Sign in with
Sign up | Sign in
Your question

Anything faster I can run on Thinnet?

Last response: in Networking
Share
Anonymous
February 18, 2005 1:25:23 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

Trying to patch a bad situation for a few months until we can get a
fiber contractor into our semi-rural mfg facility. Is there anything
faster than 10 Mb that I can run over Thinnet cable? It doesn't have
to be standard or even Ethernet as I only need to connect 2 ends for a
short period of time.

Thanks.

sPh

More about : faster run thinnet

Anonymous
February 18, 2005 3:37:32 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

> > Trying to patch a bad situation for a few months until we can get a
> > fiber contractor into our semi-rural mfg facility. Is there
anything
> > faster than 10 Mb that I can run over Thinnet cable?
>
> Can you go wireless? You didn't say what the distance was, or if
there
> were obstructions that would make this impossible.

Good thought, but I truely doubt wireless would work in that
facility/environment.

sPh
Anonymous
February 18, 2005 6:19:41 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

sphealey wrote:
> Trying to patch a bad situation for a few months until we can get a
> fiber contractor into our semi-rural mfg facility. Is there anything
> faster than 10 Mb that I can run over Thinnet cable? It doesn't have
> to be standard or even Ethernet as I only need to connect 2 ends for a
> short period of time.
>
> Thanks.
>
> sPh
>

Can you go wireless? You didn't say what the distance was, or if there
were obstructions that would make this impossible. Depending on
distance and many other factors, you might be able to get real-world
speeds better than 10Mpbs using 802.11a or 802.11g.
You can use Ethernet-to-wireless bridges for non-wireless-capable
equipment too.

Good luck,
Jonathan
Related resources
Anonymous
February 18, 2005 7:56:42 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

sphealey wrote:

>> > Trying to patch a bad situation for a few months until we can get a
>> > fiber contractor into our semi-rural mfg facility. Is there
> anything
>> > faster than 10 Mb that I can run over Thinnet cable?
>>
>> Can you go wireless? You didn't say what the distance was, or if
> there
>> were obstructions that would make this impossible.
>
> Good thought, but I truely doubt wireless would work in that
> facility/environment.

In principle you could with suitable fittings use the Thinnet cable to carry
the wifi signal. I've never heard of it being tried but it would be an
interesting experiment.

> sPh

--
--John
Reply to jclarke at ae tee tee global dot net
(was jclarke at eye bee em dot net)
Anonymous
February 18, 2005 8:19:28 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

"sphealey" <sphealey@worldnet.att.net> wrote:
>faster than 10 Mb that I can run over Thinnet cable?

I'd get a couple of WiFi devices and see if I could couple them to the
cable. Depending on their specs and the cable length you might want
to add some attenuators, but it might work to just get coax
adapters...
Anonymous
February 18, 2005 10:30:39 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

Begin <1108751122.984684.275860@o13g2000cwo.googlegroups.com>
On 2005-02-18, sphealey <sphealey@worldnet.att.net> wrote:
> Trying to patch a bad situation for a few months until we can get a
> fiber contractor into our semi-rural mfg facility. Is there anything
> faster than 10 Mb that I can run over Thinnet cable? It doesn't have
> to be standard or even Ethernet as I only need to connect 2 ends for a
> short period of time.

Telco equipment, like DS3, E3, or T3 linecards, but it expects 75 Ohm
(RG59) instead of 50 Ohm (RG58) cable. It is rather expensive unless you
can pick up second hand stuff cheaply, or borrow stuff. Oh, and you'll
need two cables, not just one.


--
j p d (at) d s b (dot) t u d e l f t (dot) n l .
Anonymous
February 18, 2005 10:37:08 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

William P.N. Smith wrote:

> "sphealey" <sphealey@worldnet.att.net> wrote:
>>faster than 10 Mb that I can run over Thinnet cable?
>
> I'd get a couple of WiFi devices and see if I could couple them to the
> cable. Depending on their specs and the cable length you might want
> to add some attenuators, but it might work to just get coax
> adapters...

Something I should have mentioned--be careful with the wifi boards if you do
that--some of the connectors are remarkably fragile--I've had a couple of
them pull right out of the board just hooking up an antenna
cable--fortunately it's just the ferrule--shoving it back on hard with a
little epoxy seems to take care of the problem as long as you don't put any
strain on them, although I wouldn't vouch for the impedence being right.

--
--John
Reply to jclarke at ae tee tee global dot net
(was jclarke at eye bee em dot net)
Anonymous
February 19, 2005 7:15:48 AM

Archived from groups: comp.dcom.lans.ethernet (More info?)

In article <1108751122.984684.275860@o13g2000cwo.googlegroups.com>,
sphealey <sphealey@worldnet.att.net> wrote:
>Trying to patch a bad situation for a few months until we can get a
>fiber contractor into our semi-rural mfg facility. Is there anything
>faster than 10 Mb that I can run over Thinnet cable? It doesn't have
>to be standard or even Ethernet as I only need to connect 2 ends for a
>short period of time.

back in about '91 DEC made an FDDI over coax converter. I don't know if
you could find one of those anymore.

I've got a span of coax running out of a power plant, with no way to
re-run fiber in the path (the active bus off the generator is in the same
cabletray), so if you find anything that works, let me know ;-)


--
Daniel J McDonald CCIE # 2495, CNX
Visit my website: http://www.austinnetworkdesign.com
Anonymous
February 19, 2005 2:34:27 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

On Fri, 18 Feb 2005 16:56:42 -0500, "J. Clarke"
<jclarke@nospam.invalid> wrote:

>sphealey wrote:
>
>>> > Trying to patch a bad situation for a few months until we can get a
>>> > fiber contractor into our semi-rural mfg facility. Is there
>> anything
>>> > faster than 10 Mb that I can run over Thinnet cable?
>>>
>>> Can you go wireless? You didn't say what the distance was, or if
>> there
>>> were obstructions that would make this impossible.
>>
>> Good thought, but I truely doubt wireless would work in that
>> facility/environment.
>
>In principle you could with suitable fittings use the Thinnet cable to carry
>the wifi signal.

