# Gaps in Power Chart, but no bad data?

**URL:** <https://forum.inductiveautomation.com/t/gaps-in-power-chart-but-no-bad-data/116772>\
**Category:** Ignition\
**Created:** [August 13, 2026, 12:00am UTC](https://forum.inductiveautomation.com/t/gaps-in-power-chart-but-no-bad-data/116772 "2026-08-13T00:00:17Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![eaolson](https://avatars.discourse-cdn.com/v4/letter/e/f07891/32.png) [@eaolson](https://forum.inductiveautomation.com/u/eaolson)\
**Post date:** [August 13, 2026, 12:00am UTC](https://forum.inductiveautomation.com/t/gaps-in-power-chart-but-no-bad-data/116772/1 "2026-08-13T00:00:17Z")

</div>

I've got an issue with the Power Chart that is confusing me. There are gaps in the lines, but I don't think there should be. This seems due to the "Show a break in the line when data is missing" checkbox, which I get is the same as the pens.display.breakLine property. I assume "missing" means bad quality data, but there isn't any, see below.

My colleague thinks this has to do with the fact that Historical Deadband is set to Analog, but I don't think so.

For example. Power Chart with gaps at the top, regular Time Series Chart at the bottom.

 ![image](https://us1.discourse-cdn.com/inductiveautomation/original/3X/6/4/640b81ac21f38a92bd939ed7a92bc296cf77db34.png)

There are no bad quality points in this time frame:

 ![image](https://us1.discourse-cdn.com/inductiveautomation/original/3X/3/8/384f46259f7f91fdef55a543667e6fcb1b17f3f7.png)

This is on 8.3.7.

---

<div class="post-metadata">

**Author:** ![tgarrison](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/tgarrison/32/80965_2.png) [@tgarrison](https://forum.inductiveautomation.com/u/tgarrison)\
**Post date:** [August 13, 2026, 6:33pm UTC](https://forum.inductiveautomation.com/t/gaps-in-power-chart-but-no-bad-data/116772/2 "2026-08-13T18:33:35Z")

</div>

What are the historical settings on that tag?

---

<div class="post-metadata">

**Author:** ![eaolson](https://avatars.discourse-cdn.com/v4/letter/e/f07891/32.png) [@eaolson](https://forum.inductiveautomation.com/u/eaolson)\
**Post date:** [August 13, 2026, 7:53pm UTC](https://forum.inductiveautomation.com/t/gaps-in-power-chart-but-no-bad-data/116772/4 "2026-08-13T19:53:05Z")

</div>

Include Metadata: false  
Deadband Style: Auto  
Deadband Mode: Absolute  
Historical Deadband: 100  
Sample Mode: Tag Group  
Max Time Between Samples: 2 hour  
Historical Tag Group: Default Historical

It's an OPC tag, type float. The Default Historical tag group is the out of the box 10 s one.

---

<div class="post-metadata">

**Author:** ![tgarrison](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/tgarrison/32/80965_2.png) [@tgarrison](https://forum.inductiveautomation.com/u/tgarrison)\
**Post date:** [August 13, 2026, 8:20pm UTC](https://forum.inductiveautomation.com/t/gaps-in-power-chart-but-no-bad-data/116772/5 "2026-08-13T20:20:29Z")

</div>

Max Time Between Samples is 2hrs, but you have a 20ish hour gap in your data. Believe that's your issue.

And yes, listen to your coworker and switch the Deadband Style to Discrete. Analog doesn't work like you think it does.

---

<div class="post-metadata">

**Author:** ![eaolson](https://avatars.discourse-cdn.com/v4/letter/e/f07891/32.png) [@eaolson](https://forum.inductiveautomation.com/u/eaolson)\
**Post date:** [August 13, 2026, 10:41pm UTC](https://forum.inductiveautomation.com/t/gaps-in-power-chart-but-no-bad-data/116772/6 "2026-08-13T22:41:14Z")

</div>

I think the Max Time Between Samples may have been modified since this data was recorded, probably because of this very problem. Previously, I think it was 24 hours. There are, in fact, several data points in that long gap spaced by 24 hours, they're just very hard to see. There's no line connecting them for some reason.

Analog deadband looks like it works roughly the same as PI System's compression algorithm, right? So data points are not recorded if the signal is changing along a linear trend, just if it deviates from that trend by a threhold.
