PDA

View Full Version : A diversion... Analogue Computers



Terry Porritt
09-08-2002, 10:44 PM
Poppa John in another posting mentioned he had maybe read about a vacuum 'toob', ie valve, computer circa 1938. If so this was probably an analogue computer, as they had been around from the mid 20s when valve were used to make high gain amplifiers, ie op-amps. They were further developed during the war.

One that I was involved as a team member in building during 1956/7 was at Joseph Lucas Gas Turbine, and it was quite advanced for the day. At that time you had to build your own, if you wanted performance.

From memory it 24 high gain amplifiers that could be wired as either integrators or differentiators, 10 turn pots were on the inputs, and the valves, again from memory were the fairly new EF86.

A patch panel was used for wiring the system.

In those days, complex systems were analysed using the method of linearizing for small perturbations, and so for example a jet engine fuel system /engine could be represented by an nth order differential equation.

Early jet planes like Vampires and Meteors were somewhat unstable and the pilot was continually adjusting the throttle to keep constant rpm with changing altitude and Mach number, whilst at the same time steering a delicate course between engine surging and stalling during acceleration.

Flame-outs were not uncommon, particularly in hot desert conditions.

The analogue computer we built could either be setup to behave as an engine, or as a fuel system.

Our electronics lab was directly over the basement test cells.
Several of the staff were ex-RAF so it was only natural to use ex-RAF R1155 receivers and T1154 transmitters to communicate between the two departments :)

Output from the computer was fed into a lab built very, very, high speed real time X-Y plotter, servo controlled, with the pen driven by taught Bowden cables and pulleys, again nothing like that could be bought.

Most often the output was the response of the system to a step-function input, which represented the pilot suddenly opening or closing the throttle.

Interesting days of Otis King slide rules, hand wound mechanical calculators, even electric Monroe mechanical calculators, 6 or 8 figure log tables.

How would the young lads of today cope if transported back to that time? ;)

godfather
09-08-2002, 10:57 PM
The good old days....

I cant profess to have experience that far back, but did do quite a bit of analogue servo control with valves.

Have still got quite a lot of valve equipment, never use it though.

100 watt modulator using push-pull KT88's, silver plated lecher-line tuned 144 MHz transmitter
200 watt linear amplifier using 807's with a 1.5 kV power supply. Receiver from a japanese WW2 submarine with plug-in coil units to band change
Hallicrafters SR80 Transceiver

Slightly newer,
A Toshiba T1000 laptop running MS DOS 2.2, 512kb ram, 1 FDD (720k)

Wont accept XP though, cant find a place for the CD (Parry could suggest somewhere...)

Baldy
09-08-2002, 11:01 PM
I think Poppa might have seen one of them at the Battle Of Hastings, 1066..... evidently, according to one of the other threads, he WAS there.

Mike
09-08-2002, 11:05 PM
> How would the young lads of today cope if
> transported back to that time? ;)

well... ummm... I took apart an 8088 once... and ummm... I've programmed in BASIC... ummm... <grows old and dies>

Mike.

Poppa John
09-08-2002, 11:09 PM
Terry Porritt.. Oh the memories!! Same here with the Sliderules. Log tables. Algebra & Tryg. When "Handheld" Calculaters came out, they weren't allowed in the Exam Room. I would be lost without one now.
Re the Valve Computer, maybe that was the one I was thinking about, I believe it was Massive very basic. Poppa John

godfather
09-08-2002, 11:11 PM
I'm not really THAT old, its just that I was very young when I was born.

Mike
09-08-2002, 11:14 PM
I've got this big old radio sitting on my shelf next to my 3' high computer speakers...

its big and white, got three big knobs on the front, and if you turn it around it doesn't have a back on it and inside its got lots of things that look like half-size lightbulbs.

Mike.

Mike
09-08-2002, 11:15 PM
> I was very young when I was born.

you too? so was I!

Mike.

Poppa John
09-08-2002, 11:18 PM
Baldy, You'll keep. Poppa John :D :D

godfather
09-08-2002, 11:22 PM
Poppa John is the one on the right, here (http://www.battle1066.com/index.html) with the computer operated sword.

Poppa John
09-08-2002, 11:37 PM
Goddie, You'll keep as well!!! Can't be me actually, He looks as if he is wearing a Halo!! My Ex would say it should be Horns. Have saved the site to favourites to peruse tommorow. Lived in Eastbourne (the Pommie one) from 10yrs to 15yrs & never got to Hastings. One of the things on my list of things to do when I win Lotto. Wishful Poppa John :D

exLL
10-08-2002, 01:20 AM
Hi Poppa John.

