• Please review our updated Terms and Rules here

Looking for Jerrold/General Instrument documentation and software

philpem

Experienced Member
Joined
Oct 3, 2009
Messages
61
Location
West Yorkshire, England
Hi all,

I'm looking for any software, manuals, photos of hardware - anything really - that relates to the Jerrold Communications / General Instrument analog cable TV headend equipment. (this is on topic, I promise!)

Many of their headend "addressable controllers" were built around DEC hardware -- the AH series units used a PDP-11, the Terminal Configurator ran on a PC, the ACC-4000 ran on a DEC Prioris server running SCO UNIX.

Things I'm looking for especially --

- The software from the ACC-4000 Addressable Controller (Prioris based), AI-0/AI-O or AH-4/AH-4E (PDP-11/73 based) controllers, or Terminal Configurator (PC based).
- Backup tapes from a running system (may be TK50, or DAT/DDS)
- Terminal Configurator, Message Editor (ME-1000) or "OSD Edit" software
- Any documentation
- Photos, or other details of the I/O cards (either the PC one - which may have been called ANIC - or the SCX11, SCX11E, SCX11M or SRT11 cards used in the PDP systems).

I'm trying to build an analog cable TV headend from scratch, as a bit of a preservation and "to see if I can" project.
So far I've managed to modulate a couple of channels and get a cable box to tune to them, but my two boxes have different frequency maps, and I need some way of sending an "Input Frequency Map" or channel name table to them. I've figured out part of the channel mapping process, but not the frequency mapping.

I'm hoping that someone might have inherited a bunch of backup tapes, hardware or media from a cable TV company who was migrating to digital....

Cheers
Phil
 
Since the equipment you are looking for is no longer in production, it may be difficult to find the documentation and software you need. You may want to try searching online forums, websites dedicated to vintage technology, or reaching out to former employees of Jerrold Communications / General Instrument who may have access to these resources.
 
Thanks Sophia... that was the point of this post. I'm hoping that someone who worked at Jerrold or with some Jerrold equipment stashed away comes across it and gets in contact with me to help fill in the gaps in what I've figured out (which is mostly on my website, https://www.philpem.me.uk )
 
This really came out of left field today, but there is an old General Instruments ANIC-A for sale on eBay right now that seems to have some documentation... It is very expensive, but you might be able to haggle with them.

Unfortunately, I cannot help much beyond that.. I do work in cable, and was around during the Analog-Digital transition days, but all of that ACC-3000/4000 stuff was recycled many *many* years ago. Now if you were looking for early DAC related stuff, I have that in spades.
 
I've spoken to the seller - they're unwilling to move on price. I asked about getting a Xerox of the manual and they asked for $400.

I'm not surprised the ACC stuff was recycled years ago. I'm really just hoping someone has a software disk lying around somewhere. I found a photo of what looks like the ANIC ISA card on Segaretro the other day, billed as a "universal cable interface".

The early DAC stuff might be helpful for figuring out the older analog stuff - can you please drop me a private message or an email with some info on what you have?

Thanks
 
Hi all,

I'm looking for any software, manuals, photos of hardware - anything really - that relates to the Jerrold Communications / General Instrument analog cable TV headend equipment. (this is on topic, I promise!)

Many of their headend "addressable controllers" were built around DEC hardware -- the AH series units used a PDP-11, the Terminal Configurator ran on a PC, the ACC-4000 ran on a DEC Prioris server running SCO UNIX.

Things I'm looking for especially --

- The software from the ACC-4000 Addressable Controller (Prioris based), AI-0/AI-O or AH-4/AH-4E (PDP-11/73 based) controllers, or Terminal Configurator (PC based).
- Backup tapes from a running system (may be TK50, or DAT/DDS)
- Terminal Configurator, Message Editor (ME-1000) or "OSD Edit" software
- Any documentation
- Photos, or other details of the I/O cards (either the PC one - which may have been called ANIC - or the SCX11, SCX11E, SCX11M or SRT11 cards used in the PDP systems).

