[EE] 'Alcom Bus' communication system
Roger Weichert
I'm working on some gear made by DeLaval which uses what they call the
"Alcom Bus" to communicate back to a central control.
It is a simple 2-wire system that daisy chains from one unit to the next,
and requires a 150 ohm termination resistor at each end. There is also a
comment in the operation manual about NOT using shielded cable.
Under 'Technical Data' and 'Protocol' it says 'Token Passing'
Capacity : 62.5 kBaud
Signal Levels : Tx 5.0 V p-p .. Rx 0.3 V
Max Bus length : 1000 metres
Cable Type : 2 x 1.5 mm2, twisted pair. NOTE : Shielded cable not allowed
I haven't been able to find out if it is something of their own or just some
adaptation of a more common system, despite spending ages on Google.
Has anyone heard of or know anything about the 'Alcom' system ... and how
do you go about trying to make sense of any control without investing
$thousands on a central control processor and it's very expensive software?
Regards, Roger
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
John Chung
John.
--- On Thu, 6/17/10, Roger Weichert <rwei...@bigpond.net.au> wrote:
Marcel Birthelmer
> I am curious..... WHY and WHY is shield cables NOT ALLOWED?!
>
> John.
>
Probably because shielding inevitably causes the issue of where to
ground the shield etc... and if you have different grounds in
different parts of the system, it's even trickier.
Michael Watterson
> On Thu, Jun 17, 2010 at 11:48 AM, John Chung <kra...@yahoo.com> wrote:
>
>> I am curious..... WHY and WHY is shield cables NOT ALLOWED?!
>>
>> John.
>>
>>
>
> Probably because shielding inevitably causes the issue of where to
> ground the shield etc... and if you have different grounds in
> different parts of the system, it's even trickier.
>
Mixing desks with Balanced Microphones or STP for 10/100/1G ethernet.
This provides screen and prevents ground loops as the signal is supposed
to be differential on the twisted pair only.
Coax needs both ends grounded. If the coax has a 2nd *isolated* screen
insulated from inner screen, it should only be grounded at one end.
David Duffy (AVD)
> Marcel Birthelmer wrote:
>
> Shield on twisted pair need (should) only one end grounded. Applies to
>
Bad analogy I think. Leaving the shield disconnected on a microphone
cable is just plain wrong. It's there to extend the shielding through
to the body of the mic.
Not connecting the shield at both ends of many differential data lines
can also cause problems. Would you float the shield of an RS485 line?
If you did and there was no other common earth, what reference do the
data lines use? A lot of paired transmission lines still need the
shield as they are 3 wire systems, not 2 wire.
David...
--
___________________________________________
David Duffy Audio Visual Devices P/L
Unit 8, 10 Hook St, Capalaba 4157 Australia
Ph: +61 7 38235717 Fax: +61 7 38234717
Our Web Site: www.audiovisualdevices.com.au
___________________________________________
Lee Jones
> From: Roger Weichert <rwei...@bigpond.net.au>
>
>> call the "Alcom Bus" to communicate back to a central control.
>>
>> It is a simple 2-wire system that daisy chains from one unit to
>> the next, and requires a 150 ohm termination resistor at each end.
>>
> I am curious..... WHY and WHY is shield cables NOT ALLOWED?!
>
> John.
Possibly to prevent ground conduction through a set of nodes
that can span a kilometer? If shielded cable is used, and each
connector shield is tied to each node's signal ground, you are
effectively hooking the ground of the first node to the ground
of the last node in the daisy chain.
This can be BAD. It may work just fine for months and months.
Then, in a matter of minutes, it can start blowing components.
I was involved with one site, many years ago, where 2 buildings
were 10 meters apart. There was a communications link between
the 2 buildings. Communication signal ground was tied to power
ground inside equipment at both ends. One day a storm moved
through -- nothing uncommon, just a rain storm.
All of a suddon, parts blew ... communications IC case tops
actually came off. Upon investigation, we found that building
grounds tied into power system at opposite corners -- circa
150 meters apart. I personally measured over 70 volts between
the building grounds (that had formerly had roughly 0 volts
difference). Not enough resistance in 50 meters of 22-24 AWG
wire to keep ground current flow below the multiple amps range.
After a few days, ground potentials returned to normal.
Lee Jones
Ruben Jönsson
>
> John.
>
One reason is if the circuit driving and receiving the data signal is not
floating with respect to the shield, then that circuit sees a lot higher
capacitive load from the shield which may restrict cable lengths and/or
communication speed.
