Help me , running long cables ...

[rishi]

I have a project to log the keypresses from an hall of 60m X 50m area.
There are around 500 switches in this area(actually there r 500 users
sitting in this hall, each table has one toggle switch,reset switch).

Tables are arranged as 10cols X 50 rows.

I was thinking for a simple matrix logic as its extremely simple and
would be cost effective in terms of cabling costs as well as other
hardware required.

I have a requirement of running 60 m cables with ttl signals.Is it
possible with using BUS Buffers/transreceivers?

ON the computer end i will use 60 bit digital I/O board and write the
software which will scan rows and columns continuosly.

Will be feasible or there is any better alternative...

Please , can u suggest me on this.

Thanks alot
Best Regards
Rishi Bhanot

 

[[email protected]]

I have a project to log the keypresses from an hall of 60m X 50m area.
There are around 500 switches in this area(actually there r 500 users
sitting in this hall, each table has one toggle switch,reset switch).

Tables are arranged as 10cols X 50 rows.

I was thinking for a simple matrix logic as its extremely simple and
would be cost effective in terms of cabling costs as well as other
hardware required.

I have a requirement of running 60 m cables with ttl signals.Is it
possible with using BUS Buffers/transreceivers?

Not well. Capacitive loading on cable runs that long will do you
in. And remember you are really going twice that distance.

Using 1488/1489 transmitters/receivers at a lowish rate (10's of kHz)
might be reasonable... but you will not be able to scan if someone
holds their button down.

ON the computer end i will use 60 bit digital I/O board and write the
software which will scan rows and columns continuosly.

Will be feasible or there is any better alternative...

I would suggest a "smart" button and a shared bus (two wires for signal,
a third for power) which is used to send keypresses back to the PC. Each
button has a serial number, and on a change of state it reports back the
new state and serial number.

Tim.

 

[Jeroen Vriesman]

Hi.

500 smart buttons is a lot of hardware, abd then there are collisions on
your serial bus to handle.

But I think it's possible to use one cable for 5 buttons, and put a
resistor in series which each button with a value of (2^N)*R0, (eg. 1k 2k
4k ... 16k ), and use a current source on the cable, and an AD convertor
to get an 8 bit value (computer rounding to 5 bits).

8 bit AD convertors are fast enough to sample all hundred cables, you
still need 100 cables, but only one AD convertor.

I can design and/or build it for you if needed.

Cheers,
Jeroen

joenix at gmx dot net

 

[Wim Ton]

a third for power) which is used to send keypresses back to the PC. Each

 

[Wim Ton]

I would suggest a "smart" button and a shared bus (two wires for signal,
a third for power) which is used to send keypresses back to the PC. Each
button has a serial number, and on a change of state it reports back the
new state and serial number.

Tim.

Something for a these Dallas/Maxim 1 wire devices ?

Wim

 

[Tim Wescott]

Not well. Capacitive loading on cable runs that long will do you
in. And remember you are really going twice that distance.

Using 1488/1489 transmitters/receivers at a lowish rate (10's of kHz)
might be reasonable... but you will not be able to scan if someone
holds their button down.


I would suggest a "smart" button and a shared bus (two wires for signal,
a third for power) which is used to send keypresses back to the PC. Each
button has a serial number, and on a change of state it reports back the
new state and serial number.

Tim.

I would either take the suggestion here and use a smart button, or I'd
investigate using a keyboard matrix type of circuit using balanced pairs.
For either of these ideas you should use twisted-pair wire (preferrably
shielded) and pay attention to suppressing reflections by making sure it's
properly terminated.

For the smart button idea you'd want to find the smallest Atmel AVR or
Microchip PIC with a serial interface. Use RS-485 (for which many drivers
exist) on twisted pair and terminate it at both ends with the correct
resistance (most twisted pair has an impedence of about 100 ohms, and close
is good enough). Invent a little serial protocol that is robust in the face
of collisions. You'd have to manufacture 500 smart buttons, but at that
volume a PC board should be cost effective. The nice thing about this
arrangement is that the PC will only need an external RS-232 to '422
adaptor, your system will be more expandable and the cabling will be easier.

