Re: Electronics question

[Kevin Hawkins <yahoogroupskh@xxxxxxxxxxxxxx>]

We're both really saying the same thing Neill  - just from different
perspectives... :-)

My point was that the 3.5V transmit signal on DTR is not legal and
within the RS232 spec (as a transmit output) and that surprised me given
the purpose of the Avocent device and the fact it had it's own inbuilt
PSU and so could easily be designed to supply voltages to the
specification.  It made me think that just maybe DTR was appearing on a
different pin or not being asserted correctly.   This was after all
Paul's issue - driving the Quasar 3145 sregulator.

It is only acceptable as an input signal because of the noise threshold
allowance of 2V - although this would actually be within 500mv of that
threshold .    Pass that transmit signal through a noisy electrical path
and it could be troublesome, or at least more troublesome than a 'to
spec' +/- 5V or above signal.

As I said in my post most modern RS232 ports are tolerant to TTL which
even operates in that deadband of -3 to +3V volts and is never negative.
. The RS232 spec is really a very voltage unfriendly thing in terms of
modern circuitry and hence very adapted nowadays..

K

Neil Wrightson wrote:
> Kevin,
>
> Thanks for the links. They prove my original point.
> Quote "The receiver logic levels were defined to provide a 2V
noise margin.
> As such, a high level for the receiver is defined as between +3V to
+15V,
> and a low level is between -3V to -15V."
> Exactly what I said.
>
> I was only giving this information so that people realized that a 3.5V
> signal is a legal signal.
>
> Mind you, lots of PC serial ports will work on a TTL signal i.e.
0V..5V. Not
> legal but works.
>
>
> Regards,
>
> Neil Wrightson.
>
>
>   _____
>
> From: ukha_d@xxxxxxx [mailto:ukha_d@xxxxxxx] On Behalf Of
> Kevin Hawkins
> Sent: Friday, 26 June 2009 9:53 PM
> To: ukha_d@xxxxxxx
> Subject: Re: [ukha_d] Electronics question
>
>
>
>
>
> I guess that depends on which version of the RS232 spec document
you're
> reading - and I don't have a copy of the original 1962 EIA-232 one to
> double check so I could be misinformed but I have always believed +/-
> 5V was the minimum requirement for an output and the inputs had a
> threshold spec of +/- 3V as per the Dallas Semiconductor document, or
> the Maxim one, here... This allowed for an a rather generous 2V noise
> margin.
>
> http://www.lammertb
<http://www.lammertbies.nl/download/dallas-appl-83.pdf>
> ies.nl/download/dallas-appl-83.pdf
>
> http://www.maxim-
<http://www.maxim-ic.com/appnotes.cfm/an_pk/83/>
> ic.com/appnotes.cfm/an_pk/83/
>
> . RS232 was ratified before TTL and since then there have been loads
> of unofficial interpretations/relaxations of the specification to
allow
> laptops and other low power devices to be more tolerant of the lower
> voltages used in TTL or even CMOS at 5V or 3.3V even - so indeed that
> 3.5V will likely work with most modern serial ports / devices and some
> older kit too - but it is still outside of the official RS232
> specification I believe...
>
> Regardless it's a rather mute issue as it's certainly not going to
> work with a 5V regulator ;-)
>
> K
>
> Neil Wrightson wrote:
>
>> Hi,
>>
>> Actually, 3.5 V is enough for RS232C communication. The voltage
levels
>>
> must
>
>> be between 3V and 15V.
>> That is a logic 0 is between -3V..-15V and a logic 1 is between
+3V..+15V.
>> A voltage between -3V..+3V is deemed to be in the nomans land and
an
>>
> illegal
>
>> value.
>>
>> RS232 voltages were not intended to be used as power supplies.
>> It just happens that older equipment could typically supply ~20mA
at these
>> voltages so people/companies took advantage of this.
>> This was more the case when the actual driver IC i.e. 1488 had
it's own
>> external +/-12V supplies fed to it and it just switched these
supplies.
>> IC's such as the MAX232 run directly from the +5V rail and
generate their
>> own +/- supply usually in the +8/-8 voltage region.
>> Which is still a legal voltage.
>>
>>
>> Regards,
>>
>> Neil Wrightson.
>>
>>
>> _____
>>
>> From: ukha_d@yahoogroups. <mailto:ukha_d%40yahoogroups.com>
com
>>
> [mailto:ukha_d@yahoogroups. <mailto:ukha_d%40yahoogroups.com>
com] On Behalf
> Of
>
>> Kevin Hawkins
>> Sent: Friday, 26 June 2009 1:09 AM
>> To: ukha_d@yahoogroups. <mailto:ukha_d%40yahoogroups.com>
com
>> Subject: Re: [ukha_d] Electronics question
>>
>>
>>
>>
>>
>> Could it be anything as simple as some configuration parameters
for the
>> specific Avocent port in use that causes the port to assert DTR ?
3.5V
>> is not a large enough voltage here to work with most RS232
equipment
>> -and obviously isn't enough to power a 5V regulator. The port was
>> 'open' when you made this measurement I assume.
>>
>> Ant's suggestions is neat way around this - albeit requiring a
secondary
>> PSU.
>>
>> K
>>
>> ant wrote:
>>
>>
>>> Hi Paul,
>>>
>>> Yeah you're pretty much there:
>>>
>>> 1) disconnect the DTR line from your circuit (belt and braces)
>>> 2) connect the +ve wire from a 9v DC PSU to pin 1 of the 78L05
>>> 3) connect the ground wire from the 9v DC PSU to one of the
>>> regulator's ground pins (I can't read the schematic, there
might only
>>> be one ground pin, depends on the package)
>>> 4) leave the RXD and GROUND connections to the serial port
>>>
>>> cheers
>>>
>>> ant
>>>
>>>
>>> ------------------------------------
>>>
>>> Yahoo! Groups Links
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>
>>
>>
>>
>>
>> [Non-text portions of this message have been removed]
>>
>>
>>
>> ------------------------------------
>>
>> Yahoo! Groups Links
>>
>>
>>
>>
>>
>>
>
>
>
>
>
>
>
> [Non-text portions of this message have been removed]
>
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>
>
>



------------------------------------