Another reason is potential ground loops if the shield is connected at both
ends (which it should be if it is to be of any use EMC wise). EX circuits, for
example, should not have the shield connected at both ends when the cable is
connected between different zones.
/Ruben
==============================
Ruben Jönsson
AB Liros Electronic
Box 9124, 200 39 Malmö, Sweden
TEL INT +46 40142078
FAX INT +46 40947388
ru...@pp.sbbs.se
==============================
Michael Rigby-Jones
> From: piclist...@mit.edu [mailto:piclist...@mit.edu] On
Behalf
> Of John Chung
> Sent: 17 June 2010 10:49
> To: Microcontroller discussion list - Public.
> Subject: Re: [EE] 'Alcom Bus' communication system
>
> I am curious..... WHY and WHY is shield cables NOT ALLOWED?!
>
It may be a simple capacitance problem; a shielded cable 1km long may
have enough capacitance to affect the bus.
Mike
=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================
Michael Watterson
> Michael Watterson wrote:
>
>> Marcel Birthelmer wrote:
>>
>> Shield on twisted pair need (should) only one end grounded. Applies to
>> Mixing desks with Balanced Microphones.
>>
>>
>
> Bad analogy I think. Leaving the shield disconnected on a microphone
> cable is just plain wrong. It's there to extend the shielding through
> to the body of the mic.
>
> Not connecting the shield at both ends of many differential data lines
> can also cause problems. Would you float the shield of an RS485 line?
> If you did and there was no other common earth, what reference do the
> data lines use? A lot of paired transmission lines still need the
> shield as they are 3 wire systems, not 2 wire.
> David...
>
>
have centre tapped transformers. Only a true common mode choke (huge for
audio) would inherently have no ground.
With ANY differential signalling no 3rd wire or earth interconnect is
needed. All that is required is proper design of the termination at each
end. The shield or screen in such case is best connected only at one
end, if one end is RX only, then at the "RX" end.
I don't know of any "3" wire systems for differential signalling. Three
phase Electricity needs 3 wires, it doesn't need a ground wire, that can
be re-created with transformers.
David Duffy (AVD)
> I think you misunderstand how differential signalling works and why you
> have centre tapped transformers. Only a true common mode choke (huge for
> audio) would inherently have no ground.
>
Microphones and microphone inputs do not use centre tapped transformers,
so I'm not sure why you've brought that up?
> With ANY differential signalling no 3rd wire or earth interconnect is
> needed. All that is required is proper design of the termination at each
> end. The shield or screen in such case is best connected only at one
> end, if one end is RX only, then at the "RX" end.
>
> I don't know of any "3" wire systems for differential signalling. Three
> phase Electricity needs 3 wires, it doesn't need a ground wire, that can
> be re-created with transformers.
RS485 is a 3 wire system. The data pair still needs the ground
reference to work correctly and keep the common mode voltage within limits.
Yes, there is a voltage difference between the data wires, but they're
still both referenced to the 0V line. Feel free to check out the various
RS485 reference designs as well as the DMX512 standard.
David...
--
___________________________________________
David Duffy Audio Visual Devices P/L
Unit 8, 10 Hook St, Capalaba 4157 Australia
Ph: +61 7 38235717 Fax: +61 7 38234717
Our Web Site: www.audiovisualdevices.com.au
___________________________________________
--
Olin Lathrop
> I'm working on some gear made by DeLaval which uses what they call the
> "Alcom Bus" to communicate back to a central control.
>
> It is a simple 2-wire system that daisy chains from one unit to the
> next, and requires a 150 ohm termination resistor at each end. There
> is also a comment in the operation manual about NOT using shielded
> cable.
I've never heard of "Alcom bus". My guess from your description is that
it's RS-422 with their own protocol on top of that. That fits with the
chain topology, the need for terminators at each end, and the 0-5V signal
range. RS-422 is basically a differential pair with each wire 0-5 volts.
There are off the shelf line driver and receiver chips for that, so that's
most likely what they used.
The requirement to use unshielded cable is probably to guarantee symmetry
between the two lines. If you use two conductor shielded cable, like common
audio cable, then the outer conductor picks up all the external noise, which
then becomes a differential signal. In a twisted pair, both wires pick up
the same external noise so that it becomes a common mode signal, which the
line receivers are specifically designed to ignore within a certain "common
mode range". Cable with a twisted pair inside and a third conductor as the
shield would be fine for preserving the symmetry between the two data lines,
but would introduce ground loops that they may feel installation technicians
wouldn't know how to deal with. It's also more expensive cable, so maybe
they weren't thinking of that anyway.
