[neonixie-l] Re: Helping Nixie Tubes Fire in a Darkened Room
I've done some testing with common 3mm UV LEDs that are in the 395nM range. Running them at ~5mA provides a dim purple color light that when placed under or in front of the tube actually makes the reliability of the ignition very stable. Even in a fully darkened room, the segments and decimal points light reliably. This particular tube (MG-17G) has segments with no anode screen in front of the tube, so they seem to behave a little differently than a regular nixie. I also found that if I placed fingers on the sides of the tubes while operating, the segments lit more reliably. I think that I will go for an LED under each tube but powered just enough to make the tubes work better and not to try to supply any visible base lighting. I will try 365nM LED's when they arrive and may use them if there is less noticeable, visible light. Thanks for the suggestions! Jeff On Wednesday, November 29, 2023 at 8:43:48 PM UTC-6 Jeff Walton wrote: > I've recently come across a situation where I have some tubes in a clock > that are being directly driven and are having trouble starting when the > room is darkened but light right up when a room light is turned on. These > particular tubes were probably intended for use in a calculator. They are > seven segment neon MG-17G tubes. Once the tubes have any of the segments > lit, there is really no issue with the performance. It's when the tubes go > completely dark if a space is used while scrolling a message or lighting a > dash on and off to emulate a colon. I'm wondering if others have found any > particular tricks to help convince tubes to light up. There is no > "baselighting" and the HV is ~172v. I'm considering increasing the HV by > 10-15v but don't want to over drive the tubes. Short of putting a > radioactive source in the vicinity, are there other things that anyone has > had any luck with? > > Jeff > -- 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/cfde3460-1737-4506-a58e-a1a6401ca7cen%40googlegroups.com.
Re: [neonixie-l] Scope Clock with 3LO1i
Hi The DH3/91 is a nice tube. It has "automatic focus" so there is no external connection to the focus anode. To keep the pin count down to 8 it has asymmetrical Y deflection but it is an easy low voltage tube to use. I used one in this project http://www.sgitheach.org.uk/scope1.html Grahame On 02/12/2023 20:14, Max DN wrote: @ Anders Thanks for that. Yes, I'm sure you are right. At the moment I can only get it to work well with Brightness and Focus potentiometer in parallel (it's working ok, but I want to change that and have those potentiometers in series as per standard config). Picture attached (CRT came with a center point already a bit burnt, no biggy, it'll likely extinguish itself soon anyway!). I only have 500k and 1M POTs at the moment. I'm waiting for a 100k POT for the Brightness and from my calcs that should help given my power supply (+240V and -450V). I'll update you when I receive the parts and I'll post my solution for the resistor bleeder. @Grahame. I bought a DH3/91 - super cute CRT! I cannot wait to try it. Il giorno sabato 2 dicembre 2023 alle 12:46:52 UTC Anders Mikkelsen ha scritto: The current draw for CRTs is mainly between the cathode and A2/A4 (main accelerating anode). A majority of this current flows directly between the electrodes, while a small fraction is represented by the actual beam current hitting the phosphor screen. You need to ensure that the impedance driving these electrodes is low enough that the voltages don't move around a lot when adjusting the brightness grid bias, otherwise the focus, astigmatism and brightness controls will interact in ways that make it hard to control, and brightness modulation will also affect focus. On Tuesday, November 28, 2023 at 8:49:39 PM UTC+1 Max DN wrote: I agree, that bright dot must be the beam resting on the same spot for too long. I'll check the code, where the refresh seems to be driven by the blanking pin, so with no blanking currently connected (still missing a part), the dot is not been managed. I thought that the same settings as for 3RP1A would work on 3LO1i but I guess not, they must have different physical characteristics. Il giorno martedì 28 novembre 2023 alle 17:05:12 UTC Grahame ha scritto: I would guess that the bright spot is where the beam rests after drawing the face and is waiting for the next refresh trigger. My own scope clock draws each face once every 20mS (16.66mS in 60Hz land) synchronised to the mains. If a particular face takes, say, 8mS to draw then the beam is parked somewhere and left blanked for the remaining 12mS (in this example) waiting for the next face refresh trigger. So if the beam is left on for the waiting period then it will appear as a very bright spot at the parked position. You'd have to investigate the software you're using for what triggers the face drawing and what happens in the waiting period. Grahame On 27/11/2023 22:50, Max DN wrote: I'm attaching the voltage divider that I am using and the one that I'm going to try using values calculated assuming 0.5mA current draw (which I understand makes sense for this CRT). Sharing this useful website to calculate resistor values using parameters from the datasheet, it is starting to make sense now! https://www.robkalmeijer.nl/techniek/electronica/radiotechniek/hambladen/qst/1946/12/page45/index.html Blanking pin currently disconnected (still waiting for an optocoupler). I wonder whether that will fix the bright spot (of course it will fix the smears on the display). @ gregebert - I may find this latest update useful Il giorno lunedì 27 novembre 2023 alle 09:28:02 UTC Max Di Noi ha scritto: Ok, I managed to make some improvements by changing some resistors on the voltage divider, I'll post my new layout later. Much better now, although I need to better understand how to calculate resistor values as at the moment I don't know current draws by each crt pin. I get a very bright spot on the clock image, I don't know if that's because I have not connected the blanking circuitry yet or overvoltage on A1 (I'm waiting for an optocoupler, hopefully it arrives this week). Picture attached. Getting there. Original message From: 'Grahame' via neonixie-l Date: 26/11/2023