I'm trying to build an analog cable TV headend from scratch, as a bit of a preservation and "to see if I can" project.
So far I've managed to modulate a couple of channels and get a cable box to tune to them, but my two boxes have different frequency maps, and I need some way of sending an "Input Frequency Map" or channel name table to them. I've figured out part of the channel mapping process, but not the frequency mapping.

I'm hoping that someone might have inherited a bunch of backup tapes, hardware or media from a cable TV company who was migrating to digital....

Cheers
Phil
Hi Phil,

I came across your post while doing a Google search... I may have something you might be interested in:
I've got an old DEC 486 machine that has some General Instrument hardware in it, specifically this odd controller that was sitting in one of the 5.25" drive bays:

It's labelled "ANIC" by General Instrument.

1692841956863.png 1692841978971.png 1692841995986.png

1692842015222.png

It was connected to the system via a 50-pin SCSI cable to a BusLogic BT-542B SCSI card and it has 2 F-type coax connectors/cables coming out of it and going to one of the rear slot covers of the machine. The system itself (very heavy) is loaded with RAM, 3 SCSI Hard Drives, SCSI Tape drive and 1 out of 2 486-DX50 CPU's.

The machine hasn't been powered-on, so I couldn't tell you what software it's running, etc.

-Chris
 
I was never expecting to see that ...! I think you have one of the ACC access controllers - an early one. Possibly an ACC-2000 or 3000, but it's hard to tell without booting it up (and knowing the passwords)

The ANIC is an "Asynchronous Network Interface Controller". It connects the PC to the cable network so it can send commands to the headend equipment (scramblers and modulators) and cable boxes.
I thought it was an ISA card, I never expected it to be a full embedded 286 PC with SCSI. The Xilinx chip isn't too much of a surprise, GI used those in the MVP (video scrambler).

Would you be able to dump the contents of the hard drives, and the ROMs on the ANIC -- and maybe take some photos of the full system?
I can provide a Dropbox upload link if you need it.

If you're willing to part with it, I'd be interested in buying the GI-specific hardware and drives - but I suspect the whole machine might be too expensive to ship from Canada to the UK.

Feel free to drop me a PM or email (my forum username at gmail.com works, or see https://www.philpem.me.uk/contact ) if that's easier for you.
 
Last edited:
I was never expecting to see that ...! I think you have one of the ACC access controllers - an early one. Possibly an ACC-2000 or 3000, but it's hard to tell without booting it up (and knowing the passwords)

The ANIC is an "Asynchronous Network Interface Controller". It connects the PC to the cable network so it can send commands to the headend equipment (scramblers and modulators) and cable boxes.
I thought it was an ISA card, I never expected it to be a full embedded 286 PC with SCSI. The Xilinx chip isn't too much of a surprise, GI used those in the MVP (video scrambler).

Would you be able to dump the contents of the hard drives, and the ROMs on the ANIC -- and maybe take some photos of the full system?
I can provide a Dropbox upload link if you need it.

If you're willing to part with it, I'd be interested in buying the GI-specific hardware and drives - but I suspect the whole machine might be too expensive to ship from Canada to the UK.

Feel free to drop me a PM or email (my forum username at gmail.com works, or see https://www.philpem.me.uk/contact ) if that's easier for you.
Interesting, I had a feeling it was some sort of head-end controller.

Yes, I'm willing to part with it. I can also image the three hard drives for you (not sure how well the hard drives would survive if I attempted to ship them).

The system itself is a DEC 'application DEC 400xp', loaded with RAM and a single 486DX-50 installed (dual-capable machine), Adaptec AHA-1740 SCSI controller and a DEC 4-port multiplex interface board. Unfortunately, the system is extremely heavy and there'd be no way to ship it safely without palletizing it.

