|
Vintage Computers Any vintage computer systems, calculators, video games etc., but with an emphasis on 1980s and earlier equipment. |
|
Thread Tools |
14th Oct 2004, 3:40 pm | #1 |
Hexode
Join Date: Apr 2003
Location: Higham-on-the-Hill, Leicestershire, UK.
Posts: 338
|
Thermionic Flip-Flops
An odd subject for those who don't know a bit about digital electronics I admit!
Does anyone know where I can find online, or in books, circuits for a basic Flip-Flop using thyratrons? I know these can be set to conduct then have to be reset, giving a low/high output like a flip-flop; and have been used to form decade counters, as in Colossus. I am struggling to find books in the University Library, as understandably all the old texts have been put into the stores, and I don't really know what I am looking for! I have found a little information on the thyratron rings on the various Colossus rebuild sites, but I will have a sit down at the weekend I think with a pen, paper, valve data sheets, and my electronics notes to work some stuff out! I have found a triode-based (and a variation on this for pentodes) flip-flop in a book, the Eccles-Jordan circuit. But this is conventional valve-based, and I have a load of thyratrons from Paul Sherwin I would like to use! I just hope I can get a few pointers. Once I have these, I reckon I can come up with a load of counters to produce an odd-looking clock! Sam *added* If it helps, Wynn-Williams designed the thyratron-based counter rings.
__________________
Can he lead a Normal life, Doctor? No, he will be ... an Engineer. |
14th Oct 2004, 8:31 pm | #2 |
Retired Dormant Member
Join Date: Mar 2003
Location: Bishop's Waltham, Hants, UK.
Posts: 939
|
Re: Thermionic Flip-Flops
Sam,
the thyratron counters I've seen have been of the ring type, and ISTR that it involves capacitive coupling between stages (cathode to Grid?) so that the one stage prepares the next for a pulse. The Eccles - Jordan circuit is the classic circuit that most of us are more familiar with in its transistor incarnation, using collector cross coupled to base (anode to grid). Another form of frequency divider, much beloved of Tektronix, in their time mark generators involves the use of a non-retriggerable monostable, where the input frequency triggers the monostable, and the monostables period is set to an integer division of the input. The great thing about these is that you can divide by any integer you like, using a single double triode, with maybe an added cathode follower (Tek typicallly use it in divide by 5 and divide by10). The down side is that the division rate can change as the valve ages - if I put a new valve in my time mark generator, I have to keep re-adjusting the stage until the valve " burns in " . Somewhere (I suspect in storage ), I have an old MOD training book that gives details of early valve logic circuits - I will attempt to find it. You can get the details of the Tektronix 181 time mark generator, which describes the divider circuits from: http://bama.sbc.edu/ and go to the Tektronix page, then look for 181. The 180A uses the same technology, but as the instrument has a lot more outputs, the manual is considerably thicker.... (hint - you may find it better to use the mirror site - there is a link on the BAMA homepage). Jim. |
21st Oct 2004, 1:26 pm | #3 |
Hexode
Join Date: Apr 2003
Location: Higham-on-the-Hill, Leicestershire, UK.
Posts: 338
|
Re: Thermionic Flip-Flops
Thanks for the hints Jim. I keep forgetting to reply!
If you could find that book, it would be much appreciated. Don't go to to much trouble though! So far, I am trawling the University Library, and the old Periodicals in there! (Nature, the Proceedings of the Royal Society, etc). Reading about the old computing circuits, and the computers themselves. Descriptions, not circuits though, unfortunately! I will try and see if Wynn-Williams put his counter into there. I found his paper about what he was counting though! evingar - I have a NIXIE display clock I built using the boards and description from Mike's Electric Stuff page. When I was doing it, I thought it was a shame there were so many ICs and transistors in there, and so little glass! Sam
__________________
Can he lead a Normal life, Doctor? No, he will be ... an Engineer. |
21st Oct 2004, 3:26 pm | #4 |
Rest in Peace
Join Date: Feb 2004
Location: Hexham, Northumberland, UK.
