I believe that some of the plotters projected light onto photosensitive media. There would be an XY motion system and a series of aperture shapes and sizes that would mask off the light.
My father was a machinist at the Gerber Factory in Vernon Connecticut in late 60s and early 70's. He said they made Fabric cutters and plotters. He's gone now, sorry I don't have much info. Thanks for investigating this equipment!
on card #2, the large chips with stickers are EPROMS, possibly with microcode for what looks like a TTL cpu card #4 looks like a 16 bit ALU or some sort
'The Gerber format is an open, ASCII, vector format for printed circuit board designs. It is the de facto standard used by PCB industry software to describe the printed circuit board images.' This is still used today every time a PCB is designed/built.
The chips on the GNS-1-3 card that are labeled aren't microprocessors. MM5204 is listed as a 4096-bit EPROM. Either the other chips on that card are doing addressing logic for the EPROMs, or the card uses the EPROMS to do something similar to what a PAL chip does (another example of that kind of design would be the Apple ][ Disk controller card - except that the GNS-1-3 card would be a much more complex design; if it is in fact doing something like that).
Going by the comment on table 2 of my Gerber Book, you had a choice of the following data formats over all the various models of plot controllers eia ascii / eia / bcd / ebdic / iso ascii The 6x00 controllers could only handle eia ascii or ebdic, but there is a footnote to the table that you had to choose which of the two you wanted when ordering the controller This explains all the various ROMs with the character set info on them labeled "ascii" the only data format it can handle is rs-274-d this controller is before my time as the only Gerber controller I have used was the GPC one (and that was 30 years ago)
The last time I saw something from Gerber Scientific it was a CNC engraving machine. The extra little power supply in the back is a plug-in module for powering the contactors. It was designed this way so it could be swapped very quickly by removing four screws.
The IMP-00 and IMP-16 chips on the green card together make up a multi-package 16-bit processor from National Semiconductor. The four IMP-00 Chips seem to be 4-bit bit-slice ALU chips so 4 are needed for 16-bit computation, and the IMP-16A chips seem to be microcode to provide the sequencing of what happens when the processor runs each instruction such as setting up the ALU operation to do Add, AND, OR, INCrement, etc. and where to fetch or store data, from memory or register. I found the relevant data sheets on the Internet Archive.
the green card has Solder mask on it, due to its smaller tollerances and closer solderingpoints. the others are probeblely older, and did not make use of this. That colour is therefore the color of the board itself, just the bear PCB.
I have had to do some manual editing of some older Gerber files to open them in a modern Gerber viewer application, and they are effectively nothing more than a text file with a series of commands and coordinates. Then, you have one file each per conductor layer, solder mask (per side), silkscreen, and drilling. So if you were able to work out a PC interface to the tape drive itself, the files would effectively read back as text files - depending on the the actual file format...
Aah, there's all the actual control logic. Lots of lovely ROMs too. Could you post high resolution photos of the boards so we can identify all the ICs? Btw, I invested in a Dataman 48Pro not too long ago and can dump those ROMs if you'd like.
The cards have a lot of logic ics it seens probably ttl or lsttl. A complete processor can be made with these ic”s. Also by using wire wrap as things progressed a tech could come out and Upgrade the processor right on site by adding some new ics to the card and wire wrapping them into the circuit! I saw this with my own eyes on a Data General system that was used at a travel agency!
What a find! From the green tag shown around 9:37 - The acronyms with lines over them are named signals (and the overline means inverted, so a logic "0" is active and a "1" is inactive). I can't tell if the top one is supposed to read OSOD (Output Serial Operation Done?) or OSAD. SLDY = Slave Ready (a perhipheral management signal)? OFD = Output FIFO something. Disable? OSCD = Output Shift Clock Done?? Just some guesses based on the domain and era.
@@koenlefever also each of those controllers were custom build to suit the customers requirements, so likely every one had slight differences. And at the low-ish volumes for these things wirewrap saved a shed-load of time and money
The "plotter" likely cut ruby acetate(?) that was then used as a mask on photosensitive PC boards. The board would then be developed an then etched. Big bed type "plotter".
Many of us: Gerber? That sounds like some PCB stuff, but nah, that's only a surname after all and this will be from an unrelated company under that name.
The first cards are logic cards with likely TTL chips given it was likely designed in the 70's or late 60's. The ROMs are interesting. The ones with white stickers may be EPROMs. Peel off a sticker and see if it has the window. Regardless on the green card there are two rows of 8 chips below white sticker ROMs. We don't get a good look at the chips but given that they are 2 rows of 8, these could be memory chips. Also on the green card, there is a larger chip to the side of the 6 ROM chips. We don't get a good look at that. I'd be curious if has any legible markings.
The non numbered board is likely the actual processor. The 6 chips at top would probably make the cpu, I see 4 eeproms in the lower right and under those probably the RAM chips.
