Getting a token visual pass [Highpass for 32MHz]

30 views
Skip to first unread message

Louis Ohland

unread,
Nov 15, 2021, 8:12:50 PM11/15/21
to
https://ardent-tool.com/NIC/IBM_F-Coupler.html

Folks, I have looked for the mythical "F-Connector" that enables one to
attach F-Connector video to an ICS Token-Ring MAU. I have never foundt
even -ONE-.

Can someone intelligent enough design a High pass filter, maybe 32MHz to
infinity, that will allow us to use an 8228 [or similar] to transmit
both Token-Ring -AND- video?

Tim Clarke banged out the serial control app for the Sony EVI-D70, might
as well get some use out of it.

Grant Taylor

unread,
Nov 15, 2021, 9:42:30 PM11/15/21
to
On 11/15/21 6:11 PM, Louis Ohland wrote:
> https://ardent-tool.com/NIC/IBM_F-Coupler.html

That ... is interesting.

> Folks, I have looked for the mythical "F-Connector" that enables one to
> attach F-Connector video to an ICS Token-Ring MAU. I have never foundt
> even -ONE-.

I don't think the RF goes /through/ the MAU. Instead it looks to me
like you put the F-Coupler on either side of -- what I think is called
-- the horizontal cable that runs from the wiring closet to the
workstation outlet.

Or said another way, the F-Coupler combines the broadband RF and the
baseband ICS Data over the horizontal cable. The RF still has it's own
distribution in parallel to the MAU(s).

The F-Coupler allows re-using the horizontal cables for distribution of
both baseband ICS data and broadband RF.

> Can someone intelligent enough design a High pass filter, maybe 32MHz to
> infinity, that will allow us to use an 8228 [or similar] to transmit
> both Token-Ring -AND- video?

I've not read the content of the patent (yet) -- US4885747.pdf available
from the Google Patent link you have. But it looks like figures 2 and 4
provide the schematic for the F-Coupler. The thing that I don't see on
the schematic is the value for the components.

The thing that I can't figure out is the wires. I'm used to Token Ring
being four wire. But the ICS transmitter & receiver seem to be two
wires for the baseband.

I'm also not sure what the difference is in figures 2 and 4.

There are probably a lot more details in the 6+ pages of text.



--
Grant. . . .
unix || die

Louis Ohland

unread,
Nov 16, 2021, 8:20:43 AM11/16/21
to
Stop wearing sunglasses at night...

The patent has a lot of detail on the chokes used, how to run the wires
through the six or so holes, a circuit schematic...

I've gone through it before. If you are curious, blue, I can extract the
circuit and component values. The patent for the "Improved IBM
F-Coupler" is the one you wandt.

Electrical connector US 5293298 A Improved IBM F-Coupler
https://www.google.com/patents/US5293298

https://ardent-tool.com/NIC/F-Coupler_Fig1.jpg

Shows "RF", now that I am injecting caffeine into my veins, yeah, sure
looks like the video portion is stripped out from the T/R and fed up to
a "TAP/COMBINER"



On 11/15/2021 20:42, Grant Taylor wrote:
> value for the components.

Louis Ohland

unread,
Nov 16, 2021, 9:02:43 AM11/16/21
to
Therefore, the F-coupler has three entry/exit ports with one port
providing attachment for baseband utilization devices, a second port
providing attachment for broadband utilization devices and the third
providing attachment to the network.

Must meditate.

Louis Ohland

unread,
Nov 16, 2021, 9:17:02 AM11/16/21
to
I see nothing yet that says you can't have BOTH an ICS -AND- Coax
network. Methinks the video over T/R [ICS] is possumble. Further
meditation on initial patent is indicated.

Louis Ohland

unread,
Nov 16, 2021, 10:11:53 AM11/16/21
to
Broadband and baseband LAN
https://patents.google.com/patent/US4885747

The improved communications system includes a broadband network which
transmits broadband signals and a baseband network which transmits
baseband signals. A first F-coupler circuit arrangement accepts the
baseband and broadband signals and simultaneously concentrates said
broadband signals and said baseband signals onto one end of a length of
shielded twisted pair wires. A second F-coupler is coupled to the
opposite end of the shielded twisted pair wires. The second F-coupler
separates and distributes the broadband and baseband signals to their
respective connector ports to which utilization devices are connected.

So, methinks the network graphic shows you can have a separate ICS -AND-
Coax distribution. If you look at the bottom row, the workstations AND a
TV camera are connected through an F-Coupler to a T/R network with ONLY STP.

There is NO separate "RF" circuit.