The keyboard matrix idea should also get you where you want to go with lots
more sweat on the cabling and lots less on the software. Use RS-422
transmitters and receivers on the PC, and figure out how to do
keyboard-style blocking diodes on balanced-pair lines. You'll have
reflections at the switches because the cable won't be seeing a matched
impedance. The receivers are terminated, so they'll absorb reflections on
the receive side. To do the same on the transmit side you'll want to add a
50 ohm resistor in series with each leg of the transmitter. This _should_
result in a system that will work for you, as long as you don't have
everyone in a row pushing their button at once and loading down the poor
transmitter.

 

[Luhan Monat]

I have a project to log the keypresses from an hall of 60m X 50m area.
There are around 500 switches in this area(actually there r 500 users
sitting in this hall, each table has one toggle switch,reset switch).

Tables are arranged as 10cols X 50 rows.

I was thinking for a simple matrix logic as its extremely simple and
would be cost effective in terms of cabling costs as well as other
hardware required.

I have a requirement of running 60 m cables with ttl signals.Is it
possible with using BUS Buffers/transreceivers?

ON the computer end i will use 60 bit digital I/O board and write the
software which will scan rows and columns continuosly.

Will be feasible or there is any better alternative...

Please , can u suggest me on this.

Thanks alot
Best Regards
Rishi Bhanot

Hi,

Well, I would use a pic12f629 at each location and do serial
communication in software. Run just 3 wires: power, ground, and
bi-directional signal. Each pic has a serial number (in the internal
eeprom) and is interrogated by the central PC. Button pushes are stored
in the pic's until the PC interrogates it.

No collisions and the buttons will probably cost more than the pics for
each location.

 

[Trevor Matthews]

Try this:

Use 500x DS2405 (1-Wire addressable switch) - run three (twisted) wires
to all devices (each DS2405 is in parallel to all of the others) - +5V,
GND and data.

The PIO pin is pulled to +5V via some largish value resistor, and pulled
to ground via the switch. Connect the GND and data lines to each DS2405.
Then use any uController/uProcessor/PC with a spare serial port to
communicate over the 1-wire bus. (See www.maxim-ic.com use the 1-wire
links for information on how to do this)

Code the computer, and do a "Active Search" of the 1-wire bus which will
return all serial nos (8 bytes) of all DS2405 with their PIO pin pulled
low - examine the datasheets from Maxim's site above for the details.

The biggest hassle will probably be activating each switch to determine
the serial # of the associated DS2405 - so you can determine how the
serial # relates to which switch is pressed.

HTH

Trevor

 

[Mac]

I have a project to log the keypresses from an hall of 60m X 50m area.
There are around 500 switches in this area(actually there r 500 users
sitting in this hall, each table has one toggle switch,reset switch).

Tables are arranged as 10cols X 50 rows.

I was thinking for a simple matrix logic as its extremely simple and
would be cost effective in terms of cabling costs as well as other
hardware required.

I have a requirement of running 60 m cables with ttl signals.Is it
possible with using BUS Buffers/transreceivers?

ON the computer end i will use 60 bit digital I/O board and write the
software which will scan rows and columns continuosly.

Will be feasible or there is any better alternative...

Please , can u suggest me on this.

Thanks alot
Best Regards
Rishi Bhanot

Rishi. If you just need to detect key-presses, I think the cabling is OK.
You can put a pull-up (2.2k) and a capacitor on each line, near where the
logic senses it. Then put a 100-Ohm resistor in series with the wire. The
switches can be normally-open types. (view with courier or similar)

VCC
|
/
\ 2.2k
/
\
|
/| |
/ | | 100 60m wire
< +---+----/\/\/\/-------------------------+
\ | | |
\| = C \
| \ switch
| |
+------------------------------------+
| 60m wire
GND

So when a switch is closed, the logic will detect a low.

