Mysteries of The Gallery of Old Iron Mystery #1 -- Do you know what this is? Information about any of the devices on this page is much appreciated. Please email The Curator (curator@thegalleryofoldiron.com)
The device pictured above is most likely some kind of System 360 tester. The color, light style, and SLT circuits inside would indicate 360 equipment. However, it could also be another architecture, but made around the same time. There is a manufacturing tag inside that says 1969. This device is very heavy and had two large fat cables coming out the back (both cut years ago -- one with 4 large fat wires, one with at least a hundred small signal wires). Inside is a complete backplane populated with SLT cards, so it obviously has some sophisticated logic capabilities. There are no tags on the outside identifying a model or type. Any help identifying this will be appreciated. Have we Finally Identified this Device? The box depicted here looks familiar to me. although I don’t remember a great deal about it, I’ll tell you what I do recall. I was an IBM CE working in Washington DC between late 1966 and the summer of 1970. I originally worked on hardware, and then graduated to software support (OS/360), and then went on to work as a programmer for the Field Engineering Division, in Raleigh, NC. The box in the picture is a special logic-testing box designed to try to “trap” elusive hardware bugs in most any IBM box using one of the forms of SLT logic (the technology of the 360). I never used one in the field, but I vaguely recall being introduced to the box in a special lab session at one of the CE training courses I attended. I think it may have been on the 2050 (360/50) school in Kingston, NY. They showed the box to us, as I recall, so we would be aware of its existence, just in case we ever ran into an intractable bug in the field and needed to bring in more sophisticated diagnostic hardware. As you’ve probably surmised, the box will compare a set of logic signals with a set of switch settings (forming, for example, a ROS address, or some such thing), and then it use this equal-compare to generate a sync pulse for an oscilloscope, or just use it to capture and hold some logic lines into a set of lights. As you can see, the sync pulse it produced could be tapped-into from the BNC connector on the front panel. The logic probes themselves were what was attached to the heavy cable you described as having once being attached to the backside of the box. This tool, as I recall, was not supplied to individual CE’s or even to branch offices, but to “Area Support Centers”, where “Area Specialists” were on call to work on the tough problems encountered by the regular field force. By the way, when I transferred from the field to work in Raleigh, I was assigned to a group working on an information system used for supporting customers during the System/370 era. (See the article on “RETAIN” in Wikipedia). This group was part of an FE organization called “Maintenance Technology”. In the same building where I worked, there was a small group of engineers who comprised the “Tools and Test Equipment” lab. I don’t know if they developed the box in your picture, but they did a number of other devices of the same general type. The most advanced of these was a suitcase-sized computer called the PT-2 (I believe PT stood for Portable Tester, or some such thing). The PT-2, like the box in your picture, was equipped with a big plug in the back that could be used to tap into the logic circuits of most any type of control unit that was equipped with the right sort of connector. It was, in fact, a small computer with 64K of RAM, a microprocessor code-named the “Palm” , a keyboard, display screen, and small cartridge tape drive. … and a modem with acoustic coupler. I did take a short programming course on writing code for it, in support of a project involving RETAIN that was eventually scrapped. Very few people are aware that the PT-2 was actually the precursor to the IBM 5100 desktop computer. The folks from Boca Raton came up to Raleigh, and were reportedly very resentful that some guys in the “tools and test equipment” lab had built a general-purpose computer. The FE Division was not accorded the kind of respect from product development groups that made such an affront acceptable to them. I’ve never seen any mention of this connection in any “official history” of the 5100-series of computers. Russell P. Holsclaw Boulder, Colorado And Another Lead for this Device? Hi Daniel, This looks like a production model of the Buffered TDAT (TP Diagnostic