Just a very educated 'guess' but that bank of chips at 9:50 are all EPROM chips and what look like TTL glue logic chips (I can make out some numbers like 73121 and 7474N which are a Monostable multivibrator with Schmitt-trigger and Flipflops respectively) - those stickers on the larger chips are covering up the UV Erase windows. now the Guess part of this statement is that they likely contain all the instructions like a giant lookup table, so that when the processor cards (?) at 10:57 and 11:25 can interpret the instructions on the tape and perform a lookup and convert them in to motor control signals. I'd hazard a guess that it was done this way because it requires less processing to simply bitmap instructions in ROM and perform direct operations based on what direction the input points to which vastly lowers the complexity of the overall design. Say the tape says 'step x 10' - it looks up the converted address for 'step' which contains in ROM a machine instruction or routine to address the motor, x being the motor channel and 10 being the number of steps to increment the motor. - This is just a simplification, not saying this is exactly how it works.
Wirewrap could be done by hand, but for production there were things that were basically X-Y plotters with a wrapping head. Often in the era these were run from paper tape. I think some were made by Gerber, but most were by other companies or even in-house designs at the major computer companies. Wirewrap backplanes were very easy to work with, and required zero maintanance.
Work Order and Part, as in, what part of the order. Perfectly reasonable for these fields to be blank on a production unit rather than some custom design.
Who did the heavy gauge wiring on that upper panel and didn't dress it? (The wire wrap stuff always looks like a madman's spaghetti dinner... I ain't worried about that.) Man... get out the waxed linen lacing cord and make that look original again. Get rid of all the blasted ty-raps if you want to make it look right. Those contactors were likely for small motor controls or other high current loads. They are capable of carrying (and interrupting) a whole lot of current, so their use makes a lot of sense. Those horrifically huge blue connectors were made by Amphenol (and TRW made a similar line that were gray). I'm surprised the cable shrouds are not present. With the shrouds and cam locks in place, the cable ends were watertight. What a find! I hope to see this thing start up and run!
I believe that some of the plotters projected light onto photosensitive media. There would be an XY motion system and a series of aperture shapes and sizes that would mask off the light.
My father was a machinist at the Gerber Factory in Vernon Connecticut in late 60s and early 70's. He said they made Fabric cutters and plotters. He's gone now, sorry I don't have much info. Thanks for investigating this equipment!
on card #2, the large chips with stickers are EPROMS, possibly with microcode for what looks like a TTL cpu
card #4 looks like a 16 bit ALU or some sort
'The Gerber format is an open, ASCII, vector format for printed circuit board designs. It is the de facto standard used by PCB industry software to describe the printed circuit board images.' This is still used today every time a PCB is designed/built.
The chips on the GNS-1-3 card that are labeled aren't microprocessors. MM5204 is listed as a 4096-bit EPROM. Either the other chips on that card are doing addressing logic for the EPROMs, or the card uses the EPROMS to do something similar to what a PAL chip does (another example of that kind of design would be the Apple ][ Disk controller card - except that the GNS-1-3 card would be a much more complex design; if it is in fact doing something like that).
Going by the comment on table 2 of my Gerber Book, you had a choice of the following data formats over all the various models of plot controllers
eia ascii / eia / bcd / ebdic / iso ascii
The 6x00 controllers could only handle eia ascii or ebdic, but there is a footnote to the table that you had to choose which of the two you wanted when ordering the controller
This explains all the various ROMs with the character set info on them labeled "ascii"
the only data format it can handle is rs-274-d
this controller is before my time as the only Gerber controller I have used was the GPC one (and that was 30 years ago)
RS-274-D IS the current Gerber standard for PCB fabrication, however, so it’s just as relevant now.
Very cool. Thanks for sharing.
That makes total sense. It's cool drawing a connection to this and modern EDA practices
The last time I saw something from Gerber Scientific it was a CNC engraving machine.
The extra little power supply in the back is a plug-in module for powering the contactors. It was designed this way so it could be swapped very quickly by removing four screws.
The IMP-00 and IMP-16 chips on the green card together make up a multi-package 16-bit processor from National Semiconductor. The four IMP-00 Chips seem to be 4-bit bit-slice ALU chips so 4 are needed for 16-bit computation, and the IMP-16A chips seem to be microcode to provide the sequencing of what happens when the processor runs each instruction such as setting up the ALU operation to do Add, AND, OR, INCrement, etc. and where to fetch or store data, from memory or register. I found the relevant data sheets on the Internet Archive.
the green card has Solder mask on it, due to its smaller tollerances and closer solderingpoints. the others are probeblely older, and did not make use of this. That colour is therefore the color of the board itself, just the bear PCB.
I have had to do some manual editing of some older Gerber files to open them in a modern Gerber viewer application, and they are effectively nothing more than a text file with a series of commands and coordinates. Then, you have one file each per conductor layer, solder mask (per side), silkscreen, and drilling. So if you were able to work out a PC interface to the tape drive itself, the files would effectively read back as text files - depending on the the actual file format...
Aah, there's all the actual control logic. Lots of lovely ROMs too. Could you post high resolution photos of the boards so we can identify all the ICs? Btw, I invested in a Dataman 48Pro not too long ago and can dump those ROMs if you'd like.
