# PlotJuggler and logs.px4.io disagree **URL:** \ **Category:** Uncategorized\ **Created:** [October 19, 2023, 6:09pm UTC](https://discuss.px4.io/t/plotjuggler-and-logs-px4-io-disagree/34867 "2023-10-19T18:09:18Z")\ **Posts on this page:** 9\ **Page:** 1
**Author:** ![holmgeir](https://discuss.px4.io/letter_avatar_proxy/v4/letter/h/b9bd4f/32.png) [@holmgeir](https://discuss.px4.io/u/holmgeir)\ **Post date:** [October 19, 2023, 6:09pm UTC](https://discuss.px4.io/t/plotjuggler-and-logs-px4-io-disagree/34867/1 "2023-10-19T18:09:18Z")
I ran into something strange when analyzing a log file from my px4 equipped fixed wing drone. Hardware vers PX4\_FMU\_V5 (V540) sw vers 1c8ab2a0. The graphs below show the roll angle as displayed by [logs.px4.io](http://logs.px4.io) and then by PlotJuggler. The graps are identical “most of the time” but from approx 45 to 90 seconds into the Plotjuggler graph it looks like one of them is inverted and offset by some value. (I hope the grahs are visible). Does anybody have an idea what is going on? ![image](https://discuss.px4.io/uploads/default/original/2X/5/53bbb2698a5f1f9a56b63cb7cb5d7950461db727.png) ![image](https://discuss.px4.io/uploads/default/original/2X/c/cda2ad7572b6226287655e1af65db5bf277d3691.jpeg) ---
**Author:** ![JulianOes](https://discuss.px4.io/user_avatar/discuss.px4.io/julianoes/32/30_2.png) [@JulianOes](https://discuss.px4.io/u/JulianOes)\ **Post date:** [October 19, 2023, 9:52pm UTC](https://discuss.px4.io/t/plotjuggler-and-logs-px4-io-disagree/34867/2 "2023-10-19T21:52:56Z")
That’s interesting. Could you share the link to [logs.px4.io](http://logs.px4.io), so I can have a look. I’m also copying in @bkueng. ---
**Author:** ![holmgeir](https://discuss.px4.io/letter_avatar_proxy/v4/letter/h/b9bd4f/32.png) [@holmgeir](https://discuss.px4.io/u/holmgeir)\ **Post date:** [October 20, 2023, 9:22am UTC](https://discuss.px4.io/t/plotjuggler-and-logs-px4-io-disagree/34867/3 "2023-10-20T09:22:59Z")
Thanks for your reply. Here is the log file: [3D View - Flight Review](https://logs.px4.io/3d?log=8ea9c734-3454-4b2a-8a94-0dc50a9ff6e8) ---
**Author:** ![JulianOes](https://discuss.px4.io/user_avatar/discuss.px4.io/julianoes/32/30_2.png) [@JulianOes](https://discuss.px4.io/u/JulianOes)\ **Post date:** [October 20, 2023, 10:07am UTC](https://discuss.px4.io/t/plotjuggler-and-logs-px4-io-disagree/34867/4 "2023-10-20T10:07:47Z")
Hm, what are you plotting in plotjuggler? ---
**Author:** ![holmgeir](https://discuss.px4.io/letter_avatar_proxy/v4/letter/h/b9bd4f/32.png) [@holmgeir](https://discuss.px4.io/u/holmgeir)\ **Post date:** [October 20, 2023, 10:53am UTC](https://discuss.px4.io/t/plotjuggler-and-logs-px4-io-disagree/34867/5 "2023-10-20T10:53:42Z")
This is “estimator\_attitude.00/q,01” The three “estimator\_attitude.xx” and “vehicle\_attitude/q.01” look identical ---
**Author:** ![holmgeir](https://discuss.px4.io/letter_avatar_proxy/v4/letter/h/b9bd4f/32.png) [@holmgeir](https://discuss.px4.io/u/holmgeir)\ **Post date:** [October 20, 2023, 12:45pm UTC](https://discuss.px4.io/t/plotjuggler-and-logs-px4-io-disagree/34867/6 "2023-10-20T12:45:51Z")
Another interesting bit of information: If one extracts the “vehicle\_angular\_velocity” data from the log file, and then integrates the (roll) values from the beginning, the result is the same graph as in logs… page. ![image](https://discuss.px4.io/uploads/default/original/2X/2/2d03c9d7b137c5940ebbdfdfeb7aa13413f1df26.png) ---
**Author:** ![JulianOes](https://discuss.px4.io/user_avatar/discuss.px4.io/julianoes/32/30_2.png) [@JulianOes](https://discuss.px4.io/u/JulianOes)\ **Post date:** [October 23, 2023, 6:20pm UTC](https://discuss.px4.io/t/plotjuggler-and-logs-px4-io-disagree/34867/7 "2023-10-23T18:20:47Z")
q,01 would be a quaternion and hence different. You need to convert to Euler angles, see here: [feature request: yaw\_degrees for imu quaternion · Issue #238 · facontidavide/PlotJuggler · GitHub](https://github.com/facontidavide/PlotJuggler/issues/238) ---
**Author:** ![bkueng](https://discuss.px4.io/letter_avatar_proxy/v4/letter/b/2bfe46/32.png) [@bkueng](https://discuss.px4.io/u/bkueng)\ **Post date:** [November 3, 2023, 7:56am UTC](https://discuss.px4.io/t/plotjuggler-and-logs-px4-io-disagree/34867/8 "2023-11-03T07:56:09Z")
Yes it depends on the quaternion to angle conversion you use in plotjuggler. ---
**Author:** ![holmgeir](https://discuss.px4.io/letter_avatar_proxy/v4/letter/h/b9bd4f/32.png) [@holmgeir](https://discuss.px4.io/u/holmgeir)\ **Post date:** [November 3, 2023, 10:46am UTC](https://discuss.px4.io/t/plotjuggler-and-logs-px4-io-disagree/34867/9 "2023-11-03T10:46:32Z")
Yes, indeed. De-quaternying did the trick. Thanks!