Author Topic: Any IO-Link experts out there?  (Read 128 times)

0 Members and 1 Guest are viewing this topic.

Offline mubes

  • Regular Contributor
  • *
  • Posts: 187
  • Country: gb
  • Do Not Boil
Any IO-Link experts out there?
« on: May 17, 2018, 07:58:54 am »
Folks,

I'm having terrible trouble getting an IO-Link device to talk to me. Hoping there are some experts here who can point out whatever I'm doing wrong.  The device itself is a SICK IMB12-04BPPVC0K proximity sensor, which is a IO-Link 1.0 device at COM1 (4800 baud). I've done all the merry dance required in the IO-Link spec with the result that I've now got the output shown in the scope shot below, direct from the I/Q signaling line. Some notes;

1) The WKUP pulse is exactly 80uS long. In this specific scope shot I've produced below there's not the resolution to show that, but it is.

2) If I set C/Q high (by means of putting a piece of metal in front of the sensor) I can see it go low for 100mS in response to the WKUP pulse...so I'm reasonably confident that is working correctly.

3) I'm following the spec by trying to initialize at COM3, then COM2 and finally COM1. I wait 50mS between wakeup cycle attempts (Tdwu) per the spec.  The shot shows all three speeds, with the scope set to decode COM1...you can see the 0xA2,0x00 sequence decoded which is correct (I believe) for a MinCycleTime inquiry...Ignore the 0xFD, that just the higher speed data being mis decoded.

4) I am using even parity, and the scope confirms it's being sent correctly.

5) If I physically remove the sensor, the line floats to around half L+ in between my actively driving it. The implication is that the sensor is actively, and deliberately, pulling the line down (in the light of point 2 above)....its ignoring me on purpose, not by accident. |O

So....I'm stumped. I've obviously tried things like varying the timings, just connecting at COM1 etc. etc. all to no avail. I'm pretty certain I'm just doing something wrong, but I have no idea what at the moment.  If there's anyone out there with a COM1 IO-Link device who can scope me a working version of this startup sequence I would be eternally grateful since that gives me something to compare against...other suggestions for what to try next also gratefully received!

Regards

DAVE
 

Offline mubes

  • Regular Contributor
  • *
  • Posts: 187
  • Country: gb
  • Do Not Boil
Re: Any IO-Link experts out there?
« Reply #1 on: May 17, 2018, 08:48:49 pm »
Yup, problem between screen and keyboard. Turns out it communicates at COM2. Doh. :palm:  For reference, since the lack of working examples was one of the big problems I had, a scope shot of the working initialisation is attached for when someone else needs it.
« Last Edit: May 17, 2018, 09:14:35 pm by mubes »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf