We've got a PC over there <- and a APC UPS over there-> around 10 metres
apart, don't ask!, its the way it is;!.

We need to hook up that to the PC for monitoring etc these units use USB
for that.

NE one know if its OK to extend them or daisy chain a few together for
that sort of length?..


TIA...
--
Tony Sayer

"tony sayer" <tony@bancom.co.uk> wrote in message
news:L3NBTQDMAJQFFw8i@bancom.co.uk...
The maximum is generally 5 metres.
If you need to go 10 put a hub in the middle.

In article <45409b1f$0$21511$afc38c87@news.optusnet.com.au> , Neil Green
<nrgreenNOSPAM@optusnet.com.au> writes
Duh!, hub yes of course

cheers M8!...
--
Tony Sayer

On Thu, 26 Oct 2006 22:45:08 +0100, tony sayer <tony@bancom.co.uk>
wrote:

I believe that data speed transfer between PC & that UPS would be /is/
rather slow (serial port replacement on UPS) & no need for extra
powering the client, since UPS has its own powering Usb port IMHO ...,
so IMHO would work only with quality USB cable extending a line ...
I´ve seen/read cases working with daisy chaining Usb cables up to 13m
w/o hub cases, but that could be a lottery chosing right one (could
work or even not, since its over the specs) ...

Some my Net-Pal managed to run an Usb powered WebCam (it also
does not need much DC powering; with shorting serial 22ohm resistors
on cams data lines internal circuitry) over 20m BUT using UTP5e (Lan)
cable ... so, IMHO replacing original Usb cable with that kind of
cable should work for sure for "only" 10m w/o adding a hub ...

I also "developed" a solution for longer distances (over 20m)
& power demanding devices like Usb-WLAN-Keys w/o using powered hubs
or data signal boosters (purpose: cheap DIY active [Usb] wifi antenna
connections-no RF signal loss!) using cheap common 75ohm Sat coax,
which should also work BUT I still did NOT TESTED it ! ...

/a day before popped me that idea, mine UsbKey died when performing
some other experiments ... will get a cheap replacement some day to
test myself also this last idea .../ if someone interested, here is
a schematic to test (if someone willing to do it by himself..)

http://193.189.160.28/seibert/HP/Usb...bLongCable.gif

So, if someone tests that stuff before me, give me a note (or mail)
please ...






--
Regards , SPAJKY ®
mail addr. @ my site @ http://www.spajky.vze.com
4y - "Tualatin OC-ed / BX-Slot1 / inaudible setup!"

Spajky wrote:

Your circuit will work ! It is though very dangerous !
Ignoring the fact that you have used co-axial cable outer sheath to
provide DC power to the remote end, you have no way of guaranteeing
that the ground potential is going to be the same at each USB
connection ! That is no voltage difference between the negative
connections. Just the induced voltages due to cable length will
potentially cause issues. Which is why twisted pair should be used, as
in Cat5. Also to protect the computer USB ports transformer Isolation
should be used.


--
Baron:

On Sat, 28 Oct 2006 12:19:50 +0100, Baron <me@home.com> wrote:

Well, ´till I test it, I am not 100% sure (still could be some data
signal phase problems ...)

good to have pointed to that ...

Yes, that could be the problem, when having connected not just only
one or mixed powered & non powered "Usb clients" or hubs together or
daisy chained ... but as I recall, negative (4th usb pin) & grounds on
"clients" side should not be tied/connected together as a standard
states ...

But anyway, the schematic idea was primerly ment only for single Usb
port use with single non-auto /bus/ powered power-hungry device like
WLAN UsbKey ...

Yes, that could cause some problems maybe ...

IMHO that won´t work, since Usb is practically half-duplex serial on
single twisted pair & DC part on data lines serves to switch mode of
operation too .. (its not like for ethernet) ... but IMHO
opto-couplers should do the trick in that cases .. Usually for
extending up to 1km the Usb connection, the RS-4** chips are used for
twisted pair line ... but I wanted simpliest/cheapest solution ; well,
nothing is idiotproof ... :-) , always something could go wrong ..

.... but anyway, thanks for opinions ... you know, anytime you try to
do something "special" & out of ordinary stuff, problems could arise &
a desire not to fry something :-)) is always present ...

--
Regards , SPAJKY ®
mail addr. @ my site @ http://www.spajky.vze.com
4y - "Tualatin OC-ed / BX-Slot1 / inaudible setup!"

Spajky wrote:

If you have full control over what happens at both ends ! Someone could
use your circuit to tie two disparate systems together !

Yes that would be fine !

Actually it should. A transformer would only need to carry the signal
pair ! Then isolated DC supplies could be used at the ends. It would
be similar to "Power over Ethernet" topology.

If you are into salvage ! the signal transformers from junked 56K modems
work fine !

I've fried plenty in my time ! Trouble is making an assumption that
catches you out.

--
Regards:
Baron:

On Sun, 29 Oct 2006 10:16:47 +0000, Baron <me@home.com> wrote:

good to warn them this way :-)

Thanks for the tip ! :-) (good to know such stuff) ...
--
Regards , SPAJKY ®
mail addr. @ my site @ http://www.spajky.vze.com
4y - "Tualatin OC-ed / BX-Slot1 / inaudible setup!"