-Chris
 
Digging deeper, I found a Year 2000 declaration which identifies it. It seems like it's an early ACC-4000 (called an ACC-4000/XP): http://web.archive.org/web/20010108.../year2k/html/product_compliance/ANS_plans.htm

By the software notes, it could have been updated as far as version 7 or 8, which means it can control almost everything except the CFT-2200 "advanced analog" boxes. Those are pretty cool because you could download apps onto their application processor. There was an EPG called Starsight for them, and a version of Prevue Guide.

The Buslogic BT-542B is (according to https://groups.google.com/g/comp.periphs.scsi/c/MXtee95dUzU ) compatible with the Adaptec 1542B. That seems legit, as they both use the same QNX4 driver: https://fsck.technology/software/QNX/QNX 4/QNX4/Soft/Free/os/hardware_support/scsi_support.html

I'm curious what that DEC 4-port multiplex card looks like. I guess it's something like a Rocketport card, four 16550 UARTs and line drivers.
 
What an odd thing! Had a dig around and found some old posts on classiccmp about it. Apparently under one of those labels is the text "Corollary Inc".
Digging for more on them turned up a few things:


Seems like it might be a Corollary 8X4AT terminal mux card. That gives 32 ports on a half-slot ISA card; each mini-DIN goes to a Corollary 8/TC terminal concentrator, each of which provides 8 terminal ports. Sounds like there's processing in the terminal concentrators too, so the card's probably kinda useless without one.

No idea where you'd get drivers or a terminal concentrator...
 
Good news! I got the ANIC working. It turned out to be battery leakage - there's a buffer chip next to the CMOS battery. When the battery leaked, it took out four tracks under the chip ... the buffer was for the RAM address pins. I fixed the tracks and put it back in the PC I built to run the headend software - sure enough, it came right up.
I've had a CATC SCSI analyset hooked up to the bus - and indeed, the ANIC is the initiator.

You can skip ahead if you don't care about how I figured out the buffer was the issue :)

To diagnose this mess, I partially reverse-engineered the boot ROM. One of the first things it does is a wordwise 5555/AAAA data RAM test. Absolutely useless for spotting bad address lines, but at this point I didn't know that was the problem...

After that I soldered a bunch of right-angle headers to a spare 27C512 EPROM and programmed it with a copy of the boot ROM. This was so I could hook a HP 16700A logic analyser up to the ROM - I set it to state mode, clocked from /CS and qualified by /OE. Hit run, booted the ANIC, and found it was CALLing the RAM test routine, running it (I couldn't see what the return value was), then it executed a RET and ran off into the weeds.
It was pretty clear that the RAM test was overwriting the stack - which was a bit of a hint that one or more of the upper RAM address lines was broken.

Given it's a PC with I put Ruud's diagnostic ROM in there - there's a serial port UART mapped as COM1 on the top board and Ruud's ROM can output checkpoint codes to the serial port. Sadly Ruud's original version fails (DMA test failure I think) and a later version failed because there's no video RAM (the later versions use this for the stack to pivot and test the system RAM). I was in the process of modifying this to run without VRAM when a friend convinced me to pull out the multimeter!

I pulled out the CHIPS SCATsx manual and the block diagram - the example design had a buffer on the RAM address lines. I followed the tracks back from the RAM sockets with the aid of a multimeter, and found they went to the aforementioned buffer chip... right next to the battery. Some of the tracks for the high-order address lines were fine up to the vias near the chip, but didn't get to the chip itself.

Sure enough, I desoldered the chip and found a bunch of verdigris and schmutz under the chip. Gave that a clean with a Dremel and a green rubber polishing pad (to remove the verdigris, rotten solder and solder resist), tinned the remains of the traces, and ran new traces with Roadrunner wire. After that I fitted an IC socket, did the requisite tests, and sure enough the board came back!

And that's the end of the repair story...