Posts: 2,234
|
Re: Thermionic Flip-Flops
At last! someone else who's built a nixie clock. I sent off for the PCB, and spent hours soldering in all those transistors. I was very impressed by the quality of the PCB, and luckily had most of the components (including the CMOS int***ated cir*uit*), but I must agree that it does seem a bit odd mixing nixies and CMOS. The clock was even ok'd by my wife, and it's allowed in the house. I did consider building one to my own design using the 74141s I have lying around, but the power consumption is a bit higher, meaning using a transformer instead of the resistive dropper method. Maybe I could build a clock-radio using nixies and valves! How sad is that?
Biggles. |
21st Oct 2004, 3:38 pm | #5 |
Retired Dormant Member
Join Date: Dec 2003
Location: North London, UK.
Posts: 6,168
|
Re: Thermionic Flip-Flops
Does seem a little odd mixing chips with vacuum. Difficult to believe now that the 74141 was specially designed to drive Nixie tubes. Presumably for digital clocks. I don't think 7 segment LED devices existed at the time and were very expensive when they were first introduced.
|
21st Oct 2004, 5:05 pm | #6 |
Dekatron
Join Date: Jan 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 8,194
|
Re: Thermionic Flip-Flops
Hi Gents, I have some circuits of a counter/timer that was developed at Reyrolle and used transistor ring counters (BC017 equiv) with a BSX21 as HV nixie driver in pre - IC days.
Ed |
21st Oct 2004, 6:24 pm | #7 |
Retired Dormant Member
Join Date: Mar 2003
Location: Bishop's Waltham, Hants, UK.
Posts: 939
|
Re: Thermionic Flip-Flops
Jeffrey (and Sam),
the 74141 fulfilled a huge need in the late 60's and early 70's. I think the original need was for display driveres in the early desktop calculators, but I've found them in allsorts of test equipment (Advance used a lot in their counter timers). I think the clock was a later branch of technology, when the prices started to drop. I also have some 7 segment displays which use incandescant segments in a glass envelope, looking a little like a nixie. There are also a couple of other display technologies that were tried around this time, the Digivisor, which is an analogue device - it uses a moving coil mounted mirror to direct a beam of light through a mask with the numbers on it . The driving circuit generates a staircase waveform to count (used by Venner in some of their counters). I have (and we still use at work) a few displays which have an individual bulb for each number, shining through a mask onto a ground flass front screen. This requires a decimal decoder (some of the units at work actually show the digits between one and twelve, used in remote switching applications). I have some large " digital " clocks that came from an old fighter control centre (don't ask........), in which each numeral is made by drilling holes in a thin sheet of perspex, and the sheets are stacked to form a single digit of the display. To display a numeral, the perspex sheet is edge lit, and the ligth is then thrown forward out of the sheet where the refractive index changes at the drilled holes. These require a decimal drive, in this case it was provided by a number of 20 way 60 position ledex switches (think mechanical washing machine programmer) and a few relays. The whole lot was driven by the one second output of a standard Post Office Impulse clock. I have a number of displays, and a few of the Ledex switches, and I've just about worked out the relay arrangement (not as simple as you think, as all the digits have to change together - you have to have relays to " prepare " the next stage of the counting, and others to change the display). The final popular display for early electronic clocks is the analogue impulse dial (an ordinary clock face, but driven by a solenoid - the common ones have either a one minute step, or a 30 second step) as used by Synchronome and the like. The accurate ones used a one second drive. Some clocks used a crystal referenced synchronous motor, and this is basically what you find in a current analogue quartz clock (you may find that a synchronous motor is sometimes referred to as a phonic motor in older texts). That should give you a little more to think about Jim. |
21st Oct 2004, 6:39 pm | #8 |
Retired Dormant Member
Join Date: Dec 2003
Location: North London, UK.
