# Typical Response Time to AB SLC via DHRIO

**URL:** <https://forum.inductiveautomation.com/t/typical-response-time-to-ab-slc-via-dhrio/12608>\
**Category:** Ignition\
**Created:** [January 26, 2017, 4:00pm UTC](https://forum.inductiveautomation.com/t/typical-response-time-to-ab-slc-via-dhrio/12608 "2017-01-26T16:00:06Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![george](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/george/32/14066_2.png) [@george](https://forum.inductiveautomation.com/u/george)\
**Post date:** [January 26, 2017, 4:00pm UTC](https://forum.inductiveautomation.com/t/typical-response-time-to-ab-slc-via-dhrio/12608/1 "2017-01-26T16:00:06Z")

</div>

I need to connect Ignition to an Allen Bradley SLC504 on DH+.  
To do this I’m going via a ControlLogix 1756-DHRIO module.  
It all connects ok but for this particular PLC I need to read about 1200 tags and the diagnostics screen (Ignition 7.9) is showing this as 37 requests and the “Load Factor” is over 600% ☹  
Result: the comms keeps going bad.

I’m getting a Mean Response Time of around 176ms.

If in a test lab I put the same program in an SLC505 processor directly on Ethernet then the response time is about a 10th of that and the Load Factor is around 50%.

Can anyone advise what typical response times you’d get communicating to a Rockwell PLC via a CLX DHRIO card? Is 170ms expected? Or is that far too high? PINGing the CLX rack itself takes under 1ms so the delay must be on the DH+ side of things.

---

<div class="post-metadata">

**Author:** ![pturmel](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/pturmel/32/3295_2.png) [@pturmel](https://forum.inductiveautomation.com/u/pturmel)\
**Post date:** [January 26, 2017, 4:21pm UTC](https://forum.inductiveautomation.com/t/typical-response-time-to-ab-slc-via-dhrio/12608/2 "2017-01-26T16:21:08Z")

</div>

Sounds perfectly normal for DH+. That protocol is normally only 56kbits/sec. Do you really need all of that data all of the time? Have you tried to split it up into fast and slow scan classes?  
If the DH+ pipe is simply too slow, you will almost certainly have to replace the processor.

---

<div class="post-metadata">

**Author:** ![pturmel](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/pturmel/32/3295_2.png) [@pturmel](https://forum.inductiveautomation.com/u/pturmel)\
**Post date:** [January 26, 2017, 4:22pm UTC](https://forum.inductiveautomation.com/t/typical-response-time-to-ab-slc-via-dhrio/12608/3 "2017-01-26T16:22:46Z")

</div>

Oh, yeah: Is there anything else talking on the same DH+ network? That’ll kill you.

---

<div class="post-metadata">

**Author:** ![george](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/george/32/14066_2.png) [@george](https://forum.inductiveautomation.com/u/george)\
**Post date:** [January 27, 2017, 3:29pm UTC](https://forum.inductiveautomation.com/t/typical-response-time-to-ab-slc-via-dhrio/12608/4 "2017-01-27T15:29:33Z")

</div>

Yes there is a legacy SCADA system (that I’m going to replace with Ignition).  
We did try unplugging that today but it actually seemed to have little effect.

My next step as you say is to try adjusting scan classes, as currently everything is on a default 1 second class.

---

<div class="post-metadata">

**Author:** ![greg.simpson](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/greg.simpson/32/3135_2.png) [@greg.simpson](https://forum.inductiveautomation.com/u/greg.simpson)\
**Post date:** [February 3, 2017, 8:55pm UTC](https://forum.inductiveautomation.com/t/typical-response-time-to-ab-slc-via-dhrio/12608/5 "2017-02-03T20:55:28Z")

</div>

You can also consider placing some tags in a leased scan class with a slow rate of zero. Then when the tags are not being viewed they are not subscribed to.