Страница большая – добавляю сканы со справочного листа. Техника 50-х – 80-х годов, кое где и сейчас остались – известен автомат для точечной сварки без микроконтроллера а на такой лампе. С немецкого поможет перевести немка – или наша училка.. дружить надо а не воевать. а то кто там денег просит чтобы опять пойти подраться – что история повторяется??.

Общаться надо, только только в 2009 девушка ехала одна изучать язык и смотреть чем мы занимаемся, от западных немцев через Польшу Киев в Москву и хотела в Питер, переезжала на поезде и останавливалась на недельку в городах. Может чем здесь заниматься – танцами например или ателье сделать или чем еще, про интернет спрашивала и телефоны, и про антиквариат какой то . А по теме – чешский мастер делает цифровые лампы неоновые а болгарский – газовое освещение винтажное , типа карбидки на ацетилене.

ниже проект часов на декатронах . Все очень неплохо смотрится в комнате или в доме (на даче) , еще одна особенность этих индикаторов – сейчас попробую показать – они очень шустрые, вот переключение 200 тысяч раз в секунду, второй разряд реально каждый импульс вспыхивает, первый конечно не успевает. ИН-12 они содержат немножко ртути и написано активированные что это значит расписывать здесьне буду, чтобы народ не пугать, через стекло не страшно. Скорость сиреневого декатрона до миллиона в секунду но он не очень яркий, ночью хорошо видно, оранжевого от 1400 (А101) до 20000 в секунду.

Die Tausch-, Informations- und Sammler-Seite für
Nixie-, und andere Anzeigeröhren

The information
and collector page for Nixie and other display tubes

Zurück zur Hauptseite/
Back to Main-Page

Some examples of

Abkürzungen/Shortcuts:

Abkürzung/ Shortcut	German	English
Richtg./ Direct.	Zählrichtung	Counting direction
COU	Zählröhre	Counting tube
NOM	Nomotron	Nomotron
SEL	Zähl- und Selector- Röhre	Counting- and selector tube
REG	Register- Röhre	Register Tube
BI	Richtung bidirektional	Counting direction: bidirectional
UNI	Richtung unidirektional	Counting direction: unidirectional
DP	Doppel- Impuls- Arbeitsweise	Double pulse operation
SP	Einzel- Impuls- Arbeitsweise	Single pulse operation

Zurück zur Hauptseite/Back to Main- Page

Dieter’s
K8A Dekatron Spinner
last update (and finished): November-25-2007

Back to Dieter’s Nixie Tube Page
Email to Dieter
What is a Dekatron spinner?
A Dekatron spinner displays a spinning dot by using a vintage ‘Dekatron’ style glow-transfer counting tube.
The colour of the dot depends to the gas the tube is filled.
Neon filled tubes display a red-orange dot.
Argon filled tubes display a purple dot.
The spinning frequency depends to the mains voltage the spinner is connected to and to the counting steps of the Dekatron.
If the spinner is connected 50Hz and the Dekatron is a 10-way counter (like the one here in this project) the dot spins with 5 turns per second.
If the spinner is connected to 60Hz the dot spins with 6 turns per second.

An animation of a Dekatron spinner:

You can find more information about Dekatrons here on my Dekatron page.

The main idea of this project was to develope a simple Dekatron spinner for all known Dekatrons with K8A base.
I took my B13B Deaktron spinner circuit and modified it to this purpose..

After my modifications I came to this circuit:

The circuit works with 230VAC/50-60Hz and 115VAC/50-60Hz.
All K8A Dekatrons I know will work in this circuit.
These are the following:

CV2271; Z303C; GC10B; GC10B/S; 6802; CK6802; 6909; CK6909; GC10B/L; CV6044; GC10/4B/L; CV6100; VX9194/4B; GC10/4B; CV1739; VX9194; M2465-401C; GC12/4B; GC10D; CV5143; 6482; Z302C; OG3(ОГ3); OG4(ОГ4); OG5(ОГ5); OG5-II-60(ОГ5-II-60); OG7(ОГ7); OG8(ОГ9); OG9(ОГ9); CT783.

Here you see the layout of the circuit:

Here you see the finished board (unequipped):

Here you see the spinner with different tubes:

Z303C (Adzam)

CV2271 (Mullard)

GC10B (Trigon)

GC10/4B (Hivac)

GC12/4B (ETL / Baird Atomic)

OG3 (ОГ3) (Reflector/Sovtec)

This pictures show the configuration of the board for the OG3 Dekatron as example:

OG4 (ОГ4) (Reflector/Sovtec)

CT783 (Sylvania)

GC10D (Hivac)

This pictures show the configuration of the board for the GC10D and the CV5143 Dekatron as example:

CK6802 (Raytheon)

6802(Sylvania):

6909 (Sylvania)

DK23 (Rodin)

VX9194 CV6044 (ETL)

GC10/4B (ETL/Baird Atomic)

GC10/4B (Etelco)

The bottom side of the board:

The tube is mounted in pin receptacles, so it’s no problem to replace it:

The following pictures show the tube spinning.
Of course, only one of the cathodes is glowing, but the opening time of the camera leads to that different pictures:

a middle opening time of the camera:

a short opening time of the camera:

a long opening time of the camera:

by night, with long opening time of the camera:

The wonderful purple glow of Argon filled tubes:
a long opening time of the camera:

a short opening time of the camera:

by night, with long opening time of the camera:

An animation of a spinning dekatron:

You can download the connection-instructions here.

какие есть лампы и индикаторы –

Die Tausch-, Informations- und Sammler-Seite für Nixie- und andere Anzeigeröhren
The information and collector page for Nixie and other display tubes

Die letzten Updates finden Sie unter Nixie News:
The last updates you’ll find at Nixie News:

Go to Tube-Tester.com

Email
Disclaimer
Impressum

Schlüsselwörter: nixie nixieröhre nixieröhren röhre röhren tube tubes buizen indicator numerical vfd vacuum clock nixieclock uhr readout

Лампы Никси трубки – у нас это цифровые газоразрядные индикаторы , неоновые лампы, еще есть типы газовых стабилитронов, ламп с холодным катодом и счетчики импульсов – декатроны. Другой тип – вакуумные люминесцентные индикаторы.

несколько более распространенных у нас –

А-108

Миниатюрный коммутаторный декатрон в бесцокольном исполнении, для работы в счетных и счетно-решающих устройствах.

Это самый низковольтный из приборов этого класса, его напряжение горения составляет 135+10 вольт (в то время как у большинства остальных оно лежит в пределах 200-300 вольт); и самый экономичный, с рабочим током анода вдвое меньше, чем у одноклассников.

А управляется он вообще импульсами со сверхнизкой для декатронов амплитудой, всего 18 вольт. Такое сочетание параметров должно благотворно влиять на удобство применения совместно с полупроводниковыми приборами.

Вероятно, платой за это стало невысокое быстродействие, скорость счета у А-108 не превышает 15 кГц.

Справочный лист и данные из отраслевого каталога на него. Судя по фиолетово-синему цвету свечения, наполнен он гелиево-водородной смесью.

Производитель – завод “Анод”, Дятьково (Брянская обл.).

Источники:

1. Справочник по электронным приборам. Гурлев Д.С. Изд. 5-е, исправленное и дополненное. “Технiка”, 1974
2. Згурский В. С. и Лисицын Б. Л. Элементы индикации. Справочник. М., “Энергия”, 1974.
3. Кацнельсон Б.В. и др. Электровакуумные электронные и газоразрядные приборы: Справочник / Б.В. Кацнельсон, А.М. Калугин, А.С. Ларионов; Под общ. ред. А.С. Ларионова. – 2-е изд., перераб. и доп. – М.: Радио и связь, 1985

добавлю еще А101 и А107 ОГ-3 (есть на сайте 155la3.ru). А108 не нашел пока – ИН-13 есть парочка – в индикатор громкости звука конечно. (продал я как то в середине 90-х самодельный усилитель, там ничего особенного, КТ825 – КТ827 на +- 40 вольт и по схеме из Радио 1985, она же от Американцев. Так вот в предусилителе вот такие индикаторы стояли ,VU-meter.) Сейчас их поискать..

нецелевое использование декатронов

Блог им. woddy Если вы хотите использовать аутентичную схему — вам не сюда. Здесь же я расскажу как использовать декатрон в качестве необычного индикатора.

www.adslclub.ru/upload/img/2014-02/27-22375665849.jpg

Декатроны бывают
по цвету
* неоновый наполнитель. Например ОГ4, А101 — светят оранжевым. Работают на частотах до 1-2кгц
* водородно какая-то смесь. Светят голубовато-фиолетовым. Работают на частотах до 20-50кгц.

по типу управления
* Двухимпульсные. По факту там три группы катодов, с таким и будем работать.
* Одноимпульсные. По факту там четыре группы катодов.

по фунционалу
* полиатроны (а101 например) позволяют управлять внешней нагрузкой (например газоразрядной лампой. ссылку на часы вставить). Позволяют иметь перестраиваемый коэф деления.
* обычные счетчики. ог4 например. имеют отвод только от 0 и умеют делить на 10.

Мне попались ОГ-4 (просто потому что были у мужика на рынке). В сети самое распространенное использование декатрона — бесполезная крутилка. Собирается на десятке деталей и включается в розетку. Но я решил что крутилка это не интересно. Потому будем собирать столь же бесполезные часы.

Типовые схемы включения достаточно необычные, для стабильной работы требуются времязадающие цепочки и напряжение смещения. Но для режима обычного индикатора они и не нужны. Нам для работы нужен только источник высокого напряжения. Я выбрал степап 12 -> 350в. Это близко к штатным режимам работы. Но по отзывам в сети, может быть достаточно и 310, выпрямленного 220в из сети. (будьте осторожны!!!). Я использовал step-up на mc34063, популярная схема, широко используется для питания газоразрядных индикаторов 12в -> 180в. На выходе удвоитель, поднимает напряжение до 350-360 вольт.

А для управления нам надо всего лишь 4 высоковольтных транзистора, которыми будет дергать АТМега. Я взял st13003, т.к. они были в наличии в ближайшем магазине. Можно так же использовать популярные 13001 которые стоят во многих китайских блоках питания и зарядных устройствах, или любые мосфеты на 400 вольт (поправив резисторы в цепи управления).
www.adslclub.ru/upload/img/2014-02/27-22284091971.gif

Принцип управления какой. В лампе идут по кругу катоды 123123123,… всего 30 штук, объединенных в три группы. При подаче питания на первый в произвольном месте возникает разряд: 123123123. Дальше подавая питание на соседние катоды и снимая питание с 1го мы можем «таскать» разряд в нужную сторону.
Например хотим чтоб огонек двигался вправо, значит включаем катод 2, после гасим катод 1. и огонек перескочил 123123123. Дальше аналогично включаем катод 3, после гасим катод 2: 123123123.