No. You would not.
Your thinnet cable is designed for 10 MHz operation and is
practically useless above say 100 MHz.
Wifi operates at 2400 MHz. So go figure :-)

--
Kind regards,
Gerard Bok
Anonymous
February 19, 2005 2:34:28 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

bok118@zonnet.nl (Gerard Bok) wrote:
>Your thinnet cable is designed for 10 MHz operation and is
>practically useless above say 100 MHz.
>Wifi operates at 2400 MHz. So go figure :-)

So you're saying that the OP's cable has more than (say) 120dB loss at
2.4GHz? Even without knowing the cable type or length?

Belden 9907, for instance, has 14.8dB loss per 100 feet at 1GHz...
Anonymous
February 19, 2005 2:34:28 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

Gerard Bok wrote:

> On Fri, 18 Feb 2005 16:56:42 -0500, "J. Clarke"
> <jclarke@nospam.invalid> wrote:
>
>>sphealey wrote:
>>
>>>> > Trying to patch a bad situation for a few months until we can get a
>>>> > fiber contractor into our semi-rural mfg facility. Is there
>>> anything
>>>> > faster than 10 Mb that I can run over Thinnet cable?
>>>>
>>>> Can you go wireless? You didn't say what the distance was, or if
>>> there
>>>> were obstructions that would make this impossible.
>>>
>>> Good thought, but I truely doubt wireless would work in that
>>> facility/environment.
>>
>>In principle you could with suitable fittings use the Thinnet cable to
>>carry the wifi signal.
>
> No. You would not.
> Your thinnet cable is designed for 10 MHz operation and is
> practically useless above say 100 MHz.
> Wifi operates at 2400 MHz. So go figure :-)

ROF,L. If your source for that information is a book, burn it. If it's a
teacher, drop the class. If it's your boss, be very afraid.

Thinnet cable is typically RG-58 coax, which is an RF cable _rated_ for over
a GHz and capable of carrying a good deal more than that with reasonable
attenuation.

--
--John
Reply to jclarke at ae tee tee global dot net
(was jclarke at eye bee em dot net)
Anonymous
February 19, 2005 2:34:29 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

In article <v7he11ho7ei40shpgfduqkbf4smhmjpnje@4ax.com>,
<William P.N. Smith> wrote:
>bok118@zonnet.nl (Gerard Bok) wrote:
>>Your thinnet cable is designed for 10 MHz operation and is
>>practically useless above say 100 MHz.
>>Wifi operates at 2400 MHz. So go figure :-)
>
>So you're saying that the OP's cable has more than (say) 120dB loss at
>2.4GHz? Even without knowing the cable type or length?
>
>Belden 9907, for instance, has 14.8dB loss per 100 feet at 1GHz...
>

The belden spec tops out at 1GHz. It's going to be lots gigher at
2.4Ghz. A google for generic RG58 found 1.1DB/m at 2GHz. It could be
wrong.

I looked that up for a query about a 600ft/200M run, which I think the
path loss (maybe 200DB down) rules a WiFi hack for that user.

It's possible that for some path that's not a clear shot (say indoors)
for WiFI, or slightly outside the WiFI range (say 300 Ft) the
radio-over-coax might work. A max-length TW run seems to be beyond
the reach of a WiFI radio.

Since the OP has 10Mb/sec TW ethernet running "b" wifi isn't as fast.
It would have to be "g", which is in the same RF band.



--

a d y k e s @ p a n i x . c o m

Don't blame me. I voted for Gore.
Anonymous
February 19, 2005 5:04:47 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

J. Clarke wrote:

> In principle you could with suitable fittings use the Thinnet cable to
> carry the wifi signal.  I've never heard of it being tried but it would be
> an interesting experiment.

RG-58 is very lossy at WiFi frequencies.
Anonymous
February 19, 2005 5:12:36 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

Gerard Bok wrote:

> No. You would not.
> Your thinnet cable is designed for 10 MHz operation and is
> practically useless above say 100 MHz.
> Wifi operates at 2400 MHz. So go figure :-)

RG-58 cables, which is what's used for ethernet, is usable at much higher
frequencies than 100 MHz. However, it becomes very lossy at higher
frequencies, so that it's only suitable for short distances. According to
my catalog, at 900 MHz, the loss is about 22 dB at 100 feet or almost 66 dB
at 100 metres. For rough estimates, 20 dB = 99% loss and 60dB = 99.9999%
loss. The figures will be much higher for 2.4 GHz. At WiFi frequencies,
you'd only use RG-58 for short patch cords.
Anonymous
February 19, 2005 5:17:16 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

William P.N. Smith wrote:

> Belden 9907, for instance, has 14.8dB loss per 100 feet at 1GHz...

However, it's not likely to have been used for ethernet.

BTW, Belden 9311 would be a bit better and use the same hardware as RG-58.
Anonymous
February 19, 2005 5:22:53 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

William P.N. Smith wrote:

> "sphealey" <sphealey@worldnet.att.net> wrote:
>>faster than 10 Mb that I can run over Thinnet cable?
>
> I'd get a couple of WiFi devices and see if I could couple them to the
> cable. Depending on their specs and the cable length you might want
> to add some attenuators, but it might work to just get coax
> adapters...

You might want to read some of the other posts, to find out that attenuators
are the last thing you need. WiFi over coax will not likely work as far as
WiFi through the air.
Anonymous
February 19, 2005 5:39:49 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

William wrote:
> "sphealey" <sphealey@worldnet.att.net> wrote:
> >faster than 10 Mb that I can run over Thinnet cable?
>
> Bail us out here, and tell us how long the existing cable is! 8*)
>
> [I kinda like the idea of using the cable _shield_ as an antenna!]

