[N8VEM: 17864] SBC Mark IV -- twisted pair communications

Discussion:

John Coffman

2014-04-25 03:55:16 UTC

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>

<meta http-equiv="content-type" content="text/html; charset=ISO-8859-1">
</head>
<body bgcolor="#ffffff" text="#000000">
Ref:  RS-422 twisted pair line termination impedance 
        Mark IV schematic, page 2, Communication 
 
The desired termination impedance of the twisted pair line (R11,
R12, & R13) is 120 ohms.  It has been pointed out that the
resistor combination of  1200 / 150 / 1200 does not achieve 120
ohms. 
 
In figuring the values, I calculated R11 and R13 in parallel.  In
actual fact, for AC signals, they probably need to be figured in
series.  Figuring them in series, R11 and R13 should be 300 ohms to
achieve an AC line termination impedance of 120 ohms. 
 
So, the engineering change suggested: 
 
R11, R13  =  change to 300 ohms on boards not soldered yet. 
 
R11, R13  =  1200 ohms in parallel with 400 ohms on board already
built up.  390 is close enough to 400 to yield a final network
impedance of 119.5 ohms. 
 
--John 
 
P.S.  I'd appreciate comments from any other Electrical Engineers
out there. 
 
 
 
 
</body>
</html>



-- 
You received this message because you are subscribed to the Google Groups "N8VEM" group. 
To unsubscribe from this group and stop receiving emails from it, send an email to <a href="mailto:n8vem+unsubscribe-/***@public.gmane.org">n8vem+unsubscribe-/***@public.gmane.org</a>. 
To post to this group, send email to <a href="mailto:n8vem-/***@public.gmane.org">n8vem-/***@public.gmane.org</a>. 
Visit this group at <a href="http://groups.google.com/group/n8vem">http://groups.google.com/group/n8vem</a>. 
For more options, visit <a href="https://groups.google.com/d/optout">https://groups.google.com/d/optout</a>.

Nikolay Dimitrov

2014-04-25 18:02:41 UTC

Permalink

Hi John,

For me, it's not a question of AC vs DC signals, as there are no
capacitors in the RS-485 signal path. It's more of single-ended vs
differential termination.

So in general, what are you trying to achieve with resistors R11 & R13?
People typically assume that they're for line termination, while they're
not used for
this purpose. These resistors typically bias the line, to avoid buggy
behavior when the line is not driven, and the values of these resistors
are typically
calculated for a specific voltage drop across R12, so you can get a
defined bus state when idle.

The issue with this approach is that this bias acts as a weak driver of
the line, and the chip output drivers will work against different
currents on positive /
negative sides, causing different slew rates and asymmetry of the signal
waveform, which in turn can and will cause troubles with high-speed async
communication.

Usually R12 is matched to the line impedance, and the other resistors
should be at least 10x bigger to avoid changing this termination
significantly. The
differential impedance is calculated with the sum of both resistors R11
& R12.

Kind regards,
Nikolay

PS: The board is pretty cool!

Nikolay Dimitrov

2014-04-25 18:07:00 UTC

Permalink

Oops, this should be read as: "The differential impedance is calculated
with the sum of both resistors R11 & R13."

The differential impedance is calculated with the sum of both
resistors R11 & R12.

John Coffman

2014-04-25 19:52:26 UTC

Permalink

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Nikolay, 
 
I was out hiking this morning, so I had several hours to mull over
the situation. 
 
I think your assessment is quite correct.  So I would recommend
keeping R11,R13 at 1200, and simply reducing R12 from 150 to 120
ohms.  This terminates the line with very close to 120 ohms, and
provides the bias when no one is driving the line. 
 
--John 
 
 
R12 -- change from 150 to 120 ohms 
R11,R13 -- leave unchanged at 1200 ohms each 
 
 
 
 
 
 
On 04/25/2014 11:07 AM, Nikolay Dimitrov wrote:
<blockquote cite="mid:535AA444.5030305-***@public.gmane.org" type="cite">Oops,
this should be read as: "The differential impedance is calculated
with the sum of both resistors R11 & R13."
 
 
On 4/25/2014 9:02 PM, Nikolay Dimitrov wrote:
 
<blockquote type="cite">The differential impedance is calculated
with the sum of both resistors R11 & R12.
 