Но есть еще и нулевой катод. Это как бы один из штырьков 3го катода, но не подключенный к общей шине. Он позволит нам позиционировать разряд. Чтоб переключить огонек на него, то мы должны отключить остальные катоды, а нулевой включить: 123120123123123.

В один момент времени горит только одна точка. Все спецэффекты мы получаем только гоняя её туда-обратно по кругу.

Для подключения декатрона можно купить панельку от старого лампового телевизора, или же красивую панельку для реле на дин-рейку. Она мне обошлась всего 25 рублей, называлась «база реле pf083a» (есть на алиэкспрессе)

www.adslclub.ru/upload/img/2014-02/27-22193116717.jpg

Собираем тестовую схему, подключаем к ардуине

www.adslclub.ru/upload/img/2014-02/27-22223243135.jpg

www.adslclub.ru/upload/img/2014-02/27-22224099072.jpg

www.adslclub.ru/upload/img/2014-02/27-22222415798.jpg

Несколько тестов

Другие модели декатронов если попадут в руки — обновлю пост.

За схему включения спасибо
www.diyaudio.ru/forum/index.php?topic=2961.0
За источник питания спасибо
www.vecoven.com/elec/hv_supply/hv.html

upd продолжение

Решил сделать драйвер с возможностью подключать декатрон к почти любым часам. За основу взял ардуину про (100рублей на ебай), плата с умножителем и ключами.

итоговая схема вышла примерно такая
www.adslclub.ru/upload/img/2014-03/29-18221464659.gif

на входе 8 пинов. берем любую схему со статической индикацией и вместо 155ид1 подаем сигнал на драйвер.
www.adslclub.ru/upload/img/2014-03/29-18282558544.jpg
(диапазон в двоичном виде от 0х00 до 0х60)

если мы хотим динамическую индикацию, то тоже можно попробовать. пин7 на 1, пин6 на 0, пин5 и пин4 подключаем к ключам соответствующих разрядов, а пины 0-3 подключаем ко входу дешифратора (не тестировалось, за неимением часов с динамикой).

а при подаче кодов в диапазоне от 0х60 до 0x80 можно зажигать отдельную конкретную точку

прошивка на атмега328 и сорцы под ардуину в аттаче. фьюзы
low_fuses=0xFF
high_fuses=0xDA
extended_fuses=0x05

яркость контролируется ШИМом на 62кгц, менее заметны перепады.

размер прошивки всего 2.5кбайт, можно пересобрать под атмега8 (с изменением распиновки)

видео работы

+9
27 февраля 2014, 22:40
woddy
2

Файлы в топике: dekatron schem.gif, dekatron2.zip

Комментарии (9)

RSS свернуть / развернуть Картинок не видать. 0

count_enable
28 февраля 2014, 01:17

Перезалил в другое место. Это у тебя почему-то не видно… 0

woddy
28 февраля 2014, 10:25
↑

Интересно, где такое можно достать на поиграться… 0

Trjam
28 февраля 2014, 12:23

ebay от 12 до 20 баксов (наши), до 100 баксов импортные.
на рынках от 100 рублей. на истоке от 120р + доставка.
кстати а101 (на ебай есть) по отзывам красивее но имеет другой цололь и чуть другую схему(надо соединять катоды вместе) 0

woddy
28 февраля 2014, 13:07
↑

отличная идея для олдскульного тахометра:) лучше, чем 6Е5С.
спасибо за статью!

зы. кстати, а насколько яркое свечение? при боковом солнечном свете будет чего-нибудь видно? 0

madmazy
28 февраля 2014, 12:54

неоновые декатроны — яркие. не хуже обычных индикаторов (ин-8,12,14). видео сняты при ярком освещении над столом.
водородные декатроны (голубое свечение) гораздо более тусклые. сегодня купил пару, на неделе проверю. 0

woddy
28 февраля 2014, 13:09
↑

кстати а на схеме насколько я понял ukat — это подкатоды, kat0(1) — нулевой, а kat(8)? просто произвольный? 0

madmazy
28 февраля 2014, 17:00
↑

В скобочках — номер пина декатрона на панельке чтоб не путаться. У декатрона 8пиновый сокет, у которого задействованы 5 пинов. Документация на ог4 тут
www.155la3.ru/og4.htm

kat(8) — это основной катод. kat0(1) — нулевой штырек основного катода, который не подключен к собственно нулевому катоду.

Если соединить kat(8) и kat0(1) то у нас получится три совершенно равнозначных катода: подкатод1, подкатод2 и основной катод. Для целей индикации этого достаточно. 0

woddy
28 февраля 2014, 17:46
↑

обновил пост 0

woddy
30 марта 2014, 19:13

Только зарегистрированные и авторизованные пользователи могут оставлять комментарии.

Комментирую. В блоке питания на 34063 лучше даже не удвоитель а сделать отвод от катушки где 200 витков, а там где 13 витков соединяется на 12 вольт через датчик тока который 0.2 ома, соответственно между 6 и 7 резистор, питание на 6 вывод MC34063 а контроль по току 7 вывод, туда же подсоединяется первичная обмотка. Выпрямленное напряжение контролируется, на 5 выводе микросхемы должно быть 1.25 вольта, соответственно у меня на катушке отвод от 100 витков, и от 200 витков выпрямитель и питание на анод декатронов, через 750 килоом на каждый, 400 вольт. С отвода на лампы ИН-1 берется анодное 200 вольт, оно же через делитель 470 килоом и 5.7 килоом регулируется – подается к 5 лапе микросхемы. Если надо 170 вольт на ИН 14 – то отвод сделать от 80 витков, подобрав соответственно резисторы. Может помочь схема увеличения возможных интервалов импульсов для MC34063 – она есть на странице Фонарик на светодиоде, она снизит потребление тока от 12 вольт. Ключевой режим полевика – напрямую с микросхемы можно тоже, но ее режим будет тяжеловат, 2 лампы потянет а 6 скорее нет, ток микросхемы 1.5 ампера максимально, на 1 обмотке 25 вольт импульсы, ток полевика максимально тоже 1.5 ампера.

Часы на декатронах (с программой). Источник сделан на обычном трансформаторе не импульсном, это дляобучения советую. Импульсный сложнее если не инженер конечно.

Decatron Clock Project

My current project in the form of a web-log.

Ronald Dekker

Introduction

When Jim Oostveen, a colleague of mine, had to clear away his storage space, he invited me to have a look around to see if I could find something useful. Amongst a few dozen or so of some very interesting radio tubes, I was lucky to find a nice set of Nixies and a beautiful set of Decatrons (Fig. 1). Lucky me! Until then I only knew decatrons from pictures on the web. It is not a device one encounters everyday.

Figure 1. My set of decatron tubes consisting of three GC10B, four GC10/4B and two GR10A tubes

Surfing the web, the most popular thing to make with a decatron seems to be a spinner [1,2,3]. Looking at these spinners I thought it would be a nice idea to make a clock with six (or perhaps even seven) decatrons in a row, with the decatrons showing all kinds of display effects. Instead of simply incrementing the time every new second, it would be great to have a display of forward- and backward-spinning digits, slowly rolling-out to the new time, a little bit like the reels of a slot machine.

Having never encountered them, decatrons are completely new to me. How do they work, how to interface them to a processor? Instead of working on it for a few months or so and then writing a new web-page on it, I decided to present this page in the form of a web-log. The idea is that this page is expanded every time the project has advanced a bit.

The Decatron

Like the E1T “scaling tube” the decatron preceded the nixie tube, perhaps not by invention, but certainly by application. The most likely reason is that, besides a being a display device, both the Scaling Tube and the Decatron perform the counting function themselves, whereas Nixie tubes require quite a bit of electronics to make them work. They are both small wonders of ingenuity.

The Decatron was first announced in a paper in Electronic Engineering of May 1950 entitled: “The Decatron, A New Cold Cathode Counting Tube” by R.C. Bacon and J.R. Pollard from Ericsson Telephones Research Laboratories, Nottingham, England [4]. The paper discusses the general operation, the performance, and the circuit requirements. The working of the Decatron is best explained in the words of the inventors:

A cathode glow on one of a set of inter-connected cathodes, arranged around a central common anode, is caused to transfer from one position to the next by application of controlling voltages to intermediate electrodes or “guides”. This arrangement will become clear by considering Figs. 2 and 3.

Fig. 2 shows, around the central anode disk A, a number of equi-spaced rods or wires (on a pitch circle of 20mm. diameter in the case of GC10A). These wires actually comprise thirty electrodes, of which ten are cathodes, ten are guide 1 electrodes, and ten guide 2 electrodes. Nine of the cathodes are connected internally in the sequence shown in part in Fig. 2. The tenth cathode is isolated and brought out separately, as the output or marking cathode. The ten guide 1 electrodes are connected internally, g1, g1a, g1b, etc., and in the same way the ten guide 2 electrodes marked g2, g2a, etc., are internally linked. Hence the tube has five outlets, comprising anode, output cathode, cathode ring, guide 1 ring and guide 2 ring.

Referring to Fig. 2, together with the simple circuit arrangement of Fig. 3, let it be assumed that a discharge is taking place between anode A and cathode k2. If a negative pulse of 120 volts is applied to the group of guide 1 electrodes (g1, g1a, etc.) which were originally at +60 volts with respect to the cathode k2, a transfer of the glow will take place in a clockwise direction to electrode g1a. That is to say the glow discharge will transfer preferentially one position clockwise as opposed to two positions anticlockwise to electrode g1. If it be arranged that simultaneously with the restoration of guide 1 to +60 volts, a similar pulse is applied to guide 2, the glow will move clockwise to electrode g2a. The function of the two guide systems is thus to render certain the direction of motion of the discharge. When guide 2 is restored to its original potential of +60 volts relative to the cathode, the glow moves forward to the next cathode k3. This accomplishes one complete cycle of operation: for each such cycle the illumination moves one step in a clockwise direction.

It follows from the above argument that by reversing guide 1 and guide 2 connexions the glow can be caused to rotate in an anticlockwise direction.

Figure 4. The interior of the GC10B.