Analyzer Device), which was deveopled to trouble shoot TP problems in the late 1960s in Kingston by tech ops. The original prototype was hand made and was in a leather suitcase. In 1969 I was a Instructor at the FE Education Center in Endicott, NY. My specialty was Teleprocessing, more specifically "line control protocols." Endicott was point of control for TP education and we were contacted by the FE Tools & Test department in Kingston, NY who asked us to try the "Buffered TDAT" to diagnose problems with TP equipment at our lab. Another instructor, Frank Sapio and myself were sent to Kingston to evaluate the equipment and its functionality. It was the first prototype and we were impressed with its capability. We brought it back to Endicott and used it to exercise our 1050 and 2740 terminals. It worked as advertised. By the time it made its way through the bureaucracy of Engineering it incorporated other test function, one of which was developed in the field by Jack Scully, a FE Specialist in Philadelphia, who worked for me when I left Endicott to go to Phila as a Field Manager . Jack later received a huge suggestion award. The production device was cumbersome and later the technology was integrated into the next generation of TP devices thus eliminating any real need for the tool. That's all I remember. I hope it is accurate and useful for your purposes. Regard, Frank G Ahern Another lead I believe Mr. Holsclaw is correct and this is the logic analyzer we called the 'rat trap'. I was an IBM CE and PSR in the late 60's and 70's, trained on the 360 Models 65 and 168 and had occasion to see some Area and Regional specialists use it to debug intermittent problems. It was not a commonly available tool and was very difficult to set up and use productively. Mike Tunney Charlotte NC More Mysteries Below are more devices that we either have been unable to identify at all, are unsure about, or have only limited informatin about. Click on thumbnails for larger versions. Mystery #19 follows. Gallery of Old Iron reader Michael Moran would like to identify this IBM device. It would seem to be a communications device. Are we right? Any idea of the model or function? Send ideas to us here at The Gallery (curator at-sign nc dot rr dot com), or the submitter (msmoran at-sign nycapp dot rr dot com)and thanks! Mystery #18 follows. Gallery of Old Iron reader Brennan Barber would like to identify this IBM chip. Any ideas how to decode the numbers? Send ideas to bbarber@nc.rr.com and thanks! Mystery #17 follows. This is an old stack of IBM core memory. Can you help a Gallery reader who is having trouble identifying the machine type? It's obviously quite old and from the days when these were hand made (notice each plane has handwriting). An Identification? I think that is a core storage unit from a 305 RAMAC. It held 100 bytes of data--there were 100 cores on each layer. I believe 6 data bits and a parity bit-total 7 layers. It was used as a buffer as information passed to and from peripherals and the drum storage. The 305 used for storage : drum, disk, core, relays and punch cards and a control panel. Dave Milsom, Kingston, Ontario Below is Mystery #16. This is an old bank of relays, each stamped, "IBM." The Gallery is curious to know what kinds of devices might have used such banks of relays. The only device we are aware of is the Automatic Sequence Controlled Calculator, also known as the Harvard Mark 1, which can be seen on this page. Below is Mystery #15. This is an old wooden logo plaque of unknown vintage. Any idea as to what device might this might have been attached? It is said to be from the 1939 World's Fair in New York, although The Gallery has been unable to find any photographic evidence that confirms any origin, World's Fair or otherwise. It is approximately 5 inches (13cm) in diameter and 1/2 inch (1.3cm) thick. Below is Mystery #14. This is again clearly 1950's technology. The Gallery is unsure what IBM devices might have used this style of pluggable unit. Does anyone have information on this type PU? We now believe this is Bendix computer technology, but that's about all we've found. Click the thumbnail below and solve a Mystery #13 for us. SAGE experts have given up. We believe this is a SAGE Pluggable Unit, but the configuration is odd -- 7 IBM labelled 6146 tubes and some very hefty bar resistors in the cage, each separated by a metal plate (for heat shielding? electrical interference shielding?). Can you identify what function this PU served? Below is Mystery #12. This is clearly 1950's technology. It utilizes IBM 70x series 8-tube pluggable units, plus single tube PUs. The tag inside says "CLOCK NO. 1." We believe this is a clocking signal generation circuit, possibly used to drive a series 70x CPU, but we've been unable to find anything else quite like it and nothing like this appears in any 70x photos that we have (but perhaps it is like other customer engineering panels and generally hidden from view). Then again it could be an early piece of test equipment. Below is Mystery #11. This control panel came from an estate with a lot of IBM 705 and SAGE related items. It would appear to be a maintenance panel of some kind. It might be from a SAGE installation, since it mentions 'drum', 'tape', 'card', 'print', 'punch', and 'processor,' but no 'disk' (SAGE was said to use drums instead of disks because one of the chief designers didn't trust disk storage). Or it could be from a 650 drum calculator, or a 70x series machine, or ??? Some older mysteries ... not all of which are solved! We know Mystery #2 below, called "Maintenance Device 1," could help diagnose the 3880 control unit (for 3380 disk drives), the 3480 tape cartridge drive, the 3088 CTC device, and certain DPPX machines. We believe it may have been able to diagnose the 3083 processor controller for 308X processor complexes. Can you verify? Does anyone still have any program listings for any diagnostic programs that ran on this device? Kevin O'Mara dropped us a note to say that this device was also used to test the 3800 laster printers and could also act as a host controller for bi-sync communications testing. Mystery #3 is the big core below, from a Mohawk Data Sciences Corp. System 2400, date unknown. It seems like early 70's technology. Do you have pictures of this device, or more information about it? #3 was surprisingly tough. After a long while on this page, someone who knows about this device contacted us. Thanks very much to Paul Schroeter of Australia not only for recognizing this, but also for sending the following picture of a bunch of MDS devices together in a 2400 configuration. Below is Mystery #4. Do you know what this is? Below is Mystery #5. Do you know what this is? Below is Mystery #6. Do you know what this is? Below is Mystery #7. Do you know what this is? Thanks to Uwe Seier from Germany, we know this is from a StorageTek Nearline 4410 robotic tape system. It is a fascinating device and Uwe sent great pictures! Dave Hobin also wrote to tell us that this is the rear panel from 4480 or 4490 Tape unit, which can also be used stand-alone (not connected to the 4410 robotic tape system). He went on, "Your photo [link above] of the 4410 shows a drive 44xx attached to panel 9 on the 4410. the door is panel 11 and if you count over clock wise you will see a cover panel on panel 10 the a white box/fraim attached to the 4410 on panel 9 this is the drive fraim were the rear panel you had listed as mystery #7 is located." Thanks, Dave! Below is Mystery #8. Model I, II, or III? Can you help identify The Gallery's IBM 705 as one of these submodels? We believe based on the picture of a Model III that it must be a Model I or a Model II, but we are unsure how to tell them apart by the operator console (if that's even possible. We now know the Model 705 console is from a Model I (thanks to Kevin Stumpf). Mystery #9: There was a vector processing unit (VPU), attachable to the System 360 Model 44 via Direct Data Channel, and perhaps to other 360 models as well. Can anyone identify the device by model number? A picture? Documentation? We are told that a number of these devices were sold to oil companies in Texas, presumably for analysis of seismographic data. Thanks to George Paul (who designed the FFT algoritm for it), we know that device in question is the IBM 2938 Algorithmic Processor and it has a very interesting history. Hopefully The Gallery will collect enough about it to develop a page dedicated to it (with a picture ... we hear one customer painted racing stripes on one). We'd still like to find the functional characteristics manual, stories, or other artifacts concerning this fascinating device. Mystery #10: There was another device attachable to at least certain System 360 Model models that was also (like the 2938 above) dedicated to mathmatical calculation. It was called the IBM 2937 Multiplication Summation Processor (MSP). Anyone have more information about this device, such as the functional characteristics, a picture, or stories? We've been able to find almost nothing, except mention of its existance. Return to The Gallery of Old Iron home page