There are some things you will have to be careful with. For example, if
the capacitor is not big enough, or too big, the logic may switch more
than once per press. Also, there are a bunch of issues associated with the
scan speed. If you scan fast, (>10 Hz) and they press slow, there should
be no problem.

What I don't understand is how you plan to multiplex your 60 IO lines into
500 (or 1000?) switches. Is that where you would use the bus receivers? I
guess it could work. Logically you have 10 banks of 50-line bus receivers.
In reality, each bank may be more than one chip. The banks are all
connected together, and at any given time, only one bank is active,
and the others are disabled with their drivers in a high impedance state.

The 10 remaining lines on your IO board can be configured as outputs, and
each one can select one of the banks.

It's going to be a pretty large circuit. I hope you can have a board made
for it, because I wouldn't want to prototype it with point to point wires!

The other option is trying to find an IO board with 500 IO's on it.

Mac
--

 

[Frank Bemelman]

I have a project to log the keypresses from an hall of 60m X 50m area.
There are around 500 switches in this area(actually there r 500 users
sitting in this hall, each table has one toggle switch,reset switch).

Tables are arranged as 10cols X 50 rows.

I was thinking for a simple matrix logic as its extremely simple and
would be cost effective in terms of cabling costs as well as other
hardware required.

I have a requirement of running 60 m cables with ttl signals.Is it
possible with using BUS Buffers/transreceivers?

ON the computer end i will use 60 bit digital I/O board and write the
software which will scan rows and columns continuosly.

Will be feasible or there is any better alternative...

Please , can u suggest me on this.

There have been made a number of suggestions. 500 tables
are a lot of tables. I would never use a single serial
bus to poll 500 controllers, for so many tables. While the
cabling will be cheap, you will end up in a mental hospital.
Or use one controller for ten tables, so that you only
need to poll 50 of those controllers.

A simple matrix is not a bad idea at all. Perhaps it
helps to make more smaller matrixes. 5 of 10 by 10.

Another option is a parallel addressing scheme, each table
has a small blackbox, with a dipswitch to set the address.
With 9 bits, you can address 512 units. You could use a PIC
or perhaps better to use standard cmos logic.

I'd go for a simple matrix.

 

[mike]

There have been made a number of suggestions. 500 tables
are a lot of tables. I would never use a single serial
bus to poll 500 controllers, for so many tables. While the
cabling will be cheap, you will end up in a mental hospital.
Or use one controller for ten tables, so that you only
need to poll 50 of those controllers.

A simple matrix is not a bad idea at all. Perhaps it
helps to make more smaller matrixes. 5 of 10 by 10.

Another option is a parallel addressing scheme, each table
has a small blackbox, with a dipswitch to set the address.
With 9 bits, you can address 512 units. You could use a PIC
or perhaps better to use standard cmos logic.

I'd go for a simple matrix.

Give us more detail on what you're trying to do.
For a general solution, you might want to be able to tell the
difference betwen Yes, no, abstain and unused.
Toggle switches may be a nono for unmanned positions.
You may need to latch the votes simultaneously.
You may need the capability to reset the latched vote.
You may need some resistance to "stuffing the ballot box".
You may want to be able to run the system from ANY laptop.
Bite the bullet, put a microcontroller in each switch.
Polling 500 addressable controllers should be a piece o cake. yes??
Isn't that what RS422 is all about? Or use I2C or whatever is most
compatible with the power distribution system you select.
You can break it up into a matrix and still serial poll each group.

Once you have the smart system, you can easily react to the
inevitable changes in requirements or system upgrades. You could do
a LOT of reconfiguration just by plugging each unit into a
reprogrammer. If you're clever and use a bootloader, you could
reconfigure the pods on the fly. The LAST thing you want is to discover
a flaw when you have 500 unconfigurable boxes built.