The paper beautifully describes the transfer of the glow from one cathode to another. However, it fails to explain why only one cathode ignites. For the reader in 1950 this was probably obvious, for the modern reader it is perhaps less so. Figure 3 shows that the anode is connected to the high voltage supply via a resistor. When the voltage is raised, at a certain moment one of cathode-anode gaps ionizes. Which one of the ten cathodes will ignite, depends on small differences in gap width, asymmetries, the accidental passing of an ionizing particle etc. When the first cathode has ignited, a current will start to flow that will cause a voltage drop over the anode resistor. The voltage drop will lower the anode potential to a value which is too low for the ionization of another cathode-anode gap, but high enough to maintain the glow on the first cathode.

A very simple trick is used to generate a carry signal when the tube has counted ten input pulses [4]:
When the discharge rests on the single cathode which is connected to a separate base pin, the current flowing to that electrode will give rise to a potential difference across a series resistor in the cathode lead (see fig. 3), so that it is possible to obtain an output pulse coincident with every tenth input cycle.
The separate cathode is called the “zero” cathode. Note that in contrast to the E1T scaling tube the decatron does not need any special circuitry to reset the tube to zero after it has reached nine.

The working of the decatron resembles that of a Charge Coupled Device (CCD). In a CCD device packages of charge are transported in a potential well induced by a three phase control gate system. The simplest way to drive the guides is connecting them to a sinusoidal input signal via a simple RC phase shifting network. This is the method used in the most popular decatron gadget on the web: the decatron spinner

First Experiments

Having never worked with decatron tubes before, I did not have a real “feeling” for them apart from what I learned from the datasheets [5]. I decided to do some simple experiments to see if the tube was really so simple to operate as the descriptions suggest. The circuit depicted in Fig. 5 is about the simplest circuit that can be used to test a decatron tube. When the anode voltage is applied, one of the cathodes will ignite, or “strike”. The anode potential at which one of the cathodes strikes is called the striking voltage. In my first experiments I used a high-voltage supply that “only” went up to 320 V. This was just enough to make the tube strike. I sometimes observed a delay between the application of the anode voltage and the striking of the tube. When it was dark in the room, the tube sometimes would not strike at all. Switching on the light usually ignited the tube. At page 71 in the Ericsson manual we find:
A minimum value for the H.T. supply voltage is quoted by the tube manufacturers. If a smaller supply voltage than this is used, the tube may not strike when the voltage is first applied. Since the tubes are not primed, there may be a delay of about a minute before they strike if a value of H.T. supply voltage near to the minimum recommended value is employed.
I did not have a clue as to what “primed” meant in this context, but David Forbes [9] was so kind to explain it:
There are nixie tubes and plasma display panels that have primer electrodes, which are small cathodes (typically) supplied with full ionization voltage at a very low current. The primer’s job is to generate enough ions in the envelope that the ionization of the main cathode will occur instantaneously instead of waiting for some random cosmic particle to kick it off. Some other types of tubes had a bit of radioactive krypton gas introduced to generate ionization.
The manual does not mention anything about the dependency of the striking voltage on the illumination conditions. However, the datasheet of the GC10B specifies the minimum supply voltage as 350 V under normal room illumination, so the effect must be well known. Obviously my 320 V is well below the specification for reliable operation.

Figure 5. Decatron test circuit

I now have a second H.V. supply (thanks to Wilma, Ronald and Peter), so that I can test the tube under normal conditions. Actually, there is no upper limit for H.T. supply voltage. Once the tube has ignited, the anode potential drops to the “maintaining voltage”, which is about 192 V for the GC10B. The maintaining voltage is to a large extend independent of the current flowing, so that the value of the anode resistor can be calculated by application of Ohm’s law: Ra = (Vht-192)/Ia, with Vht the H.T. supply voltage, and Ia the anode current. The datasheet specifies a minimum anode current of 250 µA (Vht=400 V and Ra=800 kohm). However, if the tube does not have to run at its maximum specified counting speed, much lower currents may be used. I prefer to use an as low as possible anode current since this will increase the life of the tube.

The life of the decatron tubes is normally many tens of thousands of hours. One of the factors which determines the life expectancy of the tube is the sputtering of material from the electrodes and the deposition of the sputtered material on adjacent cathodes [6]. This will alter the discharge characteristics of the electrodes resulting in a lower counting speed. Additionally sputtered material from the cathodes will darken the inside surface of the glass envelope. A longer life can be obtained by using a low as possible current and by regularly circulating the discharge. I found the lower limit for reliable operation to be 170 µA. For the clock I will start at a save value of 250 µA, and than decrease the current to the minimum value still yielding accurate counting.

Just for fun I measured the GC10B I-V characteristics with a Tektronix 370A curve-tracer. A curve tracer is used to characterize semiconductor components. Basically it is a combination of a variable saw-tooth generator and an oscilloscope. A saw-tooth shaped voltage is applied to a device under test and the current through the device is recorded. In order to be able to test high voltage diodes and transistors, a maximum saw-tooth voltage of 1400 V can be generated, so it is also ideal for testing tubes! Figure 6 depicts the I-V characteristic of the GC10B with cathode 0 tied to the common cathodes via a 100 kohm resistor, and the guides not connected. The curve starts at the origin. For increasing voltage, the current is basically zero until the tube strikes at 320 V (point A). The next few measurement points shoot over the I-V plane, eventually ending in point B. I have to say that I was completely surprised by this behavior. I have two possible explanations: in the first place the current and dissipation limiting circuit inside the curve-tracer can have difficulties with the sudden striking of the tube, resulting in oscillations. In the second place, it is possible that the negative impedance characteristic of the tube in combination with interconnect inductances and capacitances causes an oscillation that quickly dampens. The negative impedance behavior can be seen for the part of the curve between B and C. Here, as the current decreases the voltage over the tube increases, or in other words we have an decreasing current for an increasing voltage. Observe that under normal operation conditions (Ia=250 µA point B), the maintaining voltage is 192 V as expected. After the plasma has extinguished (point C), the current returns to zero, and remains zero going back to zero volts again. The “looping” of the current near the x-axis is caused by charging and de-charging of interconnect capacitances.

Figure 6. I-V curve of a GC10B measured on a Tektronix 370A curve-tracer.

Using the switches (Fig. 5) it was easy to move the glow around the tube. The sequence S1, S1+S2, S2 makes the glow rotate in one direction, the sequence S2, S1+S2, S1 makes it rotate in the opposite direction. Although the datasheet specifies optimal quide voltages of +Vg=60 V and -Vg=-60 V, it appeared that a reliable glow transfer could still be realized if they are reduced to +/- 20 V. This means that for interfacing the tube to a processor a normal small signal transistor like a BC550 with a BVceo of 100 V can be used. The absolute minimum values for my tube were +Vg=3 V and -Vg=-16 V.

With the switches in Fig. 5 open, there is a current of about 7 µA flowing out of the tube. Obviously when the tube has ignited, each of the guides is always adjacent to a glow. Apparently some of the ions end-up on the guide instead of on the cathode, resulting in a small leakage current. For a short moment I considered using this current to make the quides self-biasing. I even did a few tests, and although I could move the glow around, it did not function correctly.

Decatron Spinner

Certainly the most popular decatron circuit on the web must be the decatron spinner. A nice animation of a spinning decatron tube can be seen on Hans Summers site [7]. If I am not mistaken, the original spinner circuit can be found on Mike’s Electric Stuff pages [8]. The circuit only contains a few components and the effect is, certainly considering its simplicity, “striking”.

Figure 7. Simple decatron spinner circuit (adapted from the original circuit on Mike’s Electric Stuff pages [8])

In Fig. 7 I have reproduced Mike’s original circuit with some modifications to the component values. The circuit consists of two sub-circuits: a voltage doubler and a phase-shifter. The voltage doubler consisting of C1,C2, D1 and D2 rectifies and doubles the mains voltage so resulting in a DC H.T. voltage of 450 V. The phase shifter can be found around the circuit consisting of R3,R4,R5,C3 and C4. All 100 kohm resistors are actually two 47 kohm resistors in series to half the voltage across each resistor.

Figure 8. Simple decatron spinner

Please note that this circuit is directly connected to the mains! Contact with any part of the circuit can be lethal. The circuit should be build in a completely isolated box. Also remember that the capacitors will remain charged after the mains has been disconnected. With these kind of directly mains fed circuits one can nor be too careful!

Figure 9. Working of the phase shifter circuit in the decatron spinner.

To better understand the working of the phase shifter, I recorded the guide voltages on the scope (Fig. 8). For this measurement I used two transformers in cascade (240 V->6 V and 6 V->240 V) to isolate the circuit from the mains. A small 1k resistor was added in series with the cathodes so that I could see when the glow was resting on the cathodes or on the guides (lower trace in Fig. 9). When the voltage on guide 2 drops to ca. -16 V, the glow shifts from the cathode to quide 2 (point A). Due to the integrating network C4, R5 the voltage on guide 1 lags behind on the voltage on guide 2. At one point the voltage on quide 2 will increase again and with the voltage on guide 1 lagging behind, at a certain point the glow will transfer to guide 1. Finally when the voltage on guide 1 rises 16 V above the cathode potential, the glow transfers to the next cathode. Reliable and very simple! I have to say that I do not really understand the function of C3. To my opinion it can be omitted by increasing the value of R3, but I did not bother to test it.

CPU Interface

Any interface of the decatron to a CPU has to perform at least two functions. In the first place the CPU has to control the voltages on the guides so that the glow can rotate clock-wise or anti- clock-wise. Secondly, since a decatron is essentially a counting device, there must be some means to synchronize the counts on the tube to a counter in the CPU. For the synchronization there are basically two possibilities: 1. The CPU can reset the tube to zero, so that it forces the glow to a known position, 2. The CPU can detect the glow as it passes the zero cathode. Obviously, also a combination of the two possibilities is possible.

Figure 10. Idea to generate a reset pulse without the need for a negative power supply.

To reset the glow to the “zero” cathode, a negative pulse of about 100 V has to be applied to cathode zero with respect to the other cathodes. To avoid an additional negative supply voltage, the simple circuit of Fig. 10 can be used. When T1 is not conducting, C1 will be charged to 100 V. The charge current as well as the cathode current will flow via diode D to ground. When T is switched on, the positive plate of the capacitor will be connected to ground so that the negative plate will drop to -100 V below ground with D blocking. To my humble opinion this approach has two drawbacks. In the first place it requires a high voltage switching transistor. In the second place, if the reset fails, we still do not have any way of knowing on which cathode the glow rests. For these reasons I prefer to synchronize the tube by detecting the glow on cathode zero.

Figure 11. Breadboard circuit used to test the interfacing of the decatron tube to a digital environment.