As for the protocol, most likely that is proprietary.
********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014. Gold level PIC consultants since 2000.
David Duffy (AVD)
>> I am curious..... WHY and WHY is shield cables NOT ALLOWED?!
>>
> that can span a kilometer? If shielded cable is used, and each
> connector shield is tied to each node's signal ground, you are
> effectively hooking the ground of the first node to the ground
> of the last node in the daisy chain.
>
The same problem can occur with DMX512 transmission (RS485) if the
various devices in the chain are earthed and the DMX I/O sections are
not isolated from the local earth. Isolated splitters are common in the
lighting industry for that very reason.
David...
--
___________________________________________
David Duffy Audio Visual Devices P/L
Unit 8, 10 Hook St, Capalaba 4157 Australia
Ph: +61 7 38235717 Fax: +61 7 38234717
Our Web Site: www.audiovisualdevices.com.au
___________________________________________
--
Michael Watterson
> Michael Watterson wrote:
>
>> I think you misunderstand how differential signalling works and why you
>> have centre tapped transformers. Only a true common mode choke (huge for
>> audio) would inherently have no ground.
>>
>>
>
> Microphones and microphone inputs do not use centre tapped transformers,
> so I'm not sure why you've brought that up?
>
>
>> With ANY differential signalling no 3rd wire or earth interconnect is
>> needed. All that is required is proper design of the termination at each
>> end. The shield or screen in such case is best connected only at one
>> end, if one end is RX only, then at the "RX" end.
>>
>> I don't know of any "3" wire systems for differential signalling. Three
>> phase Electricity needs 3 wires, it doesn't need a ground wire, that can
>> be re-created with transformers.
>>
>
> RS485 is a 3 wire system. The data pair still needs the ground
> reference to work correctly and keep the common mode voltage within limits.
>
>
In fact adding a ground wire connected at both ends could result in
exceeding common mode limits in certain circumstances. If the common
mode voltage is high then you will have a problem of "ground " loop
current. Designing a better interface is a better solution.
Olin Lathrop
> With ANY differential signalling no 3rd wire or earth interconnect is
> needed.
Not true. Transformer isolated differential signals generally don't use a
ground because of the very large common mode range they can tolerate.
Ethernet is a good example of this.
However, there are other differential signal conventions that assume a
direct connection to electronics and therefore typically have a low 10s of
volts common mode range. This type can't be allowed to float too far. CAN
is a good example of this. All CAN systems I've seen in practise have a
ground connection (and often power) along with the two signal lines. In
some small systems the ground is directly connected to the local ground at
each point.
For larger systems entire CAN nodes float to the supplied ground when they
don't need any other local electrical connection, or the CAN interface
circuit is separately powered and isolated from the rest of the local node.
********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014. Gold level PIC consultants since 2000.
David Duffy (AVD)
> You can easily design an R485 interface that doesn't need a ground wire.
> In fact adding a ground wire connected at both ends could result in
> exceeding common mode limits in certain circumstances. If the common
> mode voltage is high then you will have a problem of "ground " loop
> current. Designing a better interface is a better solution.
>
Well, you seem to flying in the face of recommended and accepted
practice then.
As has been said previously, non isolated (from mains earth) RS485
interfaces can drag ground currents through the shield. A lot of the
time, this is low enough that the connection suffers no ill effects.
If you use RS485 as a 2 wire medium, the reference will usually be the
mains earth. Yes, you have two data wires, but that mains earth is
still providing the reference. (3rd wire)
I'm ready to be convinced if you can show a solution for RS485 that uses
only 2 wires, and no reference. Mind you, you don't get any points for
presenting a solution that uses an LED as the receiver!
David...
--
___________________________________________
David Duffy Audio Visual Devices P/L
Unit 8, 10 Hook St, Capalaba 4157 Australia
Ph: +61 7 38235717 Fax: +61 7 38234717
Our Web Site: www.audiovisualdevices.com.au
___________________________________________
--
Roger Weichert
Olin Lathrop wrote
> I've never heard of "Alcom bus". My guess from your description is that
> it's RS-422 with their own protocol on top of that.
Thanks for that, I was hoping it was going to be something like that.
> There are off the shelf line driver and receiver chips for that, so that's
> most likely what they used.
>
I have drawn out various areas of the circuit, though not the communications
section yet.
I guess that should give me more clues, especially if it does use an off the
shelf chip set.