Grant Taylor

unread,
Nov 16, 2021, 12:02:22 PM11/16/21
to
On 11/16/21 8:10 AM, Louis Ohland wrote:
> There is NO separate "RF" circuit.

I'll say it this way. The RF is super-imposed on the twisted pair
that's used to carry the baseband signal. Thus the baseband and RF
(broadband) use the same horizontal wire from the wiring closet to the
workstation outlet.

The RF and baseband are two separate logical networks that happen to
share a cable segment via the F-Couplers. Baseband data (Token Ring)
goes one to one set of equipment while RF (broadband) goes to a
different set of equipment. The only common parts of the cable plant
are the F-Couplers and the twisted pair wire in between.

At the workstation end, the typical ICS cable will come out of the
F-Coupler and go to the Token Ring workstation and standard coax will
come out of the F-Coupler and go to the TV (or other RF using
equipment). In the wiring closet, the typical ICS cable will come out
of the F-Coupler and got to a MAU and standard coax will come out of the
F-Coupler and go to the standard RF equipment; coax splitter,
distribution equipment, etc.

Kevin Moonlight

unread,
Nov 21, 2021, 3:12:14 AM11/21/21
to
I have some odd token-ring type adapters that have been floating around my desk for decades now and never quite knew what they were, I assumed they were for re-using token-ring building wiring for 10base2 ethernet or something along those lines. I cannot remember where it came from, but the highschool I attended in the 90's was an IBM case study of sorts as I was told at the time and it was excessively wired up with token-ring, and also had a television in every room linked back to a fully equipped studio that was used for live news broadcast style video morning announcements... I wonder if it used this system to share the wiring.

https://imgur.com/a/hoFZpwN





Grant Taylor

unread,
Nov 21, 2021, 3:32:46 AM11/21/21
to
On 11/21/21 1:12 AM, Kevin Moonlight wrote:
> I have some odd token-ring type adapters that have been floating around
> my desk for decades now and never quite knew what they were, I assumed
> they were for re-using token-ring building wiring for 10base2 ethernet
> or something along those lines.

I think that both 10Base2 and 10Base5 have some fairly hard requirements
on the cable / RF distance the card can be from the common bus. As such
I'm fairly certain that you couldn't have very much cable at all between
the coax bus and the transceiver.

Remember, 10Base5 had taps with active electronics directly next to ~>
on the coax bus and then completely separate AUI connection using
completely different technology on a service drop cable.

So, no, I'd bet lunch that you couldn't run 10Base2 over the adapters
that you have.

Aside: You /might/ be able to use two of them back to back as a part of
the larger 10Base2 bus, but not the lobes.

> I cannot remember where it came from, but the highschool I attended
> in the 90's was an IBM case study of sorts as I was told at the time
> and it was excessively wired up with token-ring, and also had a
> television in every room linked back to a fully equipped studio that
> was used for live news broadcast style video morning announcements... I
> wonder if it used this system to share the wiring.

Interesting.

> https://imgur.com/a/hoFZpwN

That's a BNC, not an F connector.

It also doesn't seem to have any data cable connected to it. Is it
supposed to be there? Or are the adapters that you have specifically
meant to re-use / abuse the horizontal in-wall cable explicitly for the
signal going over the BNC connector?

My assumption of the F-Coupler was that it could be used at the same
time as Token Ring data connections.

Kevin Moonlight

unread,
Nov 21, 2021, 3:38:42 AM11/21/21
to
Yeah I just took a minute to search after all these years, it seems these "Red" ones are for 3278 terminals, Blue would have been for ethernet, and there are some others yellow,green etc matching for other types.

Found a random listing on ebay while doing an image search that took me to some IBM part numbers.

https://www.ebay.com/itm/294459065378

Louis Ohland

unread,
Nov 21, 2021, 8:49:50 AM11/21/21
to
https://www.magnetika.com/history

Over the years, Magnetika has acquired numerous magnetics companies
including, but not limited to, Servomechanisms, Ferrodyne, Electro
Vector, Torwico Electronics, OPT Industries, Hyperion, Titan, Zenith and
the magnetics lines of DB Products and WEMS Electronics.

Louis Ohland

unread,
Nov 21, 2021, 8:53:44 AM11/21/21
to

Louis Ohland

unread,
Nov 21, 2021, 9:01:24 AM11/21/21
to
I waded through the website [what little there was], and I think OPT
made the transformer inside the Balun, and that's about all that it
specifically made.

OPT would make custom components. My SWAG is OPT made [a / the]
transformer, and assembled the units. Not a stock part, AFAICT.
Reply all
Reply to author
Forward
0 new messages