The circuit depicted in Fig. 11 was used to test the CPU interface. The circuit uses three supply voltages: +350 V, + 24 V and -24 V with respect to ground. The cathodes of the GC10B are connected to ground. Pull-up resistors R1 and R2 keep the guides at +24 V when they are not activated. Transistors T2 and T3 pull the quides to -24 V when activated. In this test circuit a HEF4017 Johnson counter and a HEF4049 inverter are used to simulate the CPU outputs and inputs. The digital logic is fed from an auxiliary power supply which is derived from the -24 V supply using a standard 7805. So the -24 V is the ground for the digital circuits. The maximum input voltage of a 7805 is 35 V, so if needed the supply voltages can still be increased a little. The only thing we need to keep track of is the dissipation. For an input voltage of 24 V the voltage drop over the 7805 is 24-5=19 V. The thermal resistance of a TO-220 package to ambient is something like 70 K/W. If we want to limit the temperature of the 7805 to 50C with an ambient temperature of 20 C (so a 30 C increase) we have to limit the dissipation to 30/70=420 mW. With a voltage drop of 19 V this means a maximum current of 0.42/19=22 mA. For the processor I had a 16F874 of 16F876 in mind. They consume about 10 mA at 20 MHz clock, so we are well within the safe dissipation range.

Figure 12. Breadboard test circuit of Fig. 10 in practice.

The common cathodes have all been tied to ground. As mentioned before, the “zero” cathode is used to detect the position of the glow. Transistor T1 is used in what is called a “common base” configuration. Its working can easily be understood if we first consider the collector disconnected. In that case the emitter-base junction of T1 will be forward biased, and the current flowing out of cathode zero will flow through the forward biased emitter-base junction to ground. So cathode zero “sees” a potential Vgnd (Vgnd+0.8 V to be precise). With the cathode of T1 connected, almost all the emitter current will flow out of the collector into R7 ( Ic=Ie*hFE/(hFE+1) so for large hFE, Ic=Ie). In this way current and potential have been decoupled: the cathode still sees a ground potential, while the cathode current is redirected through R7 to -24 V.

Figure 13. Current flowing out of cathode zero for the glow resting on different positions around cathode zero (Va=350 V and Ra=600 kohm).

The “zero” cathode current causes a voltage drop over R7 which is used to signal the position of the glow. The value of R7 needs some consideration. The problem is that already when the glow is approaching the zero cathode a considerable current starts to flow out of cathode zero. The current is caused by electrons from the discharge on one of the adjacent guides which accidentally hit the “zero” cathode. Figure 13 gives the “zero” cathode current for different positions of the glow, for a 350 V anode voltage and an 600 kohm anode resistor. We find that when the glow rests on one of the guides immediately adjacent to the “zero” cathode the cathode current is approximately 8 µA. The value of R7 needs to be such that this current does not result in a erroneous “1” reading by the CPU. According to the PIC datasheet the input high and low levels are defined as: “0” < 0.8 V and “1” > 2 V. If we take R7=53 kohm, a glow on one of the adjacent guides will result in a voltage drop of 8 µA*53 kohm = 0.424 V, what the PIC will read as a “0”. When the current is resting on cathode zero the voltage drop will be 132 µA*53 kohm=7 V. However, when the voltage drop increases beyond 5 V, D2 will become forward biased and limit the voltage drop to 5 V. This will protect the input of the PIC for any over-voltage. In the test circuit of Fig. 11 LED D1 will light when the glow is resting on cathode “zero”.

Figure 14. Different quide driving schemes.

When I first built the test circuit of Fig. 11, I thought to be clever, and implemented the quide driving scheme depicted in Fig. 14B. This four phase driving scheme is obtained from the modified circuit on the left. The idea was that in this way there is a well defined overlap between the quide 1 low phase and guide 2 low phase. As it turned out this driving scheme resulted in an irregular spinning of the tube. What happens is the following: during the C phase the glow is resting on the cathode. When guide 1 becomes low (G1 phase) the glow transfers to guide 1. In the next phase guide 2 also becomes low. In this phase where both guides are low, the glow sometimes stays on guide 1 and sometimes transfers to guide 2, resulting in an irregular spinning. A perfect result is obtained with the driving scheme that is implemented in Fig. 11, and depicted in Fig. 14A. In theory a small overlap between guide 1 low and guide 2 low is required to prevent the glow from turning back to the last cathode (fig. 14C). In the circuit of Fig. 11 no special precautions have been taken to guarantee such an overlap. The capacitances of the guide electrodes probably take care of that, resulting in a perfect spinning of the glow. In the final CPU driven version I will first try the simple driving scheme of Fig. 14A. If that doesn’t work I will implement something like Fig. 14C.

Figure 15. Difference a glow resting on a cathode or on a guide.

When I had the decatron spinning, I observed that the glow also partially rests on the anode. Furthermore there was a difference in the glow near the anode when the glow was resting on an anode, or on one of the guides. I have tried to capture the effect in Fig. 15. The difference disappeared when the guide driving voltages where increased to +/- 60 V (Fig. 16).

Figure 16. The pattern of the glow resting on a cathode for different guide bias values.

The Case and the Circuit

Designing and making a suitable case is always the biggest problem for every new clock I build. For this particular clock, I was looking for an original way to mount the tubes in such a way that the beautiful construction of the interior of the decatron tubes would also be visible. This triggered the idea to mount the tubes vertically! This gives an optimal view on the interior of the tube, while a mirror placed at a 45 degree angle at the front of the tube will allow for reading of the tube.

In this way the clock can simply be attached to the wall as a kind of electronic painting. This idea additionally solves another problem: my house is slowly becoming saturated with clocks. One of the few rooms that is still without one of these “wonders of technology” is the toilet. An exotic and interesting clock with all kind of visual effects will certainly make nice change for the eternal birthday calendar. A “toilet clock” has the additional advantage that the tubes can be switched off when the toilet is not occupied. A simple LDR can be used to check if the light is on or off. This will greatly increase the life of my precious decatrons.

So much for the case. In the mean time I have decided on the processor I am going to use. Being a bit of a PIC addict, I selected the 16F874, mainly because of the large number of IO pins that I will need (ca. 24).

Figure 17. Schematic diagram of the power supply

Figure 17 depicts the circuit diagram of the power supply for the clock. It is based on a nice small transformer I found at “Radio Service Twente” [10]. They have a great stock of surplus and “out of production components”, and if you are ever in Den Haag (The Hague) [11], you should visit them. The transformer has two secondary windings: 24 V and 170 V with a tap at 60 V. There was no current rating for these windings but the DC resistances are 36 ohms and 1400 ohms respectively. Lets assume for the moment that each decatron will draw ca. 300 µA. Then the total HT current consumption will be less than 2 mA, resulting in a voltage drop over the 170 V winding of 2.8 V, resulting in a resistive loss of only 5.6 mW, which should be no problem at all. I compared the 24 V winding with another 24 V transformer I had lying around. This small transformer had a secondary resistance of 140 ohm, and is rated at 60 mA. This would imply that my transformer should be rated at (140/36)*60 mA= 240 mA. To me this seems rather a lot. I will try to keep the current consumption below 50 mA.

The simple voltage doubler circuit formed by D1,D2,C1 and C2 is used to generate the high tension anode supply for the decatrons. The unloaded output voltage is 430 V. The high tension supply is switched on and off by S1, which is one of the contacts of relay Re1.

The secondary 24 V winding is used to generate the +/- 35 V guide voltages and the 5 V supply voltage for the digital ICs. Just as in the interface test circuit in one of the previous sections, the 5 V supply is positioned against the -24 V supply in such a way that the -35 V guide voltage is at the same time the Vss for the digital ICs. To avoid confusion is will call the real 0 V Gnd, the ground for the digital ICs (= -35 V guide bias) Vss, and the positive supply voltage for the ICs (ca. -30 V with respect to Gnd) Vdd. 35 V is very near or even in excess of the maximum input voltage allowed for the 7805 regulator. That is why a 12 V zener diode is used to reduce the input voltage of the 7805 to a safer 24 V. A 24 V relay uses far less current than a comparable 5 V relay. That’s why I used a 24 V relay to switch the HT on and off. The relay I use has a DC resistance of 1K2. With a 470 ohm resistor in series the current consumption by the relay is 21 mA. Since the supply voltage for the digital stuff was derived from the negative phase of the 24 V, the relay is fed from the positive phase, resulting in a more symmetrical loading of the transformer. The simple circuit around T1 and T2 interfaces the relay to the processor. To provide a more stable guide voltage, the relay has its own rectifying circuit consisting of D4 and C4.

Figure 18. Total circuit diagram of the decatron clock. Clicking on the figure will open a window with a detailed circuit diagram in pdf format.

After the discussion of the CPU interface, the circuit diagram for the clock itself will hold no surprises (Fig. 18). Just as in most of my other clocks, the 50 Hz timebase signal is derived from the mains by means of two opto-couplers and a NAND Flip-Flop. An LDR is used to check if the light in the toilet is on or off. The LDR is connected to the PICs AD converter via one of the analog inputs, so that the light on and off threshold can be chosen under software control. A single push-button is used to set the time. The distribution of the guides and cathodes of the I/O ports at first sight seems rather random. Connecting them in this way however, resulted in the simplest wiring. Figure 19 finally shows the front- and back-side of the finished PCB excluding the high voltage supply, which will be added on a separate PCB.

Figure 19. Front- and backside of decatron clock PCB

Reading the datasheet, I found out that not every combination of digital I/Os and analog inputs is allowed for PORTA. When both analog and digital I/Os are used on PORTA, the analog I/Os have to be connected to the lowest pins of PORTA. At first I had the LDR analog input connected to PORTA,5 with the other pins as digital I/O. I changed it so that the LDR is now connected to PORTA,0. Table 1 summarizes the connections of the guides and the “0” cathodes to the ports of the 16F874. Table 2 gives the connections of the LDR, the push-button switch, the 50 Hz mains frequency input and the high-voltage on/off relay.

	guide 1	guide 2	cathode
sec*1	PORTD,0	PORTD,2	PORTD,1
sec*10	PORTC,4	PORTD,3	PORTC,3
min*1	PORTC,5	PORTC,2	PORTC,1
min*10	PORTE,0	PORTE,1	PORTA,4
hrs*1	PORTB,4	PORTA,3	PORTA,2
hrs*10	PORTA,1	PORTB,5	PORTA,5

LDR	PORTA,0	A-inp.	light = low
switch	PORTC,6	D-inp.	push = “0”
50 Hz	PORTC,0	D-inp.	–
relais	PORTE,2	D-outp.	HV on = “1”

Figure 20 depicts the complete clock. Remember that the clock will be hung against the wall with the front of the tubes facing down. A small mirror will allow for simultaneous viewing of both the front as well as the interior of the tubes. For the octal bases I used six old sockets from Pirani pressure gauges. The tubes are clamped in what is known as “tool clamps”. The small PCB to the left holds the high-voltage supply and the HV on/off relay. In Fig. 20 the clock is connected via the In System Programming (ISP) connector to Wouter Ooijens ISP programmer [12].

