Searching \ for '[PIC]: RS232 voltage levels (was PIC SBC requireme' 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/microchip/ios.htm?key=rs232
Search entire site for: 'RS232 voltage levels (was PIC SBC requireme'.

Exact match. Not showing close matches.
PICList Thread
'[PIC]: RS232 voltage levels (was PIC SBC requireme'
2000\10\31@090217 by Bob Ammerman

picon face
As I remember the RS-232 spec said:

At the receiver:

Anything below -3v is a mark.
Anything over +3v is a space.

Anything in between is 'invalid'.

At the transmitter:

Send -5v (or -6v?) to -15v (or -30v? or ??) for  a mark.
Send +5v (or +6v?) to +15v (or +30v? or ??) for a space.

(does anybody know with assurance the correct range on the transmitter
side?)

Actually, I'd normally be able to live with TTL levels, but since the
application is deep in the guts of a medium-large hydroelectric plant...

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.




2000\10\31@091419 by Alan B. Pearce

face picon face
>As I remember the RS-232 spec said:

>At the receiver:

>Anything below -3v is a mark.
>Anything over +3v is a space.

Yes, but

There is also a requirement that when the input is open circuit (i.e. the plug
has been pulled out) the receiver chip should behave as though it is receiving a
constant mark signal. This stop your processor getting continuous interrupts due
to an inadvertent break condition  ;)

Implementation left up to the circuit designer. The normal means of doing this
is the way the 1489 series do it by having the threshold level at a slightly
positive voltage.

I cannot quote chapter and verse for the paragraph with this requirement
unfortunately, and do not have the time to look for it.

P.S. have you ever seen an RS232 receiver with 6V hysteresis centered on 0V?

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.




2000\10\31@104848 by David VanHorn

flavicon
face
>
>At the receiver:
>
>Anything below -3v is a mark.
>Anything over +3v is a space.
>
>Anything in between is 'invalid'.

True enough, but look at the circuit for the 1488/89 receivers.
They can only see positive voltage.

>At the transmitter:
>
>Send -5v (or -6v?) to -15v (or -30v? or ??) for  a mark.
>Send +5v (or +6v?) to +15v (or +30v? or ??) for a space.
>
>(does anybody know with assurance the correct range on the transmitter
>side?)


I don't think they spec it that way. They define a standard load, and it's
up to you to deliver the goods.

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.




2000\10\31@115732 by Bill Westfield

face picon face
> Anything below -3v is a mark.
> Anything over +3v is a space.

For this to actually WORK, your rs232->digital converters would need an
"error" output, and UARTS would need go take that into effect.  Effectively,
the spec has a requirement that is impossible to enforce, so engineers have
taken "liberties" (usually for the better.)

Standards are interesting.  Some standards are very clear about how
something is supposed to work, but not so clear on what should happen
under certain error conditions (most of the internet RFCs fall into this
catagory, for instance.)  Other standards go into excruiating detail on
all the possible error conditions, but say very little about the
variables of "correct" operation.  (like ITU X. specs, ie X.25)

BillW

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.




2000\10\31@124134 by Dan Michaels

flavicon
face
Dave VanHorn wrote:
Bob Ammerman wrote:
>>
>>At the receiver:
>>
>>Anything below -3v is a mark.
>>Anything over +3v is a space.
>>
>>Anything in between is 'invalid'.
>
>True enough, but look at the circuit for the 1488/89 receivers.
>They can only see positive voltage.
>

Ok, but if you use 0-5v levels, your noise margins are at best
5-0.7v = 4.3v and 0.7-0v = 0.7v. Do you really want to use 0.7v
NM over a 20-50' cable?

With +/-12v or +5/-12v levels, noise margins are many times
better.
===============


{Quote hidden}

I don't know the actual specs either, but seems +/-3v minimum was
the old original, and my Maxim book says MAX232 are designed to meet
EIA-232D/V.28 - which calls for +/5v minimum output levels under worst
case conditions, ie, 3K load, Vcc=4.5v, and max temperature.

Clearly, today's 115 kbps requires a little more rigor than
1970's 300 baud.

- danM

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.




2000\10\31@224059 by Lee Jones

flavicon
face
I'm quoting from American National Standard ANSI/EIA/TIA-232-E-1991.
Electronics Industry Association EIA/TIA-232-E, July 1991, titled
"Interface Between Data Terminal Equipment and Data Circuit-Terminating
Equipment Employing Serial Binary Data Interchange".  (Now you can
impress your friends at parties by knowing what DCE stands for.)

>>> At the receiver:
>>>
>>> Anything below -3v is a mark.
>>> Anything over +3v is a space.
>>>
>>> Anything in between is 'invalid'.