</blockquote>
 
</blockquote>
</body>
</html>



-- 
You received this message because you are subscribed to the Google Groups "N8VEM" group. 
To unsubscribe from this group and stop receiving emails from it, send an email to <a href="mailto:n8vem+unsubscribe-/***@public.gmane.org">n8vem+unsubscribe-/***@public.gmane.org</a>. 
To post to this group, send email to <a href="mailto:n8vem-/***@public.gmane.org">n8vem-/***@public.gmane.org</a>. 
Visit this group at <a href="http://groups.google.com/group/n8vem">http://groups.google.com/group/n8vem</a>. 
For more options, visit <a href="https://groups.google.com/d/optout">https://groups.google.com/d/optout</a>.

William R Sowerbutts

2014-04-25 18:21:18 UTC

Permalink

(This is my first post to N8VEM. Hello, world!)

I have been discussing this with John and did a bit of reading around the
subject this morning.

In an RS422 point-to-point application I think it would be safe to omit the
bias resistors R11 and R13 entirely, and use R12=100--120ohm (to match the
characteristic impedance of the cabling) providing the receiver circuit has
"fail-safe" operation. The MAX3467 specified for this board does include the
fail-safe feature.

The "fail-safe" feature simply means that when the differential receiver's
inputs are equal (bus driver is absent or tri-state) the receiver output is
defined as logic 1 rather than "undefined"; the receiver's differential
threshold is reduced from 0V to around -125mV to achieve this.

Very much looking forward to receiving my Mark IV board and assembling it!

Unrelated, but I don't suppose anyone knows where I can source an 8-slot
backplane PCB?

Very best wishes

Will Sowerbutts

Post by Nikolay Dimitrov
Hi John,
For me, it's not a question of AC vs DC signals, as there are no
capacitors in the RS-485 signal path. It's more of single-ended vs
differential termination.
So in general, what are you trying to achieve with resistors R11 &
R13? People typically assume that they're for line termination, while
they're not used for
this purpose. These resistors typically bias the line, to avoid buggy
behavior when the line is not driven, and the values of these
resistors are typically
calculated for a specific voltage drop across R12, so you can get a
defined bus state when idle.
The issue with this approach is that this bias acts as a weak driver
of the line, and the chip output drivers will work against different
currents on positive /
negative sides, causing different slew rates and asymmetry of the
signal waveform, which in turn can and will cause troubles with
high-speed async
communication.
Usually R12 is matched to the line impedance, and the other resistors
should be at least 10x bigger to avoid changing this termination
significantly. The
differential impedance is calculated with the sum of both resistors
R11 & R12.
Kind regards,
Nikolay
PS: The board is pretty cool!
--
You received this message because you are subscribed to the Google Groups "N8VEM" group.
Visit this group at http://groups.google.com/group/n8vem.
For more options, visit https://groups.google.com/d/optout.

_________________________________________________________________________
William R Sowerbutts will-***@public.gmane.org
"Carpe post meridiem" http://sowerbutts.com
main(){char*s=">#=0> ^#X@#@^7=",c=0,m;for(;c<15;c++)for
(m=-1;m<7;putchar(m++/6&c%3/2?10:s[c]-31&1<<m?42:32));}

p***@public.gmane.org

2014-04-26 09:00:03 UTC

Permalink

Hi Will,

I know that on the Marck IV schematic this circuit is labeled as RS-422, but I think that the actual intent was to use it as a bi-directional link, e.g. RS-485 (I'm making this conclusion because I don't see a second RS-422 module).

Regarding the 8-slot ECB backplane, I got mine from Andrew Lynch, you can check with him.

Regards,
Nikolay

-------------------------------------
Mail.BG: Безплатен e-mail адрес. Най-добрите характеристики на българския пазар - 20 GB пощенска кутия, 1 GB прикрепен файл, безплатен POP3, мобилна версия, SMS известяване и други. http://mail.bg

--
You received this message because you are subscribed to the Google Groups "N8VEM" group.
To unsubscribe from this group and stop receiving emails from it, send an email to n8vem+***@googlegroups.com.
To post to this group, send email to ***@googlegroups.com.
Visit this group at http://groups.google.com/group/n8vem.
For more options, visit https://groups.g