Searching \ for '[EE]: High voltage line driver part needed' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/power.htm?key=voltage
Search entire site for: 'High voltage line driver part needed'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: High voltage line driver part needed'
2002\04\25@231130 by Chris Eddy

flavicon
face
I have one for all of my electron buddies out there.

I have a long distance DC power/communications bus. I need to design a
device to attach to this line, that will receive and transmit data on
this pair. The pair will have a differential of 24VDC nom on it. The
transmitter must send out a 0v/5v balanced drive, but must withstand the
24V when not sending.

I have used an LM311 to receive in the past, which works well. I must
now add the transmit option.

Typical DS75176 parts are only rated for -7V/+12V on the pair.

Must have low ICC as well. The normal 20/30mA is too high.

Ideas?
Chris~

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email spam_OUTlistservTakeThisOuTspammitvma.mit.edu with SET PICList DIGEST in the body


2002\04\26@112946 by Dwayne Reid

flavicon
face
At 10:57 PM 4/25/02 -0400, Chris Eddy wrote:
>I have one for all of my electron buddies out there.
>
>I have a long distance DC power/communications bus. I need to design a
>device to attach to this line, that will receive and transmit data on
>this pair. The pair will have a differential of 24VDC nom on it. The
>transmitter must send out a 0v/5v balanced drive, but must withstand the
>24V when not sending.

data rate?

dwayne


Dwayne Reid   <.....dwaynerKILLspamspam@spam@planet.eon.net>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
(780) 489-3199 voice          (780) 487-6397 fax

Celebrating 18 years of Engineering Innovation (1984 - 2002)
 .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-
    `-'   `-'   `-'   `-'   `-'   `-'   `-'   `-'   `-'
Do NOT send unsolicited commercial email to this email address.
This message neither grants consent to receive unsolicited
commercial email nor is intended to solicit commercial email.

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestspamKILLspammitvma.mit.edu


2002\04\26@142801 by Chris Eddy

flavicon
face
Dwayne Reid wrote:
> data rate?

Negotiable, possibly even programmable on the job site. I picture
9600bps for a mile, or 2400bps for 10 miles.

Last night I looked over the CAN bus driver chips. The PCA82C251 is
designed to handle +/-36VDC on the bus.. could be just the ticket I was
looking for.

Chris~

--
http://www.piclist.com hint: To leave the PICList
.....piclist-unsubscribe-requestKILLspamspam.....mitvma.mit.edu


2002\04\26@155100 by Olin Lathrop

face picon face
> Negotiable, possibly even programmable on the job site. I picture
> 9600bps for a mile, or 2400bps for 10 miles.
>
> Last night I looked over the CAN bus driver chips. The PCA82C251 is
> designed to handle +/-36VDC on the bus.. could be just the ticket I was
> looking for.

The CAN collision detection protocol limits the speed*distance.  You can
have one or the other, but not both.  I haven't worked it out, but I'm
guessing that 2400 bits/sec over 10 miles is too much.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com hint: To leave the PICList
EraseMEpiclist-unsubscribe-requestspam_OUTspamTakeThisOuTmitvma.mit.edu


2002\04\26@181144 by John Dammeyer

flavicon
face
Hi Olin,

No no no no no.  8-)

The CAN protocol is a definition of how the bits behave and just like
UARTs there are devices that impliment that protocol:  SJA1000,  MCP2510
etc.

The ISO11898 specification defines the electrical bus part and the
82C250/251 series of drivers can be used for anything you like.  The CAN
protocol could be used with OC transitor drivers and it would still
work.

So now the question of whether the 11898 drivers would do the job.  It
has two bus driver output lines that are referenced to ground.  CAN_L
and CAN_H.  With a logic high input, both CAN_L and CAN_H are held at
2.5 Volts.  When the TTL input is brought low,  CAN_H moves to about
4.0V and CAN_L moves to about 1.0V.  Or in other words,  the difference
between them changes from 0V to about 3V.

Over 10 miles of wire the greatest problem to overcome will be both wire
resistance and capacitance.  The combination will slow down the slew
rate regardless of the driver used.  It's why TTY used 20ma baudot codes
but at very low bit rates and modems were invented to be used over
things like the telephone lines to bring the bit rates up.

Since the telephone line has as much as -48V on it and the ring voltage
is something like 100VAC at 20Hz one simple way to signal at 9600baud
(or even faster) on a 24VDC bus is to just use modems.  The transformers
provide the DC isolation and a 14400 is probably simple (and cheap) to
use over a 10 mile distance.

Like the discussion about jobs in [OT],  the idea is to come up with a
solution that works rather than refine a proposed solution that may not
work.  Of course, I haven't paid a lot of attention to this thread so I
may be all wet but mention CAN and my ears perk up.

<GRIN>

John Dammeyer


> {Original Message removed}

2002\04\26@221712 by Chris Eddy

flavicon
face
John Dammeyer wrote:
>
> The ISO11898 specification defines the electrical bus part and the
> 82C250/251 series of drivers can be used for anything you like.  The CAN
> protocol could be used with OC transitor drivers and it would still
> work.

Rightamundo. This is my target.

> Over 10 miles of wire the greatest problem to overcome will be both wire
> resistance and capacitance.  The combination will slow down the slew
> rate regardless of the driver used.

Granted, so the slower the rate the longer the push.