It's all smd, really compact and has a horrible thick yellowing conformal
coating which makes it hard to see through and even harder to measure
through.
> The requirement to use unshielded cable is probably to guarantee symmetry
> between the two lines. If you use two conductor shielded cable, like
> common
> audio cable, then the outer conductor picks up all the external noise,
> which
> then becomes a differential signal. In a twisted pair, both wires pick up
> the same external noise so that it becomes a common mode signal, which the
> line receivers are specifically designed to ignore within a certain
> "common
> mode range".
Ok, that makes sense.
>Cable with a twisted pair inside and a third conductor as the
> shield would be fine for preserving the symmetry between the two data
> lines,
> but would introduce ground loops that they may feel installation
> technicians
> wouldn't know how to deal with.
That's likely to be the answer. This gear is installed in dairies, so it
needs to be as robust and foolproof as possible.
> As for the protocol, most likely that is proprietary.
Ok, again that makes sense. It gives them an excuse to charge even more for
the gear. I discovered today that the processor that controls all this gear
is about $6000, so I'm not about to invest in one just so I can fix the odd
peripheral device.
Thanks for your in depth explanation Olin, and thanks to everyone else who
has contributed.
Regards, Roger
John Chung
JOhn
--- On Thu, 6/17/10, Lee Jones <l...@frumble.claremont.edu> wrote:
> From: Lee Jones <l...@frumble.claremont.edu>
> Subject: Re: [EE] 'Alcom Bus' communication system
John Chung
Regards,
John
--- On Thu, 6/17/10, Olin Lathrop <olin_p...@embedinc.com> wrote:
> From: Olin Lathrop <olin_p...@embedinc.com>
> Subject: Re: [EE] 'Alcom Bus' communication system
John Chung
through the mails on unshielded cables.
Googled more on it and found this:
http://www.maxim-ic.com/app-notes/index.mvp/id/2045
interesting read.
Regards,
John
Michael Watterson
> Thanks for the info guys. Took me about half a hour to read
> through the mails on unshielded cables.
>
> Googled more on it and found this:
> http://www.maxim-ic.com/app-notes/index.mvp/id/2045
>
> interesting read.
>
> Regards,
> John
>
>
>
>
>
>
>
>
(at least 250V AC common mode) is desirable and if the equipment is in
different buildings then one end at least should have 1kV or at least
G.P.O. standard isolation. Hence my comment about screens only connected
at one end.
I've designed communications with pulse transformer and opto isolation
techniques. For microphones and audio generally transformers with
various isolation are available.
In BBC in days when video might be sent between buildings on Base Band
they had the coax feed wound on laminated core that looked like a
1KW transformer. Enough common mode current choke to work well at
50Hz! Such common mode "chokes" are common enough on Transmitter
aerials or even as ferrite bead on a coax on Microwave gear.
With Cable TV and Communal satellite distribution in a Hotel or
Apartment block the "Cabinet" and all the amps, multiswitches etc need a
proper barrier quality earth common to wiring earth. There are various
regulations depending on country. When Cable TV was originally done in
Ireland (purely for BBC & ITV distribution, not US style cable
multichannel, though nowadays a multichannel + BB platform) it was not
well implemented. The TV aerial sockets used two isolation capacitors,
one to tuner and one to "live" chassis ( typically a single rectifier
and no SMPSU or transformer). But they also fed BBC & RTE FM Radio on
band II. Radio Tuners typically have earthed aerial sockets (All
75 Ohm, belling lee coax, not F-connector or balun 300 R).
So the TVs would make the cable "live" and you would get mains hum
connecting up to Radio tuner or even sparks. The metal of the aerial
patch cord would be unpleasant to touch!
Dwayne Reid
>I am curious..... WHY and WHY is shield cables NOT ALLOWED?!
Most likely because of the increased capacitance. This leads to
problems at high data rates.
dwayne
--
Dwayne Reid <dwa...@planet.eon.net>
Trinity Electronics Systems Ltd Edmonton, AB, CANADA
(780) 489-3199 voice (780) 487-6397 fax
www.trinity-electronics.com
Custom Electronics Design and Manufacturing
Dwayne Reid
>
>You can easily design an R485 interface that doesn't need a ground wire.
Sure, you can. But why bother?
We make DMX LED controllers - the DMX front-end is a 65lbc179 that
feeds a fast opto-isolator. The front end is powered with an
isolated DC-DC converter.
The 'lbc179 ** REQUIRES ** that you constrain its input signal to
within its common-mode range. The easiest way to do that is to
require that the installer uses shielded cable to transport the DMX
signal, and to connect the drain wire at both ends.