Sorry dudes - I had to go back and look at my own original post as I
thought I had included that information. Which I had not ;-(

Length is over 100m - about 125 I think. Longer than 10' for sure.

sPh
Anonymous
February 19, 2005 5:50:21 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

On Sat, 19 Feb 2005 09:03:03 -0500, William P.N. Smith wrote:

>bok118@zonnet.nl (Gerard Bok) wrote:
>>Your thinnet cable is designed for 10 MHz operation and is
>>practically useless above say 100 MHz.
>>Wifi operates at 2400 MHz. So go figure :-)
>
>So you're saying that the OP's cable has more than (say) 120dB loss at
>2.4GHz? Even without knowing the cable type or length?

What I am saying is, that for Wifi frequencies a Pringle can
probably outperforms any thinnet cable :-)


--
Kind regards,
Gerard Bok
Anonymous
February 19, 2005 6:41:27 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

On Sat, 19 Feb 2005 09:19:19 -0500, "J. Clarke"
<jclarke@nospam.invalid> wrote:

>Gerard Bok wrote:
>
>> On Fri, 18 Feb 2005 16:56:42 -0500, "J. Clarke"
>> <jclarke@nospam.invalid> wrote:
>>
>>>sphealey wrote:
>>>
>>>>> > Trying to patch a bad situation for a few months until we can get a
>>>>> > fiber contractor into our semi-rural mfg facility. Is there
>>>> anything
>>>>> > faster than 10 Mb that I can run over Thinnet cable?
>>>>>
>>>>> Can you go wireless? You didn't say what the distance was, or if
>>>> there
>>>>> were obstructions that would make this impossible.
>>>>
>>>> Good thought, but I truely doubt wireless would work in that
>>>> facility/environment.
>>>
>>>In principle you could with suitable fittings use the Thinnet cable to
>>>carry the wifi signal.
>>
>> No. You would not.
>> Your thinnet cable is designed for 10 MHz operation and is
>> practically useless above say 100 MHz.
>> Wifi operates at 2400 MHz. So go figure :-)
>
>ROF,L. If your source for that information is a book, burn it. If it's a
>teacher, drop the class. If it's your boss, be very afraid.

It's even worse than you imagine :-)

>Thinnet cable is typically RG-58 coax, which is an RF cable _rated_ for over
>a GHz and capable of carrying a good deal more than that with reasonable
>attenuation.

I am talking about the stuff that typically came out of the box
labeled 'thinnet cable'. Typical specs:
RG58 c/u
capacity 93.5 pf/m
attenuation at 100 MHz 174 dB/km (= 278 dB per mile)

In my opinion you cannot even characterize this stuff at 2.4 GHz
as its behavour is no longer that of a 'coaxial cable'.
I would imagine that you could even get better performance if you
just connect a single wire; either the core or the shield.


--
Kind regards,
Gerard Bok
Anonymous
February 19, 2005 6:41:28 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

Gerard Bok wrote:

> On Sat, 19 Feb 2005 09:19:19 -0500, "J. Clarke"
> <jclarke@nospam.invalid> wrote:
>
>>Gerard Bok wrote:
>>
>>> On Fri, 18 Feb 2005 16:56:42 -0500, "J. Clarke"
>>> <jclarke@nospam.invalid> wrote:
>>>
>>>>sphealey wrote:
>>>>
>>>>>> > Trying to patch a bad situation for a few months until we can get a
>>>>>> > fiber contractor into our semi-rural mfg facility. Is there
>>>>> anything
>>>>>> > faster than 10 Mb that I can run over Thinnet cable?
>>>>>>
>>>>>> Can you go wireless? You didn't say what the distance was, or if
>>>>> there
>>>>>> were obstructions that would make this impossible.
>>>>>
>>>>> Good thought, but I truely doubt wireless would work in that
>>>>> facility/environment.
>>>>
>>>>In principle you could with suitable fittings use the Thinnet cable to
>>>>carry the wifi signal.
>>>
>>> No. You would not.
>>> Your thinnet cable is designed for 10 MHz operation and is
>>> practically useless above say 100 MHz.
>>> Wifi operates at 2400 MHz. So go figure :-)
>>
>>ROF,L. If your source for that information is a book, burn it. If it's a
>>teacher, drop the class. If it's your boss, be very afraid.
>
> It's even worse than you imagine :-)
>
>>Thinnet cable is typically RG-58 coax, which is an RF cable _rated_ for
>>over a GHz and capable of carrying a good deal more than that with
>>reasonable attenuation.
>
> I am talking about the stuff that typically came out of the box
> labeled 'thinnet cable'. Typical specs:
> RG58 c/u
> capacity 93.5 pf/m
> attenuation at 100 MHz 174 dB/km (= 278 dB per mile)
>
> In my opinion you cannot even characterize this stuff at 2.4 GHz
> as its behavour is no longer that of a 'coaxial cable'.
> I would imagine that you could even get better performance if you
> just connect a single wire; either the core or the shield.

I fail to see your point. Belden Thinnet cable, purpose made, is rated for
16 dB/100 feet at 1 GHz. Their RG58 is rated 14-22 depending on which
particular variety. 1671A, a purpose-made microwave cable rated to 20 GHz,
has 19.4, 1672A, a purpose-made high-frequency video cable, has 26.

Now you can "imagine" anything you want to but in the real world cables with
that performance have been carrying signals in those frequency ranges for
decades.

--
--John
Reply to jclarke at ae tee tee global dot net
(was jclarke at eye bee em dot net)
Anonymous
February 19, 2005 6:53:05 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

James Knott <james.knott@rogers.com> wrote:
>William P.N. Smith wrote:
>> Belden 9907, for instance, has 14.8dB loss per 100 feet at 1GHz...

