Does anybody have 10base5, 1base5, starlan1, or 10broad36 still running?

I'm trying to set up a 10base5 segment at home, can I use radio amateur type rg-8 or do I have to have the yellow rg8 designed for 10base5? I have tranceivers and a piercing tap, and assume I can terminate with a resister between ground and center. I think I saw they were only terminated at ONE end, but maybe it would be better to terminate with

150 ohms at both ends? After all, 10base2 was supposed to be terminated at both ends, wasn't it (I always did)?

Also, does anyody have operating 10broad36 or any of the 1 mb/sec ethernet-like protocols?

Reply to
sqrfolkdnc
Loading thread data ...

The orange cable was terminated at both ends with 50 ohm (nominal) N connector terminators.

ISTR that a tap in 10base5 was capacitive. It was just a little antenna poked thru the coax shield. Google for "ethernet blue book dec intel xerox". You might interesting info.

Looking for info I found this blast from the past, last updated in 1994.

formatting link
Who knows what will work with a small hobbyist network but when it was curent nobody ever used anything but the orange cable, for many reasons. For one, the tap clamps wouldn't fit right on coax unless the O.D. was correct. Nobody I know tried anything but orange cable.

Putting the taps in was a big deal at first. There was a tool kit that had a battery electric drill, a jig, and the right drill bits. Later,

3rd-party taps came along that were self-tapping, they came with a hex wrench.
Reply to
Al Dykes

50 Ohms, not 150. And you terminate at both ends. You _ground_ at only one point.

Ebay is your friend and good hunting.

Reply to
J. Clarke

Why one earth would you want to take such a huge step backwards? While you could use RG-8, I believe the ethernet cable was made with specific tap points, where it was easier to connect the taps. Also, you need 50 ohm terminators at each end. Do you also have NICs that can support the transceivers?

Reply to
James Knott

It sounds like a hobbyist network. That's cool. Redirecting the thread to alt.folklore.computers might get more info. A query there might turn up orange cable and related bits of kit.

The stripes were for mandatory tap seperation.

There are lots of little adapters around to convert the AUI connector on the tap to TW or UTP.

Google for a copy of the DEC/Xerox/Intel "Blue Book" spec for the original ethernet. It's one of the clearer bits of technical writing you'll ever read.

Reply to
Al Dykes

You've sure got a laundry list of standards there. Do you have any idea how they fit in the marketpalce in their day ? I don't know what "1 mb/sec/ ethernet-like" means. Mike Padlipsky's _The Elements of Networking Style_ has an interesting chapter on how people thought that IP (which was new and designed for point-to-point circuits) couldn't work on Ethernet, a broadcast meduim.

There was _very_ little of the pre-Blue Book equipment made. It was used only at Xerox and a few universities and since the original specs allowed only a handfull of computers on a LAN there was not the tons of desktop infrastructure to be found in dumpsters. As soon as the

10mb stuff hit the market (1981 ?) people with older stuff upgraded. That was a long time ago. At this point the discussion is firmly th the alt.folklore.computers are and should be picked up there.
Reply to
Al Dykes

He's probably talking about Starlan and its kin. Thinking about it, I _may_ actually have some 1 Mb/sec Starlan hardware in the attic. If so it will be ISA NICs--I'm pretty sure I don't have a hub.

Reply to
J. Clarke

The orange cable was used for plenum applications; the more common cable was bright yellow. (In fact, it was "1968 Corvette Yellow," to be specific.) I have some of the original prototype batch in my garage.

The tap was did not use capacitive coupling. It required that the transceiver input have DC connectivity with both the center conductor and the shield. (Remember, carrier-sense and collision-detect were based on DC voltage thresholds; a capacitive coupler would not work for this purpose.)

-- Rich Seifert Networks and Communications Consulting 21885 Bear Creek Way (408) 395-5700 Los Gatos, CA 95033 (408) 228-0803 FAX

Send replies to: usenet at richseifert dot com

Reply to
Rich Seifert

Thanks to all who replied, I will try alt.folklore.computers. This was my first foray into groups since a brief encounter with Sun groups several years ago, and I came across the cisco group looking for something else.

I have ONE N-I-B self tapping vampire tap and tranceiver. I have three tranceivers (two dual and one quad) that clearly were inteded to be spliced into RG-8 cable with coax connectors, although I have been unable to find any information on the web about these. I have 14 long and 8 short AUI cables and ISA cards/hubs/switches/routers with AUI ports to connect them to. Besides 10base2, I also have a 10baseF hub (optical fiber at 10 mb/sec) running in my lab/museum.