Mysteries of The Gallery of Old Iron

Mystery #1 -- Do you know what this is?

Information about any of the devices on this page is much appreciated. Please email The Curator (curator@thegalleryofoldiron.com)

The device pictured above is most likely some kind of System 360 tester. The color, light style, and SLT circuits inside would indicate 360 equipment. However, it could also be another architecture, but made around the same time. There is a manufacturing tag inside that says 1969.

This device is very heavy and had two large fat cables coming out the back (both cut years ago -- one with 4 large fat wires, one with at least a hundred small signal wires). Inside is a complete backplane populated with SLT cards, so it obviously has some sophisticated logic capabilities. There are no tags on the outside identifying a model or type. Any help identifying this will be appreciated.

Have we Finally Identified this Device?

The box depicted here looks familiar to me. although I don’t remember a great deal about it, I’ll tell you what I do recall.
I was an IBM CE working in Washington DC between late 1966 and the summer of 1970. I originally worked on hardware, and then graduated to software support (OS/360), and then went on to work as a programmer for the Field Engineering Division, in Raleigh, NC.
The box in the picture is a special logic-testing box designed to try to “trap” elusive hardware bugs in most any IBM box using one of the forms of SLT logic (the technology of the 360). I never used one in the field, but I vaguely recall being introduced to the box in a special lab session at one of the CE training courses I attended. I think it may have been on the 2050 (360/50) school in Kingston, NY.
They showed the box to us, as I recall, so we would be aware of its existence, just in case we ever ran into an intractable bug in the field and needed to bring in more sophisticated diagnostic hardware. As you’ve probably surmised, the box will compare a set of logic signals with a set of switch settings (forming, for example, a ROS address, or some such thing), and then it use this equal-compare to generate a sync pulse for an oscilloscope, or just use it to capture and hold some logic lines into a set of lights.
As you can see, the sync pulse it produced could be tapped-into from the BNC connector on the front panel. The logic probes themselves were what was attached to the heavy cable you described as having once being attached to the backside of the box.
This tool, as I recall, was not supplied to individual CE’s or even to branch offices, but to “Area Support Centers”, where “Area Specialists” were on call to work on the tough problems encountered by the regular field force.
By the way, when I transferred from the field to work in Raleigh, I was assigned to a group working on an information system used for supporting customers during the System/370 era. (See the article on “RETAIN” in Wikipedia). This group was part of an FE organization called “Maintenance Technology”. In the same building where I worked, there was a small group of engineers who comprised the “Tools and Test Equipment” lab. I don’t know if they developed the box in your picture, but they did a number of other devices of the same general type. The most advanced of these was a suitcase-sized computer called the PT-2 (I believe PT stood for Portable Tester, or some such thing). The PT-2, like the box in your picture, was equipped with a big plug in the back that could be used to tap into the logic circuits of most any type of control unit that was equipped with the right sort of connector. It was, in fact, a small computer with 64K of RAM, a microprocessor code-named the “Palm” , a keyboard, display screen, and small cartridge tape drive. … and a modem with acoustic coupler. I did take a short programming course on writing code for it, in support of a project involving RETAIN that was eventually scrapped.

Very few people are aware that the PT-2 was actually the precursor to the IBM 5100 desktop computer. The folks from Boca Raton came up to Raleigh, and were reportedly very resentful that some guys in the “tools and test equipment” lab had built a general-purpose computer. The FE Division was not accorded the kind of respect from product development groups that made such an affront acceptable to them. I’ve never seen any mention of this connection in any “official history” of the 5100-series of computers.

Russell P. Holsclaw
Boulder, Colorado

And Another Lead for this Device?