>However, it's not likely to have been used for ethernet.

My mistake then, I went to the Belden site and typed "thinnet" into
their search tool.
Anonymous
February 19, 2005 6:57:04 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

James Knott <james.knott@rogers.com> wrote:
>William P.N. Smith wrote:

>> I'd get a couple of WiFi devices and see if I could couple them to the
>> cable. Depending on their specs and the cable length you might want
>> to add some attenuators, but it might work to just get coax
>> adapters...

>You might want to read some of the other posts, to find out that attenuators
>are the last thing you need. WiFi over coax will not likely work as far as
>WiFi through the air.

What, no attenuators for the OP's 10 feet of cable? Hunh, I'd at
least have kept an open mind, but maybe that's just me.
Anonymous
February 19, 2005 6:58:33 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

"sphealey" <sphealey@worldnet.att.net> wrote:
>faster than 10 Mb that I can run over Thinnet cable?

Bail us out here, and tell us how long the existing cable is! 8*)

[I kinda like the idea of using the cable _shield_ as an antenna!]
Anonymous
February 19, 2005 10:32:52 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

On Sat, 19 Feb 2005 11:36:04 -0500, "J. Clarke"
<jclarke@nospam.invalid> wrote:

>>>>>sphealey wrote:
>>>>>>> > Trying to patch a bad situation for a few months until we can get a
>>>>>>> > fiber contractor into our semi-rural mfg facility. Is there
>>>>>> anything
>>>>>>> > faster than 10 Mb that I can run over Thinnet cable?
>>>>>>>
>>>>>>> Can you go wireless? You didn't say what the distance was, or if
>>>>>> there
>>>>>>> were obstructions that would make this impossible.
>>>>>>
>>>>>> Good thought, but I truely doubt wireless would work in that
>>>>>> facility/environment.
>>>>>
>>>>>In principle you could with suitable fittings use the Thinnet cable to
>>>>>carry the wifi signal.
>>>>
>>>> No. You would not.
>>>> Your thinnet cable is designed for 10 MHz operation and is
>>>> practically useless above say 100 MHz.
>>>> Wifi operates at 2400 MHz. So go figure :-)

>>>Thinnet cable is typically RG-58 coax, which is an RF cable _rated_ for
>>>over a GHz and capable of carrying a good deal more than that with
>>>reasonable attenuation.
>>
>> I am talking about the stuff that typically came out of the box
>> labeled 'thinnet cable'. Typical specs:
>> RG58 c/u
>> capacity 93.5 pf/m
>> attenuation at 100 MHz 174 dB/km (= 278 dB per mile)
>>
>> In my opinion you cannot even characterize this stuff at 2.4 GHz
>> as its behavour is no longer that of a 'coaxial cable'.
>> I would imagine that you could even get better performance if you
>> just connect a single wire; either the core or the shield.

>I fail to see your point. Belden Thinnet cable, purpose made, is rated for
>16 dB/100 feet at 1 GHz.

Wich would be some 30 dB per 100 feet at 3 GHz, I guess ?
A stretch of 10Base-2 is limited to 600 feet so...

>Their RG58 is rated 14-22 depending on which particular variety.

Some browsing revealed that over 50 dB per 100 feet at 3GHz is
not an uncommon figure for commercial grade RG 58 C/U thinnet
cable. (Which actually surprises me. I didn't expect any testing
to be done on this cable in the GHz region. As I wouldn't even
expect it to be '50 ohms' at this type of frequencies.)

>1671A, a purpose-made microwave cable rated to 20 GHz,
>has 19.4, 1672A, a purpose-made high-frequency video cable, has 26.

Great! But the original poster doesn't have either 'purpose-made
microwave cable' or 'purpose-made high-frequency video cable''.
He has a cable rated as 'thinnet cable' preinstalled.

>Now you can "imagine" anything you want to but in the real world cables with
>that performance have been carrying signals in those frequency ranges for
>decades.

Sure. I don't claim that you cannot run 2.4 GHz over any coax
cable. (Although it is not always easy).
I just stated that it is very unlikely that OP can effectively
run a 2.4 GHz link over coax that is thinnet rated.

--
Kind regards,
Gerard Bok
Anonymous
February 19, 2005 10:32:53 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

Gerard Bok wrote:

> Some browsing revealed that over 50 dB per 100 feet at 3GHz is
> not an uncommon figure for commercial grade RG 58 C/U thinnet
> cable. (Which actually surprises me. I didn't expect any testing
> to be done on this cable in the GHz region. As I wouldn't even
> expect it to be '50 ohms' at this type of frequencies.)

A length of coax can greatly improve the performance of a dummy load, used
in testing UHF or microwave transmitters.
Anonymous
February 19, 2005 10:50:59 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

"sphealey" <sphealey@worldnet.att.net> wrote:
>Length is over 100m - about 125 I think.

Ah, in that case the wireless thing probably wouldn't work for you.
What kind of cable do you have installed?
Anonymous
February 20, 2005 3:41:17 AM

Archived from groups: comp.dcom.lans.ethernet (More info?)

Gerard Bok <bok118@zonnet.nl> wrote:
> What I am saying is, that for Wifi frequencies a Pringle
> can probably outperforms any thinnet cable :-)

An interesting statement.

How many dB of loss are there in closely coupled transmitting
& receiving antennae ? 100 dB in antenna efficiency? Add to
that the inverse-square losses of ~400ft of air (52db?)

-- Robert
Anonymous
February 20, 2005 4:17:13 AM

Archived from groups: comp.dcom.lans.ethernet (More info?)

