# Modbus signal not connected to slave

**URL:** <https://forum.universal-robots.com/t/modbus-signal-not-connected-to-slave/27633>\
**Category:** Robot Communication\
**Created:** [March 28, 2023, 3:01am UTC](https://forum.universal-robots.com/t/modbus-signal-not-connected-to-slave/27633 "2023-03-28T03:01:03Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![anon70625529](https://avatars.discourse-cdn.com/v4/letter/a/e495f1/32.png) [@anon70625529](https://forum.universal-robots.com/u/anon70625529)\
**Post date:** [March 28, 2023, 3:01am UTC](https://forum.universal-robots.com/t/modbus-signal-not-connected-to-slave/27633/1 "2023-03-28T03:01:03Z")

</div>

![image](https://us1.discourse-cdn.com/flex020/uploads/universal_robots/original/2X/f/f530bc52e6868cdea063d926e30394842df056f4.jpeg)  
Hi guys, these days I’ve been getting this error message and I do not know what went wrong. I can ping all my devices and all my i/o are connected correctly. Random MODBUS signal were detected to lose connection and stops the program but when I checked again its back on. Not sure what else I can do. Would appreciate for any solution given! thanks

---

<div class="post-metadata">

**Author:** ![eric.feldmann](https://avatars.discourse-cdn.com/v4/letter/e/94ad74/32.png) [@eric.feldmann](https://forum.universal-robots.com/u/eric.feldmann)\
**Post date:** [March 28, 2023, 12:26pm UTC](https://forum.universal-robots.com/t/modbus-signal-not-connected-to-slave/27633/2 "2023-03-28T12:26:58Z")

</div>

I think maybe this happened to me at one point in the past, and I think it ended up being that another device was added to the network with the same IP address.

---

<div class="post-metadata">

**Author:** ![WattsUp](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.universal-robots.com/wattsup/32/5732_2.png) [@WattsUp](https://forum.universal-robots.com/u/WattsUp)\
**Post date:** [March 28, 2023, 12:49pm UTC](https://forum.universal-robots.com/t/modbus-signal-not-connected-to-slave/27633/3 "2023-03-28T12:49:17Z")

</div>

I get this fault with my UR ↔ Click PLC modbus connection.

It seems to happen when the system is idle for long periods of time, BUT has happened once while production was running.

I assume it is the Click PLC in my configuration bc i have 3 other UR talking modbus with a Mitsubishi PLC using the exact same IO & program style without any issues at all.

To resume production, I’ve just gone to Installation\>fieldbus\> Modbus\> Refresh All. Then just play from beginning and its happy again.

---

<div class="post-metadata">

**Author:** ![dpeva](https://avatars.discourse-cdn.com/v4/letter/d/b5ac83/32.png) [@dpeva](https://forum.universal-robots.com/u/dpeva)\
**Post date:** [March 28, 2023, 1:36pm UTC](https://forum.universal-robots.com/t/modbus-signal-not-connected-to-slave/27633/4 "2023-03-28T13:36:36Z")

</div>

We get this at random intervals on several of our robots. Sometimes we’ll go a month or more without the error, then we might see it on 2 or 3 different robots (all in the same room) on the same day. Usually have to power cycle the device (Acromag XT1121 digital I/O device) to clear the problem. Thinking it’s some kind of network glitch, and have asked our IT group to see if they can monitor the connection, but no feedback so far. Our IP addresses are reserved, so there shouldn’t be any conflicts that way.  
I can replace that unit with a couple relays, so probably will end up going that route.

---

<div class="post-metadata">

**Author:** ![roguebuck](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.universal-robots.com/roguebuck/32/4848_2.png) [@roguebuck](https://forum.universal-robots.com/u/roguebuck)\
**Post date:** [March 28, 2023, 9:22pm UTC](https://forum.universal-robots.com/t/modbus-signal-not-connected-to-slave/27633/5 "2023-03-28T21:22:23Z")

</div>

setting the modbus client to “sequential mode” has resolved this for me in the past.

you have to enable “show advanced options” for it to show.

 ![image](https://us1.discourse-cdn.com/flex020/uploads/universal_robots/original/2X/7/741c2920c994770fe11d135e0bcca0037d9df2e4.png)

This forces the modbus client to wait for a response from one request before scanning the next. It shouldnt hurt in any case and will help smooth out any network hiccups.

You also might consider reducing the frequency if these signals are not timing-sensitive (such as most outputs; particularly cycle\_start)

NOTE: FYI - setting the frequency to 0 is acceptable. This effectively makes that particular signal a “one-shot” signal instead of a polled signal. Be careful using 0Hz on inputs - they wont update except for when modbus\_get\_signal\_status() is run explicitly.

---

<div class="post-metadata">

**Author:** ![WattsUp](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.universal-robots.com/wattsup/32/5732_2.png) [@WattsUp](https://forum.universal-robots.com/u/WattsUp)\
**Post date:** [March 29, 2023, 12:34pm UTC](https://forum.universal-robots.com/t/modbus-signal-not-connected-to-slave/27633/6 "2023-03-29T12:34:00Z")

</div>

My robots are all offline & direct connections to PLC so I can say the error is not only caused by network glitches.
