Re: [Emc-users] Smart-Serial cabling

2016-01-01 Thread Jack Coats 
My only suggestion is to ensure the wire you have is stranded, not
solid.  Stranded will be 'softer' and deal better with the bending
needed for use in drag-chain use.  Solid core is better for
infrastructure use (putting in walls), stranded is found in 'patch
cables' and used in wiring closets to connect the infrastructure or
from a wall to equipment in an office.

Great use.  BTW, Cat6 or 7 will have a bit more noise rejection, but
Cat 5 is cheaper and will do well.  You will find some of the higher
cat numbers have shielding, much cat 5 does not have shielding but
depends on the twisted wire pairs for noise rejection (each pair is
twisted at a different twist rate, so they don't cross contaminate
each others signal).

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Smart-Serial cabling

2016-01-01 Thread Ben Potter 
From: andy pugh [mailto:bodge...@gmail.com] 
> On 1 January 2016 at 16:16, Peter C. Wallace  wrote:
> > SSerial is differential RS-422 at 2.5 MBits/sec and requires ~100 Ohm 
> > differential cable cable impedance. This usually means twisted pairs 
> > are required for TX and RX (the 5V/GND power connections are not 
> > critical)

> The pairs are twisted inside the individual foil screen with drain wire
then there are 4 other small (0.14mm2) wires and two bigger
> (0.5mm2) wires all inside an overall braided screen.

> I suppose the thing to do is to try it and see. I have determined that the
pairs fit nicely in an RJ45 connector, but I can't decide what to do about
the 2x 5V and 2x 0V positions in the connector. Are the 0V and 5V contacts 
> commoned at the PCB ends, such that leaving two positions unpopulated in
the RJ45 is OK? I only need enough 5V power to run the 7i73 logic.

If it doesn't work, have a look at igus chainflex CFBUS.045 - cat5, should
be good for e-chains and oil from the specs. I think the link should be:
http://www.igus.co.uk/iPro/iPro_01_0013_0056_GBen.htm?ArtNr=CFBUS.045=GB
=en

I should be ordering some in a month or two for a pendant (I've been saying
that for a year or so though)

If it does work, let us all know - good test data is always good

> atp
> If you can't fix it, you don't own it.
> http://www.ifixit.com/Manifesto

Ben


--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Smart-Serial cabling

2016-01-01 Thread Rene Hopf 

> On 01 Jan 2016, at 12:23, andy pugh  wrote:
> 
> My lathe build is going to have a smart-serial device mounted in the
> lathe apron. This means that I need a CAT5 cable in the drag-chain,
> and also a 24V supply for the proximity sensors.
> Finding oil/coolant resistant CAT5 rated for drag-chain use is likely
> to be a challenge.
> I have lots of this cable:
> https://picasaweb.google.com/108164504656404380542/Holbrook#6217838146609796818
> (Lapp Ölflex drag-chain encoder cable).
> Is there any reason to think that the 2x3 screened pairs in that cable
> will be unsuitable for transporting smart-serial data (if I can
> persuade an RJ45 connector on to them)?

the cable looks ok. there are loads of drag chain compatible cat cables 
available, just google for industrial ethernet cable.

> 
> -- 
> atp
> If you can't fix it, you don't own it.
> http://www.ifixit.com/Manifesto
> 
> --
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Smart-Serial cabling

2016-01-01 Thread Bertho Stultiens 
On 01/01/2016 12:23 PM, andy pugh wrote:
> Is there any reason to think that the 2x3 screened pairs in that cable
> will be unsuitable for transporting smart-serial data

The bandwidth of rs232 style serial connections is so low that you will
not see any problems with twisted pairs. I assume that you are not
connecting half a kilometer of cable with a speed higher than 115k2 baud
(or you may run into capacitive loading problens).

Pairing TX and RX with a GND may actually improve noise immunity. But
the cable is already well shielded.


> (if I can persuade an RJ45 connector on to them)?

That is actually a hard thing to do on these cables. The isolation is
very hard and does not bend easily. Standard crimp-connectors will be
almost impossible to attach while ensuring a mechanically stable
connection. There is also a minimum bend-radius which can be rather large.

These cables are normally used as distribution cables and go into female
plug assemblies with seperate holding facilities for the cable.


-- 
Greetings Bertho

(disclaimers are disclaimed)

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Smart-Serial cabling

2016-01-01 Thread Gene Heskett 
On Friday 01 January 2016 06:23:06 andy pugh wrote:

> My lathe build is going to have a smart-serial device mounted in the
> lathe apron. This means that I need a CAT5 cable in the drag-chain,
> and also a 24V supply for the proximity sensors.
> Finding oil/coolant resistant CAT5 rated for drag-chain use is likely
> to be a challenge.
> I have lots of this cable:
> https://picasaweb.google.com/108164504656404380542/Holbrook#6217838146
>609796818 (Lapp Ölflex drag-chain encoder cable).
> Is there any reason to think that the 2x3 screened pairs in that cable
> will be unsuitable for transporting smart-serial data (if I can
> persuade an RJ45 connector on to them)?

Other than cat-5 is 4 twisted pairs with no shielding, no.  If its a real  
ethernet, the shielding would mess up the termination impedance. Even 
going thru a pair of crimp on rj45's plugged into a splicer block is 
enough of an impedance and echo causing bump that a 100megabit circuit 
can downgrade itself to a 10megabit circuit.  We've been bitten by that 
when rearranging gear in the racks at the tv station & had to add 
another 4 feet of cable.  To get the speed back Jim had to pull new 
cable.