In article <N0RRd.14838$D34.4118@newssvr12.news.prodigy.com>,
Robert Redelmeier <redelm@ev1.net.invalid> wrote:
:Gerard Bok <bok118@zonnet.nl> wrote:
:> What I am saying is, that for Wifi frequencies a Pringle
:> can probably outperforms any thinnet cable :-)

:An interesting statement.

:How many dB of loss are there in closely coupled transmitting
:& receiving antennae ? 100 dB in antenna efficiency? Add to
:that the inverse-square losses of ~400ft of air (52db?)

http://www.signull.com/fsc.php

and choose a consumer access point specification such as +15 dBm
transmission and -80 dBm sensitivity. The result is 0.1 miles
which is greater than the 0.08 miles which is 400'.

Thus, with a plain inexpensive AP of no great quality, using
only the standard omnidirectional diversity antenna, one can
go the necessary distance.

That's with omnidirectional. If one using a Pringles can antenna,
then the EIRP is much more focused and one can go considerably further.

How much further? Well, under contest conditions in Utah, up to 0.8
miles using jury-rigged antennae.

http://www.wired.com/news/culture/0,1284,64440,00.html
--
100% of all human deaths occur within 100 miles of Earth.
Anonymous
February 20, 2005 4:45:36 AM

Archived from groups: comp.dcom.lans.ethernet (More info?)

Walter Roberson <roberson@ibd.nrc-cnrc.gc.ca> wrote:
> http://www.signull.com/fsc.php
>
> and choose a consumer access point specification such as
> +15 dBm transmission and -80 dBm sensitivity. The result is
> 0.1 miles which is greater than the 0.08 miles which is 400'.

> Thus, with a plain inexpensive AP of no great quality,
> using only the standard omnidirectional diversity antenna,
> one can go the necessary distance.

I don't doubt that. What I wonder is if direct coupling
antennae feeds with coax has lower attenuation. All
sort of different numbers are floating around for
coax attenuation at 2.4 GHz.

It would be relatively interesting that a fairly decent
direct-coupled copper line has higher losses than
indirect EMF over the air.

-- Robert
Anonymous
February 20, 2005 12:26:26 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

Robert Redelmeier wrote:

> Gerard Bok <bok118@zonnet.nl> wrote:
>> What I am saying is, that for Wifi frequencies a Pringle
>> can probably outperforms any thinnet cable :-)
>
> An interesting statement.
>
> How many dB of loss are there in closely coupled transmitting
> & receiving antennae ? 100 dB in antenna efficiency? Add to
> that the inverse-square losses of ~400ft of air (52db?)

Then again, using a cable probably has fewer calories. ;-)
Anonymous
February 20, 2005 12:30:54 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

William P.N. Smith wrote:

> What, no attenuators for the OP's 10 feet of cable?  Hunh, I'd at
> least have kept an open mind, but maybe that's just me.

I don't recall any mention of 10' of cable. He did mention 10 Mb.
Anonymous
February 20, 2005 1:12:11 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

James Knott <james.knott@rogers.com> wrote:
>I don't recall any mention of 10' of cable.

Exactly my point. Without knowledge of the cable length and
characteristics, the whole discussion was degenerating into "We don't
know nothing, but it can't possibly work over 165 meters of the lowest
quality Thinnet cable, we think...."
Anonymous
February 20, 2005 3:16:16 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

On Sun, 20 Feb 2005 01:45:36 GMT, Robert Redelmeier
<redelm@ev1.net.invalid> wrote:

>> Thus, with a plain inexpensive AP of no great quality,
>> using only the standard omnidirectional diversity antenna,
>> one can go the necessary distance.

Provided you have clear line of sight :-)

>I don't doubt that. What I wonder is if direct coupling
>antennae feeds with coax has lower attenuation.

I think the first question would be: what happens if you attempt
to feed 2.4 GHz into commercial thinnet coax.
Are you actually feeding signal into a transmission line or are
you chooking your source by connecting some weird object, mainly
a huge lumb capacitor to it ?

>It would be relatively interesting that a fairly decent
>direct-coupled copper line has higher losses than
>indirect EMF over the air.

There is no direct EMF over the air.
Either you couple your output directly into a proper antenna or
you feed it into a 50 ohms transmission line :-)

Note: output data for your wifi device is only valid if it is
terminated as in one of the above situations :-)

--
Kind regards,
Gerard Bok
Anonymous
February 20, 2005 4:29:38 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

> back in about '91 DEC made an FDDI over coax converter.

I have been been able to find some mention of that technology, but no
part numbers or product line names. Would anyone happen to remember
what they were called and/or a part number?

Thanks.

Sorry about leaving the length out on my original post; I thought I had
included it. There is no usable specification information that I can
find on the cable.

sPh
Anonymous
February 20, 2005 8:10:33 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

In article <1108934978.001018.114540@o13g2000cwo.googlegroups.com>,
"sphealey" <sphealey@worldnet.att.net> wrote:

> > back in about '91 DEC made an FDDI over coax converter.
>
> I have been been able to find some mention of that technology, but no
> part numbers or product line names. Would anyone happen to remember
> what they were called and/or a part number?
>

There was an early proposal for a version of FDDI that ran over coax; it
was referred to as LDDI. DEC also made a proprietary product (pre-FDDI)
that ran on thick Ethernet coaxial cable; it was used for the back-end
connection in the original VAXCluster product.


--
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
Anonymous
February 20, 2005 10:07:00 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

In article <1108934978.001018.114540@o13g2000cwo.googlegroups.com>,
sphealey <sphealey@worldnet.att.net> wrote:
>> back in about '91 DEC made an FDDI over coax converter.
>
>I have been been able to find some mention of that technology, but no
>part numbers or product line names. Would anyone happen to remember
>what they were called and/or a part number?
>
>Thanks.
>
>Sorry about leaving the length out on my original post; I thought I had
>included it. There is no usable specification information that I can
>find on the cable.
>
>sPh
>


Wait a moment while I whip out my DEC Network Buyer's Guide 1992-1993
(several hundred pages) Here it is, Chapter 3.

