GCA50 Addressing again

RocNet(RaspberryPi), mergCBUS and LocoNet based hardware developed by Peter Giling. (GCA)
Documentation: http://wiki.rocrail.net/doku.php?id=english#hardware
Order information: http://wiki.rocrail.net/doku.php?id=gca:gca-index-en/

GCA50 Addressing again

Postby mcnaugg » 08.06.2015, 20:41

In Jan I requested help in solving a problem I had with the addressing of the GCA50 units. This was speedily resolved but I now have an additional problem related to addressing.
To keep the wiring etc tidy I have mounted my GCA units in a wall mounted cabinet (ex Aircon Cabinet)
I have 5 x GCA50 in a daisy chain and they are numbered 82/1, 83/1, 84/1, 85/1 & 86/1 leaving 81/1 address unused as advised. I then have a GCA93 Block detector connected to the 82/1 module. Until now I have only been using 82/1, the remaining GCA50s have been disconnected. This has been sufficient to allow me to run a simple DMU to and fro on a U shaped track with three stations. There is one switch in the track which is operated via a Digitrax DS52 Module. This has addresses 1 & 2 of which I am using address 1
I now want to lay more track and bring in the switch modules and additional block detectors. However I am having problems with addressing the GCA50s. Am I right in thinking that the addresses are in sequential? Once the GCA50s have been given the module addresses ie 82/1, 83/1 etc then the addresses of the 16 module ports can be any number between 1 & 2048. In my setup 82/1 is used for my original circuit. My next module is 83/1 which I want to use for 4 switches. So under “Easy Setup” I have set ports 1-8 to 17,17: 18,18: 19,19: and 20,20. I have selected Pulse operation and every second C2 box is ticked. At this stage I am not using ports 9 – 16.
I have connected LEDs with suitable resistors for 20+ V to the four pairs of outputs on the GCA77. I do not seem to get any sensible result on the LEDs when I click on the Test boxes (Easy Setup). At best only one or two Leds are pulsed and then only if I give the Test boxes a double click.
I have taken note of the requirement to power down and the power up the GCA50 after any change. I have also tried adding 4 switch graphics to my working DMU track diagram and giving them the addresses 17,18,19,20. Although the switch graphics on the track diagram show a change in direct the results are similar to the Easy Setup panel.
Any ideas what I am doing wrong?

Gareth
mcnaugg
 

Re: GCA50 Addressing again

Postby rjversluis » 11.06.2015, 14:12

Best Regards, Rob.
:!: PS: Do not forget to attach the usual files.
:!: PS: Nicht vergessen die übliche Dateien an zu hängen.
[ macOS - Linux] - [ N: CBUS - CANGCx ] - [ G: CBUS - CANGCx ]
rjversluis
Site Admin
 

Re: GCA50 Addressing again

Postby mcnaugg » 13.06.2015, 15:09

I have done some more testing and have come to the conclusion that I am using the wrong switch /point motors on the GCA77 units.
I have followed the instructions as indicated in the setup procedure as far as I can tell. As stated previously I have 5 x GCA50 in a daisy chain but since I am only using two at the moment three are disconnected. I have addressed them as 82/1 and 83/1. 82/1 is setup to control my original U-Bend layout so I have taken that out of circuit if only to ensure that I do have a track that works. I am therefore using 83/1 only whilst testing the switches / points. Since there are 16 ports on a GCA50 I have reserved addresses 1-16 for the first GCA50 (U-Bend).
I have attached a listing of the “Easy Setup” page for 83/1 showing that I have used addresses 17, 18, 19, & 20 on the first 8 ports. I have nothing connected to the second 8 ports.
Testing has now shown that if I now connect a GCA76 (High Power) board to the first ports (17-20) the switch/ point motors work correctly both from the Easy Setup page and also from the simulated switches on the track plan.
If I swop the connection over to a GCA77 (one of two) the switch operation is incorrect with at least two port addresses permanently powered and operation of the Easy Setup test buttons or the screen switches makes no impression. I have therefore concluded that the Darlington ICs (ULN2803) have “blown”. Re-reading the spec sheet for the 77 shows that it is suitable for 0.5A max. Since I am mainly using Hornby switch motors plus some Casadio units it seems that I have overloaded the 77 and “blown” the ICs. I am not sure how the “current” is worked out for the switches as it is a pulse but the Hornby units show a DC resistance of 6-7 ohms and the Casadio show a DC resistance of ± 24 ohms. Given that the supply volts to the switches is a pulse of 20vDC this would mean 3 A for the Hornby and 1A for the Casadio switches which is outside of the limits of the 77.
I will therefore order some replacement ICs and then consider ordering some new GCA76 PCBs and make these my standard switch controller.
You do not have the required permissions to view the files attached to this post.
mcnaugg
 


Return to GCA (Hardware by Peter Giling)