EIA/TIA-232-E, page 3,
 Section Two,
   2. Signal Characteristics
     2.1 Electrical Characteristics
       2.1.2 [...] The receiver on an interchange circuit shall
         be designed to withstand (not be damaged by) any input
         signal within the 25 volt limit specified in Sec. 2.1.6.

       2.1.3 For data interchange circuits, the signal shall
         be considered in the marking condition when the voltage
         [...] is more negative than -3 volts with respect to
         [signal common, aka signal ground].  The signal shall
         be considered in the spacing condition when the voltage
         is more positive than +3 volts with respect to [common].

         The region between +3 volts and -3 volts is defined as
         the transition region.  The signal state is undefined
         when the voltage is in this tranistion region.

Section 2.1.8 lists a bunch of constraints on the charactertics
of signals going through the transition region.


>> Yes, but
>> There is also a requirement that when the input is open circuit
>> (i.e. the plug has been pulled out) the receiver chip should
>> behave as though it is receiving a constant mark signal. This
>> stop your processor getting continuous interrupts due to an
>> inadvertent break condition  ;)

I don't recall reading such a requirement.  I've just skimmed
EIA/TIA-232-E again and can't find anything like this.  The
received data line is specified to be held in the marking state
(below -3 volts) except when certain modem control signals are
asserted.  See section 4.4.3 Circuit BB Received Data.

Maybe it was added to a later revision.  I think EIA/TIA-232
is up to revision F now.  But I don't have a copy.


>>> At the transmitter:
>>>
>>> Send -5v (or -6v?) to -15v (or -30v? or ??) for  a mark.
>>> Send +5v (or +6v?) to +15v (or +30v? or ??) for a space.
>>>
>>> (does anybody know with assurance the correct range on the
>>> transmitter side?)

Yes.  See EIA/TIA-232-E, page 5...

   2.1.6 The open-circuit generator [aka transmitter] voltage
     with respont to [signal common] shall not exceed 25 volts
     in magnitude.  [This is used by receiver section 2.1.2]

     Additionally, the generator design shall be such that, with
     a test load of 3000 ohms to 7000 ohms, the potential at the
     [transmitter] interface point shall not be less than 5 volts
     nor more than 15 volts in magnitude (see section 6.5).

page 28...

   6.5 Test Receivers [excerpt]

   The [transmitter output] 5 volt tolerance should therefore be
   tested with a 3000 ohm load while the 15 volt limit should be
   tested using a 7000 ohm load.

So voltage at a transmitter pin should be between -15 volts and
-5 volts or between +5 volts and +15 volts.

Also in section 2.1.6, it states "a short circuit between any two
conductors (including ground) [...] shall not result in a current
in excess of 100mA".


> I don't think they spec it that way. They define a standard
> load, and it's up to you to deliver the goods.

This is wrong.  The electrical characteristics of both the
receiver and generator (aka transmitter) end are specified.

                                               Lee Jones

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.




2000\10\31@224508 by David VanHorn

flavicon
face
>
>           The region between +3 volts and -3 volts is defined as
>           the transition region.  The signal state is undefined
>           when the voltage is in this tranistion region.

Somebody better tell motorola that the chips they've been making for the
last 25 years + won't work.. :)

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.




2000\10\31@234422 by Chunhee Song

flavicon
face
Hello,

I am always curious about the fact that logical 1 should be in between
-5V to -15V and logical 0 should be in between +5V and +15V.  Is there
any important reason about that? My question might sound stupid. But why
is this the other way around?

Chunhee Song

On Tue, 31 Oct 2000, Lee Jones wrote:

{Quote hidden}

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.





'[PIC]: RS232 voltage levels (was PIC SBC requireme'
2000\11\01@002922 by Dan Michaels
flavicon
face
Chunhee Song wrote:
>Hello,
>
>I am always curious about the fact that logical 1 should be in between
>-5V to -15V and logical 0 should be in between +5V and +15V.  Is there
>any important reason about that? My question might sound stupid. But why
>is this the other way around?
>

Probably because the first EEs were better at making inverters
than they were at making active pullup stages.

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
use spam_OUTlistservTakeThisOuTspammitvma.mit.edu?body=SET%20PICList%20DIGEST




2000\11\01@012216 by Sean H. Breheny

face picon face
I would suspect that the reason comes from one of two things:

#1) AFAIK, Old telegraphy standards were that when the line was idle,
current was flowing, and when you pressed the key, it interrupted the
circuit. This was so that a line that was cut would be detected even when
no one was trying to send. This sorta inverts things because the key down
(which is naturally "on" or 1) is actually a "low" or "off" state.