I assume the VERY FIRST thick was traditional black RG-8, the yellow and orange came later, else why would they have used the rg-8 spec? Yes, I knew the yellow cable was marked for tap points, I thought just to enforce the separation, which I figure should not matter if I am not trying to get published maximum distance. Yes, I got confused vis-a-vis grounding at one point vs terminating at one end.

The 1MB/s protocals were starlan1 and 1base5, both running on unshielded phone cable. It seems the 1 mb/s media were intended to be a local interface from desktops, implying the hubs would also have 10 mb for connedtion to the backbone, so with two hubs and a crossover, I could run it even though no current os would support any pc isa cards.

10broad36 was sharing coax with other signals (TV?) and supposed to go 3600 meters, hence the name, but some web documents said it only went 1800 meters.

If anybody can help with equpment, I would appreciate it. I'd like to get a used vampire tap so I can keep the new one "new". I should be able to get started with just the RG-8 (assuming I can make my own terminators), and if I can't find yellow/orange, I'll try some local radio amateurs I know to get a little from them (On Ebay I can buy 50 or 250 feet, and just the postage would exceed my budget).

Reply to
sqrfolkdnc

I stand corrected, thanks, but having done just a few taps myself don't see how that little pin could make a reliable connection, but It worked.

Reply to
Al Dykes

Note that 10base2 is electrically the same as 10base5, you can connect them together with BNC to N adapters. It is also easier to find BNC 50 ohm terminators, as they are commonly used in other than ethernet applications. My favorite are feed-through terminators to avoid the tee on the last connection.

RG8 or RG213 will work fine. (The maximum length will be less, but I don't expect that to cause you problems.) You can solder a 50 ohm resistor at each end if you want to. N connectors are expensive, but you may be able to find them in surplus stores.

-- glen

Reply to
glen herrmannsfeldt

There's lots of N hardware out there, and N-BNC adapters. Ask the hams and check Fair Radio Sales.

formatting link

Reply to
Al Dykes

Trust me, it took a *LOT* of mechanical and materials engineering to get that system to work reliably. Back in 1980-82, I was working on an almost daily basis with the folks at AMP in Harrisburg, PA who did the original 10 Mb/s vampire tap. The difficulty is in having the tap pin find its way through a viscous foam dielectric to achieve a solid berth in the center conductor of the cable. The project engineer described it as, "trying to nail Jello to a tree." His analogy was quite good.

-- Rich Seifert Networks and Communications Consulting 21885 Bear Creek Way (408) 395-5700 Los Gatos, CA 95033 (408) 228-0803 FAX

Send replies to: usenet at richseifert dot com

Reply to
Rich Seifert

The very first 10 Mb/s coax medium was in fact the custom-designed yellow (and later, orange) cable; it was never RG-8, or any standard cable. The electrical and mechanical requirements of the system precluded off-the-shelf designs. (I personally designed the original yellow/orange cables; they were manufactured for me by Belden (and later, many others).

The separation and the maximum distance are unrelated phenomenon. The forced spacing was to prevent lumped capacitive loads on the system, which could cause unacceptable signal reflections. The maximum cable length is a function of the resistance of the center conductor, which affects the carrier-sense and collision-detect thresholds.

If you have few taps (i.e., nowhere near the maximum of 100), the spacing is not that critical.

It all depends on how you measure it. Remember, 10BROAD36, like most CATV-style systems, uses a head-end device at the "source" of the cable. Thus, the maximum range is 1800 m (radius from the headend), or 3600 m (diameter, or maximum distance between the farthest pair of end stations); it's the same thing.

You may have problems trying to use RG-8 or any standard cable with a vampire tap. The dimensions/geometry are simply wrong. Also, the tap is designed to work only with a solid center conductor (not stranded) and a foam dielectric. The standard solid polyethylene dielectric of RG-8 is much too dense for the tap probe, and the higher dielectric constant reduces the diameter of the cable, i.e., the tap probe will be too long.

I have a large spool of an early prototype Ethernet cable (from back in the 1980 timeframe); it is dimensionally correct, but is black instead of yellow, and has only a single braid/single foil shield (rather than the double braid/double foil of the real stuff). In fact, it was tests on this cable that brought me to add the additional shields. Unless you are local in northern California, the shipping cost would exceed its value to you.

