Cromemco Cyclops 88 CCC board set reproduction

[daver2]

I am allowed a cup of coffee!

The interconnection diagram is on page 7 of the Cromemco Cyclops 88 CCC manual.

This diagram shows "pin for pin" interconnection, with the exception of pins 5, 7, and 9 which are "no connect".

Pin 5 is "optional reset".
Pin 7 is "video -".
Pin 9 is "optional clock".

Dave

 

[Hugo Holden]

I am allowed a cup of coffee!

The interconnection diagram is on page 7 of the Cromemco Cyclops 88 CCC manual.

This diagram shows "pin for pin" interconnection, with the exception of pins 5, 7, and 9 which are "no connect".

Pin 5 is "optional reset".
Pin 7 is "video -".
Pin 9 is "optional clock".

Dave

I am allowed a cup of coffee!

The interconnection diagram is on page 7 of the Cromemco Cyclops 88 CCC manual.

This diagram shows "pin for pin" interconnection, with the exception of pins 5, 7, and 9 which are "no connect".

Pin 5 is "optional reset".
Pin 7 is "video -".
Pin 9 is "optional clock".

Dave

Thanks Dave, I figured it should likely be a pin number to pin number relation from the D9 connector on the camera to the 16 pin connection on board 1 of the 88-CCC, sounds plausible as it shows on page 7, but the thing I cannot get to grips with is the connection from pin 2 of the 16 pin connector of board 1, (called "composite clock" in the document) to pin 2 of the Cyclop's D9 connector. Looking at pin 2 of the 16 pin connector on the board 1 schematic, all it has is a 560R resistor to ground and connects nowhere else in the schematic. That is what threw me off. Checking the actual pcb, it matches the schematic too, all that is connected to pin 2 of the 16 pin connector, is that 560 Ohm resistor. There must be a missing link on the board 1 pcb and on the schematic to pin 2 of the 16 pin connector to somewhere in the schematic ?

 

[daver2]

I see what you mean. I agree with you. This cannot work!

Dave

 

[hmb]

Thanks Dave, I figured it should likely be a pin number to pin number relation from the D9 connector on the camera to the 16 pin connection on board 1 of the 88-CCC, sounds plausible as it shows on page 7, but the thing I cannot get to grips with is the connection from pin 2 of the 16 pin connector of board 1, (called "composite clock" in the document) to pin 2 of the Cyclop's D9 connector. Looking at pin 2 of the 16 pin connector on the board 1 schematic, all it has is a 560R resistor to ground and connects nowhere else in the schematic. That is what threw me off. Checking the actual pcb, it matches the schematic too, all that is connected to pin 2 of the 16 pin connector, is that 560 Ohm resistor. There must be a missing link on the board 1 pcb and on the schematic to pin 2 of the 16 pin connector to somewhere in the schematic ?

There must be an error in the schematic. I've looked closely at photos of the original board set shown here:

deramp.com

deramp.com

....and it looks like pin 2 of the camera connector connects through 560 ohm resistor (R2) not to ground, but instead "most likely" to pin 8 on the 7420 located just below the camera connector. The pin 8 output of the 7420 seems to be a composite of several signals generated on Board 1 including the camera clock signal itself. I will try to contact Bill Sudbrink (he has the original board set) to confirm.

 

[Hugo Holden]

I see what you mean. I agree with you. This cannot work!

Dave

I think there is an error in the schematic. I've looked closely at photos of the original board set shown here:

deramp.com

deramp.com

....and it looks like pin 2 of the camera connector connects through 560 ohm resistor (R2) not to ground, but instead "most likely" to pin 8 on the 7420 located just below the camera connector. The pin 8 output of the 7420 seems to be a composite of several signals generated on Board 1 including the camera clock signal itself. I will try to contact Bill Sudbrink (he has the original board set) to confirm.

I agree. It will be pin 8 of the 7420.