We don't have ground-loop problems because the DMX front-end is
completely isolated from the rest of our circuitry.
dwayne
--
Dwayne Reid <dwa...@planet.eon.net>
Trinity Electronics Systems Ltd Edmonton, AB, CANADA
(780) 489-3199 voice (780) 487-6397 fax
www.trinity-electronics.com
Custom Electronics Design and Manufacturing
--
Ruben Jönsson
> feeds a fast opto-isolator. The front end is powered with an
> isolated DC-DC converter.
>
> The 'lbc179 ** REQUIRES ** that you constrain its input signal to
> within its common-mode range. The easiest way to do that is to
> require that the installer uses shielded cable to transport the DMX
> signal, and to connect the drain wire at both ends.
>
> We don't have ground-loop problems because the DMX front-end is
> completely isolated from the rest of our circuitry.
>
> dwayne
So the shield is not connected to chassie/earth?
/Ruben
==============================
Ruben Jönsson
AB Liros Electronic
Box 9124, 200 39 Malmö, Sweden
TEL INT +46 40142078
FAX INT +46 40947388
ru...@pp.sbbs.se
==============================
--
Michael Watterson
>> We make DMX LED controllers - the DMX front-end is a 65lbc179 that
>> feeds a fast opto-isolator. The front end is powered with an
>> isolated DC-DC converter.
>>
>> The 'lbc179 ** REQUIRES ** that you constrain its input signal to
>> within its common-mode range. The easiest way to do that is to
>> require that the installer uses shielded cable to transport the DMX
>> signal, and to connect the drain wire at both ends.
>>
>> We don't have ground-loop problems because the DMX front-end is
>> completely isolated from the rest of our circuitry.
>>
>> dwayne
>>
>
> So the shield is not connected to chassie/earth?
>
>
This is the point I'm making. In fact if you are going between buildings
it might even be illegal in some cases to interconnect the earths.
David Duffy (AVD)
> Ruben Jönsson wrote:
>
>>> We make DMX LED controllers - the DMX front-end is a 65lbc179 that
>>> feeds a fast opto-isolator. The front end is powered with an
>>> isolated DC-DC converter.
>>>
>>> The 'lbc179 ** REQUIRES ** that you constrain its input signal to
>>> within its common-mode range. The easiest way to do that is to
>>> require that the installer uses shielded cable to transport the DMX
>>> signal, and to connect the drain wire at both ends.
>>>
>>> We don't have ground-loop problems because the DMX front-end is
>>> completely isolated from the rest of our circuitry.
>>>
>>> dwayne
>>>
>> So the shield is not connected to chassie/earth?
>>
>>
> It would obliviate the point of isolated DMX if it was.
>
> This is the point I'm making. In fact if you are going between buildings
> it might even be illegal in some cases to interconnect the earths.
>
But that's the point we're making. You're not interconnecting the
earths. You're connecting the common reference. That reference may or
may not be mains earth. There's no requirement for it to be earth - it
just needs to be a common reference.
In the case of isolated RS485, do you have a common reference (often the
shield)? Yes! Is it mains earth? Not if both ends are truly isolated.
As long as the I/O of only one device is earthed and the rest are
isolated, there's no problem. Ideally they are all isolated from mains
earth so that a single fault (common contacting earth) will have no ill
effect.
So, we're all agreed that differential signals (RS485 for example)
travelling between places of differing earth potentials should be
isolated. That isolation doesn't negate the need for the common
reference conductor, it just means the I/O sections are floated with
respect to earth.
David...
--
___________________________________________
David Duffy Audio Visual Devices P/L
Unit 8, 10 Hook St, Capalaba 4157 Australia
Ph: +61 7 38235717 Fax: +61 7 38234717
Our Web Site: www.audiovisualdevices.com.au
___________________________________________
--
Michael Watterson
local 0V termination at both ends and "earthing" screen at one end will
work. You don't inherently need a 3rd connection. The 3rd connection, if
used carries no signal, it's just a primitive (and unsafe if the gear
is on different supplies without isolated I/O) method of limiting
common mode voltage.
Capacitance is potentially higher with a screen, but only if the twisted
pair is otherwise in free air. Assuming a correct impedance of
transmission line correct driving and termination, then the capacitance
per meter affects the distance. For longer distance you need fatter
spacing between twisted pair and screen and also fatter core copper and
fatter twisted pair insulation to reduce the loss.