Figure 20. Assembled clock connected to my “In System Programming” board.

Software

It is the task of the software (which can be downloaded at the end of this page) to keep track of the time and to control and drive the decatron tubes. To make the clock more interesting and attractive to look at, the clock can make the glow in the tubes spin or rotate. The spin and rotation actions are combined in different display effects. During rotation the glow moves at a constant speed to the new resting position. During a spin the glow starts rotating at a very high speed and then “rolls out” until it stops at the next position. Both clockwise (forward) and counter clockwise (reverse) spinning and rotation is possible.

Figure 21. Flow-diagram of the main program (A) and the interrupt service routine which is called every 200 µs on a timer0 roll-over (B). The names refer to the subroutines called.

The main program (Fig. 21A) is very simple and performs the non-time critical tasks. The main program is interrupted by a roll-over (transition from FFH to 00H) from timer0. The clock input of timer0 is connected to the internal processor clock running at 20 MHz/4=5 MHz, with a cycle time of 1/(5 MHz)=200 ns, via the prescaler set at 4. A roll-over of timer0 thus occurs every 200ns*4*256=204.8 µs. The interrupt service routines (Fig. 21B) perform a number of time critical task such as the actual control of the tubes.

The main program first-of-all initializes the special function registers in routine “init_sfr”, followed by the program variables in “init_var”. Next the tubes are initialised or rather synchronised. Remember that when the tubes ignite, the position of the glow is unknown. By rotating or spinning the glow at least one full circle, the glow passes “cathode zero” which is detected by the program so that from then on the position of the glow is known. This procedure is discussed in detail below in the description of the spin and rotate routines. In the main program the synchronisation of the tubes is done in “init_tube”.

Next, the clock is set to the correct time in the routine “set_time”. For this the one push-button procedure is used that I use in all my clocks. When the push-button is pressed, the minutes start advancing at a pace of one per second. When the button is pressed again the minutes stop advancing and now the hours start advancing. Pressing the button again stops counting of the hours. When the button pressed for the fourth time the clock is switched on.

The main program now enters a loop that takes care of three tasks: switching of the tubes when it is dark (“ldr”,”onoff”), initiating the different display effects for every new minute (“do_displ”), and detection of the pressing of the push-button (“chk_key”). When the push-button is pressed the clock is reset.

In normal operation the seconds tubes count in the normal fashion from 0 to 59. When the seconds advance from 59 to 0, the S10, M1, M10, H1 and H10 tubes move to their new position while displaying a particular spin of rotation sequence. There are 10 different sequences. The routine “do_displ” initiates these display effects. The routine checks the auxiliary new seconds flag “newsec2”. When a new second is detected, the glow position of the seconds tubes is adjusted. However, on the transition to a new minute the routine initiates the display effects by setting up the appropriate control bytes for the spin/rotate routines of the interrupt service routines. A separate subroutine “show_1, show_2, …..” is used for each of the ten display effects. Which routine is called depends on the value of the lower nibble of variable “MIN”

Tube Saver

As mentioned before, the idea is to hang this clock on the wall of the toilet. This means that the tubes only need to be switched on when there is actually somebody using the toilet. Since people normally switch the light on when they use the toilet, a simple Light Dependent Resistor (LDR) can be used as sensor. As the name suggests, the resistance of an LDR depends on the light on the LDR. In the dark the resistance is usually several mega-ohms. When illuminated the resistance drops to a few-hundred ohms, depending on the light intensity.

The LDR in combination with Rxx form a voltage divider which is connected to PORTA,0. When it is light, the resistance of the LDR is low and the voltage on PORTA,0 low. When it is dark, the resistance of the LDR is high, and so is the voltage on PORTA,0. With Rxx = 47 kohm the threshold between light and dark was found to be somewhere around 1V. PORTA,0 is switched as analog input and internally connected to the AD converter of the PIC processor.

Figure 22. Distribution of the 10-bit AD result over the bytes ADRESH and ADRESL depending on the value of bit ADFM

The result of the AD conversion is a 10-bit value. To make things a little bit simpler I have used a trick to reduce the resolution of the AD converter to 8-bits. The ADFM bit is used to determine how the 10-bit result from the AD conversion is distributed over the two AD converter result bytes ADRESH and ADRESL (Fig. 22). When ADFM=0 the result is what is called “Left Justified” meaning that the eight most significant bits from the AD conversion are stored in ADRESH, while the remaining two least significant bits are stored in ADRESL. We now ignore the data in ADRESL, and only use the data in ADRESH.

When the input (PORTA,0) is 0V, ADRESH=00H, and when the input is 5V, ADRESH=FFH. So every count from ADRESH represents 5/256=0.01953125 V. To build in an hysteresis between light on and light off, all input voltages below (5/256)*16=0.312 V are defined as “light”, while input voltages higher than (5*255)*64=1.25 V are regarded as “dark”. This definition makes checking for these values very easy since values less than 16 satisfy (11110000.AND.ADRESH)=0, and values higher than 64 satisfy (11000000.AND.ADRESH)<>0.

For input voltages less than 0.312 V (light) the “LIGHT” flag is set. For input voltages higher than 1.25 V the “DARK” flag is set. For values between those two limits both flags are cleared. All these actions are performed by routine “ldr”.

Figure 23 Flow diagram of the routine “onoff” which switches the tubes on and off.

Knowing my youngest son Daan (7), I guessed that he would like to sit in the dark and see the tubes spin. Therefore a delay was built in between the LDR becoming dark and the tubes being switched off. The delay (in seconds) is the program constant “SECDRK” which is set at 180 (3 min). Also a delay between the LDR becoming illuminated and the tubes being switched on was included. This delay (“SECLHT”) is set at one second.

The on- and off-switching of the tubes is performed by routine “onoff”. Figure 23 depicts the flow diagram for routine “onoff”. It is pretty straight forward, and uses the additional new second flag “newsec3”. Counter “tubecnt” is used to count the seconds in the specified delays.

The Interrupt Service Routines

The beating heart of the program is the interrupt service routine “tmrint” (Fig.21B) which is called every 200 µs on an interrupt generated by a roll over from timer0 (20 MHz X-tal, prescaler=4). After saving the working registers W, STATUS and PCLATH in routine “push”, the interrupt service routine in turn calls another set of routines which perform tasks which are time critical: “filt50”, “subsec”, “inctime”, “key”, “s1_spin”, “s10_spin”, “m1_spin”, “m10_spin”, “h1_spin”, “h10_spin”.

The routine “filt50” polls the 50 Hz mains input frequency and filters out any transients and other irregularities. When a new and valid 50 Hz period is found the “NEW50” flag is set.

The routine “subsec” checks, and if required resets the “NEW50” flag, and counts down the 50 Hz periods into seconds. When a new second is found, the new second flags: “newsec1”, “newsec2” and “newsec3” are set. These flags are used by the time keeping routine “inctime”, the display controller “do_displ” and the tube saver “onoff” routines respectively.

The routine “inctime” checks the “newsec1” flag, and if set increments the time as kept in variables “HRS”, “MIN” and “SEC”. The routine only starts incrementing the time when the “GOT” (GO Time) flag is set. For the decimal incrementation of the time subroutine “incwdec” is used. This routine performs a decimal incrementation of W.

The subroutine “key” processes and debounces the push-button. When the push-button is pressed during “KEYREL” consecutive calls to “key”, the “NEWKEY” flag is set. The program “KEYREL” is set to 030H corresponding to a debounce time of 48*200 µs ca. 10 ms. The “NEWKEY” flag has to be reset by the main program.

Furthermore the spin routines which control the individual tubes are a part of the interrupt dervice routine. These routines will be described in the next section.

The Spin and Rotation Routines

The spinning and rotation of the glow in the tubes is controlled by a separate routine for each tube. The routines are: “s1_spin”, “s10_spin”, “m1_spin”, “m10_spin”, “h1_spin”, “h10_spin”. Each of these routines is called every 200 µs. The “do_displ” routine in the main program communicates with these routines via a set of variables. For each routine an identical set of six variables is used. As an example the variables for the routine that controls the seconds*10 tube (“s10_spin”) are: “S10_NOW”, “S10_NXT”, “S10_Z”, “S10_CNT”, “S10_REL”, “S10_BIT”. “S10_NOW” Holds the current cathode position of the glow. Before initialisation of the tubes, the current cathode position is unknown. The value of S10_NOW is then set to 0FFH by the initialisation routine. “S10_NXT” The next position (cathode) for the glow. When the “GOR” flag of the routine is set by the main program, the routine spins or rotates the glow to this position. “S10_Z” An internal counter used to keep track of the glow within one count. Z=0 glow on cathode, Z=1 glow on guide 1 and Z=2 glow on guide2. “S10_REL” Reload value for internal counter “S10_CNT”. The value in “S10_REL” determines the rotation speed and the start speed of spinning. A value of 010H corresponds to fast(est), and FFH corresponds to slowest rotation/spinning. After a spin operation the initial value of “S10_DEL” is lost since the routine increments its contents to 0FFH. “S10_CNT” Internal counter used to count down the number of calls to “s10_spin” (every 200µs) before the next glow movement is performed. “S10_CNT” has to be set to “S10_REL” during every new initialisation of the spin routine. “S10_BIT” Holds the flags which control the spin routine:

“GOR”, bit 0, when set starts the spin/rotation operation. The routine clears “GOR” upon completion of the operation.
“FR”, bit 1, 1=forward and 0=reverse spin/rotation.
“SR”, bit 2, 1=spin and 0=rotation.
“QUE”. bit 3, when set the routine automatically sets the GOR flag of the next digit (in this case m1_spin) after completion of the current operation. This allows for cascading of the spin/rotation events from right to left.