You have the DECbridge 500/600/700 (FDDI ring to 10mb 802.3 ethernet)
and the DECconcentrator 500 (FDDI ring to FDDI devices)

You'd need two cables because FDDI requires a ring, or better
yet 4 for a dual ring.




--

a d y k e s @ p a n i x . c o m

Don't blame me. I voted for Gore.
Anonymous
February 21, 2005 9:50:16 AM

Archived from groups: comp.dcom.lans.ethernet (More info?)

> Wait a moment while I whip out my DEC Network
> Buyer's Guide 1992-1993 (several hundred
> pages) Here it is, Chapter 3.

I don't know which is worse: that you have that in your possession, or
that I envy you for having it when I don't! Must .... hide ......
sickness ... from .... spouse. Must .... hide .......

sPh
Anonymous
February 21, 2005 10:45:30 AM

Archived from groups: comp.dcom.lans.ethernet (More info?)

Do you have any phone lines between the buildings?

Long range ethernet operates at up to 15Mbps full duplex
(well 15 is more than 10). Speed though depends on the
*cabling* and *distance*.

Have a quick look at
http://www.cisco.com/nobel/initiative/docs/NNI_CP_02.pd...
Then look up the costs.

Why do you need more than 10Mbps?

If you for example have an old network using repeaters
you might be able to get by with installing a switch
or router to control the traffic, eliminate unwanted
broadcasts, whatever. Maybe you can use data compression?

You can get laser links that will do the job however
they are prone to interruption by birds, rain, have
a read, just google on [laser ethernet] for example.

In the UK non-omni antenna are illegal for 802.xyz (I
think anyway) however there may well be commecial
solutions available in the US.

google [802.11 directional antenna]

http://www.hyperlinktech.com/web/antennas_2400_out_dire...
There you go! I would thing a couple of these would
sort you out.

http://www.fab-corp.com/
HAve amplifiers too - I would go for antennas first.

I am sure that one of the previous contributors to this
thread can quickly work out which one of these beasts
would do the job for you. I might be able to attempt
it but it would take me hours or days and I don't
know yet how much confidence I might have in the result:) 

Wireless can be affected by weather (maybe other posters
can contribute?) so I wouldn't go for the mimimum possible
solution.

You can always keep the coax link on 10M as a backup.
Anonymous
February 21, 2005 1:04:44 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

In article <1108997416.774125.309870@z14g2000cwz.googlegroups.com>,
sphealey <sphealey@worldnet.att.net> wrote:
>> Wait a moment while I whip out my DEC Network
>> Buyer's Guide 1992-1993 (several hundred
>> pages) Here it is, Chapter 3.
>
>I don't know which is worse: that you have that in your possession, or
>that I envy you for having it when I don't! Must .... hide ......
>sickness ... from .... spouse. Must .... hide .......
>
>sPh
>


Everything shows up on ebay, eventually.

--

a d y k e s @ p a n i x . c o m

Don't blame me. I voted for Gore.
Anonymous
February 21, 2005 2:48:17 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

On Sun, 20 Feb 2005 17:10:33 -0800, Rich Seifert
<usenet@richseifert.com.invalid> wrote:

>In article <1108934978.001018.114540@o13g2000cwo.googlegroups.com>,
> "sphealey" <sphealey@worldnet.att.net> wrote:
>
>> > back in about '91 DEC made an FDDI over coax converter.
>>
>> I have been been able to find some mention of that technology, but no
>> part numbers or product line names. Would anyone happen to remember
>> what they were called and/or a part number?
>>
>
>There was an early proposal for a version of FDDI that ran over coax; it
>was referred to as LDDI. DEC also made a proprietary product (pre-FDDI)
>that ran on thick Ethernet coaxial cable; it was used for the back-end
>connection in the original VAXCluster product.

But the crucial question hete is: would this product run more
than 10 Mbps over 'thinnet coax' ? :-)

--
Kind regards,
Gerard Bok
Anonymous
February 21, 2005 2:48:18 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

In article <4219c80b.541526@News.Individual.NET>,
Gerard Bok <bok118@zonnet.nl> wrote:
>On Sun, 20 Feb 2005 17:10:33 -0800, Rich Seifert
><usenet@richseifert.com.invalid> wrote:
>
>>In article <1108934978.001018.114540@o13g2000cwo.googlegroups.com>,
>> "sphealey" <sphealey@worldnet.att.net> wrote:
>>
>>> > back in about '91 DEC made an FDDI over coax converter.
>>>
>>> I have been been able to find some mention of that technology, but no
>>> part numbers or product line names. Would anyone happen to remember
>>> what they were called and/or a part number?
>>>
>>
>>There was an early proposal for a version of FDDI that ran over coax; it
>>was referred to as LDDI. DEC also made a proprietary product (pre-FDDI)
>>that ran on thick Ethernet coaxial cable; it was used for the back-end
>>connection in the original VAXCluster product.
>
>But the crucial question hete is: would this product run more
>than 10 Mbps over 'thinnet coax' ? :-)
>
>--
>Kind regards,
>Gerard Bok


The FDDI-ethernet version had 10mb ethernet interfaces so you'd
get no speed increase on any single system.

The FDDI ring to fddi computer version requires a FDDI card for one of
yoru systems (at both ends). It's possible that the standards are
good enough that you could get any PCI fddi card and hook it up.

And, as I said before, FDDI requires a ring so you'd need two TW
cables to form one.