If everything is done properly a screened twisted pair is superior for
long distance as there is less crosstalk with other cables and less
noise/radiation. The reason phone lines and typical Cat5e isn't
screened is cost. Cat5 bundled with other function cables or outside
cable trays/trunking (earthed) is best done with STP rather than USP.
Less noise pickup, radiation to Radios and crosstalk. Ideally only earth
screen at "central" patch panel, never both ends.
David Duffy (AVD)
> David Duffy (AVD) wrote:
>
>> So, we're all agreed that differential signals (RS485 for example)
>> travelling between places of differing earth potentials should be
>> isolated. That isolation doesn't negate the need for the common
>> reference conductor, it just means the I/O sections are floated with
>> respect to earth.
>>
>>
> local 0V termination at both ends and "earthing" screen at one end will
> work. You don't inherently need a 3rd connection. The 3rd connection, if
> used carries no signal, it's just a primitive (and unsafe if the gear
> is on different supplies without isolated I/O) method of limiting
> common mode voltage.
>
You just described the 3rd conductor we're talking about. It may not be
a piece of wire that you can point to, but another point that is common
to both ends of the cable.
The whole reason "it works with screen connected at one end" is that you
have a secondary common reference path. (mains / ground earth)
Also, I'm not sure why the "centre tapped" reference keeps getting
dragged out. The I/O we're talking about doesn't use transformers.
> Capacitance is potentially higher with a screen, but only if the twisted
> pair is otherwise in free air. Assuming a correct impedance of
> transmission line correct driving and termination, then the capacitance
> per meter affects the distance. For longer distance you need fatter
> spacing between twisted pair and screen and also fatter core copper and
> fatter twisted pair insulation to reduce the loss.
>
Using the correct cable with RS485 will yield a good result in
practice. If you use mic cable for DMX (250K RS485) then yes, expect
problems.
David...
--
___________________________________________
David Duffy Audio Visual Devices P/L
Unit 8, 10 Hook St, Capalaba 4157 Australia
Ph: +61 7 38235717 Fax: +61 7 38234717
Our Web Site: www.audiovisualdevices.com.au
___________________________________________
--
Olin Lathrop
> So, we're all agreed that differential signals (RS485 for example)
> travelling between places of differing earth potentials should be
> isolated. That isolation doesn't negate the need for the common
> reference conductor, it just means the I/O sections are floated with
> respect to earth.
Twisted pair ethernet is a example of isolated differential signals where
there is not common reference conductor.
********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014. Gold level PIC consultants since 2000.
David Duffy (AVD)
> David Duffy (AVD) wrote:
>
>> So, we're all agreed that differential signals (RS485 for example)
>> travelling between places of differing earth potentials should be
>> isolated. That isolation doesn't negate the need for the common
>> reference conductor, it just means the I/O sections are floated with
>> respect to earth.
>>
>
> Twisted pair ethernet is a example of isolated differential signals where
> there is not common reference conductor.
>
Yes it is. It's also a transformer based design, which is not was I was
referring to.
David...
--
___________________________________________
David Duffy Audio Visual Devices P/L
Unit 8, 10 Hook St, Capalaba 4157 Australia
Ph: +61 7 38235717 Fax: +61 7 38234717
Our Web Site: www.audiovisualdevices.com.au
___________________________________________
--
alan smith
Google doesn't seem to find it...tried most the places (IR, Vishay,etc)
Anyone ever heard of this part?
Olin Lathrop
> I have a design (clients) calling out FET part number BUZ172H but
> can't seem to find it anywhere.
I haven't looked, but are you sure the Z is not a 2?
********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014. Gold level PIC consultants since 2000.
RussellMc
seem to find it anywhere.
Ignoring the H suffix to start
Infineon & Siemens datasheets - see below
This will get you well on way to a substitute depending on what H means
(HV?) and whether gate capacitance etc matter tyo you.
Can you describe the application>
TO220 MOSFET P Channel
Vds -100V
Id -5.5A
Rdson 0.6 ohm max at
VGs max +/- 0 V
Pdis 40W
Rthjc 3.1 K/W
Vgsth 2.1 / 3 / 4 min typ max
This path here provides real data sheets
http://www.datasheetcatalog.net/de/datasheets_pdf/B/U/Z/1/BUZ172.shtml
<http://www.datasheetcatalog.net/de/datasheets_pdf/B/U/Z/1/BUZ172.shtml>Quite
a compfrehensive MOSFET cross reference guide here.
