Today is...
Friday, January 18, 2019
Welcome to, the global online
community of automation professionals.
Featured Video...
Featured Video
Wiring and programming your servos and I/O just got a lot easier...
Our Advertisers
Help keep our servers running...
Patronize our advertisers!
Visit our Post Archive
Pros and Cons of IOLink?

Interested to hear comments about IOLink from people who have deployed it in systems. I've read a few manufacturers white papers- would be helpful to hear pros/cons from users.

Thank you

By James Ingraham on 27 November, 2018 - 5:09 pm
1 out of 1 members thought this post was helpful...

I love IO-Link. I much prefer it to analog. That said, there are always trade-offs.

-You get values in some kind of engineering unit, rather than having to figure out what 6.7mA translates to.

-You know if you're connected. (Yes, 4-20mA gives you this to.)

-It's only 3 wires to get as much data as possible. Power is two of those wires.

-Depending on the device and the controller, you can change parameters on the fly.

-Depending on the device, you can have bi-directional communication.

-Depending on the device, you can get diagnostic data as well as process data.

-Even without real-time diagnostic data, you can do things like check the device number, helping to make sure you've got the right thing plugged in.

-Generally easier to set up than devices that take an IP Address.

-One IP address can have many IO-Link devices, saving addresses and switch ports.

-Network war independent (i.e. masters are available on EtherNet/IP, Profinet, etc.)

-Devices are generally cheaper than comparable Ethernet devices.

-Generally uses standard M12 4-pin connectors.

-You can add an analog to IO-Link or IO-Link to analog adapter. (Although this is a bit of the worst of both worlds.)

-You can add a few digital I/O if you need to.

-There's an inductive wireless option for some admittedly niche applications.

-You need the IODD file, or else enough documentation to figure out how to set it up as a generic device and parse the data.

-IO-Link is a "bag of bits" protocol. The IODD helps, but you still end up mashing bytes together, and endian matters.

-There's no where to put a meter. On a 0-10V, it's very easy to hook up a meter and see what you're reading. On a 4-20mA it's only slightly harder. You can't do that with IO-Link. (It should be noted that there are tools you can hook up to a laptop to check on the device, but that's orders of magnitude harder than checking voltage with a meter)

-Sometimes there's not an IO-Link version of a device, so you're stuck with analog anyway.

-Can be more costly than analog, and is almost certainly more costly than digital I/O, such as photo-eyes. Yes, you get the parameterization and diagnostics, but money is money.

-Point to point only, so you need an IO-Link port for every device. (You can also argue this is a good thing, since you don't get the complexity that comes from networks or fieldbuses when at the device level.)

-Speed is generally "fast enough", but may be a factor in some applications.

-James Ingraham
Sage Automation, Inc.

Does anybody know if power and ground can be bussed with IO-Link? In other words do they wink the 24V line or something to start transmission? All of the IO-Link modules I've seen have connections for all 3 wires which leads me to believe the power is not exactly passive.


By James Ingraham on 29 November, 2018 - 12:32 pm

I'm pretty sure that the power can be bused, and it's just power. I think the main reason that the modules all have 3 points is just because most of them are anticipating 3 wires off of 4-pins on an M12. However, the 1734-IOL 4-port IO-Link module only has two terminals for power (and 2 for common), and I believe they're internally jumpered.

-James Ingraham
Sage Automation, Inc.