The cards have a lot of logic ics it seens probably ttl or lsttl. A complete processor can be made with these ic”s. Also by using wire wrap as things progressed a tech could come out and Upgrade the processor right on site by adding some new ics to the card and wire wrapping them into the circuit! I saw this with my own eyes on a Data General system that was used at a travel agency!
What a find! From the green tag shown around 9:37 - The acronyms with lines over them are named signals (and the overline means inverted, so a logic "0" is active and a "1" is inactive). I can't tell if the top one is supposed to read OSOD (Output Serial Operation Done?) or OSAD. SLDY = Slave Ready (a perhipheral management signal)? OFD = Output FIFO something. Disable? OSCD = Output Shift Clock Done?? Just some guesses based on the domain and era.
Is it ironic that a machine designed to make PCBs is built using wire wrap?
It is not ironic; it is a snapshot of an industry busy in the process of pulling itself up by its bootstraps.
@@koenlefever also each of those controllers were custom build to suit the customers requirements, so likely every one had slight differences. And at the low-ish volumes for these things wirewrap saved a shed-load of time and money
My logic is that's how they programmed it.
they made this so they could start making pcbs it all had the start somewhere
@@koenlefever That doesn't change the irony . . .
The "plotter" likely cut ruby acetate(?) that was then used as a mask on photosensitive PC boards. The board would then be developed an then etched. Big bed type "plotter".
On the gns 1-3 card, the eeproms are labeled L and U for upper and lower bits, so they act as pairs.
The eeproms would be individually burned, likely by hand. So the labels indicate what code is burned in each and that date.
Many of us: Gerber? That sounds like some PCB stuff, but nah, that's only a surname after all and this will be from an unrelated company under that name.
Crazy how much history is in this machine we are still using today!
The first cards are logic cards with likely TTL chips given it was likely designed in the 70's or late 60's. The ROMs are interesting. The ones with white stickers may be EPROMs. Peel off a sticker and see if it has the window. Regardless on the green card there are two rows of 8 chips below white sticker ROMs. We don't get a good look at the chips but given that they are 2 rows of 8, these could be memory chips. Also on the green card, there is a larger chip to the side of the 6 ROM chips. We don't get a good look at that. I'd be curious if has any legible markings.
The non numbered board is likely the actual processor. The 6 chips at top would probably make the cpu, I see 4 eeproms in the lower right and under those probably the RAM chips.
So what you have there is an early CAD/CAM interface.
As soon as you said the printed circuit board bit
Gerber made products for the sign industry up until today. Although I'm pretty sure Gerber vinyl is now just rebranded 3M
Just a very educated 'guess' but that bank of chips at 9:50 are all EPROM chips and what look like TTL glue logic chips (I can make out some numbers like 73121 and 7474N which are a Monostable multivibrator with Schmitt-trigger and Flipflops respectively) - those stickers on the larger chips are covering up the UV Erase windows. now the Guess part of this statement is that they likely contain all the instructions like a giant lookup table, so that when the processor cards (?) at 10:57 and 11:25 can interpret the instructions on the tape and perform a lookup and convert them in to motor control signals.
I'd hazard a guess that it was done this way because it requires less processing to simply bitmap instructions in ROM and perform direct operations based on what direction the input points to which vastly lowers the complexity of the overall design.
Say the tape says 'step x 10' - it looks up the converted address for 'step' which contains in ROM a machine instruction or routine to address the motor, x being the motor channel and 10 being the number of steps to increment the motor. - This is just a simplification, not saying this is exactly how it works.
skunkworks' u2 program?
That wire wrap looks like it was a nightmare to assemble and worse to diagnose problems
Wirewrap could be done by hand, but for production there were things that were basically X-Y plotters with a wrapping head. Often in the era these were run from paper tape. I think some were made by Gerber, but most were by other companies or even in-house designs at the major computer companies. Wirewrap backplanes were very easy to work with, and required zero maintanance.
... so that's why "gerbers" are called "gerbers"... woah! Perhaps the "g" in "G. code" also stands for "Gerber"????
My understanding is...g-code is shortened from g&m code. G meaning geometry and M meaning misc.
@@countryguy828 Meh! I was looking for simple answers to complex questions... not surprised I was wrong.
On the label, W.O. number would be Work Order number. Not sure about PT number
Work Order and Part, as in, what part of the order. Perfectly reasonable for these fields to be blank on a production unit rather than some custom design.
30amps at 240volts that thing would produce some amount of heat.
gbr is used by my laser etcher
That's awesome we are still using a format that's been around that long!
Who did the heavy gauge wiring on that upper panel and didn't dress it? (The wire wrap stuff always looks like a madman's spaghetti dinner... I ain't worried about that.) Man... get out the waxed linen lacing cord and make that look original again. Get rid of all the blasted ty-raps if you want to make it look right. Those contactors were likely for small motor controls or other high current loads. They are capable of carrying (and interrupting) a whole lot of current, so their use makes a lot of sense. Those horrifically huge blue connectors were made by Amphenol (and TRW made a similar line that were gray). I'm surprised the cable shrouds are not present. With the shrouds and cam locks in place, the cable ends were watertight.
What a find! I hope to see this thing start up and run!