Hi Daniel,
This looks like a production model of the Buffered TDAT (TP Diagnostic Analyzer Device), which was deveopled to trouble shoot TP problems in the late 1960s in Kingston by tech ops. The original prototype was hand made and was in a leather suitcase.
In 1969 I was a Instructor at the FE Education Center in Endicott, NY. My specialty was Teleprocessing, more specifically "line control protocols." Endicott was point of control for TP education and we were contacted by the FE Tools & Test department in Kingston, NY who asked us to try the "Buffered TDAT" to diagnose problems with TP equipment at our lab. Another instructor, Frank Sapio and myself were sent to Kingston to evaluate the equipment and its functionality. It was the first prototype and we were impressed with its capability. We brought it back to Endicott and used it to exercise our 1050 and 2740 terminals. It worked as advertised.
By the time it made its way through the bureaucracy of Engineering it incorporated other test function, one of which was developed in the field by Jack Scully, a FE Specialist in Philadelphia, who worked for me when I left Endicott to go to Phila as a Field Manager . Jack later received a huge suggestion award. The production device was cumbersome and later the technology was integrated into the next generation of TP devices thus eliminating any real need for the tool.
That's all I remember. I hope it is accurate and useful for your purposes.

Regard,
Frank G Ahern

Another lead

I believe Mr. Holsclaw is correct and this is the logic analyzer we called the 'rat trap'. I was an IBM CE and PSR in the late 60's and 70's, trained on the 360 Models 65 and 168 and had occasion to see some Area and Regional specialists use it to debug intermittent problems. It was not a commonly available tool and was very difficult to set up and use productively.

Mike Tunney
Charlotte NC

More Mysteries

Below are more devices that we either have been unable to identify at all, are unsure about, or have only limited informatin about. Click on thumbnails for larger versions.

Mystery #19 follows. Gallery of Old Iron reader Michael Moran would like to identify this IBM device. It would seem to be a communications device. Are we right? Any idea of the model or function? Send ideas to us here at The Gallery (curator at-sign nc dot rr dot com), or the submitter (msmoran at-sign nycapp dot rr dot com)and thanks!

Mystery #18 follows. Gallery of Old Iron reader Brennan Barber would like to identify this IBM chip. Any ideas how to decode the numbers? Send ideas to bbarber@nc.rr.com and thanks!

Mystery #17 follows. This is an old stack of IBM core memory. Can you help a Gallery reader who is having trouble identifying the machine type? It's obviously quite old and from the days when these were hand made (notice each plane has handwriting).

An Identification?

I think that is a core storage unit from a 305 RAMAC. It held 100 bytes of data--there were 100 cores on each layer. I believe 6 data bits and a parity bit-total 7 layers. It was used as a buffer as information passed to and from peripherals and the drum storage.

The 305 used for storage : drum, disk, core, relays and punch cards and a control panel.

Dave Milsom,
Kingston, Ontario

Below is Mystery #16. This is an old bank of relays, each stamped, "IBM." The Gallery is curious to know what kinds of devices might have used such banks of relays. The only device we are aware of is the Automatic Sequence Controlled Calculator, also known as the Harvard Mark 1, which can be seen on this page.

Below is Mystery #15. This is an old wooden logo plaque of unknown vintage. Any idea as to what device might this might have been attached? It is said to be from the 1939 World's Fair in New York, although The Gallery has been unable to find any photographic evidence that confirms any origin, World's Fair or otherwise. It is approximately 5 inches (13cm) in diameter and 1/2 inch (1.3cm) thick.

Below is Mystery #14. This is again clearly 1950's technology. The Gallery is unsure what IBM devices might have used this style of pluggable unit. Does anyone have information on this type PU?

We now believe this is Bendix computer technology, but that's about all we've found.

Click the thumbnail below and solve a Mystery #13 for us. SAGE experts have given up. We believe this is a SAGE Pluggable Unit, but the configuration is odd -- 7 IBM labelled 6146 tubes and some very hefty bar resistors in the cage, each separated by a metal plate (for heat shielding? electrical interference shielding?). Can you identify what function this PU served?