--

a d y k e s @ p a n i x . c o m

Don't blame me. I voted for Gore.
Anonymous
February 21, 2005 2:48:18 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

In article <4219c80b.541526@News.Individual.NET>,
bok118@zonnet.nl (Gerard Bok) wrote:

> On Sun, 20 Feb 2005 17:10:33 -0800, Rich Seifert
> <usenet@richseifert.com.invalid> wrote:
>
> >There was an early proposal for a version of FDDI that ran over coax; it
> >was referred to as LDDI. DEC also made a proprietary product (pre-FDDI)
> >that ran on thick Ethernet coaxial cable; it was used for the back-end
> >connection in the original VAXCluster product.
>
> But the crucial question hete is: would this product run more
> than 10 Mbps over 'thinnet coax' ? :-)

Since the impedance of 'thinnet coax' is identical to that of thick
Ethernet coax, clearly any system that will run on one will run on the
other; the only question is the reduction in length for acceptable
levels of operation.

(Remember, other than the connector used to attach to the medium, a
'thinnet' transceiver is *IDENTICAL* to a thick Ethernet transceiver.)


--
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
Anonymous
February 21, 2005 7:30:53 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

On Mon, 21 Feb 2005 07:43:07 -0800, Rich Seifert
<usenet@richseifert.com.invalid> wrote:

>In article <4219c80b.541526@News.Individual.NET>,
> bok118@zonnet.nl (Gerard Bok) wrote:
>
>> On Sun, 20 Feb 2005 17:10:33 -0800, Rich Seifert
>> <usenet@richseifert.com.invalid> wrote:
>>
>> >There was an early proposal for a version of FDDI that ran over coax; it
>> >was referred to as LDDI. DEC also made a proprietary product (pre-FDDI)
>> >that ran on thick Ethernet coaxial cable; it was used for the back-end
>> >connection in the original VAXCluster product.
>>
>> But the crucial question hete is: would this product run more
>> than 10 Mbps over 'thinnet coax' ? :-)
>
>Since the impedance of 'thinnet coax' is identical to that of thick
>Ethernet coax, clearly any system that will run on one will run on the
>other; the only question is the reduction in length for acceptable
>levels of operation.

Sorry, but I have to disagree here.
Basically, both cables should at least comply to the same specs.
That, I agree to.

But from a practical point of view:
A typical thicknet cable is Belden 9880.
6.9 dB per 100 feet attenuation at 1000 MHz, 11.5 dB at 2.5 GHz
(So it would be suitable for 'Wifi')

A typical thinnet cable is Belden 9907
14.8 dB per 100 feet at 1000 MHz, 2.5 GHz not even specified :-)

As you see: size does matter, at least if you want to run
something else over an existing cable.

--
Kind regards,
Gerard Bok
Anonymous
February 21, 2005 7:30:54 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

In article <421a0b11.3798725@News.Individual.NET>,
Gerard Bok <bok118@zonnet.nl> wrote:
>On Mon, 21 Feb 2005 07:43:07 -0800, Rich Seifert
><usenet@richseifert.com.invalid> wrote:
>
>>In article <4219c80b.541526@News.Individual.NET>,
>> bok118@zonnet.nl (Gerard Bok) wrote:
>>
>>> On Sun, 20 Feb 2005 17:10:33 -0800, Rich Seifert
>>> <usenet@richseifert.com.invalid> wrote:
>>>
>>> >There was an early proposal for a version of FDDI that ran over coax; it
>>> >was referred to as LDDI. DEC also made a proprietary product (pre-FDDI)
>>> >that ran on thick Ethernet coaxial cable; it was used for the back-end
>>> >connection in the original VAXCluster product.
>>>
>>> But the crucial question hete is: would this product run more
>>> than 10 Mbps over 'thinnet coax' ? :-)
>>
>>Since the impedance of 'thinnet coax' is identical to that of thick
>>Ethernet coax, clearly any system that will run on one will run on the
>>other; the only question is the reduction in length for acceptable
>>levels of operation.
>
>Sorry, but I have to disagree here.
>Basically, both cables should at least comply to the same specs.
>That, I agree to.
>
>But from a practical point of view:
>A typical thicknet cable is Belden 9880.
>6.9 dB per 100 feet attenuation at 1000 MHz, 11.5 dB at 2.5 GHz
>(So it would be suitable for 'Wifi')
>
>A typical thinnet cable is Belden 9907
>14.8 dB per 100 feet at 1000 MHz, 2.5 GHz not even specified :-)
>
>As you see: size does matter, at least if you want to run
>something else over an existing cable.
>
>--
>Kind regards,
>Gerard Bok


The spec ifor TW is primarily impedance and mechanical dimensions
which just have to match the BNC connector used on each end (N for
thickwire cable).

The loss specification is either "don't care" for small networks or
"lower the better" if you are pushing the distance limits. These
belden part numbers are used for applications other than Thinwire and
designers need to know loss numbers. I don't imagine there is a big
market for TW, or thickwire, these days.

You are not talking about pulling new wire. You've gotta use what
you've got which, ISTR you've said has no identifying info on it. I
hope it didn't come from Rat Shack.

Your only hope to use your cable, ISTM, is to buy a pair of "g" wifi
APs that have screw-on antenna connectors and coax adapters
(SMC-to-BNC, or whatever) to connect to the cable and go for it. It
might work. This is not big bucks and if the test fails you can sell
the APs to someone. I guess you set them up as a mesh bridge.

IMO, on a good day, "g" will be modestly faster than 10mb ethernet,
measured by actual thruput.








