[etherlab-dev] Problems with PDO mapping on EL4132

[Ralf Rösch]

Hello Thomas,

sometimes ago we also used the EL4132 for testing purposes.

cat master0/slave006/info on our test system shows:

Ring position: 6
State: OP (OP)
Flags: online, ok

Data link status:
  Port 0: Phy 0 (EBUS), Link down, Loop open, Signal detected
  Port 1: Phy 3 (unknown), Link down, Loop open, Signal detected
  Port 2: Phy 0 (EBUS), Link down, Loop closed, No signal
  Port 3: Phy 0 (EBUS), Link down, Loop closed, No signal

Identity:
  Vendor ID: 0x00000002 (2)
  Product code: 0x10243052 (270807122)
  Revision number: 0x00000000 (0)
  Serial number: 0x00000000 (0)

Mailboxes:
  RX: 0x1800/246, TX: 0x18F6/246
  Supported protocols: CoE, FoE

Current consumption: 180 mA

General:
  Group: AnaOut
  Image: TERM_AO
  Order number: EL4132
  Name: EL4132 2K. Ana. Ausgang +/-10V

Sync managers / PDO mapping:
  SM0: addr 0x1800, size 246, control 0x26, enable
    No PDOs mapped.
  SM1: addr 0x18F6, size 246, control 0x22, enable
    No PDOs mapped.
  SM2: addr 0x1000, size 4, control 0x24, enable
    PDO mapping information from CoE.
    RxPdo 0x1600 "RxPDO 001 mapping"
      0x3001:1 "Value", 16 bit
    RxPdo 0x1601 "RxPDO 002 mapping"
      0x3002:1 "Value", 16 bit
  SM3: addr 0x1100, size 0, control 0x20, enable
    No PDOs mapped.

Available PDOs from SII:
  RxPdo 0x1600 "Channel 1", default mapping: SM2.
    0x6411:1 "Output", 16 bit
  RxPdo 0x1601 "Channel 2", default mapping: SM2.
    0x6411:2 "Output", 16 bit

As you can see there are two different mapping schemes one comes from
SII and one comes from CoE.
If you start the master and wait some seconds the mapping from CoE takes
effect (PDO Remap).

In this case following #define statements should solve your problem:

#define Beckhoff_EL4132_PDO_Output1_CoE 0x3001, 1
#define Beckhoff_EL4132_Output1 Beckhoff_EL4132,
Beckhoff_EL4132_PDO_Output1_CoE

Regards
Ralf


Thomas Elste schrieb:
> Hello again,
>
> I've some problems while addressing one particular kind of
> slave. The EL4132 (analog output) gives me some headache with
> the PDO mapping:
>
> {"8", Beckhoff_EL4132_Output1,              &r_ana_out}
>
> This is the line I've been using in the domain configuration.
> This works very well and as expected under one condition: The
> task module has to be loaded directly after the master has done
> its first bus scan.
> If I wait some time (3-5 seconds) after the master comes up
> some PDO remapping is done and then the loading of the task
> module fails with "PDO entry 0x6411:1 is not mapped in slave 8.".
> It is in fact not mapped _anymore_ as some remapping is done
> somewhere in master/fsm_coe_map.c after some seconds since the
> master becomes operational.
>
> Anyone else experienced something similiar, or any ideas how
> to fix or avoid this?
>
>
> Best regards
> Thomas
>
> _______________________________________________
> etherlab-dev mailing list
> etherlab-dev at lists.etherlab.org
> http://lists.etherlab.org/mailman/listinfo/etherlab-dev
>