Below is Mystery #12. This is clearly 1950's technology. It utilizes IBM 70x series 8-tube pluggable units, plus single tube PUs. The tag inside says "CLOCK NO. 1." We believe this is a clocking signal generation circuit, possibly used to drive a series 70x CPU, but we've been unable to find anything else quite like it and nothing like this appears in any 70x photos that we have (but perhaps it is like other customer engineering panels and generally hidden from view). Then again it could be an early piece of test equipment.

Below is Mystery #11. This control panel came from an estate with a lot of IBM 705 and SAGE related items. It would appear to be a maintenance panel of some kind. It might be from a SAGE installation, since it mentions 'drum', 'tape', 'card', 'print', 'punch', and 'processor,' but no 'disk' (SAGE was said to use drums instead of disks because one of the chief designers didn't trust disk storage). Or it could be from a 650 drum calculator, or a 70x series machine, or ???

Some older mysteries ... not all of which are solved!

We know Mystery #2 below, called "Maintenance Device 1," could help diagnose the 3880 control unit (for 3380 disk drives), the 3480 tape cartridge drive, the 3088 CTC device, and certain DPPX machines. We believe it may have been able to diagnose the 3083 processor controller for 308X processor complexes. Can you verify? Does anyone still have any program listings for any diagnostic programs that ran on this device?

Kevin O'Mara dropped us a note to say that this device was also used to test the 3800 laster printers and could also act as a host controller for bi-sync communications testing.

Mystery #3 is the big core below, from a Mohawk Data Sciences Corp. System 2400, date unknown. It seems like early 70's technology. Do you have pictures of this device, or more information about it?

#3 was surprisingly tough. After a long while on this page, someone who knows about this device contacted us. Thanks very much to Paul Schroeter of Australia not only for recognizing this, but also for sending the following picture of a bunch of MDS devices together in a 2400 configuration.

Below is Mystery #4. Do you know what this is?

Below is Mystery #5. Do you know what this is?

Below is Mystery #6. Do you know what this is?

Below is Mystery #7. Do you know what this is?

Thanks to Uwe Seier from Germany, we know this is from a StorageTek Nearline 4410 robotic tape system. It is a fascinating device and Uwe sent great pictures!

Dave Hobin also wrote to tell us that this is the rear panel from 4480 or 4490 Tape unit, which can also be used stand-alone (not connected to the 4410 robotic tape system). He went on, "Your photo [link above] of the 4410 shows a drive 44xx attached to panel 9 on the 4410. the door is panel 11 and if you count over clock wise you will see a cover panel on panel 10 the a white box/fraim attached to the 4410 on panel 9 this is the drive fraim were the rear panel you had listed as mystery #7 is located." Thanks, Dave!

Below is Mystery #8. Model I, II, or III?
Can you help identify The Gallery's IBM 705 as one of these submodels? We believe based on the picture of a Model III that it must be a Model I or a Model II, but we are unsure how to tell them apart by the operator console (if that's even possible.

We now know the Model 705 console is from a Model I (thanks to Kevin Stumpf).

Mystery #9:
There was a vector processing unit (VPU), attachable to the System 360 Model 44 via Direct Data Channel, and perhaps to other 360 models as well. Can anyone identify the device by model number? A picture? Documentation? We are told that a number of these devices were sold to oil companies in Texas, presumably for analysis of seismographic data.

Thanks to George Paul (who designed the FFT algoritm for it), we know that device in question is the IBM 2938 Algorithmic Processor and it has a very interesting history. Hopefully The Gallery will collect enough about it to develop a page dedicated to it (with a picture ... we hear one customer painted racing stripes on one). We'd still like to find the functional characteristics manual, stories, or other artifacts concerning this fascinating device.

Mystery #10:
There was another device attachable to at least certain System 360 Model models that was also (like the 2938 above) dedicated to mathmatical calculation. It was called the IBM 2937 Multiplication Summation Processor (MSP). Anyone have more information about this device, such as the functional characteristics, a picture, or stories? We've been able to find almost nothing, except mention of its existance.

Return to The Gallery of Old Iron home page