--

a d y k e s @ p a n i x . c o m

Don't blame me. I voted for Gore.
Anonymous
February 21, 2005 7:30:54 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

Gerard Bok wrote:

>>Since the impedance of 'thinnet coax' is identical to that of thick
>>Ethernet coax, clearly any system that will run on one will run on the
>>other; the only question is the reduction in length for acceptable
>>levels of operation.
>
> Sorry, but I have to disagree here.
> Basically, both cables should at least comply to the same specs.
> That, I agree to.
>
> But from a practical point of view:
> A typical thicknet cable is Belden 9880.
> 6.9 dB per 100 feet attenuation at 1000 MHz, 11.5 dB at 2.5 GHz
> (So it would be suitable for 'Wifi')
>
> A typical thinnet cable is Belden 9907
> 14.8 dB per 100 feet at 1000 MHz, 2.5 GHz not even specified :-)
>
> As you see: size does matter, at least if you want to run
> something else over an existing cable.

1) Last I heard, ethernet over coax does not use frequencies anywhere near
2.5 GHz.

b) I guess you missed the point "the only question is the reduction in
length for acceptable levels of operation".
Anonymous
February 21, 2005 7:30:55 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

James Knott wrote:

> Gerard Bok wrote:
>
>>>Since the impedance of 'thinnet coax' is identical to that of thick
>>>Ethernet coax, clearly any system that will run on one will run on the
>>>other; the only question is the reduction in length for acceptable
>>>levels of operation.
>>
>> Sorry, but I have to disagree here.
>> Basically, both cables should at least comply to the same specs.
>> That, I agree to.
>>
>> But from a practical point of view:
>> A typical thicknet cable is Belden 9880.
>> 6.9 dB per 100 feet attenuation at 1000 MHz, 11.5 dB at 2.5 GHz
>> (So it would be suitable for 'Wifi')
>>
>> A typical thinnet cable is Belden 9907
>> 14.8 dB per 100 feet at 1000 MHz, 2.5 GHz not even specified :-)
>>
>> As you see: size does matter, at least if you want to run
>> something else over an existing cable.
>
> 1) Last I heard, ethernet over coax does not use frequencies anywhere near
> 2.5 GHz.

Ethernet over coax doesn't. It was suggested that 802.11a, b, or g be
tried. Those run at 2.5 or 5 GHz.

> b) I guess you missed the point "the only question is the reduction in
> length for acceptable levels of operation".

--
--John
to email, dial "usenet" and validate
(was jclarke at eye bee em dot net)
Anonymous
February 22, 2005 1:19:05 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

sphealey wrote:
>>>Trying to patch a bad situation for a few months until we can get a
>>>fiber contractor into our semi-rural mfg facility. Is there
>
> anything
>
>>>faster than 10 Mb that I can run over Thinnet cable?
>>
>>Can you go wireless? You didn't say what the distance was, or if
>
> there
>
>>were obstructions that would make this impossible.
>
>
> Good thought, but I truely doubt wireless would work in that
> facility/environment.
>
> sPh

Well don't dismiss it too quickly. If you know you have lots of 2.4Ghz
noise, for example, use 802.11a (5.8Ghz). Or vice-versa.

Considering you can get clean, LOS ranges of several miles outdoors with
commonly-available antennae, I bet there's a chance you can make this
work indoors. What's the approximate distance?

-Jonathan
Anonymous
February 22, 2005 1:29:00 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

sphealey wrote:
> William wrote:
>
>>"sphealey" <sphealey@worldnet.att.net> wrote:
>>
>>>faster than 10 Mb that I can run over Thinnet cable?
>>
>>Bail us out here, and tell us how long the existing cable is! 8*)
>>
>>[I kinda like the idea of using the cable _shield_ as an antenna!]
>
>
> Sorry dudes - I had to go back and look at my own original post as I
> thought I had included that information. Which I had not ;-(
>
> Length is over 100m - about 125 I think. Longer than 10' for sure.
>
> sPh
>


125m is definitely a possibility for wireless, as long as you have LOS.
Even if it's somewhat noisy, I think you stand a good chance of being
able to get over 10mbps. You will need some good Pringles antennas. :-)

-Jonathan
Anonymous
February 22, 2005 3:41:32 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

> 125m is definitely a possibility for wireless, as long
> as you have LOS. Even if it's somewhat noisy, I think
> you stand a good chance of being able to get over 10mbps.
> You will need some good Pringles antennas.

This is a factory environment with changes of levels, metal grating,
steel roofing, and a welding shop with what amounts to a faraday cage
around it in between! No LOS here I am afraid. That is why we are
putting in fiber, but finding fiber contractors willing to do
industrial quality work in rural locations for less than the book value
of the entire facility is difficult.

I think we are going to have to call this one dead and buried. Anyone
know where I can go to fiber splicing school?

sPh
Anonymous
February 22, 2005 8:14:01 PM

Archived from groups: comp.dcom.lans.ethernet (More info?)

sphealey wrote:

>> 125m is definitely a possibility for wireless, as long
>> as you have LOS. Even if it's somewhat noisy, I think
>> you stand a good chance of being able to get over 10mbps.
>> You will need some good Pringles antennas.
>
> This is a factory environment with changes of levels, metal grating,
> steel roofing, and a welding shop with what amounts to a faraday cage
> around it in between! No LOS here I am afraid. That is why we are
> putting in fiber, but finding fiber contractors willing to do
> industrial quality work in rural locations for less than the book value
> of the entire facility is difficult.
>
> I think we are going to have to call this one dead and buried. Anyone
> know where I can go to fiber splicing school?

If I might make a suggestion, _try_ two pulls of UTP, one for your longest
run and one that you believe to be your worst case for interference, and
see what happens. You may be pleasantly surprised. And if you aren't you
haven't spent much.

If the regs where you are located allow you to pull your own fiber, then one
option would be to measure out the pulls and order connectorized cable in
appropriate lengths, and just be careful not to pull it too hard, bend it
too tight, or bust the connectors. Don't know if it's still the case but
once you needed about a thousand bucks worth of tooling to put connectors
on fiber.

> sPh

--
--John
to email, dial "usenet" and validate
(was jclarke at eye bee em dot net)
!