>> would be invisible during the day) and then is bright during the day. >>>> >>>>> On Tuesday, July 13, 2021 at 6:28:29 AM UTC+1 owen.cra...@gmail.com wrote: >>>>> Makes perfect sense. Thank you for the explanation. Think I'll be going >>>>> down the direct drive route for my designs. >>>>> >>>>>> On Tuesday, 13 July 2021 at 14:43:30 UTC+10 gregebert wrote: >>>>>> If you use 1 decoder for 6 tubes, the duty-cycle is 1/6 and it will >>>>>> require more current to get the same brightness as a tube with a higher >>>>>> duty-cycle. So, if you use 2 decoders, the duty-cycle is 1/3 and the >>>>>> peak current will be less vs 1/6 duty cycle. >>>>>> >>>>>> Personally, I never multiplex tubes because it requires more current per >>>>>> tube, and the wearout mechanism for tubes (sputtering), increases >>>>>> exponentially with current. In my opinion, IC's are cheap and readily >>>>>> available whereas nixies are much more expensive and getting harder to >>>>>> find, so it makes the most sense to drive the tube as gently as >>>>>> possible, ie direct-drive, to prolong their life. >>>>>> >>>>>>> On Monday, July 12, 2021 at 8:00:07 PM UTC-7 owen.cra...@gmail.com >>>>>>> wrote: >>>>>>> A little but of a noob question, though looking around at different >>>>>>> multiplex clock designs, I see that some setups opt for 2 x K155ID1and >>>>>>> others only have 1. >>>>>>> >>>>>>> A classic example with 2 x drivers would be: >>>>>>> http://www.arduinix.com/Main/Gallery.htm >>>>>>> >>>>>>> Then I find others with just 1 x drivers: >>>>>>> https://www.instructables.com/simple-user-adjustable-DIY-Nixie-Clock/ >>>>>>> And: https://www.nixieclock.biz/StoreClassicRev5.html >>>>>>> >>>>>>> Why is this the case? I have also seen similar sketch codes to drive >>>>>>> both single and double drivers, so I am wondering what the decision >>>>>>> process around this is and if it's even necessary to have 2 drivers to >>>>>>> begin with? > > -- > 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+unsubscr...@googlegroups.com. > To view this discussion on the web, visit > https://groups.google.com/d/msgid/neonixie-l/7410c5f2-0be9-4ba4-8669-9752f772cc05n%40googlegroups.com. -- 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+unsubscr...@googlegroups.com. To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/4AA8E96E-1B13-4C8F-AB38-0D91DF47F3F4%40gmail.com.
