GOOGLE AI CONTINUED TO SPEW INCORRECT INFORMATION
I asked Google AI to tell me the connector size for the head cable connection to the J10 PCB in the Diablo Disk drive. It responded authoritatively and confidently but embarrassingly incompetently. It was describing pin connections that made no sense.
I called the software on one of the more egregious errors, claiming that each head had two erase coil windings. It apologized, said I was correct and updated its advice. Still woefully wrong.
It asserted that the connection was two 6 pin connectors, one above the other. It still claimed that two leads came out from the erase coil and three leads for the read/write coil (two ends and a center tap Select line). Here is a picture of the cable from each head - only four pins just as is shown on the Diablo schematics. The erase coil has one end tied to the same Select line that is the center tap of the read/write coil, but the AI didn't grasp that.
It said the connectors were standard 2.54mm pin headers. It gave pin numbers that don't match the Diablo schematics. The AI pinout had no ground connection and no mention of shields.
I went to the workshop and examined the connector and the J10 PCB. I did some signal tracing as well. The connector is a standard 14 pin DIP socket. Not 6 on the top and 6 on the bottom. Further, it has four pins that grounding the shields of four tiny coaxial cables that carry the four head lines.
I have tried using Google AI for some coding assists. It did a decent job giving me sample code to do new user interface tasks and using unfamiliar Python libraries. It seems to work adequately in that domain. However, for researching vintage computing hardware it emits AI slop.
CORRECT PINOUT USED FOR MY DESIGN
I now know the pin assignments, which do match the pin numbers from the Diablo schematic of board J10. If you were to number a DIP socket in reverse, that is start with 1 at the top left, go to the right across the top row to 7, continue below with 8 and go back to the left bottom as pin 14, then the numbers match perfectly.
Pins 1, 2, 13 and 14 are connected to the board ground. Each pin on the connector has the shield of one of the four signal cables, two per head. Since there is no connection on the other end, there is no continuity to observe here but the PCB shows the pins tied to the ground plane.
Pins 13 and 12 do nothing, they are unassigned. That leaves four pins each for the upper and lower heads. Each head has a connection to the common Select that ties the center tap of the read/write coil with one side of the erase coil. Each head has a line from the other side of the erase coil. There are two more lines, tied to the two ends of the center tapped read/write coil.
4 - one side of the read/write coil that feeds Head Bus A
5 - the other side of the read/write coil, feeding Head Bus B
6 - Select, the common point
7 - the other end of the erase coil
8 - the other end of the erase coil on the second head
9 - Select for the second head
10 - one side of the read/write coil on second head, feeding Head Bus B
11 - other side of read/write coil, feeding Head Bus A
DESIGNED A SMALL PCB WITH A DIP 14 SOCKET AND DIP 14 MALE PINS
I whipped up a PCB that has pins to plug into the DIP-14 socket on the J10 PCB. It is marked with an up arrow much like the cable connector. Above that, it mounts a DIP-14 socket into which the head cable connector will plug.
The erase coil of one head is connected to Head Bus A. The erase coil of the second head is connected to Head Bus B. The two Select lines for the heads are connected to their normal places. The other ends of the read/write coils are unconnected since this adapter uses the erase coil for alignment signal capture.
SENT TO JLCPCB FAB FOR QUICK TURNAROUND
I expect the PCB to be built in two days and shipped back to me in about a week total time. It was fastest just to ship this off to JLCPCB who I had been using, in spite of the kind sponsorship from PCBWay.com since it adds time to arrange for them to pay for each sponsored PCB.
FIY this is what copilot says. The google KI is only using a fraction of the tokens that other models use.
ReplyDeletewhat is the connector size for the head cable connection to the J10 PCB in a Diablo Disk drive ?
Copilot said:
Copilot
For the Diablo (Series 30 / 31 / 33 / 40) disk drives, the head cable connection to the J10 PCB is not a small header or IDC style connector as one might initially assume—it is quite specific and a bit unusual:
✅ Actual connector type at J10
14‑pin dual‑in‑line (DIP) style socket
Essentially the same physical format as a 14‑pin IC socket (2×7 pins, 0.1″ / 2.54 mm pitch)
The mating cable side uses a corresponding 14‑pin plug/connector assembly
➡️ In other words:
Standard 0.1″ (2.54 mm) pitch, DIP‑style 14‑pin connector footprint
✅ Supporting details
The head assembly cable itself carries:
4 signal lines (two per head, via coax cores)
4 shield connections (each coax shield grounded at the PCB)
Several pins are tied to ground for shielding
A few pins are unused depending on option configuration
A direct physical inspection report of a Diablo drive confirms:
“The connector is a standard 14 pin DIP socket … with pins for grounding the shields of four tiny coaxial cables.”
blogspot
And the official parts documentation also lists:
“Connector, 14‑pin” / mating plug/socket for the head wiring harness
wang2200
✅ Practical implications
If you're recreating or repairing the cable/interface:
You can treat it mechanically like:
14‑pin IC socket spacing (0.3″ row spacing, 2.54 mm pitch)
Electrically and mechanically, however:
It is not just a logic connector—coax shielding and analog signal integrity matter
The cable uses miniature coax pairs, not ribbon wire
✅ Quick summary
Property Value
Connector type DIP-style socket
Pin count 14 pins (2×7)
Pitch 2.54 mm (0.1″)
Typical part style IC socket / discrete mate
Special aspect Coax signal + shield grounding
If you want, I can map the exact pinout (per head: RW coil, erase, common/select, shields) or help you design a modern adapter PCB—this connector is actually very workable for that.
following: Show the full pinout diagram
ReplyDeleteBelow is a reconstructed full pinout diagram for the Diablo J10 head connector (14‑pin DIP style), based on schematic consistency + verified physical tracing of the actual board and cable.
