Relic Information query matches

Matching on query: Unique id = 7/97, Name = any, Associated machine = any, class = any


Unique id/year of acquisition: 7/97
Name: Elliot tape reader
Elliot tape reader
Dimensions: 257x210x161
Class: peripheral
Machine: n/a
Notes:
DW: This is a tape reader which was made by Elliot which was copied from a tape reader which was designed in the Computer Lab. The original version was made in the Engineering Lab from a prototype which was made in the Lab. The prototype only used five-hole tape which is all that was used on the EDSAC 2. The principles of the tape reader are really very simple but it went considerably faster than other tape readers of the time. There was a continuously running roller and underneath there was a pinch roller operated by electromagnet and when these two closed the tape was drawn through a tape reader. The essence of the design was this: there was a slab of metal resting on top of the tape which could be attracted by an electro magnet underneath and when this was attracted it applied a frictional force and the tape stopped essentially immediately. So, this could run at a 1000 characters a second, it could stop within half a character distance which is a twentieth of an inch and these types were used in the Lab perhaps for 2 decades. They were far better than other tape readers of similar age. The design was such that you could load the tape - so it was very easy to load. Another part of the design was a push button which caused the rest of the tape to be ejected at high speed from the tape reader. In this later model there was an adjustment to the width of the tape although the original one used only 5-holed tape and an interesting part of the design is the brakes. When it was originally tried it was discovered that under certain conditions it set off enormous vibrations in the tape reader and so didn't work. It was also discovered that by laying a finger on the top it would then work. The finger is replaced in this design by a little slab of foam between the two parts of the brake which serves the same purpose. These tape readers were very reliable and one occasionally had bad tape such that the tape became translucent where it shouldn't but such failures were rather rare. Another interesting part of the design is that the photocell detectors beneath the tape had to be dropped considerably below the surface. If the photo detectors were adjacent to the tape, the tape was translucent and so all the light came through whether there was a hole or not. By dropping the photodetector, essentially down a hole the translucent part dispersed the light and the signal to noise became acceptable.
Q:It says 1959 - does that seem about right?
DW: They were developed in the Lab from about 55/56 I think. They were certainly used in the early EDSAC. This is a later model which is made by Elliot so 1959 is probably correct. The Elliot ones were used by other computers made in England.
Q: And you said these were used for a couple of decades after that?
DW: Yes, I can't remember the exact date, but the latest computer we made had at least one of these attached which was enough for the purpose. Part of the design of this isn't shown here - you would need some means of collecting the tape which is spinning out at high speed for which large empty cardboard boxes proved suitable. You also need a tape dispenser and the ones we had was a simple slot which was at least a foot deep. So that if you suddenly jerk the tape it wouldn't spin out of the slot. A foot was sufficient and they were also, even for EDSAC 1 - part of the taperoom equipment were little converted grinding wheel which we used to roll up tape with a gear so it was quite fast and easy.
Q: Elliot was a British company ?
DW: Yes, Elliots of Borehamwoods were an electronics company which later became I think part of Marconi and GEC. I think the factory is still there.
See also: 17/97
See also: 71/99


Unique id/year of acquisition: 7/97
Name: Elliot tape reader
Elliot tape reader
Group id: 2
Other nos on object: none
Inscription: ELLIOT TAPE READER TYPE No D/23 SER No731 (plate at top) MADE IN ENGLAND (plate at bottom) BRIT PAT No 884935(plate at bottom)
Dimensions: 257x210x161
Description: Green metal case on 4 rubber feet. at rear are two large round plugs, a vent showing a fan beneath and a series of holes marked 1 to 9. At the front is a long lever, a mechanism of hinged pieces and a roller. There is also an on/off switch and a silver push-button
Class: peripheral
Machine: n/a
Condition: mostly good (dent)
Notes:
DW: This is a tape reader which was made by Elliot which was copied from a tape reader which was designed in the Computer Lab. The original version was made in the Engineering Lab from a prototype which was made in the Lab. The prototype only used five-hole tape which is all that was used on the EDSAC 2. The principles of the tape reader are really very simple but it went considerably faster than other tape readers of the time. There was a continuously running roller and underneath there was a pinch roller operated by electromagnet and when these two closed the tape was drawn through a tape reader. The essence of the design was this: there was a slab of metal resting on top of the tape which could be attracted by an electro magnet underneath and when this was attracted it applied a frictional force and the tape stopped essentially immediately. So, this could run at a 1000 characters a second, it could stop within half a character distance which is a twentieth of an inch and these types were used in the Lab perhaps for 2 decades. They were far better than other tape readers of similar age. The design was such that you could load the tape - so it was very easy to load. Another part of the design was a push button which caused the rest of the tape to be ejected at high speed from the tape reader. In this later model there was an adjustment to the width of the tape although the original one used only 5-holed tape and an interesting part of the design is the brakes. When it was originally tried it was discovered that under certain conditions it set off enormous vibrations in the tape reader and so didn't work. It was also discovered that by laying a finger on the top it would then work. The finger is replaced in this design by a little slab of foam between the two parts of the brake which serves the same purpose. These tape readers were very reliable and one occasionally had bad tape such that the tape became translucent where it shouldn't but such failures were rather rare. Another interesting part of the design is that the photocell detectors beneath the tape had to be dropped considerably below the surface. If the photo detectors were adjacent to the tape, the tape was translucent and so all the light came through whether there was a hole or not. By dropping the photodetector, essentially down a hole the translucent part dispersed the light and the signal to noise became acceptable.
Q:It says 1959 - does that seem about right?
DW: They were developed in the Lab from about 55/56 I think. They were certainly used in the early EDSAC. This is a later model which is made by Elliot so 1959 is probably correct. The Elliot ones were used by other computers made in England.
Q: And you said these were used for a couple of decades after that?
DW: Yes, I can't remember the exact date, but the latest computer we made had at least one of these attached which was enough for the purpose. Part of the design of this isn't shown here - you would need some means of collecting the tape which is spinning out at high speed for which large empty cardboard boxes proved suitable. You also need a tape dispenser and the ones we had was a simple slot which was at least a foot deep. So that if you suddenly jerk the tape it wouldn't spin out of the slot. A foot was sufficient and they were also, even for EDSAC 1 - part of the taperoom equipment were little converted grinding wheel which we used to roll up tape with a gear so it was quite fast and easy.
Q: Elliot was a British company ?
DW: Yes, Elliots of Borehamwoods were an electronics company which later became I think part of Marconi and GEC. I think the factory is still there.
See also: 17/97
See also: 71/99


Number of matches = 2 Copyright University of Cambridge Computer Laboratory, 1999. All rights reserved.