This is the thing that can happen where boards are reproduced and auto-routed from a schematic alone, because often there are errors in a schematic. It is better to make the track layout manually exactly match the original vintage board, when there are photos of it available, and have every visible trace accounted for and cross referenced on the schematic. There could be more to find.

Also on the reproduction board it needs the 560 Ohm pullup on board 2 added, or it could be added to board 1, to pull up the "T" Connection, pin 3 of the 16 pin connector, because that is an open collector output from the video output transistor on the ACC3 board in the camera.Though it could work without it as the pullup current would come from the TTL inputs.

It s not the only trap though that a circuit router, human or otherwise can fall into that design rules in autorouters can miss, the other one I call an "H-C error" after the now famous error made by Howard Cantin (not his fault), the designer of Atari's Arcade Pong pcb in 1972 and also the early Apple pcb's. He was given a defective schematic for the Pong board where simply two pins on a multi-input gate arrangement had their pin numbers switched on the diagram. When he routed the tracks he faithfully followed that. It switched the least significant bits of the two player's paddle data causing interactive effects between the player's paddle interactions with the ball and caused spooky effects. It is called the Ghost in the Machine Bug (I named it that).

I was looking around on the schematics for anything that could have caused an H-C error, I found one so far that could have but did not, on board 2; IC12 a 7493 on the schematic has a pin numbering error where pin 3 got labelled as pin 1, as is the clock pin 1, luckily it must have been picked up because the actual IC is wired properly.

On the actual board 1 Deramp photos, there appears to be a mod on the back of the board on U45, I'm not sure if that was to make a mod to a defective pcb to make it conform to the proper schematic, or a mod because of a schematic error that created a pcb error, or a mod to a normal system for some other reason.

 

[hmb]

On the actual board 1 Deramp photos, there appears to be a mod on the back of the board on U45, I'm not sure if that was to make a mod to a defective pcb to make it conform to the proper schematic, or a mod because of a schematic error that created a pcb error, or a mod to a normal system for some other reason.

That mod pulls the input for the A15 bus driver high. it must have been done to meet the specific requirements of the board's original owner.

 

[Hugo Holden]

The 16 pin connector at the board 1 end is best suited to a 16 way ribbon cable & plug. There are 7 conductors there that cannot be used, so simply at the camera plug end, I cut them off but folded them over and placed heatshrink over each one and grouped them together with a larger piece of heatshrink sleeve, so they cannot short to anything.

I am using an 8 foot 16 way ribbon cable.

I think Cromemco did the right thing with the low Z terminations for the composite clock signal in the differential arrangement to feed the "Osclloscope Cylclops" because as they pointed out, it would mean a long camera cable would work. For example, an 8 foot length of Ribbon cable, the capacitiance of one conductor to the others, can be as high as 170pF, that is not insignificant.

There are more other suitable connectors than I had on hand at the camera end, but I decided to use a male connector on the camea. The reason is that I have a variant of it where I can plug the wires from the camera boards individually onto its pins. So I can modify connections or remove the boards from the camera, without having to unsolder the 9 pin connector.

In any case I have attached a wiring diagram I used. It seemed absurd to me that as there were 3 additional n/c pins on the 16 pin socket of board 1, then to just rely on just the one conductor in the 16 way ribbon cable for the earth seemed far from adequate. So at the board 1 end I simply connected the pins 16, 9 & 5 to ground with link wires and used 3 addititional conductors in the cable to feed the unused pins 7, 9 & 5 on the 9 pin Camera onnector so as to get a better earth. Otherwise the connections conform to page 7 of the 88-CCC document. It seemed a shame to waste those conductors when they were there in the cable anyway and could be deployed for better earthing and I don't intend yet to make the O'scope cyclops. The attached diagram may help others with the cable construction.

 

[hmb]

The 16 pin connector at the board 1 end is best suited to a 16 way ribbon cable & plug. There are 7 conductors there that cannot be used, so simply at the camera plug end, I cut them off but folded them over and placed heatshrink over each one and grouped them together with a larger piece of heatshrink sleeve, so they cannot short to anything.

I am using an 8 foot 16 way ribbon cable.