Re: [neonixie-l] Scope Clock with 3LO1i
The current draw for CRTs is mainly between the cathode and A2/A4 (main accelerating anode). A majority of this current flows directly between the electrodes, while a small fraction is represented by the actual beam current hitting the phosphor screen. You need to ensure that the impedance driving these electrodes is low enough that the voltages don't move around a lot when adjusting the brightness grid bias, otherwise the focus, astigmatism and brightness controls will interact in ways that make it hard to control, and brightness modulation will also affect focus. On Tuesday, November 28, 2023 at 8:49:39 PM UTC+1 Max DN wrote: > I agree, that bright dot must be the beam resting on the same spot for too > long. I'll check the code, where the refresh seems to be driven by the > blanking pin, so with no blanking currently connected (still missing a > part), the dot is not been managed. > > I thought that the same settings as for 3RP1A would work on 3LO1i but I > guess not, they must have different physical characteristics. > > > Il giorno martedì 28 novembre 2023 alle 17:05:12 UTC Grahame ha scritto: > >> I would guess that the bright spot is where the beam rests after drawing >> the face and is waiting for the next refresh trigger. >> >> My own scope clock draws each face once every 20mS (16.66mS in 60Hz land) >> synchronised to the mains. If a particular face takes, say, 8mS to draw >> then the beam is parked somewhere and left blanked for the remaining 12mS >> (in this example) waiting for the next face refresh trigger. So if the beam >> is left on for the waiting period then it will appear as a very bright spot >> at the parked position. You'd have to investigate the software you're using >> for what triggers the face drawing and what happens in the waiting period. >> >> Grahame >> On 27/11/2023 22:50, Max DN wrote: >> >> I'm attaching the voltage divider that I am using and the one that I'm >> going to try using values calculated assuming 0.5mA current draw (which I >> understand makes sense for this CRT). >> >> Sharing this useful website to calculate resistor values using parameters >> from the datasheet, it is starting to make sense now! >> >> https://www.robkalmeijer.nl/techniek/electronica/radiotechniek/hambladen/qst/1946/12/page45/index.html >> >> Blanking pin currently disconnected (still waiting for an optocoupler). I >> wonder whether that will fix the bright spot (of course it will fix the >> smears on the display). >> >> @ gregebert - I may find this latest update useful >> >> Il giorno lunedì 27 novembre 2023 alle 09:28:02 UTC Max Di Noi ha scritto: >> >>> Ok, I managed to make some improvements by changing some resistors on >>> the voltage divider, I'll post my new layout later. >>> >>> Much better now, although I need to better understand how to calculate >>> resistor values as at the moment I don't know current draws by each crt >>> pin. >>> >>> I get a very bright spot on the clock image, I don't know if that's >>> because I have not connected the blanking circuitry yet or overvoltage on >>> A1 (I'm waiting for an optocoupler, hopefully it arrives this week). >>> >>> Picture attached. Getting there. >>> >>> >>> Original message >>> From: 'Grahame' via neonixie-l >>> Date: 26/11/2023 11:21 (GMT+00:00) >>> To: neoni...@googlegroups.com >>> Subject: Re: [neonixie-l] Scope Clock with 3LO1i >>> >>> Hi >>> >>> £55 is expensive - you should be able to get them for £30. >>> >>> Have you tied the deflection plates to A2 for testing purposes or left >>> them floating? They should be tied to A2. >>> >>> Can you turn the brightness pot down to fully extinguish the dot? If not >>> you need to be able to. >>> >>> Grahame >>> On 25/11/2023 18:28, Max DN wrote: >>> >>> @Grahame >>> Thanks for the suggestion. I didn't know about CV2302 DH3-91, I see >>> Langrex is selling it for £55 on eBay. >>> >>> Yes, my voltages were off. I have made some changes: >>> Cathode is now -350V to ground >>> A1 is +30V above cathode >>> A2 is +150V above ground, that is +500V above cathode >>> I'm having some problems with Grid voltage, I'm guessing... >>> >>> >>> @gregebert >>> This is the pinout I'm using: >>> Pins 1-14 heater >>> Pin 2 Cathode >>> Pin 3 Grid G1 >>> Pin 4 A1 (focus anode) >>> Pin 5 n/c >>> Pin 6 Grid G2 >>> Pin 7 X1 >>> Pin 8 X2 >>> Pin 9 A2 (acceleration anode) >>> Pin 10 Y1 >>> Pin 11 Y2 >>> Pin 12 Grid G3 >>> Pin 13 n/c >>> Pins 3-6-12 are all internally connected, other than on mine 3 is no >>> longer soldered on the grid. >>> >>> Pin 3 or 6 or 12 (Grid) is connected to -525V (that's 40V above -485V >>> negative voltage doubler in Cathode Corner's schematic), still not good >>> (pic attached). I connected a few resistors in series 500k, then 1M to >>> bring voltage down up to -250V, with not much difference. So it's clear I >>> don't understand how to connect the Grid pin!! >>> >>> I did find a schematic here (suggests to reduce m