Is this the computer you were thinking of? Atanasoff - Berry 1937-1939, about 300 or so valves used.

valve computer (http://www.ieee.org/organizations/history_center/milestones_photos/atanasoff.html)

Also, a while back you asked about the National Bank advertising music, The Four Seasons by Vivaldi. Did you know that the Warehouse have the CDs on sale for 89cents at present? They are in a cardboard cover. The number on the back of the cover is 3516.2038-2

I hope I got the above web address right, I usually seem to muck it up.

Cheers.

ExLL

Terry Porritt
10-08-2002, 10:08 AM
That's really interesting exLL. It's really amazing just how many famous concepts were first drawn up on table knapkins :)
Springing to mind is Mitchell, sketching out the Spitfire design on one, I think Barnes Wallis also used one to outline one of his designs, and also Niels Bohr with an atomic problem.

The backs of envelopes must also have had special properties, as there used to be a saying about a problem being capable of working out on the back of an envelope.

Graham L
10-08-2002, 03:21 PM
There's a bit of argument about Athanasoff + Berry (did it ever calculate anything?). There was a German who made a few digital computers in the late 30s and early 40s. The Nazis decided it wasn't useful, so he got no money. He used second-hand movie film as storage (punched holes). He got up to a version 4, I think. The US Govt put a lot of money into digital computers (for ballistic table calculations) , but not one of them was finished before the end of the war. Maybe the Germans were right about the development time scale.

A lot of the Bletchley "computing" was done with punched card equipment --- basically sorting --- with selection controlled by plug panels. People forget that after Hollerith and the US Census, card data processing was very highly developed. The early Enigma decoding was done with real mechanical computers (stacks of cardboard sheets with holes punched in each at appropriate places. Light shone through if there was a matchg.) The Bombes were built from telephone exchange hardware --- relays and uniselectors. The various Collossi were valve computers, but very simple. I don't think they were really "programmable" except with plug panels.
They had pretty hairy mechanical peripherals too: 1000 cps paper tape readers. They had some trouble with the paper tapes catching fire.

Graham L
10-08-2002, 04:33 PM
And if you want to see an analogue computer, here is the Meccano Differential Analyser (http://www.cl.cam.ac.uk/Relics/archive_photos.html) from 1938, (and some other early computers). One of the 14 Meccano DAs built and used for artillery prediction tables in WW2 came to NZ in 1950. It is now at the MOTAT.

Neil McC
10-08-2002, 04:56 PM
Don't know if you guys are into valves!
http://www.netcomuk.co.uk/~wwl/electric.html
Sorry,haven't found the fancy link thing.

-=JM=-
10-08-2002, 08:12 PM
The Otis King Pocket Calculator is a very powerful tool. Read about them here (http://dicklyon.home.mindspring.com/OK/OtisKing.html) and here (http://chris.gillings.com/collect/slide/otis_a9123/).

They even built one in to the handle of a walking cane at one stage.

They can handle powers, roots, multiplication, division, logs and the rest of it.


I really would like to get one for my self.

Poppa John
10-08-2002, 08:59 PM
exLL. No not that one either, the one I am thinking of was huge. It filled one or two rooms of the building it was in. It was 1938 tho, which for some reason was a very good year, when a fantastic boy baby was born. Cannot think of his name at the moment!!!!]:) Poppa john ]:)

Terry Porritt
10-08-2002, 09:26 PM
Thats more interesting info, JM, I would like to have one also, those in the pictures are really collectors items going back a bit.

The other slide rule manufacturer that made very long length slide rules was Fowler. These were really big, mounted on a stand like a mortar. Cant remember the scale length, but probably about 6 feet or more, ie 2 metres, for the young lads. They could be read to about 6 figures or so depending on the position along the logarithmic scale. I have a small circular pocket one about 120mm dia. with a 50 inch scale length. (note the bilingual dimensional language) :)

Refering to the differential analysers mentioned by Graham, as an aside, the simplest mechanical integrator is the 'Planimeter' used for measuring the area of a closed curve, and used to be found in most decent drawing offices.

Another simple device for solving 2 dimensional field problems such as diffusion, electric field, heat flow and fluid flow was the use of conducting paper, or 'Teledeltos' paper, with the boundaries drawn in conducting siver ink. For 3 dimensional problems, electrolytic tanks were used, but they were cumbersome and not all that practical.
A voltage was applied to the source boundary, zero volts to the sink boundary, intermediate potentials were analogous the pressure or temperature field acording to the problem.