I think Cromemco did the right thing with the low Z terminations for the composite clock signal in the differential arrangement to feed the "Osclloscope Cylclops" because as they pointed out, it would mean a long camera cable would work. For example, an 8 foot length of Ribbon cable, the capacitiance of one conductor to the others, can be as high as 170pF, that is not insignificant.

There are more other suitable connectors than I had on hand at the camera end, but I decided to use a male connector on the camea. The reason is that I have a variant of it where I can plug the wires from the camera boards individually onto its pins. So I can modify connections or remove the boards from the camera, without having to unsolder the 9 pin connector.

In any case I have attached a wiring diagram I used. It seemed absurd to me that as there were 3 additional n/c pins on the 16 pin socket of board 1, then to just rely on just the one conductor in the 16 way ribbon cable for the earth seemed far from adequate. So at the board 1 end I simply connected the pins 16, 9 & 5 to ground with link wires and used 3 addititional conductors in the cable to feed the unused pins 7, 9 & 5 on the 9 pin Camera onnector so as to get a better earth. Otherwise the connections conform to page 7 of the 88-CCC document. It seemed a shame to waste those conductors when they were there in the cable anyway and could be deployed for better earthing and I don't intend yet to make the O'scope cyclops. The attached diagram may help others with the cable construction.

Two comments:

Your revised schematic for the camera connector on board 1 shows the composite clock signal going to pin 2, with a 560 ohm load resistor to ground. The way I think this is wired on the original board is with the composite clock signal signal connected in series with the 560 ohm resistor which is connected to pin 2 of the camera connector, ie no termination to ground.

My plan is to run a ribbon cable from the board 1 camera connector to a DE9 connector on the computer rear panel. The DE9 connector on the back panel would connect to the DE9 connector on the Cyclops camera with a standard straight-through 9 conductor cable, which I think would be more robust than using a long ribbon cable.

 

[Hugo Holden]

Two comments:

Your revised schematic for the camera connector on board 1 shows the composite clock signal going to pin 2, with a 560 ohm load resistor to ground. The way I think this is wired on the original board is with the composite clock signal signal connected in series with the 560 ohm resistor which is connected to pin 2 of the camera connector, ie no termination to ground.

My plan is to run a ribbon cable from the board 1 camera connector to a DE9 connector on the computer rear panel. The DE9 connector on the back panel would connect to the DE9 connector on the Cyclops camera with a standard straight-through 9 conductor cable, which I think would be more robust than using a long ribbon cable.

Yes I have to agree in this case, thank you for spotting it, from the original board photos it does look like the 560R resistor is in series. with pin 2, despite the Dot on the schematic showing it grounded, so that makes two schematic errors, the Dot and a missing link.

On the other issue with the connector, I agree with you about having the D9 connector on the rear panel for your computer, unfortunately I cannot do that without damaging the rear panel on the SOL-20, the best I can do is run the ribbon cable under the gap in the lower rear cover/panel, like I do with the Disk drive ribbon cables.

I have attached the corrected diagram, it did mean I had to cut a trackm I generally use a small drill rathe than a knife, its less messy.

 

[Hugo Holden]

I have been slowly assembling the camera head. I cut out a stainless steel mirror plate to help reflect the light from the LED's back onto the sensor chip, probably better than Alfoil. I used the single connectors to plug onto the D9 connector's pins, because that way I can easily take the 3 boards out of the housing for diagnostics servicing etc, without having to unsolder the wires from the connector Still waiting for a number of parts, including the Tap to make the thread for the lens.

 

[Hugo Holden]

Finally the Tap arrived, I managed a successful D mount thread. I did not have a Tap handle nearly big enough for it, so I simply put it in a large Chuck to hold onto it. It was not wonderfully easy though, the Tap had a fairly small tapered section which is never helpful and it took a moderate force to get it started and careful alignment to keep it perpendicular to the surface.