Posts: 6,168
|
Re: Thermionic Flip-Flops
Jim
Thanks for the excellent summary of the technology. ISTR a lot of early digital clocks used a mechanism that looked a bit like a Rolodex. Each " card " had a number and the deck was flipped every minute. Not really electronic, more electromechanical. I also remember an early digital voltmeter that was sold at Harpenden many years ago. This used uniselectors, one per decade, to operate a Kelvin-Varley voltage divider. This took a fraction of the unknown voltage and compared it with a reference. It counted up from zero and when the voltages balanced it stopped. The uniselctor positions were indicated, I think, on nixies. Ingenious but rather noisy! Jeffrey |
21st Oct 2004, 6:49 pm | #9 | |
Octode
Join Date: Oct 2003
Location: Newbury, Berkshire, UK.
Posts: 1,770
|
Re: Thermionic Flip-Flops
Quote:
If anyone wants the circuit, I can e-mail it. I may even have the Altera .POF file lurking somewhere still. You don't need anything other than an easy to build pod (containing a 74HC244 and a few resistors) and some free software from the Altera site to programme these wonderful little devices from your PC parallel port. I am in no doubt Jeffrey will tell everyone how much more wonderful Xilinx are though. How are the Xilinx shares going Jeffrey? My Altera ones are doing very nicely. Chris
__________________
Chris Last edited by Station X; 24th Jan 2005 at 2:20 pm. Reason: Import |
|
21st Oct 2004, 7:08 pm | #10 |
Retired Dormant Member
Join Date: Mar 2003
Location: Bishop's Waltham, Hants, UK.
Posts: 939
|
Re: Thermionic Flip-Flops
Jeffrey,
most of the " flip-over " digital displays were driven by synchronous motors, but I have seen one or two impulse models (don't have an impulse one in the collection though ). I would have liked to have seen the voltmeter - these things are fascinating to watch - a successive approximation technique. One of my clocks uses uniselectors to make a 7 day time switch, but it gets a little anti-social at midnight on Sunday. The uniselector is a very versatile device, and has been used for all sorts of storage and counting applications, including pulse generators (S and Z pulses for clear down in telephone exchanges), and multi digit serial storage (telephone pulse regenerators). In fact, you could build a clock with them fairly easily. Jim. |
22nd Oct 2004, 5:05 pm | #11 |
Hexode
Join Date: Apr 2003
Location: Higham-on-the-Hill, Leicestershire, UK.
Posts: 338
|
Re: Thermionic Flip-Flops
Well, I finally found Wynn-William's paper on his first counters. In the Proceedings of the Royal Society of London, Series A, Volume 132 (1931), if anyone is interested!
Looks like I can create a conter ring reasonably simply. Just need to get the components and valve bases! I may need 45 thyratrons to make a clock though! And a 1Hz pulse generating circuit! Think I will play a bit first before I try the clock! Sam
__________________
Can he lead a Normal life, Doctor? No, he will be ... an Engineer. Last edited by Station X; 24th Jan 2005 at 2:23 pm. Reason: Import |
22nd Oct 2004, 5:36 pm | #12 | |
Octode
Join Date: Oct 2003
Location: Newbury, Berkshire, UK.
Posts: 1,770
|
Re: Thermionic Flip-Flops
Quote:
__________________
Chris |
|
22nd Oct 2004, 5:57 pm | #13 |
Retired Dormant Member
Join Date: Mar 2003
Location: Bishop's Waltham, Hants, UK.