Specs but no equivalent for you part
http://www.datasheetarchive.com/pdf/Datasheet-025/DSA00442391.pdf
RussellMc
BUZ is a common European suffix
Michael Watterson
> alan smith wrote:
>
>> I have a design (clients) calling out FET part number BUZ172H but
>> can't seem to find it anywhere.
>>
>
> I haven't looked, but are you sure the Z is not a 2?
>
>
Marechiare
>
> BUZ is a common European suffix
prefix?
Michael Watterson
> Michael Watterson wrote:
>
>> David Duffy (AVD) wrote:
>>
>>
>>> So, we're all agreed that differential signals (RS485 for example)
>>> travelling between places of differing earth potentials should be
>>> isolated. That isolation doesn't negate the need for the common
>>> reference conductor, it just means the I/O sections are floated with
>>> respect to earth.
>>>
>>>
>>>
>>>
>> However if both ends are isolated, you will find that a centre tapped to
>> local 0V termination at both ends and "earthing" screen at one end will
>> work. You don't inherently need a 3rd connection. The 3rd connection, if
>> used carries no signal, it's just a primitive (and unsafe if the gear
>> is on different supplies without isolated I/O) method of limiting
>> common mode voltage.
>>
>>
>
> You just described the 3rd conductor we're talking about. It may not be
> a piece of wire that you can point to, but another point that is common
> to both ends of the cable.
>
> The whole reason "it works with screen connected at one end" is that you
> have a secondary common reference path. (mains / ground earth)
>
It's not inherently a 3 wire system. There are many ways of avoiding
excessive common mode. A 3rd wire is the least safe method between
buildings. Fine in one room.
> Also, I'm not sure why the "centre tapped" reference keeps getting
> dragged out. The I/O we're talking about doesn't use transformers.
>
>
however termination resistors can go to 0V rather than "only" across
the pair.
Michael Watterson
>> I have a design (clients) calling out FET part number BUZ172H but can't
>>
> seem to find it anywhere.
>
> Ignoring the H suffix to start
>
> Infineon & Siemens datasheets - see below
>
> This will get you well on way to a substitute depending on what H means
> (HV?) and whether gate capacitance etc matter tyo you.
>
>
you to optionally cut middle pin and mount on a TO3 heatsink using the
two outer pins to existing holes and one of the TO3 mechanical holes to
"tab" (usually Drain or Collector).
Looking at PCB/Heatsink should reveal if this the case.
alan smith
Michael Watterson
>>> I haven't looked, but are you sure the Z is not a 2?
>>>
>> BUZ is a common European suffix
>>
>
> prefix?
>
Dwayne Reid
> > We make DMX LED controllers - the DMX front-end is a 65lbc179 that
> > feeds a fast opto-isolator. The front end is powered with an
> > isolated DC-DC converter.
> >
> > The 'lbc179 ** REQUIRES ** that you constrain its input signal to
> > within its common-mode range. The easiest way to do that is to
> > require that the installer uses shielded cable to transport the DMX
> > signal, and to connect the drain wire at both ends.
> >
> > We don't have ground-loop problems because the DMX front-end is
> > completely isolated from the rest of our circuitry.
> >
> > dwayne
>
>So the shield is not connected to chassie/earth?
Nope. The shield is connected to the Vss point
of the DMX receiver front-end. That Vss point is
completely isolated from the remainder of the
system (which most often IS connected to chassis / earth).
Ground-loop current is one of the most common
causes of data corruption as well as physical
failures in widely-distributed RS-422 or RS-485 systems.
dwayne
--
Dwayne Reid <dwa...@planet.eon.net>
Trinity Electronics Systems Ltd Edmonton, AB, CANADA
(780) 489-3199 voice (780) 487-6397 fax
www.trinity-electronics.com
Custom Electronics Design and Manufacturing
Dwayne Reid
>
>However if both ends are isolated, you will find that a centre tapped to
>local 0V termination at both ends and "earthing" screen at one end will
>work. You don't inherently need a 3rd connection. The 3rd connection, if
>used carries no signal, it's just a primitive (and unsafe if the gear
>is on different supplies without isolated I/O) method of limiting
>common mode voltage.
Good point here.
For what its worth, the transmitter (lighting-console or whatever) in
DMX lighting systems IS normally connected to mains earth. It is
only the receivers that are isolated.
Same goes for any RS-422 or RS-485 systems that I design. The comms
at the master station is connected to earth ground, the slaves all
have isolated receivers.
However, you are wrong when you say that the 3rd connection carries
no signal. The reference that it provides ensures that the
differential receiver is operated within its rated common-mode range.
