[Info-vax] grounded MMJ cables

[chris]

On 07/18/21 12:08, Phillip Helbig (undress to reply) wrote:
> In article<sd10f5$pd6$1 at gioia.aioe.org>, chris
> <chris-nospam at tridac.net>  writes:
>
>> On 07/18/21 09:35, Phillip Helbig (undress to reply) wrote:
>>> In article<sd0n91$i5o$1 at tncsrv09.home.tnetconsulting.net>, Grant Taylor
>>> <gtaylor at tnetconsulting.net>   writes:
>>>
>>>> On 7/17/21 11:22 PM, Phillip Helbig (undress to reply) wrote:
>>>>> I'm testing out some MMJ cables (useful for console cables) and
>>>>> notice that some (all of which are round rather than flat) have an
>>>>> additional wire outside of the plug at each end, which looks like it
>>>>> might be intended as ground.  What's the purpose of that?
>>>>
>>>> How long is it?  I'd expect it to be 3-9 inches if it's meant to be a
>>>> ground wire that is attached to something else on the chassis.
>>>
>>> Yes, about 5 inches.  It has an eyelet at the end, so one could connect
>>> it to the chassis via a screw.
>>>
>>>> I would also have significant worries about ground loops and / or sneak
>>>> currents.  Both of which can be quite dangerous from an electrical and
>>>> safety perspective.  As in more amperage than the wire can safely handle
>>>> turning the wire into a heater that can melt insulation and start fires.
>>>
>>> But is that an issue when using a console cable?
>>
>> Not usually from an interference to the cable  point of view and iirc,
>> the added wire is at one end only, which would not enable earth loops.
>
> I have seen those, but the two I checked today have it at both ends.

Ok, but the one I have has just one, but maybe that was third party
item.

The frame or chassis ground and signal grounds are separate pins
on a 25 pin RS232 layout, even though they may be joined inside the
equipment. If they joined, it's often with  a low value resistor, say
100 ohms, to prevent excessive earth loop currents. Don't know the 
circuitry for a VT220, for example, but DEC would have done the job
right, both from an emc and safety points of view. They were obsessive
about standards...




>
>> The answer is more likely to lower any possible emc emissions from the
>> cable, not to it. Typical terminal baud rates are not high enough and
>> the rise and fall times are usually controlled, but fast edges could
>> still be coupled into the cable from the terminal or machine hardware...
>
> OK.
>