Figure 24. Flow diagram for the s10_spin routine. The other spin routines are identical

Figure 24 depicts the flow diagram for the spin routines. Upon call the routine first checks if something has to be done at all (“GOR”=1). If so, the routine decrements the delay counter “S10_CNT”. When the counter has reached zero, it is reloaded with “S10_REL” and the glow is advanced one position. This is done by routine “s10_adv”. The routine “s10_adv” advances the glow by one position (a cathode or guide) in forward or reverse direction depending in flag “FR”. Additionally the routine performes a number of other tasks which will be discussed later in the discussion of the “s10_adv” itself. Next the routine checks if this was a rotation or spinning operation. Lets assume for the moment is was a rotation. in that case the routine first checks if at this moment the glow is resting on a cathode. If so it checks if the current cathode position corresponds to “S10_NXT. If this is the case, the rotation operation is finished. This means that the “GOR” flag is cleared, and that if the “QUE” flag is set, the “GOR” flag of the next digit is set. If it was not a rotation but a spin operation, the routine checks if the delay reload value has already reached 0FFH (slowest speed). Only when it has reached 0FFH the routine checks if the glow is resting on the next cathode position. If the reload value has not reached 0FFH yet, “S10_REL” is incremented.

Figure 25. Flow diagram for the s10_adv routine. The other advance routines are identical

The flow diagram for the advance routines is given in Fig. 25. The seconds*10 digit is again used in the example. The main purpose of the routine is to advance, either in forward or in reverse direction, the glow one electrode. For a forward advance this means from cathode->guide1, guide1->guide2 and guide2->cathode. For a reverse “advance” this means from cathode->guide2, guide2->guide1 and guide1->cathode. The variable “S10_Z” keeps track of the location of the glow: “S10_Z”=0 cathode, “S10_Z”=1 guide1, “S10_Z”=2 guide2. The variable “S10_NOW” holds the actual number being displayed on the tube.

The routine is split into two almost identical parts: one for the forward motion, the other for the reverse motion. If we follow the left branch of the flow-diagram in Fig. 25, we see that the routine first checks if the glow is resting on a cathode. If the glow rests on a cathode, it next checks the “cathode input” to see if the glow is accidentally resting on cathode 0. IF so “S10_NOW” is initialized. In this way the routine performs its other important task, namely the synchronisation of the software position counter “S10_now” with the actual position of the glow. Remember that after switching on the tube, the glow will ignite on a random, unknown cathode. In this clock I did not implement a reset circuit. Instead, by rotating the clock at least one full circle during initialisation after a power-up of the tubes, “S10_NOW” is synchronised with the glow position. In fact every time the glow passes cathode 0 the counter is updated.

When the routine finds that the glow is resting on guide 2 (Z=2), the glow is moved to the next cathode by making guide 2 low again and “S10_NOW” is incremented, but only if the most significant bit of “S10_NOW” is zero. Why is that ? To explain this lets first recall the initialisation procedure: first the high voltage is connected to the decatrons. This will cause a random ignition of one of the cathodes. The software next performs a spin operation on all tubes. This will certainly initialise the counter “S10_NOW”. The problem with this procedure is that, as mentioned before, it sometimes may take some time for a tube to ignite, especially if it is rather dark and the tube is cold. So worst case the spin operation is over before the tube has ignited. In that case “S10_NOW” will not be properly initialised. A simple trick is used to solve this problem. “S10_NOW” is given the initial value of 0FFH. By only incrementing “S10_NOW” when the highest significant bit of “S10_NOW” is zero, or in other words when “S10_NOW” has been initialised because the glow has moved over cathode 0, the spin routine keeps on spinning until the glow has ignited and passed cathode 0 once! Simple but effective. Variable Z is cleared by first making it -1 (0FFH) and then incrementing it to zero. It is important to realize that cathode 0 not necessarily needs to correspond with the “zero position”. If for instance for practical reasons cathode six is facing upward, the reset value of “S10_NOW” can be selected such that the software thinks that cathode 6 represents zero.

The Case

With the circuit and the software finished, the remaining task is to finish the case. The most important element still missing is the mirror that will be placed at a 45 degree angle at the top (now the bottom) of the tubes to allow for reading of the digits when the clock is in vertical position. The problem with a normal mirror is that it gives a double reflection, one at the front side of the glass and one at the actual mirror layer. A better result is obtained with a mirror which only reflects on the frontside of the glass. Such a mirror can “easily” be made by evaporating a thin layer of aluminium on a clean plate of glass. A readout dial can then easily be made by etching the numbers in the thin aluminium layer. So this will be the job for the next weeks !

Software Download

The Software for the decatron clock as described on this page can be downloaded here.
The compressed file contains the following files: DECATRON.PIC The assembler input file of the final decatron clock software. The assembler file was assembled with the MPASM assembler running under DOS. DECATRON.HEX The assembled decatron clock software in INTEL-HEX format.

References and Web Links

[1] http://www.electricstuff.co.uk/dekatron.html
[2] http://www.tube-tester.com/sites/nixie/different/spinner-K8A/k8a.htm
[3] http://www.hanssummers.com/electronics/misc/dekatron/index.htm
[4] R.C. Bacon, J.R. Pollard, “The Decatron,” Electronic Engineering, May 1950, pp.173-177
[5] http://frank.pocnet.net/
[6] http://www.tube-tester.com/sites/nixie/datdekat/GC10B_ericsson/gc10b-eric.htm
[7] http://www.hanssummers.com/electronics/misc/dekatron/index.htm
[8] http://www.electricstuff.co.uk/dspin.html
[9] http://www.cathodecorner.com/
[10] http://www.radiotwenthe.nl./
[11] http://www.denhaag.com/
[12] http://www.voti.nl/wisp628/index.html

Mark Ayres version of the dacatron clock waiting for a suitable case

The decatron spinner as made by Dusan from Germany.

готовые наборы на радиолампах ( Да хоть на Алиэкспресс – не секрет что у нескольких Российских производственников есть фабрика – офис в Китае, и они представлены как продавцы на Али.)

Чем интересны такие комплекты для сборки – дешевле, не требуется пайка и настройка на заводе.

Изучение принципов работы – студентам, и получение навыков по сборке – настройке.

Красиво смотрятся, усилители индикаторы и не только имеют лучшие характеристики либо надежность, если выполнены с применением ламп. Совмещены и с более современной электроникой где это лучше сделать. * Усилитель, регулятор громкости с газоразрядными индикаторами, индукционная печь и нагреватель, сварочный аппарат с выбором режима, точечная сварка, радиолокатор, эхолот, сонар.

Related Images:

Download Attachments

1 DECATRON
программа для часов не декатронах - ассемблер
File size: 14 KB Downloads: 189
2 dekatron2
File size: 4 KB Downloads: 178

Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	Z302C		K8A	COU	10	1	BI	2	SP
	OG4 (ОГ4)	Russian made	K8A	COU	10	2	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled.
	OG9 (ОГ9)	Russian made	K8A	COU	10	2	BI	2	DP
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	GC10B	Trigon, Ericsson, Hivac, Hivac/Baird Atomic, Etelco, ETL/Baird Atomic, ETL (click on the manufacturer to view tube details!)	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled. Same as GC10B/S, CV2271, Z303C.
	Z303C	Adzam	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled. Same as GC10B, GC10B/S, CV2271.
	CV2271	Mullard	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled. Same as GC10B, GC10B/S, Z303C.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	Z303C CV2271	Mullard	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled. Same as GC10B, GC10B/S, CV2271, Z303C.
	GC10B/S CV2271	Mullard	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled. Suffix “/S” stands for “services”, a British military-spec designation. Same as GC10B/S, CV2271, Z303C.
	GC10B/S CV2271	ETL	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled. Suffix “/S” stands for “services”, a British military-spec designation. Same as GC10B/S, CV2271, Z303C.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	GC10B/L CV6044	Mullard	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse long life decimal counting tube. Suffix ‘L’ for long life. Neon filled. Same as CV6044, VX9194, GC10B/L.
	VX9194 CV6044	ETL	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse long life decimal counting tube. Neon filled. Same as GC10B/L, CV6044, VX9194.
	VX9194	ETL	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse long life decimal counting tube. Neon filled. Same as GC10B/L, CV6044.
	CV6044	ETL	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse long life decimal counting tube. Neon filled. Same as GC10B/L, VX9194.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	GC10/4B	Etelco	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube with intermediate outputs. Same as GC10B, except that four main cathodes are brought out to separate base pins
	GC10/4B	Hivac	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube with intermediate outputs. Same as GC10B, except that four main cathodes are brought out to separate base pins
	GC10/4B	Hivac/Baird Atomic	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube with intermediate outputs. Same as GC10B, except that four main cathodes are brought out to separate base pins
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	GC10/4B	ETL	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube with intermediate outputs. Same as GC10B, except that four main cathodes are brought out to separate base pins
	GC10/4B	ETL/Baird Atomic	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube with intermediate outputs. Same as GC10B, except that four main cathodes are brought out to separate base pins
	GC10/4B CV1739	ETL	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube with intermediate outputs. Same as GC10B, except that four main cathodes are brought out to separate base pins
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	M2465-401C		K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube with intermediate outputs. Same as GC10B, except that four main cathodes are brought out to separate base pins
	GC10/4B/L	ETL	K8A	COU	10	4	BI	2	DP	Suffix ‘L’ for long life. Bi-directional double-pulse decimal counting tube with intermediate outputs. Same as GC10B/L, except that four main cathodes are brought out to separate base pins
	CV6100	ETL	K8A	COU	10	4	BI	2	DP	Long life type. Bi-directional double-pulse decimal counting tube with intermediate outputs. Same as GC10B/L, except that four main cathodes are brought out to separate base pins
	VX9194/4B		K8A	COU	10	4	BI	2	DP	Long life type. Bi-directional double-pulse decimal counting tube with intermediate outputs. Same as GC10B/L, except that four main cathodes are brought out to separate base pins
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	6802	Philips	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled. Same as CK6802 but only 4kHz.
	6802	Sylvania	K8A	COU	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled. Same as CK6802 but only 4kHz.
	CK6802	Raytheon	K8A	COU	10	5	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled. Same as 6802 but 5kHz.
	6482	Sylvania	K8A	COU	10	4	BI	2	DP
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	GC10/A	Ericsson	K8A	COU	10	20?	BI	2	DP	Bi-directional single-pulse decimal counting tube. Same as the CVX2199. Argon filled.
	CVX2199	Ericsson	K8A	COU	10	20?	BI	2	DP	Bi-directional single-pulse decimal counting tube. Same as the GC10/A. Argon filled.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	GC10D	ETL	K8A	COU	10	20	BI	3	SP	Bi-directional single-pulse decimal counting tube. Same as the CV5143. Argon filled.
	GC10D	Hivac	K8A	COU	10	20	BI	3	SP	Bi-directional single-pulse decimal counting tube. Same as the CV5143. Argon filled.
	GC10D CV5143	ETL	K8A	COU	10	20	BI	3	SP	Bi-directional single-pulse decimal counting tube. Same as the GC10D and CV5143. Argon filled.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	OG3 (ОГ3) metal base	Анод, Дятьково, Россия (Anode, Dyatkovo, Russia)	K8A	COU	10	20	BI	3	SP	Bi-directional single-pulse decimal counting tube. Argon filled.
	OG-3 (ОГ-3) Bakelite base	Анод, Дятьково, Россия (Anode, Dyatkovo, Russia)	K8A	COU	10	20	BI	3	SP	Bi-directional single-pulse decimal counting tube. Argon filled.
	DK23 / 4CG-10A-2	Rodin	K8A	COU	10	20	BI	3	SP	Bi-directional single-pulse decimal counting tube. Neon filled.
	OG7 (ОГ7)	Russian made	K8A	COU	10	50		2		Argon fill
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	6909	Sylvania	K8A	COU	10	100	BI	2	DP	High speed bi-directional double-pulse decimal counting tube. Same as CK6909.
	CK6909	Raytheon	K8A	COU	10	100	BI	2	DP	High speed bi-directional double-pulse decimal counting tube. Same as 6909.
	CT783	Sylvania	K8A	COU	10	100?	BI	3	SP	Bi-directional single-pulse decimal counting tube. Neon filled.
	OG8 (ОГ8)	Russian made	K8A	COU	10	100		1		Argon fill
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	GC12/4B	ETL	K8A	COU	12	4	BI	2	DP	Bi-directional, double-pulse, 12-step counting tube with intermediate outputs.
	GC12/4B	ETL / Baird Atomic	K8A	COU	12	4	BI	2	DP	Bi-directional, double-pulse, 12-step counting tube with intermediate outputs.
	OG5 (ОГ5)	Russian made	K8A	COU	10	?
	OG5-II-60 (ОГ5-II-60)	Russian made	K8A	COU	10	?
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	6476	Sylvania	B12E	SEL	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled.
	6476A	Raytheon, Sylvania, Westinghouse	B12E	SEL	10	4(5)	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled.
	CK6476	Raytheon	B12E	SEL	10	4(5)	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	GS10C	ETL	B12E	SEL	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled.
	GS10C	Etelco	B12E	SEL	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled.
	GS10C	Hivac	B12E	SEL	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled.
	GS10C	Baird-Atomic/ Hivac	B12E	SEL	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	GS10C/S	Hytron	B12E	SEL	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled. Suffix “/S” stands for “services”, a British military-spec designation.
	GS10C/S	ETL/Baird Atomic	B12E	SEL	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled. Suffix “/S” stands for “services”, a British military-spec designation.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	GS10C/S CV2325	ETL	B12E	SEL	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled. Suffix “/S” stands for “services”, a British military-spec designation.
	GS10C/S CV2325	Mullard	B12E	SEL	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled. Suffix “/S” stands for “services”, a British military-spec designation.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	Z502S Mullard	Mullard	B12E	SEL	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled.
	Z502S Valvo	Valvo	B12E	SEL	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled.
	M2465-402		B12E	SEL	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	GC10C/S		B12E	SEL	10	4	BI			Suffix “/S” stands for “services”, a British military-spec designation.
	GS10D	ETL	B12E	SEL	10	10(rect) 20(sine)	BI	2	DP	Bi-directional double-pulse decimal counting tube. Argon filled.
	GS10E	ETL	B12E	SEL	10	10	BI	2	DP	Bi-directional double-pulse decimal counting tube. Argon filled.
	M2465-403C		B12E	SEL	10	10(rect) 20(sine)	BI	2	DP	Bi-directional double-pulse decimal counting tube. Argon filled.
	CK6910	Raytheon	B12E	SEL	10	100	BI	2	DP	Bi-directional double-pulse decimal counting tube. Argon filled.
	6910	Sylvania	B12E	SEL	10	100	BI	2	DP	Bi-directional double-pulse decimal counting tube. Argon filled.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	GS12D	ETL	B12E + 2 flying leads	SEL	12	4	BI	2	DP	Bi-directional double-pulse duodecal (12-position) counting/selector tube. Neon filled.
	GS12D	Hivac	B12E + 2 flying leads	SEL	12	4	BI	2	DP	Bi-directional double-pulse duodecal (12-position) counting/selector tube. Neon filled.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	GS12C	ETL	16 solder tags on phenolic tube	SEL	12	4	BI	2	DP	Bi-directional double-pulse duodecal (12-position) counting/selector tube. Rare maintenance type! Same as CV1740.
	CV1740	ETL	16 solder tags on phenolic tube	SEL	12	4	BI	2	DP	Bi-directional double-pulse duodecal (12-position) counting/selector tube. Rare maintenance type! Same as GS12C.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	6167	Western Electric	B13B plus 3 pins in center	SEL	10	1	UNI	1 split in 2	SP	Unidirectional glow-transfer decimal counting tube. Neon fill. Nipple front. Unusual spring shaped cathodes.
	8353	Sylvania	B13B	SEL	10	4	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled.
	Z562S	RFT	B13B	SEL	10	4-5	BI	2	DP	Bi-directional double-pulse decimal counting tube. Neon filled.
	7978	Raytheon	B13B	SEL	10	5	BI	2	DP
	CK7978	Raytheon	B13B	SEL	10	5	BI	2	DP
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	Z504S	Mullard	B13B	SEL	10	5	BI	2	DP	Bi-directional double-pulse decimal counting/selector tube. Neon fill. Same as ZM1070 and 8433
	Z504S	Philips	B13B	SEL	10	5	BI	2	DP	Bi-directional double-pulse decimal counting/selector tube. Neon fill. Same as ZM1070 and 8433
	ZM1070	Valvo	B13B	SEL	10	5	BI	2	DP	Bi-directional double-pulse decimal counting/selector tube. Neon fill. Same as Z504S and 8433
	8433	Valvo	B13B	SEL	10	5	BI	2	DP	Bi-directional double-pulse decimal counting/selector tube. Neon fill. Same as Z504S and ZM1070
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	Z564S	RFT	B13B	SEL	10	25	BI	2	DP	Bi-directional double-pulse decimal counting/selector tube. Neon filled.
	Z505S	Amperex	B13B	SEL	10	50	BI	2	DP	Bi-directional double-pulse decimal counting/selector tube. Argon fill. Same as ZM1060
	ZM1060	Philips, Valvo	B13B	SEL	10	50	BI	2	DP	Bi-directional double-pulse decimal counting/selector tube. Argon fill. Same as Z505S
	8035	Sylvania	B13B	SEL	10	100	BI	2	DP	High speed decade counting tube. Same as the CT-4251.
	CT-4251	Sylvania	B13B	SEL	10	100	BI	2	DP	High speed decade counting tube. Same as the 8035.
	8353	Sylvania	B13B	SEL	10		BI	2	DP
	8262		B13B	SEL	10	100				Argon fill
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	A101	Анод, Дятьково, Россия (Anode, Dyatkovo, Russia)	RSH30 (РШ30)	SEL	10	1	BI	2	DP	Bi-directional double-pulse decimal counting/selector tube. Neon fill.
	A102	Анод, Дятьково, Россия (Anode, Dyatkovo, Russia)	RSH30 (РШ30)	SEL	10	20	BI	2	DP	Bi-directional double-pulse decimal counting/selector tube. Neon fill.
	A103	Анод, Дятьково, Россия (Anode, Dyatkovo, Russia)	RSH30 (РШ30)	SEL	10	50	BI	2	DP	Bi-directional double-pulse decimal counting/selector tube. Argon fill.
	A106	Анод, Дятьково, Россия (Anode, Dyatkovo, Russia)	RSH30 (РШ30)	SEL	10	100	BI	2	DP	Bi-directional double-pulse decimal counting/selector tube. Argon fill.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	6846	Sylvania	“Miniature Button 7-Pin” = B7G ?	FLIP						Neon binary glow-transfer counting tube, function similar to “T” flip-flop.
	10SG1B2		?	COU						Russia/U.S.S.R. ‘Dekatron’ style glow-transfer decimal counting tube. One known example with 1957 date code marked “PROBATIONARY” (“ОПЫТНАЯ”).
	CH-1047	Chatham	?	COU	10					‘Dekatron’ style glow-transfer decimal counting tube.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	N3	Pressler	8p Stahl- röhren- sockel	COU			UNI
	N4	Pressler	8p Stahl- röhren- sockel	COU			UNI			‘Dekatron’ style glow-transfer decimal counting tube.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	GS10J	ETL	B17A	SEL	10	1	BI	2	DP	Bi-directional double-pulse decimal counting/selector tube with 2 routing guides.
	GS10H	ETL	B17A	SEL	10	5	BI	2	DP	Bi-directional double-pulse decimal counting/selector tube with 2 routing guides. Equivalent to CV8630.
	CV8630	ETL	B17A	SEL	10	5	BI	2	DP	Bi-directional double-pulse decimal counting/selector tube with 2 routing guides. Equivalent to GS10H.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	Z572S	RFT	B27A	SEL	10	5	BI	2	DP	Bi-directional double-pulse decimal counting tube and Nixie tube driver. Neon filled. This very special tube can drive a Nixie tube directly with its 10 switching anodes.
	GS10G		B26A or B27A	SEL	10
	GS10K		B27A	SEL	10
	GCA10G	ETL	B27A	COU	10					‘Dekatron’ style glow-transfer counting tube. Has auxiliary anodes which directly drive a Digitron (nixie-style) tube.
	GSA10G	ETL	B27A	SEL	10					‘Dekatron’ style glow-transfer counting/selector tube. Has auxiliary anodes which directly drive a Digitron (nixie-style) tube.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	6879	Raytheon, Sylvania	B7G	COU	10					‘Dekatron’ style glow-transfer decimal counting tube. Nipple front.
	7155	Raytheon, Sylvania	B7G	COU	10					‘Dekatron’ style glow-transfer decimal counting tube. Nipple front.
	GC10/2P	ETL	B7G	COU	10					‘Dekatron’ glow-transfer decimal counting tube. Nipple front.
	GP10/2P		B7G	COU	10
	Z565C		B9D	COU	10					‘Dekatron’ style glow-transfer decimal counting tube.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	EZ10A	Elesta	ZB13	SEL	10	300	UNI	1 split in 2	SP	‘Dekatron’ style unidirectional glow-transfer decimal counting tube. Argon fill. Nipple front.
	EZ10B	Elesta	ZB13	SEL	10	1000	UNI	1 split in 2	SP	‘Dekatron’ style unidirectional glow-transfer decimal counting tube. Hydrogen fill. Nipple front.
	ECT100	Elesta	ZB13	SEL	10	1000	UNI	1	SP	‘Dekatron’ style unidirectional glow-transfer decimal counting tube. Argon fill. Nipple front.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	G10/240E	STC	B12E	NOM	10	20	UNI	1	SP	The Nomotron-Queen! Very, very early ‘Nomotron’ style unidirectional, single pulse, decimal glow-transfer counting tube, with metal shell covering cathode pins inside tube. Shell has holes for viewing glow position and an additional transparent mica cover with hand drawn lines and one arrow, indicating the position. Neon filled.
	G10/241E CV2223	STC	B12E	NOM	10	20	UNI	1	SP	Very early ‘Nomotron’ style unidirectional, single pulse, decimal glow-transfer counting tube, with metal shell covering cathode pins inside tube. Shell has holes for viewing glow position, projected on a transparent cover with hand written numbers 0-9. Neon filled. Rare!
	CVX2223	STC	B12E	NOM	10	20	UNI	1	SP	Same as G10/241E and CV2223
	11TU7	Tesla	B12E	NOM	10	25	UNI	1	SP	Same as G10/241E CV2223 but 25 kHz counting speed.
	DZ10	Cerberus	B14A	NOM	10	3	UNI	1	SP	‘Nomotron’ style unidirectional, single pulse, decimal glow-transfer counting tube, with metal shell covering cathode pins inside tube. Shell has holes for viewing glow position, stamped with numbers 0-9.
	EZ10	Elesta	ZB13	NOM	10		UNI	1	SP	‘Nomotron’ style unidirectional, single pulse, decimal glow-transfer counting tube, with metal shell covering cathode pins inside tube. Shell has holes for viewing glow position.
Bilder/ Pictures	Röhrentyp/ Tube Type	Hersteller/ Brand	Sockel/ Base	Art/ Sort	Count. steps	Speed [kHz]	Richtg./ Direct.	No. of Guides	Oper- ation	Hinweise/ Remarks
	GR10A	ETL	B12A	REG	–	–	–	0	–	‘Register Tube’ similar in appearance to a ‘Dekatron’ style glow-transfer counter, but without guide electrodes. Does not count, but provides display closely resembling appearance of Dekatrons. Same as CV5291 and Z503M.
	CV5291	ETL	B12A	REG	–	–	–	0	–	‘Register Tube’ similar in appearance to a ‘Dekatron’ style glow-transfer counter, but without guide electrodes. Does not count, but provides display closely resembling appearance of Dekatrons. Same as GR10A and Z503M.
	Z503M	Mullard	B12A	REG	–	–	–	0	–	‘Register Tube’ similar in appearance to a ‘Dekatron’ style glow-transfer counter, but without guide electrodes. Does not count, but provides display closely resembling appearance of Dekatrons. Same as GR10A and CV5291.
	Z565M	RFT	B13B	REG	–	–	–	0	–	‘Register Tube’ similar in appearance to a ‘Dekatron’ style glow-transfer counter, but without guide electrodes. Does not count, but provides display closely resembling appearance of Dekatrons.