Maximum reliable transmission speed is inversely related to cable length.
They did get pictures back from the outer solar system decades ago.
That was some LONG cable...

Give a lot of thought to failure modes. The system should still work
if you short (or open) one of the cables to a pod. You need diagnostic
modes to pinpoint failures.

mike

--
Bunch of stuff For Sale and Wanted at the link below.
laptops and parts Test Equipment
Honda CB-125S
TEK Sampling Sweep Plugin and RM564
Tek 2465 $800, ham radio, 30pS pulser
Tektronix Concept Books, spot welding head...
http://www.geocities.com/SiliconValley/Monitor/4710/

 

[CWatters]

Probably too expensive...

http://www.banxia.com/impact/wireless-keypads.html

 

[Frank Bemelman]

Hi,

Well, I would use a pic12f629 at each location and do serial
communication in software. Run just 3 wires: power, ground, and
bi-directional signal. Each pic has a serial number (in the internal
eeprom) and is interrogated by the central PC. Button pushes are stored
in the pic's until the PC interrogates it.

No collisions and the buttons will probably cost more than the pics for
each location.

What do you do, after everything is installed, and it doesn't work
as expected?

Imagine, there is more noise than anticipated. You decide to lower
the baudrate and add some redundancy in the protocol.

for(n=0; n<499; n++)
{ MoveYourLaptopToTable(n);
PullPlug(n);
for(n2=0; n2<1; n2++)
{ UnscrewMounts(n2);
}
for(n2=0; n2<4; n2++)
{ UnscrewCoverScrew(n4);
}
ConnectISPtoBox(n);
DownloadSoftware(n);
ReprogramSerialNumber(n);
VerifyBox(n);
DisconnectISPfromBox(n);
for(n2=0; n2<4; n2++)
{ ScrewCoverScrew(n4);
}
for(n2=0; n2<1; n2++)
{ ScrewMounts(n2);
}
InsertPlug(n);
}

I hope the new software works. I also hope
you never experience a zapped driver that
shorts the serial bus:

for(n=0; n<499; n++)
{ MoveToTable(n);
DiveUnderTable(n);
PullPlug(n);
if(YourBuddySaysTheBusIsFine())
{ GetSpareUnit();
ProgramSerialNumberIntoSpareUnit();
MountSpareUnit();
InsertPlug(n);
if(YourBuddySaysTheBusIsFine())
{ GetFromUnderTableWithoutBangingYourHead();
break;
}
}
InsertPlug(n);
GetFromUnderTableWithoutBangingYourHead();
}

Untested code!

 

[Watson A.Name - Watt Sun, Dark Remover]

[snip]

What I don't understand is how you plan to multiplex your 60 IO lines into
500 (or 1000?) switches.

What I don't understand is why you would want to use so much wiring
when you could use a single pair of wires and RS-485, and poll each
box with a unique address. This seems to me to be a place where a PIC
would show great promise as a solution.

Mac

--

--
@@F@r@o@m@@O@r@a@n@g@e@@C@o@u@n@t@y@,@@C@a@l@,@@w@h@e@r@e@@
###Got a Question about ELECTRONICS? Check HERE First:###
http://users.pandora.be/educypedia/electronics/databank.htm
My email address is whitelisted. *All* email sent to it
goes directly to the trash unless you add NOSPAM in the
Subject: line with other stuff. alondra101 <at> hotmail.com
Don't be ripped off by the big book dealers. Go to the URL
that will give you a choice and save you money(up to half).
http://www.everybookstore.com You'll be glad you did!
Just when you thought you had all this figured out, the gov't
changed it: http://physics.nist.gov/cuu/Units/binary.html
@@t@h@e@@a@f@f@l@u@e@n@t@@m@e@e@t@@t@h@e@@E@f@f@l@u@e@n@t@@

 

[Mac]

[snip]

What I don't understand is how you plan to multiplex your 60 IO lines into
500 (or 1000?) switches.

