Does anyone know where I can find out how to design a calculator with vacuum tubes, or where I can find an already-made schematic? It's for a science fair.
- posted
15 years ago
Does anyone know where I can find out how to design a calculator with vacuum tubes, or where I can find an already-made schematic? It's for a science fair.
That might be too challenging for a vanity fair (sorry, science fair). You would need around 5000 to 7000 tubes and 200 to 300 Amps to run it. It might take up the entire fair ground as well. How much money do you have?
For a simple calculator, less than 200 very small tubes, which fits in a small portable case, are sufficient:
-- Frank Buss, fb@frank-buss.de http://www.frank-buss.de, http://www.it4-systems.de
uch
tems.de
That doesn't seem to have any schematics... and linnix (like the name, use linux all the time): I really don't think so. To clarify: I have built a "calculator" from CMOS SSI. It is a 6-bit adder. I am building an 8-bit "calculator" from TTL logic, also SSI (with 2 74181's). It's got 32 functions, of which about 4 are real and essential math. The others are mostly logic. I also want to do it only in binary, as my other calculators have been.
Any more help would be appreciated. Does anyone know where I can find a book on digital logic using vacuum tubes, or possibly general vacuum tube theory? It needs to have schematics.
They are just simple adding machines.
Multiplication and Division need hundreds to thousands transistors/ tubes.
Eh? Eh? All required logic is the one bit adder with carry, plus a pair of shift registers. With the associated control and I/O, it takes 200...300 valves.
Vladimir Vassilevsky DSP and Mixed Signal Design Consultant
| > > linnix wrote: | > > > That might be too challenging for a vanity fair (sorry, science | > > > fair). You would need around 5000 to 7000 tubes and 200 to 300 Amps | > > > to run it. It might take up the entire fair ground as well. How much | > > > money do you have? | >
| > > For a simple calculator, less than 200 very small tubes, which fits in a | > > small portable case, are sufficient: | | They are just simple adding machines. | | > I have built a "calculator" from CMOS SSI. It is | > a 6-bit adder. I am building an 8-bit "calculator" from TTL logic, | > also SSI (with 2 74181's). It's got 32 functions, of which about 4 are | > real and essential math. | | Multiplication and Division need hundreds to thousands transistors/ | tubes.
I see Mul and Div keys on the Anita
Cheers
The text says that it can do multiplication and division. If you use a serial algorithm, you don't need many logic elements for multiplication or division. I think a good idea would be to build a computer with a small and simple instruction set. Maybe you can get a magnetic core memory from ebay, this would be even more vintage :-) and it would save some tubes.
-- Frank Buss, fb@frank-buss.de http://www.frank-buss.de, http://www.it4-systems.de
Yes, you guys are right. But these are specially made counter tubes, which are the equivalence of many basic tubes and/or transistors. They will not be easy to make or buy.
The HP9100 programmable calculator used about 20 SIP-potted discrete-transistor flipflops in its core and a lot of ROM of various types. Program and data storage was magnetic cores, and the display was 7-segment, vector characters on a CRT.
I should think that a "microcoded" (simple state machine with big diode ROM) calculator could be done with about 20 or 30 dual triodes and some sort of storage mechanism, capacitors maybe. But it would be a heap of work.
John
If someone has the money and time to do it, a better show case is to build with transistor/led unit. A blue/red led should show the state of the corresponding transistor state. It would be interesting to look at.
schematics, please. and if neccesary, I have nixie tubes, however, they would require a demux and only 3 bits (they aren't hex and I ain't doin' BCD or converting to decimal).
Look, if someone can simply either tell me or give me a web page with the schematics for: a.) a half adder b.) a full adder and possibly c.) 1/2 of a 7474 equivalent (D-type flip-flop with preset and clear)
I can do bare transistor logic. I just need to be able to do it with tubes. A book on BN or Amazon would be fine. FYI, I THINK a 12Ax7 (where x is T, U, V, X or Y) makes a 2-input AND, OR, NAND or NOR gate.
This show you how to build a full adder:
You would need 5 tubes per bit or 40 tubes for a 8-bit adder.
Might be interesting/amusing to design and build an analog calculator using valves (gotta love the Brits).
ith
for
us
sbe
te
maybe I should clarify more.I need a true schematic, showing how to get plate voltage, how to get heater voltage, and showing all resistor, capacitors and basically every single discrete component.
This help?
adder:
Sorry, I am a transistor guy.
Perhaps someone else can help you with tubes.
Go serial and you only need one a one-bit ALU.
start here:
You can still find dekatrons laying around, as the folks who like pretty glows in the dark sell them on ebay.
A couple of dekatrons and a adder might get you to CORDIC.
really, since the folks at ISEF look for innovation, this is poor choice of project.
I've mentored/sponsored a lot of students for ISEF, I know what I'm talking about.
Now if you made a micropatterned tube with a cold field emission cathode using carbon nanotubes , that would get noticed at ISEF.
So unless you can come up with something mathematically novel to reduce the number of tubes, this is a no go.
What you need to win, aka Scientific Method: You need a hypothesis, a experiment, and a conclusion to win,. not I built a toy. Find a mentor, work on a ISEF project, get a scholarship and something to put on the resume doing it. That is your goal.
This is joke, but it sorta shows the size of a vac tube cpu equvalent to a modern laptop.
Steve Roberts
Tubes are history. Transistors are still in use, just in different scale.
If the OP can build the 4004 using 2300 transistors, I would give him an A.
that may be a good point. but I barely understand this stuff to begin with. I could never make a huge innovation until I know more about, uh... what needs to be innovated on!
Sure, I understand some. I can do CMOS (and now beginning TTL) SSI logic circuits. That's about it. I can do bare transistors, too, but that's pretty much useless.
But for the science fair I want to show a REAL timeline of stuff I've built, representing various time periods in computing. The earliest is going to be a small VT adder. Next - CMOS SSI adder and TTL MSI adder after that, I'll show some REAL calculators Then my laptop, at the end.
It's a nice project in my mind, what are your thoughts?
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.