Re: Display station connection question/issue? -- MIDRANGE-L

From Al Macintyre hoping that at least one of my ideas is deemed constructive.

Some of my BS in front - some mixed in with quotes - "aha" - all through here.

When you get to < snip > you not got to end of my BS.

Been a few years since I messed with this stuff and remember it can be a pain 
to diagnose ... you have 4 addresses on the line ... it can support more 
logical addresses ... try using some logical address other than those so far. 
 We have had conditions in which a particular logical address is non-usable, 
or a certain combination is non-usable, at the controller level (a bad chip 
inside the 400).  Is this local or remote?  How close to the ceiling of 
allowable logical addresses on the controller are you?

How are you defining end of the line?  With cable thru if it is not defined 
correctly, you can get what we call an antenna ... a cable that brings in 
interference to the collection & some devices are more sensitive than others 
to this.

Some stations need to be connected with direct screw in & some need pigtail 
... there are several different ways that connection is possible but those 
that are correct are a narrower spectrum than those that appear to be 
possible & I have no idea where this is documented.  I just know from 
experience that a wrong connector in one place can mess up ability to 
communicate at a different location in the data stream.

Perhaps a bug in how we have our autoconfig setup or the fact that M/36 is in 
our picture, but when we are only dealing with display stations, autoconfig 
works A-Ok, but when a printer is in the picture, we have to delete its 
config before auto config correctly recognizes what is at that address.

first picture

>  DSP01-----Twinax cable-----PRT01--Balun--network
>  cable--Balun--DSP02--Balun--network cable--Balun--DSP03
>  
>  What happens is that DSP03 gets a signon but DSP02 does not, it stays Vary
>  On Pending... All other devices are ok.

This tells me that the connection is Ok to end of line from the 400, the 
signal is getting there all the way, so the problem may be whether you have 
the right connectors at intermediate points for the kind of devices, or if 
there is a problem with DSP02 hardware.  Do you have any splices in addition 
to the list shown above ... did you know there are ceilings both on numbers 
of splices and total length of line & distances various types of cabling can 
be from other things like power lines & other cabling?
  
>  I have double and triple checked addresses, they are ok.  Have swapped
>  network cables with ones I know work and same result. Swapped baluns with
>  ones I know work, no diff....

Do you have enough twinax cable to see if it works without the network cable?

later post

>  QINTER defined to all (If I swap the phsical position of DSP02 and DSP03
>  then DSP02 (last one in the chain) comes up.  The last one in the chain is
>  the one that displays.  The terminal with the Passthru (DSP02 in original)
>  is the only one that does not come up.

Aha ... the last one in the chain displays ... the problem is with the 
physical address not with hardware that is at that physical address.  Can you 
switch vs PRT01 so that a cable bypassing the physical address in trouble & 
have everything else working, then connect that physical address by different 
logic?  Perhaps much longer cable.

We also had a problem once with the power supply ... a bad wall plug.  The 
power supply to the display station was enough to get some indicators that it 
seemed to be functioning, but not enough to function correctly.  This was 
proven by running a LONG EXTENSION CORD & changing nothing else in the setup.

We have also on occasion run a LONG TWINAX CABLE along the floor to connect 
some device as a check on whether the problem is with the twinax in the 
ceiling.

You almost need one person at each device including the 400 on a 
teleconference circuit & a tape recorder.  Person X announces what they are 
doing & various other speakers say whether or not it makes a difference to 
them.  In our case we can have multiple cooks not communicating effectively 
with each other.

MacWheel99@aol.com (Alister Wm Macintyre) (Al Mac)
AS/400 Data Manager & Programmer for BPCS 405 CD Rel-02 mixed mode (twinax 
interactive & batch) @ http://www.cen-elec.com Central Industries of 
Indiana--->Quality manufacturer of wire harnesses and electrical 
sub-assemblies - fax # 812-424-6838

+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---