Posts: 939
|
Re: Thermionic Flip-Flops
Sam,
you can easily build a pulse generator with another thyratron - you connect a resistor from anode to HT+, and a capacitor from anode to ground. Cathode is grounded. this will generate a ramp, reseting as the capacitor voltage reaches the thyratron Ionising voltage. Some degree of frequency control can be exercised by varying the grid potential (effectivly altering the ionising voltage of the thyratron). This circuit will generate a ramp waveform (it was often used as a timebase in early TV sets and scopes). If you differentiate the output, you will get a sharp pulse when the thyratron triggers. The time constant of the circuit is set by the resistor and capacitor combination (t=CR), but you will need to use the more complex formula for the voltage across a capacitor after time, as you need to work out the time to reach the ionising voltage, which may not be the same (t=CR gives the time to 2/3 of the applied voltage). Also be aware that the capacitor may not discharge to 0V when the thyratron triggers, so you will have to calculate the time for the change between two voltages (you may need you first year notes at this point ). Finally, beware that the discharge current of the capacitor may be to much for the thyratron to handle. In this case, use two resistors between HT and anode, with the capacitor connected to the junction of the resistors. This will generate a sawtooth wave. The charge time is governed by the resistor connected to HT, and the discharge by the resistor connected to the anode. The capacitor can now only discharge to a voltage calculated from the drop across the anode load and the valve. At this point calculation of the period of the wave gets complicated (I can reccommend Excel for this kind of thing). Regards Jim. |
23rd Oct 2004, 5:37 pm | #14 |
Retired Dormant Member
Join Date: Mar 2003
Location: Bishop's Waltham, Hants, UK.
Posts: 939
|
Re: Thermionic Flip-Flops
|
10th Nov 2004, 8:34 am | #15 | |
Dekatron
Join Date: Nov 2003
Location: Near Leeds, West Yorkshire, UK
Posts: 4,609
|
Re: Thermionic Flip-Flops
Quote:
Frequency dividers. The Bird and other electronic organs used neon tubes as relaxation oscillators, triggered by the pulse from the previous one - 5 components per divider! Played up when the neon tubes aged, though.
__________________
Mike. Last edited by Paul Stenning; 29th Dec 2004 at 11:31 am. |
|
10th Nov 2004, 8:48 pm | #16 |
Retired Dormant Member
Join Date: Mar 2003
Location: Bishop's Waltham, Hants, UK.
Posts: 939
|
Re: Thermionic Flip-Flops
Mike,
do you have a circuit for the dividers? I would be interested in seeing that. Jim. |
11th Nov 2004, 8:29 am | #17 |
Dekatron
Join Date: Nov 2003
Location: Near Leeds, West Yorkshire, UK
Posts: 4,609
|
Re: Thermionic Flip-Flops
Jim
You have a PM
__________________
Mike. |
27th Mar 2005, 9:52 pm | #18 |
Retired Dormant Member
Join Date: Mar 2003
Location: Bishop's Waltham, Hants, UK.
Posts: 939
|
Re: Thermionic Flip-Flops
Sam,
I just found this info about neon lamps in logic circuits, including ring counters: http://computer-refuge.org/classiccmp/neon_lamp_logic/ Jim. |
28th Mar 2005, 1:26 am | #19 |
Triode
Join Date: Jan 2005
Location: Magherafelt, County Londonderry, UK.
Posts: 28
|
Re: Thermionic Flip-Flops
Although digital electronics isnt my bag, I had wondered while reading this thread some time ago if neons could be used in this way.
The idea came to mind when I recalled repairing a pair of scottish terriers! Yes, you read it correctly, A PAIR OF DOGS. The more senior members will remember this advertising gimmick for 'Black & White' scotch whiskey. The delph dogs had neon lamps in their eyes which flashed in a random manner. This was done by a sort of multivibrator circuit using the variation in the tolerance of the caps and firing voltages of the neons to make it appear random, the circuit was powered by a 90volt battery. michael |
28th Mar 2005, 10:31 am | #20 |
Hexode
Join Date: Apr 2003
Location: Higham-on-the-Hill, Leicestershire, UK.
Posts: 338
|
Re: Thermionic Flip-Flops
Thanks for that link Jim. I had heard mantion of neons used as switching elements in counter/logic circuits. I got a book on cold-cathode devices out of the Uni library for a little light reading over easter!
If I can get a few neons I may try experimenting with them - will be a bit easier than the Thyratrons I should think! Sam
__________________
Can he lead a Normal life, Doctor? No, he will be ... an Engineer. |