-- Rich Seifert Networks and Communications Consulting 21885 Bear Creek Way (408) 395-5700 Los Gatos, CA 95033 (408) 228-0803 FAX

Send replies to: usenet at richseifert dot com

Reply to
Rich Seifert

Any idea why it is it referred to as RG-8 all over the internet? I wondered if thick ethernet had a special dielectic so as to be easier to cut a hole into than normal cable would be, and whether my vampire tap would be able to make it into the center conductor of normal RG-8.

Where does that leave the tranceivers I have which have a pass through path for something the size of RG-8 using PL259 coax connectors, and were sold to me as thicknet tranceivers?

If thicknet is larger diameter than RG-8, does that mean the PL-259 coax won't fit the thicknet? Were their special coax connectors? Or are the three tranceivers I have that have a pass through connection appearing to be for RG-8 actually a variation on thinnet? Or did later installations use REAL RG-8 and these tranceivers but not use vampire taps? The more I learn the more ignorant I become...

Reply to
sqrfolkdnc

Probably because most hams consider any (roughly) 0.4" diameter, 50 ohm cable to be "RG-8". Don't believe everything you see or read on the Internet. ;^)

We took great pains to make the dielectric soft enough to allow easy penetration, yet firm enough to prevent movement of the center conductor when the cable is bent, which would affect the characteristic impedance. The foam dielectric makes the cable somewhat thicker than RG-8.

One can always build a transceiver that, instead of attaching through a vampire tap, requires cutting the cable and inserting the transceiver through standard connectors. The advantage is that the complexities of the vampire tap are avoided; the disadvantage is that the network must be brought down to install each new station.

PL-259 (UHF) connectors were never used, at least not according to the Ethernet standard. PL-259s are not constant-impedance connectors; they are fine for non-demanding applications, e.g. ham radio. Ethernet uses Type N connectors; these are more expensive, but specified at 50 ohms impedance. They are also quite a bit more rugged.

As a ham myself (KE1B), I would have liked PL-259s, since I could have gotten all I wanted as vendor samples from Ethernet cable manufacturers! Instead, I started using N connectors myself for amateur applications, and never went back. PL-259s are simply awful at GHz frequencies.

No standards-compliant installation used RG-8, or PL-259 connectors. Some commercial products used Type N inline connections, but most used the vampire tap.

By the way, there really is no single standard for "RG-8"; this is one of the problems we wanted to avoid. There is RG-8, RG-8/A RG-8/C, and a variety of others. Lots of manufacturers make "RG-8 style" cable; all of these vary in dimensions, attenuation, shielding effectiveness, etc. Ethernet installation would have been a nightmare if we just said "Use RG-8 cable".

-- Rich Seifert Networks and Communications Consulting 21885 Bear Creek Way (408) 395-5700 Los Gatos, CA 95033 (408) 228-0803 FAX

Send replies to: usenet at richseifert dot com

Reply to
Rich Seifert

You will find an article I wrote in the January 1990 edition, chronicling the 10th anniversary of the Ethernet ("Blue Book") Specification. A group of the original designers got together in my home for a reunion party, which resulted in the article.

-- Rich Seifert Networks and Communications Consulting 21885 Bear Creek Way (408) 395-5700 Los Gatos, CA 95033 (408) 228-0803 FAX

Send replies to: usenet at richseifert dot com

Reply to
Rich Seifert

I'd've thought that a sprung system would be necessary. Forked vampire tip to maintain contact, maybe with some spring in the fork-tip (ala RJ solid IDC) or a slight curve in the vampire.

Jell-O can be very easily nailed to a tree. Put it in a can. In this case, a long (6"?) circumferential compression clamp around the yellow garden hose.

-- Robert

Reply to
Robert Redelmeier

Another one here, VE3ZU. I wonder how many people here, realize that many of the first computer hobbyists were hams? The first popular computer magazine, Byte, was originally published by Wayne Green, who also published

73 Magazine, among others, for amateur radio. There were many computer articles in 73, along with the ham radio stuff.

Incidentally, I have every issue of Byte on the shelves behind me. I bought the first three issues in person, from Wayne, at the 1975 Radio Society of Ontario Convention, in Ottawa.

Reply to
James Knott

(snip)

The first 10base2 installations I saw used those transceivers with N to BNC adapters on them. At one point I did use one at the end of a thick ethernet cable with type N connectors.

If you want to add to your collection, I have some 100baseT4 transceivers.

-- glen

Reply to
glen herrmannsfeldt

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.