> Since the telephone line has as much as -48V on it and the ring voltage
> is something like 100VAC at 20Hz one simple way to signal at 9600baud
> (or even faster) on a 24VDC bus is to just use modems.  The transformers
> provide the DC isolation and a 14400 is probably simple (and cheap) to
> use over a 10 mile distance.

But the frequencies are much higher, IE 3000Hz, and I am concerned about
how far they will go. Plus, this system is a classic half duplex polled
network, and a 'trained' modem will not apply.

I have worked with other stuff that went 10 miles, but with;
1       Galvanic isolation (DC/DC's and optos)
2       SLOW SLOW SLOW at 300bps
3       Funky drivers like DS8831 (obsolete) for higher drive.
I know the old stuff worked, so I am hoping I can achieve a similar feat
with new technology.
I plan to run a synchronous comms protocol. I plan to apply 24VDC
alternating on the line (AC prevents corrosion of wires).
I hope that the CAN transceiver will recognize the wide swing of the
+/-24V.
The can drive will have to make it all the way back to the main line
driver/master.
Each device is line powered and communicates, thus are galvanically
isolated. This helps with ground loops and induced noise to a degree.
Nice low current loads allow me to run a hundred nodes at a mere amp.
Distant nodes will operate even at delta of 8VDC at the far end.

Chris~

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestspamspam_OUTmitvma.mit.edu


2002\04\26@231030 by John Dammeyer

flavicon
face
Hi Chris,

I've forwarded this email with the more detailed network description to
a CAN ISO 11898 chip designer. He can tell you if you will exceed or
cause problems with the signals you intend to superimpose.

Good luck.

John Dammeyer


> {Original Message removed}

2002\04\27@071828 by Olin Lathrop

face picon face
> > The CAN collision detection protocol limits the
> > speed*distance.  You can
> > have one or the other, but not both.  I haven't worked it out, but I'm
> > guessing that 2400 bits/sec over 10 miles is too much.
>
> No no no no no.  8-)
>
> The CAN protocol is a definition of how the bits behave and just like
> UARTs there are devices that impliment that protocol:  SJA1000,  MCP2510
> etc.
>
> The ISO11898 specification defines the electrical bus part and the
> 82C250/251 series of drivers can be used for anything you like.

True, but I was talking about CAN, not something that happens to use CAN
hardware.  We recently moved to a bigger office and my CAN documentation is
still in a box somewhere.  If I remember right, CAN uses collision detection
and backoff to arbitrate who gets to talk on the bus, much like ethernet.
There is a finite time window during which a sender writes a 0 bit to the
bus and detects some other sender writing a 1.  This dictates the minimum
time a bit edge must be able to travel the maximum length of the bus, go
thru a little bit of hardware processing, then travel back the full length
of the bus in time to be properly interpreted.  There are also a few other
situations where a response must be received within a time limit, but I
don't remember the details.  However, I don't think anything has tighter
timing than the collision detection.

There is an NMEA standard for communication between devices on a ship that
is layered on CAN.  They have done all this analysis for the type of cable
they specify and even show a table of maximum CAN bit rate versus maximum
bus length.  The numbers are conservative, but of course that's because they
want it to actually work in all real world situations.  I sorta remember
that the longest distance they showed was 500m and the bit rate for that was
100Kbits/sec or lower.

Again, this is from memory so the numbers may be off but I'm quite sure of
the general issue.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\04\27@133415 by John Dammeyer

flavicon
face
Hi Olin,

> > The ISO11898 specification defines the electrical bus part and the
> > 82C250/251 series of drivers can be used for anything you like.
>
> True, but I was talking about CAN, not something that happens
> to use CAN hardware.

OK.  Got it. Thought you meant what the subject line stated.

We recently moved to a bigger office
> and my CAN documentation is still in a box somewhere.  If I
> remember right, CAN uses collision detection and backoff to
> arbitrate who gets to talk on the bus, much like ethernet.

It's called CSMA/CR rather than Ethernet's CSMA/CD.  Carrier Sense,
Multiple Access, with, Collision Detection (Ethernet) or Collision
Recovery (CAN).


> There is a finite time window during which a sender writes a
> 0 bit to the bus and detects some other sender writing a 1.

You have it correct

> There is an NMEA standard for communication between devices
> on a ship that is layered on CAN.

There is also DeviceNet, SDS from Honeywell, CANOpen from CiA, CAN
Kingdom and many others more proprietary.
>
> Again, this is from memory so the numbers may be off but I'm
> quite sure of the general issue.

You've done well from memory.  There are CAN bit calculators either for
sale or reduced capabilty ones available off the net to determine not
only how fast for a given distance but will generate the correct init
values for the internal timing registers.

Just Google Search On Controller Area Network to find out more about the
world of CAN.

Opto Isolators or other Galvanic isolators all reduce the
distance/speed.  The packet ID, (11 bit or 29 bit) determines who wins
arbitration if two messages appear on the bus at exactly the same time.

What's most important for the PIC world, when it comes to CAN, is making
sure you can process the messages that arrive back to back worst case
before a new message ovewrites the unprocessed one.  That's the real
limitation to network speed;  not bus length or bit rate.

Regards,

John Dammeyer

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


More... (looser matching)
- Last day of these posts
- In 2002 , 2003 only
- Today
- New search...