🔌 Diablo Disk Drive – J10 Head Connector Pinout
📐 Orientation
Top view of J10 PCB socket (as mounted on board):
┌───────────────────────┐
│ J10 PCB │
│ │
│ 1 2 3 4 5 6 7
│ o o o o o o o
│
│ o o o o o o o
│ 14 13 12 11 10 9 8
│
└───────────────────────┘
⚠️ Important:
Numbering here follows the actual Diablo schematic convention (which is “mirrored” vs normal IC numbering, as noted from tracing). [rescue1130...ogspot.com]
🧠 Functional grouping
Upper head (Head 0) → 4 active signals
Lower head (Head 1) → 4 active signals
4 shield grounds (coax shields)
2 unused pins
Each head uses:
Read/Write coil (center-tapped)
Erase coil (shares common/select node)
📊 Full pinout table
Pin Function
1Shield ground (coax)
2Shield ground (coax)
3Upper head – RW coil end A
4Upper head – RW coil end B
5Upper head – COMMON (RW center tap + erase)
6Upper head – ERASE coil return
7Unused
8Lower head – ERASE coil return
9Lower head – COMMON (RW center tap + erase)
10Lower head – RW coil end B
11Lower head – RW coil end A
12Unused
13Shield ground (coax)
14Shield ground (coax)
🔍 Key electrical insight
Each head has exactly 3 functional nodes:
RW A
RW B
COMMON (center tap + one side of erase coil)
The erase coil’s other end is the “erase return” pin
This matches real hardware:
4 wires per head
4 separate coax shields grounded at PCB only [rescue1130...ogspot.com]
🧵 Cable structure (important for reproduction)
Each head cable bundle contains:
2 × coax for RW differential
1 × coax for erase/common node
1 × dedicated erase return conductor
Shields tied to J10 ground pins (not looped at head)
⚠️ Practical notes (from real teardown experience)
Ground pins are interleaved for signal integrity
Shields are not optional → necessary for analog read channel
Two pins are intentionally no-connect
Mechanical mating is DIP‑14 (2.54 mm pitch) but:
Electrically closer to RF harness than logic cable
✅ Quick visual grouping
[Upper Head] [Lower Head]
3 ─ RW A 11 ─ RW A
4 ─ RW B 10 ─ RW B
5 ─ COMMON 9 ─ COMMON
6 ─ ERASE 8 ─ ERASE
[Shields]
1, 2, 13, 14 → GND
[Unused]
7, 12
If you want next step
I can:
Draw a schematic-level equivalent of the head coils
Help you build a modern adapter (SMA breakout / instrumentation amp interface)
Or map this to actual Diablo 31/44 schematics (signal names like HSEL, ERS, etc.)
Just tell me 👍
That is very close to what I discovered on my board, however the unused pins are 3 and 12. The upper head signals I found are on 4, 5, 6 7 rather than 3, 4, 5, 6.
DeleteThis matches the schematics from the Diablo documentation as well.
The physical details are exactly as you described.
The adapter PCB is already 1/3 through manufacturing.
I only need this to do the head alignment in the standard method, where the adapter connects the erase coil to the read preamplifiers, taking advantage of the wide spacing of the erase coil poles to give very clear nulls on the scope. For normal use, including archiving, no board is needed.
Because the Copilot responses were considerably better than Google AI, I gave Copilot a try and asked some questions about the format of data on a 2315 cartridge used with an IBM 1130.
ReplyDeleteI had to iterate about six times correcting significant errors before what it generated was correct. Each time it would tell me I was absolutely correct and fix the item I mentioned, only to fail on a further point.
For example, it claimed that the 2315 sector format had no ECC. When I pointed out the four check bits appended to each 16 bit word, it then claimed this was a Hamming code that could correct signal bit and detect double bit errors - again, not correct.
After I covered how the check bits worked, it went on to state that the check bit scheme used by IBM was derived from the Diablo 30/31, although that has the historical sequence reversed.
It stated that a Diablo drive could not read or write compatibly with an IBM 1130. I then had to educate it on the use by third party manufacturers with custom developed controller to use a Standard density 720KHz Diablo drive as additional drives on 1130 system.
I still see these AI products as generating information with bold confidence that is not correct, then gracefully accepting each correction only to blunder on again. A human in the same situation would couch their statements as "I believe" or "I don't have full details, however I assume". If the AI answers projected a confidence level (other than 100%) I would be more willing to rely on them.
Hmmm, maybe my responses were better because I am using copilot pro from work. The license costs ~250€ per user per month. For this amount of money it better be good. If you want, send me some prompts and we can see if copilot pro does anything better.
ReplyDeleteAs for the hallucinations you are right. Every time I am using it and for a long time everything works great and I begin to build confidence, I hit a point where AI is telling absolute nonsense but with the same 100% confidence as always. It makes up commands that never existed or part numbers that are wrong. On programming tasks, for Fortran IV for example it got wrong how multi line statements work. I told it to look at another multi line statement to correct its own, which worked - only for AI to forget it some iterations later and do the exact same error again.
As said, if you want to know if copilot pro is better, send a prompt - I am curious…