For serial, where signalling is single ended, I'd make sure that txdata 
and rxdata were in separate shielded pairs.  I'd assume the protocol is 
similar to the common 7 wire that is usually terminated in db9 
connectors?  That works well for the baud rates used 20 years ago, but 
no first-hand experience above 56kbaud, and I expect this is 10x that. 
460k or better.  At those speeds termination is very important to 
control edge reflections and ringing.

Any idea what the baud rate is, Andy? These are tires I've never 
kicked. :(

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Smart-Serial cabling

2016-01-01 Thread Bertho Stultiens 
On 01/01/2016 05:16 PM, Peter C. Wallace wrote:
>> The bandwidth of rs232 style serial connections is so low that you will
>> not see any problems with twisted pairs. I assume that you are not
>> connecting half a kilometer of cable with a speed higher than 115k2 baud
>> (or you may run into capacitive loading problens).
>>
>> Pairing TX and RX with a GND may actually improve noise immunity. But
>> the cable is already well shielded.
> 
> SSerial is differential RS-422 at 2.5 MBits/sec and requires ~100 Ohm
> differential cable cable impedance. This usually means twisted pairs
> are required for TX and RX (the 5V/GND power connections are not critical)

Then the cable should be just fine. Each differential pair takes a
twisted pair. Most Ethernet cabling has a characteristic impedance of
about 115 Ohm.

The crux is then to have appropriate drivers and correct impedance
matching (termination) at the end of the cable pairs.

-- 
Greetings Bertho

(disclaimers are disclaimed)

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Smart-Serial cabling

2016-01-01 Thread andy pugh 
On 1 January 2016 at 16:16, Peter C. Wallace  wrote:
> SSerial is differential RS-422 at 2.5 MBits/sec and requires ~100 Ohm
> differential cable cable impedance. This usually means twisted pairs
> are required for TX and RX (the 5V/GND power connections are not critical)

The pairs are twisted inside the individual foil screen with drain
wire  then there are 4 other small (0.14mm2) wires and two bigger
(0.5mm2) wires all inside an overall braided screen.

I suppose the thing to do is to try it and see. I have determined that
the pairs fit nicely in an RJ45 connector, but I can't decide what to
do about the 2x 5V and 2x 0V positions in the connector. Are the 0V
and 5V contacts commoned at the PCB ends, such that leaving two
positions unpopulated in the RJ45 is OK? I only need enough 5V power
to run the 7i73 logic.

-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Smart-Serial cabling

2016-01-01 Thread Peter C. Wallace 
On Fri, 1 Jan 2016, andy pugh wrote:

> Date: Fri, 1 Jan 2016 16:59:12 +
> From: andy pugh <bodge...@gmail.com>
> Reply-To: "Enhanced Machine Controller (EMC)"
> <emc-users@lists.sourceforge.net>
> To: "Enhanced Machine Controller (EMC)" <emc-users@lists.sourceforge.net>
> Subject: Re: [Emc-users] Smart-Serial cabling
> 
> On 1 January 2016 at 16:16, Peter C. Wallace <p...@mesanet.com> wrote:
>> SSerial is differential RS-422 at 2.5 MBits/sec and requires ~100 Ohm
>> differential cable cable impedance. This usually means twisted pairs
>> are required for TX and RX (the 5V/GND power connections are not critical)
>
> The pairs are twisted inside the individual foil screen with drain
> wire  then there are 4 other small (0.14mm2) wires and two bigger
> (0.5mm2) wires all inside an overall braided screen.
>
> I suppose the thing to do is to try it and see. I have determined that
> the pairs fit nicely in an RJ45 connector, but I can't decide what to
> do about the 2x 5V and 2x 0V positions in the connector. Are the 0V
> and 5V contacts commoned at the PCB ends, such that leaving two
> positions unpopulated in the RJ45 is OK? I only need enough 5V power
> to run the 7i73 logic.

Yes, they are commoned at the sserial remote and FPGA side RS-422 interface



>
> -- 
> atp
> If you can't fix it, you don't own it.
> http://www.ifixit.com/Manifesto
>
> --
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.


--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Smart-Serial cabling

2016-01-01 Thread Peter C. Wallace 
On Fri, 1 Jan 2016, Bertho Stultiens wrote:

> Date: Fri, 1 Jan 2016 14:55:47 +0100
> From: Bertho Stultiens <ber...@vagrearg.org>
> Reply-To: "Enhanced Machine Controller (EMC)"
> <emc-users@lists.sourceforge.net>
> To: "Enhanced Machine Controller (EMC)" <emc-users@lists.sourceforge.net>
> Subject: Re: [Emc-users] Smart-Serial cabling
> 
> On 01/01/2016 12:23 PM, andy pugh wrote:
>> Is there any reason to think that the 2x3 screened pairs in that cable
>> will be unsuitable for transporting smart-serial data
>
> The bandwidth of rs232 style serial connections is so low that you will
> not see any problems with twisted pairs. I assume that you are not
> connecting half a kilometer of cable with a speed higher than 115k2 baud
> (or you may run into capacitive loading problens).
>
> Pairing TX and RX with a GND may actually improve noise immunity. But
> the cable is already well shielded.

SSerial is differential RS-422 at 2.5 MBits/sec and requires ~100 Ohm
differential cable cable impedance. This usually means twisted pairs
are required for TX and RX (the 5V/GND power connections are not critical)

>
>
>> (if I can persuade an RJ45 connector on to them)?
>
> That is actually a hard thing to do on these cables. The isolation is
> very hard and does not bend easily. Standard crimp-connectors will be
> almost impossible to attach while ensuring a mechanically stable
> connection. There is also a minimum bend-radius which can be rather large.
>
> These cables are normally used as distribution cables and go into female
> plug assemblies with seperate holding facilities for the cable.
>
>
> -- 
> Greetings Bertho
>
> (disclaimers are disclaimed)
>
> --
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.


--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users