TenTec
[Top] [All Lists]

Re: [TenTec] Paragon II RS-232 noise

To: tentec@contesting.com
Subject: Re: [TenTec] Paragon II RS-232 noise
From: "Dr. Gerald N. Johnson" <geraldj@weather.net>
Reply-to: geraldj@weather.net, Discussion of Ten-Tec Equipment <tentec@contesting.com>
Date: Wed, 15 Dec 2010 15:23:19 -0600
List-post: <tentec@contesting.com">mailto:tentec@contesting.com>
Two other things that might help involve the RS-232 drivers. In the 
classic RS-232 standard the rate of rise and fall was limited (which 
effectively limited serial port speeds sometimes below 1200 baud) by 
driver current limiting according to the standard and a certain amount 
of signal path bypassing was allowed to slow the rise and fall. These 
limits were intended to reduce RFI to receivers.

I ran into a problem with that using an Apple II+ and a Racal-Vadic 
triple modem (300 baud, and two versions of 1200 baud) that would work 
OK at 300 baud, but was unreliable at 1200 baud. Turned out Apple 
believed the RS-232 standard and built discrete component drivers that 
were properly current limited, but R-V didn't believe the standard and 
dropped a large capacitor on the serial line to keep down hash. And the 
IC drivers from TI and Motorola that have been universally copied didn't 
believe in current limiting. So now those will run 128K or 256K data 
speeds, at the cost of making hash.

The things are to add series R or L to the outputs of the drivers and 
some shunt C at the connector so the R or L and C make a low pass filter 
slowing the data rise and fall to something like 10% of the bit time. 
I'd start with 470 ohms and 1000 pf in a male to female connector shell.

For sure grounding the shield to the cases is important, just like pin 1 
on the microphones and since there has been a pin 1 problem on the 
microphones its most likely the serial port has exactly the same problem 
grounding to the board that's not grounded well.

Cat 5 is OK, but I'd prefer coax for the single ended RS-232 signal to 
get better shielding. Cat 6 might be better because the twisted pairs 
are always shielded.

73, Jerry, K0CQ

On 12/15/2010 2:44 PM, Jim Brown wrote:
> On 12/15/2010 10:16 AM, Dr. Gerald N. Johnson wrote:
>> How about checking the control board grounding, as well as the cable
>> shield grounding?
>
> Yes, cable shield connection (NOT GROUNDING) is quite important. It MUST
> go to the shielding enclosure on both ends.  DO NOT CONNECT CABLE
> SHIELDS TO THE SO-CALLED "GROUND" PIN ON THE DB CONNECTOR -- IT USUALLY
> GOES TO THE CIRCUIT BOARD, NOT THE CHASSIS.  CONNECT SHIELDS TO THE
> CONNECTOR SHELL!
>
> I also recommend use of a good twisted pair, like CAT5, for serial
> connections.  See the chapter in
> http://audiosystemsgroup.com/RFI-Ham.pdf  about Solving Problems in the
> Shack, also shown in the Power Point
> http://audiosystemsgroup.com/HamInterfacing.pdf
>
> 73, Jim Brown K9YC
>
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec

<Prev in Thread] Current Thread [Next in Thread>