On Sun, 29 Oct 2006 05:44:19 +0100, Spajky <Spajky@##volja.net> wrote:

I tested it this weekend & it WORKS like a charm ... for over 20m
away :- )))
--
Regards , SPAJKY ®
mail addr. @ my site @ http://www.spajky.vze.com
4y - "Tualatin OC-ed / BX-Slot1 / inaudible setup!"

Spajky wrote:

See I told you so...... ;-)

--
Baron:

On Tue, 21 Nov 2006 20:05:42 +0000, Baron <me@home.com> wrote:

yes, but I did also some small corrections on that schematics :-)

.... will draw also another simplier for self powered devices too next
days ... will post the link here also ...
--
Regards , SPAJKY ®
mail addr. @ my site @ http://www.spajky.vze.com
4y - "Tualatin OC-ed / BX-Slot1 / inaudible setup!"

On Wed, 22 Nov 2006 05:04:15 +0100, Spajky <Spajky@##volja.net> wrote:

http://193.189.160.28/seibert/HP/Usb...LongCable2.gif

.... good for currents up to 100mA of consumption ...
(or 200mA for 10,5m for example + 2x half standard Usb shielded cable
added at the ends...)
--
Regards , SPAJKY ®
mail addr. @ my site @ http://www.spajky.vze.com
4y - "Tualatin OC-ed / BX-Slot1 / inaudible setup!"

Well, here is something for all of you. Maybe you can shed some light
with this.

First: I am CHEAP! Point blank! So I look for the cheapest route to do
things.

First-a: I have little patience to some things. (like pcb creating -
not so simple)

Second: I have a box of Cat5e cable (250 feet) here that I have had for
the longest. [it's not going anywhere until I use every single
millimeter of it.]

Third: I have a USB - 2 foot extender cable ( NOTHING special, just the
connectors and wire. )

So the CHEAPEST thing for me to do was to cut the USB extender cable in
half and pull about 30 feet of the cat5e out. Connect 2 pair of the
cat5e on each side to one part each of the cut in half extender.

What next? I connected my Logitech USB Wheel Mouse (M-BD69) to the
'new' extended cable I just made up and it works. I haven't noticed any
lag or mis-clicks. I have NO relays or "active" components in between
the mouse and my laptop.


*** ASCII Diagram:

|Laptop|--[USB-A]---{30 foot cat5e cable}---[USB-A]--|Mouse|


wait, give me a sec to try something out........

Ok. My 6Gb External Bafo 2.5 inch Hard Disc Drive was registered as a
non-high speed device. Actually. What it says is that the device can
perform faster if connected to a high speed USB 2.0 port. Which it
doesn't say when connected without my 'extension' cause my ports ARE
2.0.

So what ideas do you guys have on that? Maybe it only matters on what
the device you are using is? Like for instance the mouse and the HDD?
Or maybe different types of keyboards? Size of the device? Data
transmitted? Power consumed? All of those factors can be either
beneficial or burdens to an extension? What say you?

On 5 Dec 2006 19:33:28 -0800, igounfazed@gmail.com wrote:

for low power or self powered devices may the Cat5e cable work Ok;
also for lower speeds like mouses or keyboards. Usb2.0 refers to
inside packets data rate, but the actual data speed may be reduced
because of CRC errors & repeating packets data sending. So the way of
knowing if it truely works faster is only to measure by copying data
from to HD & viceversa. Under certain conditions/lenght/data speed
using FTP cable works & is really cheapest & easyest solulution ...
So if it work for your needs, thats Ok ! :-)
--
Regards , SPAJKY ®
mail addr. @ my site @ http://www.spajky.vze.com
4y - "Tualatin OC-ed / BX-Slot1 / inaudible setup!"

Yeah. I see what you are saying. I am having no problems with what I
intend to use the extension for. I am only going to be using the
extensions (2 of them) for a mouse and a keyboard.

Basically, I am making a room to room extension with the ports on wall
outlets. I am putting an LCD monitor in the kitchen and the keyboard
and mouse. So in my walls I will have an extension of VGA cable and
cat5e for two usb ports. And I am thinking of running another 2-wire
for an audio extension in case I ever upgrade the LCD to one with
speakers.

Not a bad idea if I can use the cat5e for the usb ports.

And I am having NO problems with it yet. So I am sure to use the cat5e.

I am going to do a couple of more tests with this setup.

I have played video games (halo for the pc) with both the mouse and the
keyboard extended and have suffered no response errors or input
latency. ( I still get good kill scores, lol. )

I will post some of my findings on this post for any future people
interested (if you guys don't mind that is).

tony sayer wrote:
i vaguely recall like 1.8M for USB and 5M for ps2 But that;s probably
wrong. Nevertheless, it's limited.

USB booster cables are probably closest to chaining them together. Next
would be USB hub. A cat5 USB extender isn't chaining them together

options are
USB booster cables
USB hubs
Cat5 USB extender

jameshanley39@yahoo.co.uk wrote:
<snip>

has other names besides 'booster'. Like Active and Repeater .
that may help trying to find it