There are ways to eliminate the need for that 3rd wire but they are
messy and can impact the line termination if done improperly. Its
simply easier and MUCH more reliable to use shielded twisted-pair
cable and to use the shield as the 3rd wire.
dwayne
--
Dwayne Reid <dwa...@planet.eon.net>
Trinity Electronics Systems Ltd Edmonton, AB, CANADA
(780) 489-3199 voice (780) 487-6397 fax
www.trinity-electronics.com
Custom Electronics Design and Manufacturing
--
Dwayne Reid
>Twisted pair ethernet is a example of isolated differential signals where
>there is not common reference conductor.
Completely agreed. But: they use transformers. That's completely
different from using something like the 65lbc176 differential
transceivers that I have been talking about.
One of the main differences is that you can transmit very low
frequency (or DC) signals using the differential transceivers I
mentioned. The transformers used in twisted-pair Ethernet won't pass
low frequency or DC signals.
But - I think that we are talking apples and oranges here. The
original poster was, I'm pretty sure, talking about standard RS-485
or possibly RS-422 type communications.
dwayne
--
Dwayne Reid <dwa...@planet.eon.net>
Trinity Electronics Systems Ltd Edmonton, AB, CANADA
(780) 489-3199 voice (780) 487-6397 fax
www.trinity-electronics.com
Custom Electronics Design and Manufacturing
--
Dwayne Reid
>A 3rd wire is the least safe method between buildings. Fine in one room.
Boy, oh boy.
I sure disagree with you on this one.
I'm not sure how to address this issue.
First: lets establish a baseline. I'm assuming that we are talking
about a distributed RS-485 data system such as that used by door
access systems: card-readers and door lock controls. Note: it
doesn't HAVE to be that specific example, its just that I'm pretty
familiar with those systems.
It is common, at least here in Edmonton, AB, Canada to have the
master controller for the building access system to be in one
building - even in a complex that has multiple
buildings. Communications is done using a distributed RS-485 system.
The master controller has multiple ports but all of the ports are
essentially identical and the signals going to and from one port are
the same as all other ports. The controller has many ports simply so
that one segment of the network being damaged doesn't take down the
whole system. The fact that having many ports help keep the device
count on any one port down to less than the recommended limit of 32
or 256 nodes total is sort of irrelevant - this is a nice benefit but
not the main reason for having all those ports.
Each building is tied to one or more ports in the master
controller. Every node that ties to a port in the master controller
has an isolated RS-485 front end. This is done either by simply
using an isolated power-supply (so that the device isn't tied to
earth ground) or by using a 'real' RS-485 isolator.
All wiring is done with 110R or 120R shielded twisted-pair
cable. The end point of any run is terminated with the specified
termination resistor.
Most buildings have the line from the master controller fed into a
RS-485 splitter box. That box has an isolated input but all of its
outputs are referenced to the local earth ground. Again, each device
that connects to that splitter box is isolated from earth ground.
You know: the systems JUST WORK. I get called in to help when they
don't work - that's when I find out what shortcuts were taken when
the system was installed. But I have them wire up the system as I've
described and then they don't have problems anymore.
Its common for me to measure many volts of 60Hz AC between the shield
wire of the line coming into a particular building and that
building's local earth ground. That's because of inadequate
grounding at the local power utility's step-down transformer that
feeds that building. But: I simply don't care! That several volts
of difference just doesn't affect the system's operation because of
the isolated input at the RS-485 splitter box.
I'm going to change tracks now.
I spend a LOT of time dealing with professional audio - mostly
broadcast audio but also some distributed audio systems.
My favorite tool for dealing with noise / hum problems is with
high-quality audio transformers. In particular, I generally use
unshielded twisted-pair cable (telephone cable (25 pair plus) or
cat-3 (3-pair) or cat-5 (4-pair)) when running audio signals between
buildings. No drain wires, no shields. The audio transformers have
incredibly good isolation between the primary and secondary windings
and the noise problems simply vanish when I use them.
The reason I bring this up is that we are talking about what I think
are completely different systems.
Multi-drop RS-485 systems require you to do wiring in a certain way
to achieve maximum reliability. Distributed audio systems can use a
different wiring scheme to achieve high-quality transfer of audio signals.
Neither system is wrong.
dwayne
--
Dwayne Reid <dwa...@planet.eon.net>
Trinity Electronics Systems Ltd Edmonton, AB, CANADA
(780) 489-3199 voice (780) 487-6397 fax
www.trinity-electronics.com
Custom Electronics Design and Manufacturing
--