Similar problems could often be solved on paper using Southwells Relaxation method, if the boundaries could be defined reasonably.

This was before digital computErs, began to be widely used for number crunching. Up until then, computOrs were generally rows of lowly, menial girls (not PC at all!) working away on their mechanical calculators, consuming vast quantities of analysis pads.

exLL
10-08-2002, 09:58 PM
Hi Neil, I found that very interesting, especially the ball-bearing non-magnetic motor. It was weird. (and useless as a motor!)

Re the *fancy link thing* you mentioned, godfather helped me with it a few days ago. Although the explanation is all there in the formatting options box and seems easy once you know how, I found it a bit hard to follow when I first tried to understand it. Here is my explanation of how to do it. (Purposely long winded for the benefit of any absolute newbies who may find it useful also):

I will use these brackets { } so that I can demonstrate it without activating anything. When you try it, use these ones [ ] so that it will activate.

When typing a reply, to see what options are available, click on *Formatting Options* at top left of the reply box.

Another box will appear showing the various options.

If you look at the bottom item you will see the details of how to enter web addresses etc.

Basically it goes like this: Remember to use [ instead of {

Firstly, type a left bracket like this {

Then type URL=

Then enter the web address

Then close the web address with a right bracket}

At this point you type in the name that you want to substitute for a web address, for example; VALVES

You now close the instructions by typing {/URL}

The whole line will now look like this:

{URL=the web address will show here}VALVES{/URL}

Note that there are no spaces before or after any of the brackets, but you can put spaces between the words in the name/s that you choose to use, eg: GLASS RADIO VALVES, instead of just: VALVES

You can now click on Preview and it will show a clickable link named VALVES underlined in blue.

If you have finished your message click on Post Message, if not, click on go back/edit to return to the message to make changes.

-=JM=-
10-08-2002, 10:02 PM
>>>Thats more interesting info, JM, I would like to have one also, those in the pictures are really collectors items going back a bit.

I've done quite a bit of research into them and have played around with one for about a month. They are really quite easy to use once you get used to them. Only problem is when the paper is starting to peel off and isn't lined up properly.

The first one I saw was when my Year 12 (6th Form) maths teacher showed the class one. He'd found it when he was working on a farm in Scottland somewhere. He had no idea what it was.

Last I looked they went for around $60-110 here in NZ.

-=JM=-
10-08-2002, 10:08 PM
Also a different model was also found in one of the old math supplies cupboards at my school. Looked as if they had them as the school calculators (upgraded the abacus).

Also have a look here (http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=2127463518)

pporritt
11-08-2002, 10:01 AM
How about this...
http://www.slideruleguy.com/fwst.htm

Terry Porritt
11-08-2002, 12:56 PM
I see my 'lad' joined in.

I have been getting Fowler confused with Fuller slide rules.
The mortar-like cylindrical one is the Fuller and its scale length is 41 feet.

http://www.hpmuseum.org/srcyl.htm

Fowler were noted for their circular slide rules and the one I have is a Magnum just as illustrated here (http://www.gemmary.com/instcat/09/p18-114-09.html), in picture B

pporritt
11-08-2002, 02:52 PM
Thanks, Dad...

All this talk about early computers, and the person who really deserves more recognition is overlooked - Charles Babbage. You can read all about him at http://www.cbi.umn.edu/exhibits/cb.html.
Apparently, the University of Auckland has a collection of some of his material and manuscripts

Graham L
11-08-2002, 03:14 PM
Well, Babbage originated the standard government computer contract syndrome: delivered "late, over budget, and not working." (But it was capable of working --- he just never made it go before "improving" it.) The Science Museum (Kensington?) built enough of it according to his designs to prove that it was possible. Though the difference engine was not "really" a general purpose computer.

E.ric
11-08-2002, 08:02 PM
Do you know how valves were invented?
OK I will tell you,

The first electric lamps deposited soot on the glass, my guess is that this bugged some people as after a while the lamp would fog up with soot and then you would turn on a black lamp, so some clever person thought if they put a electrode inside the lamp and made it positive in respect to the heater the shoot might get attracted to the electrode and not the glass, well I think that failed, so then they made the electrode negative, Then they noticed when the electrode was positive current flowed, when the electrode was negative no current would flow, it took someone a few years after that to think of a application which turned out to be the valve. Not sure how they invented invisible soot but it works really well after a while the build up is so great it snaps the heater.