The whole thing is quite over-engineered... an entire embedded 12MHz 80286 PC-AT with a serial port (which is totally unused) and SCSI host controller, mated to a Xilinx XC3000 FPGA which is used to implement the biphase communications controller (a fancy UART). It makes more sense when you consider that the host PC was a DEC ApplicationDEC 400XP with a 50MHz 486 CPU -- the previous addressable controllers (the ACC-3000 and ACC-2000) were based around desktop PCs.
I guess that may be why they went with SCSI for the interface too -- the ANIC can spend its life requesting commands from the PC, and the job of telling it that nothing's available falls to the AHA-1540B HBA (in target mode). When there is data to send, the PC just needs to drop it in a buffer and let the HBA deal with the rest.
This goes even further - the ANIC can send "loops", which are commands addressed to a series of boxes in the system. All the PC has to do is specify the start and stop address and command, and the ANIC does the rest. It's all pretty wild.

I didn't bother tracking down a Corollary 8x4 mux and terminal concentrator - instead I picked up a Brainboxes Quad RS232 card and configured it on a spare IRQ. That needed some fiddling - it turns out the 3Com 3C509B Ethernet cards need a configuration/activation write to port 110h, which is the default for the Quad RS232... moving the serial port up to 140h got things working again. Thankfully all four serial ports can share an IRQ, or I'd have a problem. (the two motherboard serial ports can't be configured to share an IRQ - if you have COM1 on 3F8h and COM2 on 3E8h, both on IRQ4, neither port will work).

But it's all working now, and that's the important thing! Interactive Unix 4.1 is quite snappy when it's running from a BlueSCSI attached to an Adaptec PCI SCSI card. The 400MHz K6 processor and 96MB RAM probably helps!
 
I've done some more digging ... There's an ACC-4000 manual on Sega Retro which lists several newer models:
  • ACC-4000/PL Model 5500 - a Compaq Proliant 5500.
  • ACC-4000/PL Model 800 - a Compaq Proliant 800
  • ACC-4000/HX Model 6200 - Digital Prioris HX 6200
Here are some hints for identifying an ACC-4000:
  • Proliant 5500 or Proliant 800 units will be pretty much stock standard. One to three hard drives, Ethernet, and a tape drive (likely DDS). They won't have an ANIC installed - they use an external rackmount ANIC-A connected over Ethernet.
  • The Prioris HX 6200 variant may have an ANIC installed in a 5.25in drive bay, but this isn't guaranteed. More likely they'll be set up like the Proliants.
  • OS for either will be SCO UNIX, likely SCO Openserver 3. If you have a really old one it'll be Interactive Unix - though I think the V6.3 software I have is the latest for that platform.
  • If there's a mouse and keyboard plugged in, it should boot into X11 with a forms-based user interface in a Motif-style window. It looks a lot like a 3270 terminal.
  • If it drops you at a login prompt instead - the machine hostname will likely be some variant of "ACC4000". (this is hardcoded in the software). Try the login "acc4000" with password "ippv4000" (the defaults if they haven't been changed) - you should get a second password requester, or a shell. If not, try the same password for the "root" user and see if you get a shell. Username "jerrold" and password "rsx11m" is another one worth trying.
  • If it complains about Accell/Unify or Sybase licenses, set the system clock back to some time before 2020. The licensing in the UI still uses two-digit dates and thinks 2020 is 1920.
If anyone has a machine which behaves like this and can take an image of the hard drives - please get in touch. (If you're not a forum user and don't want to sign up, my email address is on https://philpem.me.uk/contact )

I'm also happy to share what I know about these systems and explain how to build up a basic system. The minimum GI kit you need is the software, the machine, and either an ANIC or ANIC-A so it can talk to the outside world.
After that you just need an RF signal generator or FM modulator to generate the side-channel data burst, a cheap TV modulator, and an RF (F-connector) splitter run in reverse (as a combiner) - and of course an old analog cable box.


I've been working on my machine this evening -- I'm having issues with the network stack crashing. Still debugging it, but I'm starting to suspect my 3C509 card might be iffy.
 
Back
Top