 

[nullvalue]

Hi @nullvalue:

I have been assembling more things on board 2 and was about to install the port jumpers according to the Cromemco manual. But I discovered on checking the jumper arrangements, on your board, they are completely diffierent to the Cromemco arrangement. I had to study that setup for a while to figure out how to set them.

Can you inspect the attached photo and see if you agree, that the jumpers are set for port A = 10h, port B = 11h and port C = 12h, just in case I have fouled it up.

Yes I was aware of this and it left it as one of those bridges we'd have to cross! Unfortunately I was unable to match the layout to the original since I don't have the foils. I think I'd prefer @daver2 look the ports over the port configuration to confirm.

Finally the Tap arrived, I managed a successful D mount thread. I did not have a Tap handle nearly big enough for it, so I simply put it in a large Chuck to hold onto it. It was not wonderfully easy though, the Tap had a fairly small tapered section which is never helpful and it took a moderate force to get it started and careful alignment to keep it perpendicular to the surface.

That looks like it turned out great!

 

[Hugo Holden]

That looks like it turned out great!

One of the tricks s to make the hole perfectly on center. I started out with a 1.5mm Pilot hole and the most accurate center marking I could create ,and went up in 1mm steps as it approached 14mm then a final step to 15mm.

You will find as you go step by step, for a number of reasons if you are doing it in a drill press and not clamped up in a milling machine, the hole center will wander. So what I did to correct for this was between each 1mm step in four places as it aproached 14mm, I measured the hole perimeter's edge to the edge of the housing with a dial micrometer, then I used a half round file to correct the centering error before the next drill step.

This way the final step was bang on center. If you use some sort of hole cutter or step drill, the center will wander off much more than you might think, so its better not to "hope" it will be ok but check it along the way with each small increment in the hole size.

 

[Hugo Holden]

Yes I was aware of this and it left it as one of those bridges we'd have to cross! Unfortunately I was unable to match the layout to the original since I don't have the foils. I think I'd prefer @daver2 look the ports over the port configuration to confirm.

Do you have an image of the foils on this board, so that Dave can inspect the wiring from the jumper area to the IC's, to see if what I posted makes sense ? Or does Dave have a set of your boards to check physically with a meter ?

 

[daver2]

I have neither.

But if someone could point me to where they are...

Dave

 

[hmb]

I have neither.

But if someone could point me to where they are...

Dave

All files related to this board can be found here:

GitHub - nullvalue0/Cromemco-Cyclops-88-CCC-Reproduction: Kicad reproduction of the 2-board set "88 CCC", which are S-100 boards meant for use with the Cromemco Cyclops digital camera and Dazzler video board

Kicad reproduction of the 2-board set "88 CCC", which are S-100 boards meant for use with the Cromemco Cyclops digital camera and Dazzler video board - nullvalue0/Cromemco-Cyclops-88-CCC-...

github.com

 

[daver2]

Cheers.

Dave

 

[Hugo Holden]

Dave: I have put your comments into the source file to suit the SOL/8080 CPU.

I decided to keep the initial attempt as simple as possible to minimise the variables since the hardware is not proven yet. Once it is working, then put in the refinements and keyboard interactions later.

Can you check it to see if I have messed anything up ?

It is good to have this all commented, Cromemco were a little remiss not doing it.

 

[Hugo Holden]

I made an error, in that I had changed the PORT A output from 80H ot 0A0H (to light the bias lights) but I typed in 0AH by mistake into the file.

I attempted assembly, it completes but there is something up with port A (see attached) I had wondered if it was the close association of and input and output instructions for the same port, so I shuffled those apart, but that made no difference.

 

[Hugo Holden]

Figured out what I did wrong, I used INP instead of IN. It now assembles without error.
In the case of the SOL-20, there are very limited card slots. The 2 Dazzler boards and the two 88-CCC boards use 4 slots, and one for my memory board, using up the whole 5 slots (unless I use the awkward vertical one)
So I am going to have to remove my N* disk drive card to free up a slot, not use CP/M for anything and convert the .Hex file into a .ENT file and serial port it to the SOL.