Beispiel-Bilder/ Picture Examples	German	English
	Nixie-Röhren seitliche Anzeige mit Ziffern oder alphanumerischer Anzeige	Nixie tubes side viewed with numerals or alphanumeric display
	Nixie-Röhren seitliche Anzeige mit Symbolen und Buchstaben	Nixie-Tubes side viewed with symbols and letters
	Nixie-Röhren frontale Anzeige mit Ziffern	Nixie-Tubes end viewed with numerals
	Nixie-Röhren frontale Anzeige mit Symbolen und Buchstaben	Nixie-Tubes end viewed with symbols and letters
	Dieter’s Nixie-Uhren	Dieter’s Nixie-Clocks
	Temp’n’Glow das IN-16 Nixie Thermometer	Temp’n’Glow the IN-16 Nixie Thermometer
	Dieter’s Nixie Videos	Dieter’s Nixie Videos
	Nocrotec FIP-Clock	Nocrotec FIP-Clock
	Dieter’s Nixie Röhren Daten Archiv	Dieter’s Nixie Tube Data Archive
	Numitrons, Minitrons, Glühfaden-Röhren	Numitrons, Minitrons, Incandescent filament readout tubes
	VFD-Röhren (Vacuum Fluorescent Display Röhren)	VFD Tubes (Vacuum Fluorescent Display Tubes)
	Dieter’s VFD-Uhren	Dieter’s VFD-Clocks
	Sperry, Babcock, Beckman. Flache Neon-Gas ‘Panaplex’-Displays	Sperry, Babcock, Beckman. Planar neon gas discharge ‘Panaplex’ readout displays
	Dekatrons, Register-Röhren, Selectrons, Zählröhren, Nomotrons	Dekatrons, Register tubes, Selector tubes, Glow-transfer counting tubes, Nomotrons
	Dekadische Zählröhren aus der E1T-Kategorie (Beam Deflection Decade Counter Tubes)	Beam Deflection Decade Counter Tubes from the E1T-Categorie
	Nocrotec E1T Clock	Nocrotec E1T Clock
	Gemischtes: Pixie-Röhren, Doppelpunkt-Röhren, und sonstige Nixie-Röhren Teile (74141 Treiber-IC’s usw.)	Misc. Stuff: Pixie-Tubes, Colon-Tubes, and other Nixie-Tube-stuff (74141 driver-IC’s aso.)
	Sockel und Fassungen	Sockets and bases
	Nocrotec.com – meine kommerzielle Nixie-Homepage	Nocrotec.com – my commercial Nixie-webpage!
	Nixie Mammoth – das ultimative Nixie Clock Netzteil	Nixie Mammoth – The Ultimate Nixie Clock Power Supply
	Dieter’s Nixie Tube Tester Projekt	Dieter’s Nixie Tube Tester Project
	Dieter’s Nixie Tube Pin Straightener	Dieter’s Nixie Tube Pin Straightener
	Dieter’s B9A Single Nixie Tube Tester & Regenerator	Dieter’s B9A Single Nixie Tube Tester & Regenerator
	Dieter’s B9A Nixie Tube Tester & Regenerator	Dieter’s B9A Nixie Tube Tester & Regenerator
	Dieter’s 74141 & 7441 Tester Projekt	Dieter’s 74141 & 7441 Tester Project
	Dieter’s B13B Dekatron Spinner Projekt	Dieter’s B13B Dekatron Spinner Project
	Dieter’s K8A Dekatron Spinner Projekt	Dieter’s K8A Dekatron Spinner Project
	Dieter’s B12E Dekatron Spinner Projekt	Dieter’s B12E Dekatron Spinner Project
	Dieter’s EZ10B Dekatron Spinner Projekt	Dieter’s EZ10B Dekatron Spinner Project
	Wissenswertes: Cathode Poisoning-Hilfe-Seite	Knowledge: Cathode Poisoning-Help-Page
	Dieter’s Nixie-Tube Link-Page	Dieter’s Nixie-Tube Link-Page
	Alle Updates auf meinen Seiten!	All updates on my pages!

(В наличии с пластмассовым цоколем)	ОГ-4 – декатрон двухимпульсный счётный для работы в счётных и счётно-решающих устройствах. Цвет свечения оранжево-красный. Индикация осуществляется через купол баллона. Выводы декатрона ОГ-4: 1 Катод нулевой 3 Анод 4 Подкатод первый 6 Подкатод второй 8 Катод 2,5,7 не подключены Лампа ОГ-4 выполнена в стеклянном корпусе с 8-ю выводами. Цоколь октальный металлический РШ5-1. Катоды холодные. Лампа работает в любом положении.
Цоколёвка декатрона ОГ-4:	Чертёж декатрона ОГ-4:
Параметры декатрона ОГ-4: Коэффициент пересчета 10 Напряжение источника питания 425-475В Напряжение возникновения разряда <375В Напряжение поддержания разряда 130-170В Напряжение смещения на подкатодах 30-40В Амплитуда импульса напряжения сброса >150В Рабочий ток 0,3-0,45мА Скорость счёта 0,01-2000Гц Амплитуда управляющего импульса 135-200В Длительность импульса 160-500мкс Длительность фронта 10-30мкс Масса, не более 70г Наработка 2000ч

	А-101– декатрон двухимпульсный коммутаторный для работы в счётных и счётно-решающих устройствах. Цвет свечения оранжево-красный. Индикация осуществляется через купол баллона. Выводы декатрона А-101: 1 К0 0-й катод 2 К9 9-й катод 3 К8 8-й катод 4 К7 7-й катод 5 1ПК Первый подкатод 6 К6 6-й катод 7 К5 5-й катод 8 А Анод 9 К4 4-й катод 10 2ПК Второй подкатод 11 К3 3-й катод 12 К2 2-й катод 13 К1 1-й катод Лампа А-101 выполнена в стеклянном корпусе с 13-ю выводами. Цоколь типа РШ30. Катоды холодные неактивированные. Лампа работает в любом положении.
Цоколёвка декатрона А-101:	Чертёж декатрона А-101: Панель ПЛ30-2П для А-101
Параметры декатрона А-101: Цвет свечения оранжево-красный Коэффициент пересчёта 10 Напряжение источника питания >450В Напряжение возникновения разряда: в темноте на свету <420В <375В Напряжение поддержания разряда между анодом и нулевым катодом 130-170В Напряжение смещения на подкатодах 36-44В Амплитуда импульса сброса 135-165В Амплитуда управляющего импульса 135-200В Амплитуда выходного импульса <20В Рабочий ток анода 0,3-0,45мА Ток перегрузки <1,5мА Время перегрузки 5мин(макс.) Скорость счёта 0,01-1000Гц Длительность импульса 200-1000мкс Длительность фронта 10-30мкс Время готовности <5c Масса, не более 50г Размеры, не более Ø31х76мм Наработка 2000ч	Ссылки: Декатроны и их применение (В.М.Липкин) Декатрон (Википедия) A-101 Dekatron Spinner (tube-tester.com) Указания по эксплуатации А-101: Напряжение смещения на управляющих электродах (подкатодах) устанавливать в отсутствие импульсов управляющего напряжения. Величину рабочего тока контролировать в цепи катода. После работы при скорости счёта <1Гц декатроны полезно тренировать при скорости счёта >100Гц; время тренировки не менее 20мин. При включении анодного напряжения в декатроне может наблюдаться одновременное горение двух индикаторных катодов, которое должно исчезать после подачи импульса напряжения сброса.
Декатронный вращатель (Dekatron Spinner) на А-101: ( www.tube-tester.com)

counting tube декатроны