#2) For many types of transducers, such as photocells, it is easiest to
just hook up a resistor to them and read the voltage between the load
resistor and the sensor. This voltage goes low when the sensor detects
light, or whatever it is detecting, because of the additional current
causing a voltage drop, pulling the voltage down. When they made
punched-card readers, perhaps it required less circuitry if they just
considered a 1 to be low.

Sean



At 09:42 PM 10/31/00 -0700, you wrote:
{Quote hidden}

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
use .....listservKILLspamspam@spam@mitvma.mit.edu?body=SET%20PICList%20DIGEST




2000\11\01@042630 by Alan B. Pearce

face picon face
>> There is also a requirement that when the input is open circuit
>> (i.e. the plug has been pulled out) the receiver chip should
>> behave as though it is receiving a constant mark signal. This
>> stop your processor getting continuous interrupts due to an
>> inadvertent break condition  ;)

Ok, I am prepared to say I may have it wrong, but I was always under the
impression that this was so, and it makes good engineering sense to do it like
this. can you imagine the performance of your PC if it was handling continuos
break characters???

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
use listservspamKILLspammitvma.mit.edu?body=SET%20PICList%20DIGEST




2000\11\01@090653 by Olin Lathrop

flavicon
face
> Ok, I am prepared to say I may have it wrong, but I was always under the
> impression that this was so, and it makes good engineering sense to do it
like
> this. can you imagine the performance of your PC if it was handling
continuos
> break characters???

I don't see how this should cause more than one "break" condition.  Most
UARTs will only report a single framing error for such a constant line -
unless the software resets it and tries again.  I doubt much of anything
would happen, especially if there wasn't app software trying to read the
port.

Also, most receivers implement the -3 to +3 volt undefined region with
hysterisis.  Most of these devices will initialize at power up as if the
voltage was low when it is within the hysterisis limits.


*****************************************************************
Olin Lathrop, embedded systems consultant in Devens Massachusetts
(978) 772-3129, .....olinKILLspamspam.....cognivis.com, http://www.cognivis.com

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
use EraseMElistservspam_OUTspamTakeThisOuTmitvma.mit.edu?body=SET%20PICList%20DIGEST




2000\11\01@093002 by Alan B. Pearce

face picon face
>Also, most receivers implement the -3 to +3 volt undefined region with
>hysterisis.  Most of these devices will initialize at power up as if the
>voltage was low when it is within the hysterisis limits.

well this is what I was saying (as the 1489 and like chips do it), and I
believed it was in the RS232 standard, but others have just been looking in the
standard, and cannot find this as a requirement. It appears that it has become a
thing of convenience by the chip makers. As regards UARTS and interrupts, I
think you will find a lot of older UART chips will just keep on producing break
characters and reporting framing errors if the input is kept in the space
condition. The more modern "smart" UARTs may well stop after receiving one
framing error.

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
use listservspamspam_OUTmitvma.mit.edu?body=SET%20PICList%20DIGEST




2000\11\01@103653 by Don Hyde

flavicon
face
I have long been of the belief that these voltages were chosen because they
were convenient and common logic voltages in vacuum-tube circuits.  Remember
that this standard is that old.

> {Original Message removed}

2000\11\01@150703 by w. v. ooijen / f. hanneman

picon face
> >
> >           The region between +3 volts and -3 volts is defined as
> >           the transition region.  The signal state is undefined
> >           when the voltage is in this tranistion region.
>
> Somebody better tell motorola that the chips they've been making for the
> last 25 years + won't work.. :)

Undefined just means that the receiver can interpret it as mark, as space,
or (when the transition is not monotonous) as a sequence of mark and/or
space. A receiver that interprets > +3 as high (I always forget whether
that is mark or space) and < +3 as low is perfectly
within secs.

Wouter

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
use @spam@listservKILLspamspammitvma.mit.edu?body=SET%20PICList%20DIGEST




2000\11\01@175405 by Barry Gershenfeld

picon face
>Undefined just means that the receiver can interpret it as mark, as space,
>or (when the transition is not monotonous) as a sequence of mark and/or
>space. A receiver that interprets > +3 as high (I always forget whether
>that is mark or space) and < +3 as low is perfectly
>within secs.
>
>Wouter

Yeah, and some of the early 1489's (without the schmitt
trigger) would generate a nice clock signal if you were
between 3 and -3.  :-)

Barry

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
use KILLspamlistservKILLspamspammitvma.mit.edu?body=SET%20PICList%20DIGEST




2000\11\02@041144 by Alan B. Pearce

face picon face
>Yeah, and some of the early 1489's (without the Schmidt
>trigger) would generate a nice clock signal if you were
>between 3 and -3.  :-)

They must have been real early ones, I dealt with 1489 chips in the late 1970's
and they definitely had a Schmidt trigger. It wasn't manufacturer dependant?

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




More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...