What I don't understand is why you would want to use so much wiring
when you could use a single pair of wires and RS-485, and poll each
box with a unique address. This seems to me to be a place where a PIC
would show great promise as a solution.

Mac

--

Well, it's not my problem, and I just tried to answer the poster's
question, but I would venture to say that keeping to a minimum the parts
that get multiplied by 500 might be a major design goal.

I like a lot of the solutions proposed, but I guess the OP will have to
make the final decision. I surmise that cost is a factor, so it would be
interesting to try to assess which of the proposed solutions is cheapest.

Mac
--

 

[Robert Monsen]

Hi,

Well, I would use a pic12f629 at each location and do serial
communication in software. Run just 3 wires: power, ground, and
bi-directional signal. Each pic has a serial number (in the internal
eeprom) and is interrogated by the central PC. Button pushes are stored
in the pic's until the PC interrogates it.

No collisions and the buttons will probably cost more than the pics for
each location.

The ID on a PIC12F629 isn't accessible to the processor while its running.

Regards,
Bob Monsen

 

[Robert Monsen]

Probably too expensive...

Not to mention they might walk off...

 

[CWatters]

Perhaps try something like this. It's a one wire serial addressable I/O
device (each chip has it's own unique 64 bit address burnt in). It has 8 I/O
lines so each switch box could support a number of switches or a small
matrix.

http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3818/ln/en

 

[Watson A.Name - Watt Sun, Dark Remover]

[snip]

What I don't understand is how you plan to multiplex your 60 IO lines into
500 (or 1000?) switches.

What I don't understand is why you would want to use so much wiring
when you could use a single pair of wires and RS-485, and poll each
box with a unique address. This seems to me to be a place where a PIC
would show great promise as a solution.

Mac

--

Well, it's not my problem, and I just tried to answer the poster's
question, but I would venture to say that keeping to a minimum the parts
that get multiplied by 500 might be a major design goal.

I like a lot of the solutions proposed, but I guess the OP will have to
make the final decision. I surmise that cost is a factor, so it would be
interesting to try to assess which of the proposed solutions is cheapest.

Mac
--

A whole bunch of wireless keyfob type buttons might save a whole lot
on wiring.


--
@@F@r@o@m@@O@r@a@n@g@e@@C@o@u@n@t@y@,@@C@a@l@,@@w@h@e@r@e@@
###Got a Question about ELECTRONICS? Check HERE First:###
http://users.pandora.be/educypedia/electronics/databank.htm
My email address is whitelisted. *All* email sent to it
goes directly to the trash unless you add NOSPAM in the
Subject: line with other stuff. alondra101 <at> hotmail.com
Don't be ripped off by the big book dealers. Go to the URL
that will give you a choice and save you money(up to half).
http://www.everybookstore.com You'll be glad you did!
Just when you thought you had all this figured out, the gov't
changed it: http://physics.nist.gov/cuu/Units/binary.html
@@t@h@e@@a@f@f@l@u@e@n@t@@m@e@e@t@@t@h@e@@E@f@f@l@u@e@n@t@@

 

[Pat Ford]

[snip]

What I don't understand is how you plan to multiplex your 60 IO lines into
500 (or 1000?) switches.

What I don't understand is why you would want to use so much wiring
when you could use a single pair of wires and RS-485, and poll each
box with a unique address. This seems to me to be a place where a PIC
would show great promise as a solution.

Mac

--

Well, it's not my problem, and I just tried to answer the poster's
question, but I would venture to say that keeping to a minimum the parts
that get multiplied by 500 might be a major design goal.

I like a lot of the solutions proposed, but I guess the OP will have to
make the final decision. I surmise that cost is a factor, so it would be
interesting to try to assess which of the proposed solutions is cheapest.

Mac

What about a dallas 1 wire network?
Pat